Posts by Richard Haselgrove
|
1)
Message boards :
Questions and problems :
how to login
(Message 113916)
Posted 5 days ago by Richard Haselgrove Post: Science United is your problem, it takes over your machine and dictates which projects you’re allowed to run.And - by the way it's been created - actively prevents you from logging in to the message board system of any of the projects your computers do useful work for. Computers connected to a project by Science United appear on their systems as belonging to Science United - and none of us is allowed to know the password used by Science United in setting them up. |
|
2)
Message boards :
Questions and problems :
how to direct BOINC to use a specific disk drive
(Message 113893)
Posted 11 days ago by Richard Haselgrove Post: All that follows assumes you are using Windows. If not, stop reading here. BOINC uses two separate directories: a program directory (for its own programs), and a data directory (for everything else, including project science applications). Before you start, make a note of what the current location of each of these is - you can find them near the beginning of the Event Log each time BOINC starts). There's probably no need to change the program directory. Leave that alone unless you feel an unstoppable urge to move it. Stop the current version of BOINC completely, including any running projects. Find the current location of the data directory. Move it to your new chosen drive - again, completely. You can't pick and move subdirectories individually. Also a good idea to take a backup while you're at it. If you're already using the current version of BOINC - v7.24.1, if you're using a 64-bit version of Windows - you'll need to uninstall it from the Windows control panel. If there's a newer version available (see the 'download all' page), you can install it directly - otherwise, you can re-use the copy you downloaded last time. Step through the installer stages carefully. You'll reach a page which shows the directory locations you used last time - probably the defaults. That page has an 'Advanced' button in the lower right hand corner. Click it. You can then change either or both of the directory locations. Again, I'd suggest you leave the program directory untouched, but change the data directory to match the location you moved it to earlier. Double-check that the match is exact. And that's it. Finish the installer normally, and start BOINC running. It should pick up and continue where it left off. |
|
3)
Message boards :
Questions and problems :
Suspend not working properly
(Message 113885)
Posted 12 days ago by Richard Haselgrove Post: Looking at the Applications page for Milkyway (https://milkyway.cs.rpi.edu/milkyway/apps.php), they currently have just two programs running, and both are multithreaded: one "with Orbit Fitting" and the other without. You need to identify which variant is refusing to suspend, and notify the Milkyway@home project of that extra detail. The most BOINC can do is to send a message to the application saying that a suspension has been requested: it's up to the project's programmer for the application in question to ensure that every application receives and responds to these messages. |
|
4)
Message boards :
Questions and problems :
BOINC Mannager completely empty
(Message 113850)
Posted 26 days ago by Richard Haselgrove Post: Keep an eye on those new tasks - make sure they complete with a "successful" outcome, and within the time limit. BOINC servers will normally only refuse to send new tasks to a computer which is returning work late or with errors. |
|
5)
Message boards :
Questions and problems :
Boinc and Rendering/Gaming GPUs
(Message 113777)
Posted 17 Mar 2024 by Richard Haselgrove Post: The use of CUDA within BOINC was first announced at the very end of 2008 (source), although that was admittedly only the end of a lengthy period of development under NDA. Later, support for a second programming language - OpenCL - was also added. BOINC projects generally tend to use one or the other, but not both. You are best off installing drivers directly supplied by the manufacturers of your graphics device(s) - Nvidia, AMD, or Intel. The drivers included in your operating system may not support the full range of languages. |
|
6)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 113687)
Posted 26 Feb 2024 by Richard Haselgrove Post: No. I have a: Microsoft Windows 11The code for detecting that was added in July 2021, so you probably need to be running something like BOINC v7.16.20 or later. The machine I quoted is running v7.24.1 |
|
7)
Message boards :
Projects :
News on Project Outages
(Message 113649)
Posted 21 Feb 2024 by Richard Haselgrove Post: Issued On Thursday, November 23, 2023 at 6:16:36 PM Expires On Wednesday, February 21, 2024 at 6:16:35 PM |
|
8)
Message boards :
Projects :
CPDN sign up
(Message 113576)
Posted 9 Feb 2024 by Richard Haselgrove Post: That all looks OK. "Couldn't resolve host name" indicates a DNS problem on the internet connection - that would happen at the operating system level, not within BOINC or CPDN. |
|
9)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 113560)
Posted 6 Feb 2024 by Richard Haselgrove Post: I ran a couple from there yesterday. They were uploading a 2.5 MB file about every 5 minutes. If your bored band was taking 18 minutes, that'll be a big backlog. |
|
10)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 113512)
Posted 30 Jan 2024 by Richard Haselgrove Post: ERR_CONNECTION_REFUSED We need another carrier pigeon! |
|
11)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 113485)
Posted 28 Jan 2024 by Richard Haselgrove Post: Yes.Something we're all prone to as we get older!Speaking for yourself? Remember what you ate last night? ;-)Also yes, but I did forget to turn on the dishwasher before going out yesterday. |
|
12)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 113483)
Posted 28 Jan 2024 by Richard Haselgrove Post: ... no one has the root password anymore.David Anderson definitely had an admin password while the project was running, and IIRC has posted on progress in the post-closedown analysis since the hibernation. David has been active on BOINC coding this weekend since the message appeared. I think it's probably simple forgetfulness - somebody initiated a normal backup run and got distracted before they had time to turn the site back on. Something we're all prone to as we get older! |
|
13)
Message boards :
Projects :
News on Project Outages
(Message 113402)
Posted 20 Jan 2024 by Richard Haselgrove Post: Project front page news item - server (Ubuntu) upgrade to latest version yesterday. Some fallout. |
|
14)
Message boards :
BOINC Manager :
Suspend not working on Linux
(Message 113391)
Posted 18 Jan 2024 by Richard Haselgrove Post: Which project(s) is/are you running? This report is consistent with a badly-implemented science application failing to respond in a timely fashion to a 'suspend' message from BOINC. BOINC itself hardly uses the CPU, and certainly not the GPU. It's the science applications which do all the heavy lifting. |
|
15)
Message boards :
Projects :
lhc-dev site seems to have certificate problem
(Message 113379)
Posted 15 Jan 2024 by Richard Haselgrove Post: Mine tells me: Common Name (CN) lhcathomedev.cern.ch |
|
16)
Message boards :
Questions and problems :
when adding a second machine/client , can I just copy the project config files from the first client
(Message 113273)
Posted 13 Dec 2023 by Richard Haselgrove Post: The file controlling which projects you added and all is client_state.xml, you can copy this file to another computer and that computer will then immediately have the projects added that the first computer has.Not a good idea. client_state.xml also contains Host IDs and RPC sequence numbers: duplicating them will trigger BOINC's anti-cheating mechanisms. |
|
17)
Message boards :
Projects :
Dedekind@Home, anyone ?
(Message 113152)
Posted 21 Nov 2023 by Richard Haselgrove Post: I think a reasonable analogy would be the NumberFields@Home Project. That, again, is the research interest of a single, now retired, mathematician - Eric Driver. He seems to manage the steady-state operations of the project (submitting new batches of work, for example) by himself, and has time to read and respond to the (few) issues raised by volunteers on the project website. I think the biggest hurdle for a small-scale project like that is the initial set-up. There are a huge number of different moving parts, each of which would probably be regarded as a separate discipline in a traditional installation: the BOINC data storage and management tools, the website, and the communications and security regime, to name but three. Professional support is probably needed at this point - I think Eric's former employers at the University of Arizona were vital here, and have continued to host his server in the university's infrastructure. I think anyone thinking of setting up a similar project could learn a lot by seeking out the early posts on their website, and reading about the problems encountered and solutions found. |
|
18)
Message boards :
The Lounge :
Let's talk politics
(Message 113092)
Posted 13 Nov 2023 by Richard Haselgrove Post: +10 |
|
19)
Message boards :
The Lounge :
Let's talk politics
(Message 113089)
Posted 13 Nov 2023 by Richard Haselgrove Post: Cruella has ...... been sacked. Good riddance. She wasn't even allowed the weasel's way out of resigning. |
|
20)
Message boards :
Android :
Resetting project "debt"
(Message 113062)
Posted 10 Nov 2023 by Richard Haselgrove Post: The devs don't currently use the term "debt" (although it was present in earlier versions). They use a concept called REC instead. Users of the desktop versions of BOINC have access to a configuration option: <rec_half_life_days>X</rec_half_life_days>I don't think that one's included in the limited number of configuration tools made available in Android. You might get a better response if you phrase it as a leveling-up request: "Why can't I do what other users can do?" |
|
21)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 113007)
Posted 29 Oct 2023 by Richard Haselgrove Post: At least with buses, they have the excuse that pesky passengers keep asking them to stop, so they can get on and off. |
|
22)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 113002)
Posted 28 Oct 2023 by Richard Haselgrove Post: Grumble, grumble. These blasted computer algorithms, can't even arrange a delivery. Ordered some computer gubbins on Thursday - special offer, next day delivery. Next step, carrier pickup and tracking. Nothing happened until Friday evening - despatched, but not going to be delivered until Monday. Saturday morning, I get an email - 1 hour delivery timeslot this afternoon, high value, can't be left unattended, time can't be altered. I re-arrange my day, and wait in. One minute after the slot ended, they change it - will now take to pick-up venue, will be available tomorrow. 12 minutes after that - now available to collect in shop. Drive down, let him scan the email - but we agree that because I've known him for 15 years, and I visit the shop six days a week, we can do without the photo ID. If they can lie to us, we can lie to them. Trouble is - carrier's computer is treating the van driver as a fellow robot. I was #73 on his day's drop-off list - and the weather is foul, and the roads are full of halloween shoppers. Or maybe they're all driving to their pickup points. The driver didn't have a hope in hell of meeting his schedule. |
|
23)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 112984)
Posted 23 Oct 2023 by Richard Haselgrove Post: It's the same as last year, £200 normal winter fuel, plus £300 COL.But last year, everyone with an electricity account - including pensioners - was getting support through their electricity companies. That's been withdrawn, but our extra hasn't. That's why I think it's special treatment for people the Tories think are likely to vote for them. Completely erroneously in my case, of course. |
|
24)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 112978)
Posted 23 Oct 2023 by Richard Haselgrove Post: LOL. The envelope would probably burn for longer, but I'll recycle them both - trying to keep down my carbon emissions. |
|
25)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 112975)
Posted 23 Oct 2023 by Richard Haselgrove Post: The General Election bribery campaign is under way. Just got a letter to tell me that my 'Winter Fuel Payment' will be £500 this year - well up from previous years. It "includes extra money ... known as a 'Pensioner Cost of Living Payment'". Roll on the gravy train. |
|
26)
Message boards :
Questions and problems :
(k)ubuntu 23.10 upgrade causes 'invalid client RPC password. Try re-installing boinc manger' error message
(Message 112952)
Posted 18 Oct 2023 by Richard Haselgrove Post: See my comments on this specific error message in https://github.com/BOINC/boinc/issues/4460 |
|
27)
Message boards :
Questions and problems :
(k)ubuntu 23.10 upgrade causes 'invalid client RPC password. Try re-installing boinc manger' error message
(Message 112949)
Posted 18 Oct 2023 by Richard Haselgrove Post: We (I) think this is a false error message: the password is likely to be unchanged and correct, but the OS upgrade has changed and tightened the security regime. The Manager runs in user mode, but the client (which does all the work and generates the event log messages) runs as a system service. What's probably happening is that the Manager is unable to read the 'gui_rpc_auth.cfg' file which contains the password. The first thing to try is adding your user name for the system, to the 'boinc' user group: sudo usermod -a -G groupName userNameReboot the system to make the change effective. Edit: If that doesn't work, it's possible that your system isn't set to auto-start the 'boinc-client' service at startup. Try sudo systemctl start boinc-clientUse the other systemctl options - stop, disable, enable - to control BOINC from there on. |
|
28)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 112920)
Posted 13 Oct 2023 by Richard Haselgrove Post: I've come across retailers that have an actual status, "waiting for collection by courier". |
|
29)
Message boards :
Questions and problems :
Location of cc_config.xml
(Message 112914)
Posted 13 Oct 2023 by Richard Haselgrove Post: I'm a pretty recent explorer in the world of Linux, but I've always assumed that it's a by-product of the security protocols coupled to the systemd installation model. BOINC is pretty tightly sandboxed in that scenario - BOINC programs can't write out to files outside it's private playpen, and applications launched by any other user can't modify files inside it. My installations have, I think, four user-configurable files accessed via that sort of wormhole into the shared world of /etc/. |
|
30)
Message boards :
Questions and problems :
Improving implementation of components associated with the function “RPC_CLIENT::get_reply”
(Message 112857)
Posted 8 Oct 2023 by Richard Haselgrove Post: The first test is automated, and simply checks that the code is syntactically correct and compiles cleanly on all four supported platforms. It runs immediately after the code changes are uploaded in the required format (GitHub Pull Request). This is a change from when Markus was last active on the BOINC project, 8-12 years ago. Functional testing of the compiled binaries is delayed until the "release candidate" stage, though this is still called 'Alpha testing'. That is when the completed packages, complete with installation tools where appropriate, are tested for functionality. I personally feel that this is wrong: that functional, assembled package, testing should be called 'Beta testing', and there should be an alpha test phase of each PR - after submission, but before packaging - to ensure that proposed changes are compatible with project servers old and new. That can be done already, because the automated syntax and other code checks run right through to producing a compiled binary, which anyone interested can download and run. But that probably won't help Markus. The automated builds are compiled in 'debug' mode, with the hooks to invoke the associated symbol tables for debugging. This means that their efficiency will not be a guaranteed guide to the efficiency of the final 'release' builds. But that is not considered to be a significant problem for BOINC itself: BOINC is an information and management tool, and puts vastly less stress on the volunteers' computers the the projects' "scientific research" applications. They are where efficiency considerations become important. |
|
31)
Message boards :
Questions and problems :
Improving implementation of components associated with the function “RPC_CLIENT::get_reply”
(Message 112845)
Posted 8 Oct 2023 by Richard Haselgrove Post: Would you ever like to take another look at published contributions?If you have ever posted a link to your own published contributions, showing how they would fit into the BOINC environment, I've missed it. Please re-post. |
|
32)
Message boards :
Questions and problems :
Improving implementation of components associated with the function “RPC_CLIENT::get_reply”
(Message 112843)
Posted 8 Oct 2023 by Richard Haselgrove Post: It is just challenging to integrate further development ideas so that more users can benefit also from corresponding software evolution.You simply have to explain clearly what improvements you have made, and provide the actual code you have written in a form which can be tested automatically for compatibility. That's how collaborative enterprises operate. |
|
33)
Message boards :
BOINC client :
Reconsidering the display of percent values in task summaries
(Message 112823)
Posted 7 Oct 2023 by Richard Haselgrove Post: But we do need to add that to the User Manual - thanks for reminding me.Done. |
|
34)
Message boards :
BOINC client :
Reconsidering the display of percent values in task summaries
(Message 112820)
Posted 7 Oct 2023 by Richard Haselgrove Post: boinccmd --get_task_summary c % Done --- 59.74% 28.47% 28.13% 59.65% --- 47.51%On my University course, 50 years ago, I was firmly taught: "A number is meaningless unless the units are stated" But we do need to add that to the User Manual - thanks for reminding me. |
|
35)
Message boards :
Questions and problems :
Reconsidering run time characteristics of the function call “XOpenDisplay(":0")”
(Message 112797)
Posted 3 Oct 2023 by Richard Haselgrove Post: Markus was referred to this message board from Github, where he joined a rwo-year-old conversation started by someone else. We had a long conversation, involving several developers, but we failed to solve his problem - largely because Markus was reluctant to supply any hard data, such as Event Logs and system configuration, which might help us to diagnose the cause of the problem he's experiencing. I got as far as this: As @CharlieFenton said, error -102 is a file read error.As far as I can tell, the file in question is 'gui_rpc_auth.cfg', which contains the password required for client RPC authentication - but that's my interpretation, not Markus'. For a client RPC to work,
|
|
36)
Message boards :
Questions and problems :
Service install.
(Message 112744)
Posted 24 Sep 2023 by Richard Haselgrove Post: You don't have to "delete" anything. Just uninstall the current programs from the control panel, then re-run the installer and make the 'service' change in the advanced settings page. It'll use all your existing projects and settings without change. |
|
37)
Message boards :
Questions and problems :
Service install.
(Message 112742)
Posted 24 Sep 2023 by Richard Haselgrove Post: Yes, the manager will connect and display the same information as before. But you won't be able to detect or use any GPU in the system. |
|
38)
Message boards :
Questions and problems :
GPU computing suspended Computer still downloads GPU tasks
(Message 112721)
Posted 21 Sep 2023 by Richard Haselgrove Post: Suspending any task for a project disables the work requests for that project. You don't need to set NNT for that project.The problem with that method is that suspending a task suspends all work requests for that project - it's not just work for the matching device. I think Dave was trying to suspend GPU requests, while still allowing CPU requests - particularly relevant for a multi-application project like WCG, which offers both types of task. Differential NNT is only available via a project website - when accessible. Edit - the only local solution for this problem I can think of is to use a 'device exclusion' option in cc_config.xml while the project website is broken. <exclude_gpu> |
|
39)
Message boards :
Questions and problems :
GPU computing suspended Computer still downloads GPU tasks
(Message 112718)
Posted 21 Sep 2023 by Richard Haselgrove Post: I think it's the 'snooze' action from right-clicking the system tray icon that's limited to 60 minutes, and I think that the 'suspend' action from the activity menu in the full display is permanent until consciously changed. |
|
40)
Message boards :
BOINC Manager :
BOINC Manager won't display from Taskbar
(Message 112709)
Posted 19 Sep 2023 by Richard Haselgrove Post: I think the full answer to the problems with the system tray icon and the Manager full display will depend on your version of Windows, and the precise 6-monthly patch upgrade that has been applied to it. Under Windows 7, the system tray icon is displayed, and double-clicking on it restores the running copy of the Manager to display in the foreground - with v7.24.1, as well as all previous versions. But starting with Windows 10, Microsoft wants us to start referring to this bit of the screen as the "notification area", and the way they manage it seems to change every six months. To start with, there was a very clear and accessible configuration window which allowed you to choose between "Only show notifications" (default), or "Show icon and notifications" (the old way). But over the years, that configuration page has been doing a fine impression of the Cheshire Cat - vanishing slowly, a bit at a time, and becoming harder and harder to find. Because there are so many variations, it's hard to give good generic advice, but many suggesteions have been posted in internet fora and help pages. If you give us your full Windows version number and patch level, somebody could try and dig one of them out for you. |
|
41)
Message boards :
GPUs :
Not all GPUs running. 3 GPUs only 2 running.
(Message 112655)
Posted 8 Sep 2023 by Richard Haselgrove Post: A helpful tip for future reference: Always check the placement of configuration commands and review your existing settings in the cc_config.xml file if you encounter similar issues.Another helpful tip is to use Options | Event Log options... to make a minor change to your Event Log settings. [I add 'cpu_sched' and 'sched_op_debug' tags, which print useful information without bloating the number of messages too severely]. When you save the changes, BOINC writes a complete fresh cc_config.xml file, in alphabetical order. It includes any user options already in use, and default values for all other possible choices. You never have to add your own tags after that, just find the existing one in the list. The mistake in this case was to add the new "use all..." tag in the 'p' section, just above "proxy_info", rather than to carry on reading to 'u', below the proxy section. The formatting of the file at that point is unhelpful, but just plough on through. The effect in this case was that the second (default) value countermanded the earlier (optional) choice. |
|
42)
Message boards :
GPUs :
Not all GPUs running. 3 GPUs only 2 running.
(Message 112591)
Posted 30 Aug 2023 by Richard Haselgrove Post: Well, it says "ignored by config". so that's the place to start looking. Double-check (proof-read) your cc_config.xml file.
and so on. |
|
43)
Message boards :
Questions and problems :
Systemd timeout stopping boinc-client
(Message 112570)
Posted 21 Aug 2023 by Richard Haselgrove Post: I would like to look at the boinc logs after the systemctl stop to see it they show anything useful, but in the boinc manager, Tools->Event Log is grayed out after the systemctl stop.You can look at the 'back numbers' from the Event Log with a command like: journalctl -b --unit=boinc-clientAs it stands, that will go back as the last reboot and show you everything since then - you may want to redirect the output to a file. Journalctl, like all Linux commands, has an enormous number of other options, but that should be enough to get you started. |
|
44)
Message boards :
Questions and problems :
Returning, now connection problem?
(Message 112564)
Posted 20 Aug 2023 by Richard Haselgrove Post: The schannel message is a complete red herring. It's the default solution to the old problems of BOINC (central) not noticing that one or more of the SSL certificates in boinc's ca-bundle.crt file had expired and needed to have been replaced at the last update. Your next action should have been to run the 'attach to project' wizard, and follow the steps with the email and password appropriate to the project - not your BOINC name/password. Unfortunately, those steps are not reported in the BOINC message log until the whole process is complete. We are completely blindsided here, and you'll just have to note the steps you've taken and do your best to describe them (and the responses) afterwards. Or screenshots (of the wizard's window, not the whole darn cinema screen - Alt-prtscrn) of anything you want help with. |
|
45)
Message boards :
Projects :
News on Project Outages
(Message 112538)
Posted 15 Aug 2023 by Richard Haselgrove Post: The few tasks that got caught (11) have all uploaded without interruption, and reported as complete. I re-enabled work fetch, and immediately got a small allocation, as requested. So it seems to be going smoothly so far, but I'll keep an eye on it. |
|
46)
Message boards :
Projects :
News on Project Outages
(Message 112520)
Posted 11 Aug 2023 by Richard Haselgrove Post: Einstein@Home will be shut down on Monday, Aug 14 at about 4 AM UTC due to necessary maintenance work on the power system of the data center. We expect to be back online roughly 36h later on Tuesday, Aug 15, at the end of the working day (2 PM UTC).Einstein's own news forum says that the project expects to restart on "Tuesday, around 5 PM UTC." That sounds more sensible for the end of the working day in Europe. |
|
47)
Message boards :
Questions and problems :
Missing GPU due to running remote desktop under Windows 11
(Message 112463)
Posted 6 Aug 2023 by Richard Haselgrove Post: This problem has been with us ever since Windows Vista and Windows 7 - it ain't going to go away. It's to do with a fundamental security decision by Microsoft, concerning the way which video drivers interact with the rest of the operating system. In particular, the Microsoft Remote Desktop Protocol disables the 3rd-party NVidia video drivers, and replaces them with Microsoft's own internal, more limited, versions. Various 3rd party remote control systems are available which don't suffer from this restriction - tightVNC is one I remember from early experiments. |
|
48)
Message boards :
GPUs :
"Hardware-Accelerated GPU Scheduling" setting in Display settings -> Graphics -> Change default graphic settings.
(Message 112456)
Posted 5 Aug 2023 by Richard Haselgrove Post: Another setting that's been suggested for multi-use computers is to turn off "Use hardware acceleration when available" or equivalent in the web browser settings. |
|
49)
Message boards :
BOINC Manager :
Boinc won't start on Linux Mint
(Message 112430)
Posted 2 Aug 2023 by Richard Haselgrove Post: I went through exactly the same process some time ago. I was running Mint 20.3, with a late version of BOINC 7.20.x from Gianfranco Costamagna's PPA. The upgrade to Mint 21 went smoothly, but Mint 21 rejected the PPA (all PPAs) because of a change in security policy. Instead, it installed v7.18.1 from it's own repositories - and in my case, it ran. I can only surmise that there is a difference in one of the support libraries between the Mint repository and the PPA: since the problem only affects the Manager component, the usual finger of suspicion points at the wxWidgets GUI component library. We're in a bit of a transitional phase at the moment. The other main platforms (Android, Apple, Windows) have transitioned to v7.22.2, but the Linux repository maintainers have been markedly slow to follow suit - even Gianfranco hasn't got that one out yet. The BOINC developers are working on an in-house Linux installer, but again incredibly slowly. I think it would be helpful if you could escalate this - and your excellent error log - to GitHub, where most issues are thrashed out these days. You could either reopen my https://github.com/BOINC/boinc/issues/4930 (give me a shout if you can't open it, and I'll open it for you), or start a whole new issue. This has been going on for far too long, and a fresh voice on git might dispel the prevailing mood of "Oh no - not him/that again." In the meantime, I've had some success with doing a binary file replacement on Mint 21 - you can download 'artifacts' automatically generated by CI testing on GitHub for any recent pull request. They are statically linked, so you should just need the boincmgr binary. Again, I can guide you through that process. Or Dave can guide you through the steps for a self-build of the Manager on your own system. |
|
50)
Message boards :
Questions and problems :
Systemd timeout stopping boinc-client
(Message 112406)
Posted 29 Jul 2023 by Richard Haselgrove Post: To me, that looks as if systemd tries first with sigterm - I think that's the 'polite' one - and waits 10 seconds. But BOINC doesn't respond, so it moves on to the aggressive one - sigkill. Is hsgamma_FGRP5_1 one of the Einstein@Home applications? They're usually pretty well behaved. It might be worth trying again with a longer delay - see if any more of the process becomes visible in the log. |
|
51)
Message boards :
Questions and problems :
Suspending Computation message meaning
(Message 112393)
Posted 27 Jul 2023 by Richard Haselgrove Post: You haven't mentioned what operating system you're running. Try to locate a tool appropriate for your OS (e.g. Task Manager for Windows) which can give you an independent view of the processes running inside your computer, and the amount of CPU utilisation shown for each one - with some utilities you can sort the columns so it's easy to see the heavy users at the top. It may be that your machine is running a rogue program which is wasting your resources, or alternatively you may have uncovered a flaw in the boinc client making it mis-read the signals the OS is providing. It would be good to know if either of those possibilities is true. |
|
52)
Message boards :
Questions and problems :
Systemd timeout stopping boinc-client
(Message 112390)
Posted 27 Jul 2023 by Richard Haselgrove Post: Quite likely. My initial installation/setup was done from Gianfanco's PPA, which I think sets up the full package of hooks into the host operating system. I've done binary upgrades since then, but just by replacing individual files - not fiddling with the package structure.I have both on my Linux machine. (That is on both the native Linux client and the Windows one running under WINE.)Just wondering if that is anything to do with compiling my own client rather than using the packaged one? |
|
53)
Message boards :
Questions and problems :
Systemd timeout stopping boinc-client
(Message 112385)
Posted 26 Jul 2023 by Richard Haselgrove Post: If he's using systemd, the old logs will be in the system journal, not a text file. Like many Linux functions, there's a hugely complex set of options, too many to list here. I refer to somewhere like https://man7.org/linux/man-pages/man1/journalctl.1.html - you need to filter it down to the boinc unit, and choose a tight time-frame. But it's all there. |
|
54)
Message boards :
Questions and problems :
Systemd timeout stopping boinc-client
(Message 112360)
Posted 25 Jul 2023 by Richard Haselgrove Post: BOINC (internally) has two modes for stopping the client; The "polite" one, where it issues a 'request' to the client to shut itself down. That allows any exit dialogs to be shown, all output files to be flushed to disk and closed, log files likewise, and so on. Only then does the client report back to the calling program that all is complete, and that it's safe to continue with the closedown without risk of data loss or damage. The other mode is much ruder (it has been described by a researcher/system admin as "terminate with extreme prejudice"). It forces the client to stop immediately, without regard to what it's working on. I would keep this one for emergencies only. Having said that, it's unusual for the 'polite' call to delay things by as much as a minute. It might be a 'feature' of a science project you're running (aka 'bug'): while it's closing down, the client goes through a similar call-and-response with each of the science apps it's started. Boinc will only shut itself down after every science app has reported a safe closedown. If just one of the science apps is badly behaved, and doesn't hear the call to close, that might have the effect you're describing. |
|
55)
Message boards :
Questions and problems :
Issue with transfers tab on 7.22.2
(Message 112307)
Posted 18 Jul 2023 by Richard Haselgrove Post: Yes, we got that fixed in 2011: - client: problem: suppose a project has 2 data servers,Even if uploads are backed off, perhaps because of a broken or inaccessible server, every new upload is tried anyway. If the other server is working and accepts the file, that's it dealt with as normal. (SVN 24339) |
|
56)
Message boards :
Questions and problems :
Issue with transfers tab on 7.22.2
(Message 112305)
Posted 18 Jul 2023 by Richard Haselgrove Post: I think the 'project backoff' message only clears when the second server upload is complete and acknowledged. There are some modes of server failure where an upload proceeds to 100% as normal, and only then reports an error: if that happens, the 'project backoff' flag isn't (and arguably shouldn't be) cleared |
|
57)
Message boards :
Questions and problems :
Manager not responding to mouse or KB.
(Message 112274)
Posted 8 Jul 2023 by Richard Haselgrove Post: Could be new with v7.22 I've seen that when different values are set for 'use at most ... % of the CPUs' in the "in use" and "not in use" sections. |
|
58)
Message boards :
Questions and problems :
Boinc 7.18.1+dfdg-4
(Message 112262)
Posted 7 Jul 2023 by Richard Haselgrove Post: The strange thing is, I had it running (briefly) from the Linux Mint 21.0 repository - so it's not completely broken. The client_release/7/7.18 branch was, as usual, a complete copy of the master branch at the time - but the time in question followed an intensive period of activity by the Android developer community, while the Windows/Linux developers were largely absent or not paying attention. The machine it worked on had been running Gianfranco's PPA under Mint 20.3. I upgraded the OS in situ, so not a clean install - it inherited many settings and installed software. But it rejected the PPA, because PPA security was (is?) handled differently in Mint 21. The upgrader was clever enough to recognise that BOINC had been installed previously, and replace it with its own version - and it ran. I haven't yet got to the bottom of how to work out exactly what a Linux package installer contains/does - there's no equivalent (that I know of) to the Windows "/a" administrative install switch, which unpacks all the contents for inspection or deployment. But I think the PPA deploys a matching version of WxWidgets, and the systemd runtime framework. The observed problems with v7.18.1 may be due to faults in one or other of those components, rather than the core BOINC programs themselves. |
|
59)
Message boards :
Questions and problems :
Linux Won't Start Agent
(Message 112196)
Posted 28 Jun 2023 by Richard Haselgrove Post: sudo systemctl enable boinc-clientis the way out of this particular snafu. I do the converse: sudo systemctl disable boinc-clientbefore major system upgrades, like new kernels. That gives time to check that nothing has gone wrong - and also time for the GPU drivers to load before BOINC queries them. I suspect that the BOINC Manager wasn't given any revised 'enable client' code when systemd was introduced - and probably couldn't be, if it needs elevated privileges like 'sudo'. That would be a nice exercise for someone. |
|
60)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 112145)
Posted 26 Jun 2023 by Richard Haselgrove Post: Grumble. Tool-free cases which aren't after n years. Compact cases which need you to remove three covers, two cables, and an expansion card in a PCI-e slot before you can get any leverage on the battery socket. Child-safe packaging which requires the dexterity and ingenuity of an 8-year old to open. And a complex configuration (mine - but I'd forgotten) which requires you to chose which BIOS mode, which boot disk, and which display output you want to run. It took me 40 minutes to complete the replacement, but at least I got the chance to give it a good hoovering. |
|
61)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 112144)
Posted 26 Jun 2023 by Richard Haselgrove Post: Probably even cheaper at Screwfix.And another glory for Screwfix. The power cuts last night reminded me (the hard way) that one of my machines has a flat CMOS battery (flat in the sense of discharged, rather than the form factor). Quick search on screwtix, click and collect, job done. Their local depot is less than a quarter of a mile beyond my newsagent - and I walk there anyway every day. Identify, confirm it's a battery, assistant picks it, handover, walk out. As I left, I passed the next customer who had driven into the car park just before I arrived - coming in. It was as quick as that. And four brand-name CR2032 batteries for £4, no delivery or collection charge. It's not worth buying second-hand non-rechargeable batteries on ebay! |
|
62)
Message boards :
Questions and problems :
Notice from BOINC A new version of BOINC is available (7.22.2). Download 6/26/2023 2:16:41 AM
(Message 112141)
Posted 26 Jun 2023 by Richard Haselgrove Post: Yup, that's one of the ones I tried, and got a successful download from at 08:04 UTC. |
|
63)
Message boards :
Questions and problems :
Notice from BOINC A new version of BOINC is available (7.22.2). Download 6/26/2023 2:16:41 AM
(Message 112139)
Posted 26 Jun 2023 by Richard Haselgrove Post: All the links that I would expect to be involved in that action (whether you upgrade manually or direct from the notice) are working for me this morning. It's possible that Berkeley were doing some maintenance work at that time (your message is timed at about 22:30 Sunday night, in Berkeley time) - something might have been switched off temporarily. Try again later today: if it still fails, can you give us the full URL the notice is trying to lead you to? |
|
64)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 112128)
Posted 25 Jun 2023 by Richard Haselgrove Post: Reason why landline failed is that it is being switched off throughout the country.Be careful with that. I've been living with VOIP for 18 months now, and a side effect of VOIP is no phone service at all - not even 999 - if there's a power cut (because it relies on the router in your home). Make sure you have an alternate kept fully charged. I've just had two (momentary) cuts to foreshadow the thunderstorms predicted for this evening. All the computers had to be restarted, but I've still got two electric clocks to reset :-( |
|
65)
Message boards :
Projects :
Universe Certification date not valid
(Message 112076)
Posted 18 Jun 2023 by Richard Haselgrove Post: They've renewed it (or the automation has kicked in): 
|
|
66)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 112013)
Posted 5 Jun 2023 by Richard Haselgrove Post: Depends where you're reading "estimated". There was a server bug (#4151) which did exactly that, which is now fixed - we fixed it for the forthcoming CPDN MT tasks - but most servers won't yet have the fix. Bit the client - I think - has always recorded elapsed time accurately, and once the initial ropey estimates have worked their way out of the system, they should be accurate, too. Even the "[sched_op] estimated total CPU task duration: 12221 seconds" in the Event Log should be OK. |
|
67)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 111973)
Posted 2 Jun 2023 by Richard Haselgrove Post: Grumble: The font size on the forums and this site in general got bigger. I was perfectly happy with the old one.I've turned it down to 80% so I can see the whole thread list and scan for changes. But - grumble - the font size is even smaller while composing a new message. There'll be more typos... |
|
68)
Message boards :
Questions and problems :
No screen.
(Message 111972)
Posted 2 Jun 2023 by Richard Haselgrove Post: Depends what operating system you run, and how BOINC was installed. If BOINC was installed "as a service" - usual under Linux, an optional choice under Windows - yes, it'll run. If BOINC was installed in "user" mode - possible under Linux, normal under Windows - it won't. I can't advise about Apple Macs. But if you can temporarily borrow a screen from another machine, and use that to log on, BOINC will keep running even if you disconnect the borrowed screen and return it. |
|
69)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111970)
Posted 2 Jun 2023 by Richard Haselgrove Post: BTW, I should have mentioned yesterday that during testing I verified that both my Linux machines can't detect mouse or keyboard activity - so they never enter the 'computer is in use' state. So everything has to be controlled using the 'not in use' settings. |
|
70)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111967)
Posted 1 Jun 2023 by Richard Haselgrove Post: LOL. Having gone through some slow, methodical testing, I think I'm less worried about the new 'in use' / 'not in use' preferences then I was this morning. But I find the new interface non-intuitive and confusing, especially with the font used by my version of Linux. Here are the new ones. Windows 7:  Linux Mint:  For comparison, this is the old version under Windows:  The Linux version has swallowed several words, doesn't allow enough space for all possible numbers, and has horribly uneven vertical alignment. It's a mess. And the old version separated out 'Usage limits' in the top box (they are always applied immediately), from 'When to suspend' in the second box (which require '... based on preferences' in the Activity menu). The new one jumbles them all together. But it does seem to work. This morning, I thought I saw a problem when using a 'new' Manager to modify the settings on a'old' client remotely, but I was mistaken. |
|
71)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111964)
Posted 1 Jun 2023 by Richard Haselgrove Post: I think I'm going to (temporarily) park the specific MT aspects of this discussion - the other ones I saw this morning worry me more. The MT issues - and there are many - should be sorted out with a comprehensive re-think: when to fetch, when to run, how many to run concurrently, what else to run alongside, how many threads to use initially, how to set that figure, whether and how to make that variable during the run. That'll do for an agenda .... But those decisions should be made by a collective design process, ideally including the projects using or thinking of using the MT model. It shouldn't be done by David alone tinkering with careless patches to his own careless code. |
|
72)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111962)
Posted 1 Jun 2023 by Richard Haselgrove Post: But in the extract I quoted, the numbers are different - 8 for 'in use', 6 for 'not in use'. That's prompted me to look at this new feature in more detail, and I think I'm seeing some more serious problems. I need to get my head awake, clear and focussed: then I'll do a comprehensive round-up here, for checking and comment, with a view to starting a new issue on github. But I get we have to get a move on, and keep DA on topic - otherwise he'll wander off for another two months (or three, or four), leaving the release in limbo. |
|
73)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111960)
Posted 1 Jun 2023 by Richard Haselgrove Post: Thu 01 Jun 2023 07:14:23 BST | | - When computer is in useI see you've changed the 'in use' settings, but not the "new this time" not in use settings. Could that be the problem - your machine is one of the ones that has difficulty detecting mouse and keyboard activity? If the subsequent changes from 8 towards 16 were also applied the same way, it would behave as you describe. That raises significant questions for the v7.22 release: if people are going to use the new preferences (and they will), we have to ensure that the detection problems are also fixed. Correction: it's the 'in use' figures which are new this time, but the basic distinction remains. You have to change the settings that match what your computer thinks is happening, not what you think is happening! I've been changing both, because I'm not sure whether my machines handle the detection: I suppose I'd better find out sometime. |
|
74)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111950)
Posted 31 May 2023 by Richard Haselgrove Post: That would be great - I'll keep an eye on things when I get back from lunch. |
|
75)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111948)
Posted 31 May 2023 by Richard Haselgrove Post: I wonder why I didn't see it - what's the difference? I wouldn't mind looking through a complete cycle of cpu_sched_debug, if you could put one where I can see it? Compare it with the one I posted on GitHub for DA, on the different issue I saw. |
|
76)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111932)
Posted 30 May 2023 by Richard Haselgrove Post: Yes, it's always 'next integer below'. |
|
77)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111927)
Posted 29 May 2023 by Richard Haselgrove Post: Can't help, I'm afraid - I just installed whatever was needed from the CERN cerncvm repository - can't find the installation guide right now. computezrmle (a few posts ago) is an expert in this field. |
|
78)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111924)
Posted 29 May 2023 by Richard Haselgrove Post: Just looked the Einstein ones that crashed were all gravity wave 4CPU + GPUThat's odd - I didn't think they had such a beast. I've only ever run tasks designed as 1 GPU plus fractional CPU, although I've usually controlled them with app_config, often to run two per GPU. The latest Gravity Wave GPU tasks have a high GPU memory demand, so I can't run two per card - although I can run 1 GW plus 1 Gamma-ray. They seem to be running down the supply of GW tasks at the moment, so I'm only getting resends: I expect that's in preparation to start clean with a new batch of data. |
|
79)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111922)
Posted 29 May 2023 by Richard Haselgrove Post: Currently, I've only run Amicable in MT mode. The Einstein tasks are defined as GPU-only, but I've allocated them a full CPU core via app_config.xml because of the OpenCL overhead. I run NumberFields as a simple, lightweight, CPU-only project. Have you seen #5257? Several bits of fine-tuning in there. |
|
80)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111920)
Posted 29 May 2023 by Richard Haselgrove Post: LHC also has a dev site with a 'native-mt' (non-Vbox) ATLAS app, though I think it still needs a bespoke comms layer adding in Linux. Their dev site doesn't routinely have work for that app, but the team were most helpful in generating a daily small batch of quick-running test data while I was testing changes to the MT server code prior to release of v1.4.2. You also need an invitation code to join the dev project. I can give you contact details if you want to reach out to them. |
|
81)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111916)
Posted 28 May 2023 by Richard Haselgrove Post: See edit below. |
|
82)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111914)
Posted 28 May 2023 by Richard Haselgrove Post: Do you know if I need the cuda drivers for the Nvidia tasks from Einstein? With the driver I have installed, they just crash as do the amicable numbers GPU tasks.No, I think Einstein is exclusively an OpenCL project. But you do need those drivers, included in the standard NV package. Mine has a 'NVIDIA driver metapackage', described in Synaptic as 'nvidia-driver-470' (older and newer numbers are available) - that one works and is good enough. NVidia enthusiasts discuss driver versions endlessly. |
|
83)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111911)
Posted 28 May 2023 by Richard Haselgrove Post: It snagged another one while I was away from the machine. I tried to follow your problem exactly, but got Sun 28 May 2023 19:20:22 BST | | Reading preferences override file Sun 28 May 2023 19:20:22 BST | | Number of usable CPUs has changed from 5 to 6. Sun 28 May 2023 19:20:22 BST | | [cpu_sched_debug] Request CPU reschedule: Number of usable CPUs has changed Sun 28 May 2023 19:20:22 BST | Einstein@Home | [cpu_sched_debug] reserving 1.000000 of coproc NVIDIA Sun 28 May 2023 19:20:22 BST | Einstein@Home | [cpu_sched_debug] add to run list: LATeah4021L08_1132.0_0_0.0_1708623_1 (NVIDIA GPU, FIFO) (prio -2.979101) Sun 28 May 2023 19:20:22 BST | Einstein@Home | [cpu_sched_debug] reserving 1.000000 of coproc NVIDIA Sun 28 May 2023 19:20:22 BST | Einstein@Home | [cpu_sched_debug] add to run list: LATeah4021L08_1124.0_0_0.0_15781647_0 (NVIDIA GPU, FIFO) (prio -3.041361) Sun 28 May 2023 19:20:22 BST | Amicable Numbers | [cpu_sched_debug] add to run list: amicable_10_21_13911_1685282102.567599_62_0 (CPU, FIFO) (prio -0.001657) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] add to run list: wu_sf3_DS-16x271-21_Grp898638of1000000_0 (CPU, FIFO) (prio -0.019242) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] add to run list: wu_sf3_DS-16x271-21_Grp898642of1000000_0 (CPU, FIFO) (prio -0.019362) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] add to run list: wu_sf3_DS-16x271-21_Grp898879of1000000_0 (CPU, FIFO) (prio -0.019482) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] add to run list: wu_sf3_DS-16x271-21_Grp898643of1000000_0 (CPU, FIFO) (prio -0.019603) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] add to run list: wu_sf3_DS-16x271-21_Grp898644of1000000_0 (CPU, FIFO) (prio -0.019723) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] add to run list: wu_sf3_DS-16x271-21_Grp906352of1000000_0 (CPU, FIFO) (prio -0.019843) Sun 28 May 2023 19:20:22 BST | | [cpu_sched_debug] final job list: Sun 28 May 2023 19:20:22 BST | Einstein@Home | [cpu_sched_debug] 0: LATeah4021L08_1132.0_0_0.0_1708623_1 (MD: no; UTS: yes) Sun 28 May 2023 19:20:22 BST | Einstein@Home | [cpu_sched_debug] 1: LATeah4021L08_1124.0_0_0.0_15781647_0 (MD: no; UTS: yes) Sun 28 May 2023 19:20:22 BST | Amicable Numbers | [cpu_sched_debug] 2: amicable_10_21_13911_1685282102.567599_62_0 (MD: no; UTS: yes) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] 3: wu_sf3_DS-16x271-21_Grp898638of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] 4: wu_sf3_DS-16x271-21_Grp898642of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] 5: wu_sf3_DS-16x271-21_Grp898879of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] 6: wu_sf3_DS-16x271-21_Grp898643of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] 7: wu_sf3_DS-16x271-21_Grp898644of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] 8: wu_sf3_DS-16x271-21_Grp906352of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 19:20:22 BST | Einstein@Home | [cpu_sched_debug] scheduling LATeah4021L08_1132.0_0_0.0_1708623_1 Sun 28 May 2023 19:20:22 BST | Einstein@Home | [cpu_sched_debug] scheduling LATeah4021L08_1124.0_0_0.0_15781647_0 Sun 28 May 2023 19:20:22 BST | Amicable Numbers | [cpu_sched_debug] scheduling amicable_10_21_13911_1685282102.567599_62_0 Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] scheduling wu_sf3_DS-16x271-21_Grp898638of1000000_0 Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched_debug] all CPUs used (6.00 >= 6), skipping wu_sf3_DS-16x271-21_Grp898642of1000000_0 Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched] Resuming wu_sf3_DS-16x271-21_Grp898638of1000000_0 Sun 28 May 2023 19:20:22 BST | NumberFields@home | [cpu_sched] Resuming task wu_sf3_DS-16x271-21_Grp898638of1000000_0 using GetDecics version 400 (default) in slot 3That's as it should be, and doesn't repro yours. So what's the difference? |
|
84)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111909)
Posted 28 May 2023 by Richard Haselgrove Post: See https://github.com/BOINC/boinc/issues/5255 - I've changed my analysis from that initial comment. It only happens when the client contacts Einstein, but it happens every time, not just when work is being requested. |
|
85)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111907)
Posted 28 May 2023 by Richard Haselgrove Post: No project specific preferences are set. If I change %CPUs before downloading tasks the multithread tasks will use the number of available cores. but what I have been doing is increasing the number of cores form 40% to 50% via the manager but BOINC doesn't recognise this and behaves as if only six instead of 8 cores are available. Shutting down the client and restarting does not change this behaviour.My first amicable task has finished, and a single-core task has (re-)started in its place - as it should. That doesn't contradict Dave's first sentence above, but I'll have to try again to reproduce. That'll involve changing the CPU count when there isn't a spare amicable task to run - that's what went wrong last time. Sun 28 May 2023 15:41:05 BST | Amicable Numbers | Computation for task amicable_10_21_2426_1685252702.324398_984_1 finished Sun 28 May 2023 15:41:05 BST | NumberFields@home | [cpu_sched] Restarting task wu_sf3_DS-16x271-21_Grp883275of1000000_0 using GetDecics version 400 (default) in slot 3With regard to the edit: yes, that's my observation too. It might be possible by changing client_state.xml, but I suspect you's have to delve into the slot directory as well. And I think I've found another one - whenever new work is fetched, the Event Log shows that the client re-reads the preferences override file, and prints all the settings. I think that's new - it's normally only read and displayed on request, or when a change is made. |
|
86)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111905)
Posted 28 May 2023 by Richard Haselgrove Post: I see you've reminded David of the original problem in the issue - fair comment. It might be useful if you indicated exactly how you're increasing the core count while running - which preference are you tweaking? That can't be done in the emulator - all files, including prefs and prefs_override, are defined at the start and can't be changed. |
|
87)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111903)
Posted 28 May 2023 by Richard Haselgrove Post: Confirmed in the client emulator - Scenario 201, Simulation 1. Reported on GitHub. |
|
88)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111901)
Posted 28 May 2023 by Richard Haselgrove Post: I've re-opened Dave's issue, to keep it in the public eye while we continue investigating. I think the next move will have to be moving on to the latest available artifact, just to confirm if it's still there. If it is - simulator. Edit - before doing that, I checked yesterday that the single amicable task was using three threads as directed - it was. CPU time was almost exactly three times elapsed time. Today, the second amicable task to start has clocked up 60 minutes CPU in 30 minutes elapsed - so it's starved of cycles. |
|
89)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111899)
Posted 28 May 2023 by Richard Haselgrove Post: Back to business. Sun 28 May 2023 11:09:50 BST | NumberFields@home | Computation for task wu_sf3_DS-16x271-21_Grp890573of1000000_0 finished Sun 28 May 2023 11:09:50 BST | Amicable Numbers | [cpu_sched] Starting task amicable_10_21_2426_1685252702.324398_984_1 using amicable_10_21 version 300 (mt) in slot 2 Sun 28 May 2023 11:10:50 BST | NumberFields@home | [cpu_sched] Preempting wu_sf3_DS-16x271-21_Grp883275of1000000_0 (left in memory) Sun 28 May 2023 11:10:50 BST | NumberFields@home | [cpu_sched] Preempting wu_sf3_DS-16x271-21_Grp881899of1000000_0 (left in memory)Amicable is running on 3 cores, as yesterday. Note the one-minute time delay before pausing the two other tasks. And this is definitely bonkers, and a bug. Sun 28 May 2023 11:20:24 BST | | Number of usable CPUs has changed from 5 to 6. Sun 28 May 2023 11:20:25 BST | Amicable Numbers | Starting task amicable_10_21_2426_1685252702.324398_998_1 Sun 28 May 2023 11:20:25 BST | Amicable Numbers | [cpu_sched] Starting task amicable_10_21_2426_1685252702.324398_998_1 using amicable_10_21 version 300 (mt) in slot 5I'm now running on 8 cores - 2x3 for Amicable, 1 each supporting the 2 GPUs. cpu_sched_debug... And here's the evidence. Sun 28 May 2023 11:25:55 BST | NumberFields@home | [cpu_sched_debug] all CPUs used (8.00 >= 6), skipping wu_sf3_DS-16x271-21_Grp883275of1000000_0 Sun 28 May 2023 11:25:56 BST | | [cpu_sched_debug] Request CPU reschedule: application exited Sun 28 May 2023 11:25:56 BST | | [cpu_sched_debug] final job list: Sun 28 May 2023 11:25:56 BST | Einstein@Home | [cpu_sched_debug] 0: LATeah4021L07_940.0_0_0.0_34824363_2 (MD: no; UTS: yes) Sun 28 May 2023 11:25:56 BST | Einstein@Home | [cpu_sched_debug] 1: LATeah4021L08_1116.0_0_0.0_6637911_0 (MD: no; UTS: no) Sun 28 May 2023 11:25:56 BST | Amicable Numbers | [cpu_sched_debug] 2: amicable_10_21_2426_1685252702.324398_984_1 (MD: no; UTS: yes) Sun 28 May 2023 11:25:56 BST | Amicable Numbers | [cpu_sched_debug] 3: amicable_10_21_2426_1685252702.324398_998_1 (MD: no; UTS: yes) Sun 28 May 2023 11:25:56 BST | NumberFields@home | [cpu_sched_debug] 4: wu_sf3_DS-16x271-21_Grp883275of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 11:25:56 BST | NumberFields@home | [cpu_sched_debug] 5: wu_sf3_DS-16x271-21_Grp881899of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 11:25:56 BST | NumberFields@home | [cpu_sched_debug] 6: wu_sf3_DS-16x271-21_Grp890082of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 11:25:56 BST | NumberFields@home | [cpu_sched_debug] 7: wu_sf3_DS-16x271-21_Grp889806of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 11:25:56 BST | NumberFields@home | [cpu_sched_debug] 8: wu_sf3_DS-16x271-21_Grp890572of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 11:25:56 BST | NumberFields@home | [cpu_sched_debug] 9: wu_sf3_DS-16x271-21_Grp898646of1000000_0 (MD: no; UTS: no) Sun 28 May 2023 11:25:56 BST | Einstein@Home | [cpu_sched_debug] scheduling LATeah4021L07_940.0_0_0.0_34824363_2 Sun 28 May 2023 11:25:56 BST | Einstein@Home | [cpu_sched_debug] scheduling LATeah4021L08_1116.0_0_0.0_6637911_0 Sun 28 May 2023 11:25:56 BST | Amicable Numbers | [cpu_sched_debug] scheduling amicable_10_21_2426_1685252702.324398_984_1 Sun 28 May 2023 11:25:56 BST | Amicable Numbers | [cpu_sched_debug] scheduling amicable_10_21_2426_1685252702.324398_998_1 Sun 28 May 2023 11:25:56 BST | NumberFields@home | [cpu_sched_debug] all CPUs used (8.00 >= 6), skipping wu_sf3_DS-16x271-21_Grp883275of1000000_0 Sun 28 May 2023 11:25:56 BST | Einstein@Home | Starting task LATeah4021L08_1116.0_0_0.0_6637911_0 Sun 28 May 2023 11:25:56 BST | Einstein@Home | [cpu_sched] Starting task LATeah4021L08_1116.0_0_0.0_6637911_0 using hsgamma_FGRPB1G version 128 (FGRPopencl2Pup-nvidia) in slot 1 Sun 28 May 2023 11:25:56 BST | | [cpu_sched_debug] enforce_run_list: end |
|
90)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111898)
Posted 28 May 2023 by Richard Haselgrove Post: I started my course in 1973, after this happened: 1969 Move to new building on an adjacent site. Titan airlifted by crane ('the computing service is suspended').(quotes from the official history. Upgrades, eh?) According to Wikipedia, Algol W was a key language on the IBM 360 range, and I presume 370 as well. That's the one I used. |
|
91)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111897)
Posted 28 May 2023 by Richard Haselgrove Post: OK, back to business - prepping up. This is a 6-core i5-9600KF, which I think is capable of hyperthreading to 12 cores, but I have that turned off in hardware. So, first prep job is to turn that down in preferences, so I can turn it back up later. Sun 28 May 2023 10:31:18 BST | | Number of usable CPUs has changed from 6 to 5. Sun 28 May 2023 10:31:18 BST | | max CPUs used: 5 Sun 28 May 2023 10:31:19 BST | NumberFields@home | [cpu_sched] Preempting wu_sf3_DS-16x271-21_Grp890573of1000000_0 (left in memory)Note that I have activity set to 'run always', not according to preferences, but that preference is acted on anyway. With the 2-week old artifact, Sun 28 May 2023 10:48:22 BST | | Starting BOINC client version 7.23.0 for x86_64-pc-linux-gnu Sun 28 May 2023 10:48:22 BST | | This a development version of BOINC and may not function properly Sun 28 May 2023 10:48:23 BST | | - max CPUs used: 5Now off to find an MT task... Got a couple. Now waiting for something to finish so MT starts up. |
|
92)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111895)
Posted 28 May 2023 by Richard Haselgrove Post: Likewise! I learned Algol 60 at the back of my mother's classes in the school holidays, and we used Algol W as the main language on my diploma course. Algol W was a Stanford University product, so Berkeley probably doesn't recognise it. |
|
93)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111893)
Posted 28 May 2023 by Richard Haselgrove Post: That really does look like a problem and a smoking gun. I've pulled down a 2-week old artifact as well, and I'll deploy that when my caffeine levels have reached optimum - see if I can reproduce it. Then, it's probably off to the simulator - that will be David's first response. I did get a change made to MT handling (#4992) for CPDN/IFS, but that should have been for the server only - and it looks like it is (sched/sched_send.cpp is a server file). The conversation in that PR rather tails off, but I think we got it tested in the end, thanks to LHC. |
|
94)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111890)
Posted 28 May 2023 by Richard Haselgrove Post: Can you show us the bit of the log that says what those 6 cpus are busy with, please? Is that your current Amicable ncpus setting? |
|
95)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111886)
Posted 27 May 2023 by Richard Haselgrove Post: v7.22/23 is all about a whole new set of global preferences, separate ones for 'not in use'. #5251 was a late afterthought by David, who thought a late and undocumented change - which he couldn't remember why he put in - might lead to "Otherwise BOINC will stop computing after 60 minutes of idleness." That would be stopping everything, not just one project, but it's not a million miles from the question you're asking. That's why I want to go on poking and prodding until we understand exactly what's going on, and why. |
|
96)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111884)
Posted 27 May 2023 by Richard Haselgrove Post: OK, I've pulled down the artifacts for #5251 - 2 days old, so probably pretty close to yours. I'll take them for a spin tomorrow. |
|
97)
Message boards :
BOINC client :
Multi core tasks alongside single core tasks.
(Message 111882)
Posted 27 May 2023 by Richard Haselgrove Post: I got involved at github because Vitalii's response includes a suggested resolution from me, and we need to do some proper testing on the latest code (Dave's running v7.23.0 self-build from master). I've joined Amicable, and found I could set the number of CPUs to use in Project Preferences, as well as the workround I suggested. I chose 3, and got a task allocated with: <avg_ncpus>3.000000</avg_ncpus>
<plan_class>mt</plan_class>
<cmdline>--nthreads 3</cmdline> so that bit's working as documented. I'll monitor what it gets up to when running.But this is with a pretty well-tested v7.20.5 from the PPA. Looks like I might have to do some building, too. Edit: And now my 6-core machine is running 3 cores for AM, 1 single-core CPU task, and 2 GPU tasks with a full core assigned to each (ugh - OpenCL). That's what I expect and want, but it took a while for BOINC to respond and pause the other two single-core tasks, so for a while I was running 8 cores. |
|
98)
Message boards :
BOINC client :
BOINC 7.22 Change Log
(Message 111867)
Posted 23 May 2023 by Richard Haselgrove Post: v7.22.2 for Windows has now been added to the 'download all' page for testing. |
|
99)
Message boards :
BOINC client :
BOINC 7.22 Change Log
(Message 111861)
Posted 22 May 2023 by Richard Haselgrove Post: A new test version 7.22.2 for Apple Mac computers is available on the 'download all' page, and should be made available for Windows machines in the next few hours. The only significant change is to correct a bug which prevented the BOINC screensaver running on those two platforms. |
|
100)
Message boards :
Questions and problems :
Where is the tabs on top of the Simple view of BOINC on macOS?
(Message 111844)
Posted 18 May 2023 by Richard Haselgrove Post: Don't know about the Mac but on my Linux BOINC CTRL Shift V takes me to basic view and CTRL Shift A to advanced view.Ditto Windows |
|
101)
Message boards :
Questions and problems :
Is there a limit on sched_request_URL.xml file ?
(Message 111836)
Posted 18 May 2023 by Richard Haselgrove Post: Yes, '0' is a placeholder for "no limit" in that case. I'd leave it like that, and only set an actual limit - and a much lower one - if the server continues to have problems processing a large number at once. |
|
102)
Message boards :
Questions and problems :
Is there a limit on sched_request_URL.xml file ?
(Message 111831)
Posted 17 May 2023 by Richard Haselgrove Post: Not in general, but you can set your own limit in the client configuration file (cc_config.xml) if you are having problems or for testing. <max_tasks_reported>N</max_tasks_reported>(I'd suggest fewer) There can be problems if you are trying to report a large number of tasks after an outage. Such problems are more commonly caused by the size of the request file, or the length of time it takes to process it, rather than the sheer number of tasks reported. Another tip: if the server is being really sluggish, set 'no new tasks' for the time being - the server has less to do if it isn't looking for new tasks at the same time as processing the old ones. |
|
103)
Message boards :
Projects :
connecting to CPDN.
(Message 111814)
Posted 15 May 2023 by Richard Haselgrove Post: And here on Mint 21.1 |
|
104)
Message boards :
Projects :
connecting to CPDN.
(Message 111798)
Posted 14 May 2023 by Richard Haselgrove Post: That's what I got on Friday night too, and emailed to you both. The section that worries me is TLSv1.2 (OUT), TLS alert, unknown CA (560)That feels to me like a server certificate problem on climateprediction.net (only) - either a mixed-up installation, or a damaged file. As Jean-David Beyer pointed out on the CPDN site, these certificate files are actually incredibly simple: just chunks of plain ASCII, concatenated one after the other. If you break then up into single certificate files, even a home version of Windows (sorry!) will decode and display the details - provider, period of validity, and so on. |
|
105)
Message boards :
Questions and problems :
High priority mode?
(Message 111791)
Posted 12 May 2023 by Richard Haselgrove Post: Incidentally, I think Amicable numbers is guilty of overestimating the time tasks will take, certainly on their multi-CPU tasks where it looks to me like the estimate is for if the task is running on just on processor rather than the six I am using.See https://github.com/BOINC/boinc/issues/4151 and subsequent links. Should be fixed in server release v1.4.2 (we went through through some heavy testing on LHC-dev on CPDN's behalf, but I'm not trawling through all the entrails). But if they don't update their server, it'll go on happening. |
|
106)
Message boards :
Questions and problems :
"Use at most N % CPU time" does not seem to have any effect
(Message 111759)
Posted 9 May 2023 by Richard Haselgrove Post: I'll install and check 7.22.1 later today or tomorrow - compare the behaviour with the underlying data files modified by the GUI. We need to check this thoroughly. |
|
107)
Message boards :
Questions and problems :
Effect of suspending and limiting on checkpoints
(Message 111753)
Posted 9 May 2023 by Richard Haselgrove Post: It depends on: The quality of the project's scientific programming. Other settings under your control. Not every project can produce valid checkpoint files, or successfully read them back for a restart after suspension. Not much you can do about that. For CPU tasks (only), leaving applications in memory when suspended should cause minimal data loss on suspension. If the application is removed from memory - as it always will be if BOINC or the computer is restarted - time spent processing since the last checkpoint will be lost and wasted. That could be up to several minutes, again depending on the project. GPU tasks are always removed from memory, so the wastage is unavoidable. |
|
108)
Message boards :
BOINC client :
BOINC 7.22 Change Log
(Message 111745)
Posted 9 May 2023 by Richard Haselgrove Post: v7.22.1 for Android and Windows has crawled onto the 'download all' page. Release notes haven't been updated in the Wiki yet, but a version is available at Client/Manager Release 7.22.1. So far, this is a test/evaluation release only: treat with caution, and watch out for bugs. |
|
109)
Message boards :
Questions and problems :
High priority mode?
(Message 111714)
Posted 3 May 2023 by Richard Haselgrove Post: Someone above my pay grade will have to explain "why?": I just try to explain the "what?" |
|
110)
Message boards :
Questions and problems :
High priority mode?
(Message 111712)
Posted 3 May 2023 by Richard Haselgrove Post: So now we have:* is the problem. It's the only one of the tasks you mention that - on the basis of the information that BOINC has been given by PrimeGrid - is at risk of missing its deadline. The most important policy directive that BOINC has in this situation is "avoid missing deadlines, at all costs". So the 50 day estimate task * gets first stab at the machine's resources, and only other tasks which can fit around it will be run. Before a task has started running, BOINC is only given two pieces of information about it: the number of floating point operations that will be needed to complete the task (estimated by the project staff in advance), and the speed of the device it will be running on (estimated by BOINC from the observed speed of previous tasks run by the same application on the same device). Nothing else. PrimeGrid could possibly make a better attempt to get those estimates right ... |
|
111)
Message boards :
Questions and problems :
High priority mode?
(Message 111710)
Posted 3 May 2023 by Richard Haselgrove Post: The word "priority" is used in many different places, and with many different meanings, according to context - in both computing in general, and BOINC in particular. Please give us an example or two of the context for this particular usage. |
|
112)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 111678)
Posted 28 Apr 2023 by Richard Haselgrove Post: ¡Demasiado caliente! A week on, four cities later, and it's hot, hot, hot. All the time, everywhere. I booked this trip to get some warmth into my bones after what felt like a long, dismal, cold English winter, but I didn't expect an unrelenting, record-breaking, week long heatwave. Nearly 38 degrees C officially yesterday, I've seen 39 on a town centre display, and it goes on into the evening. With nowhere to sit down in the shade. I've retired to an air-conditioned hotel room. And my tour company have booked me into that hotel, with that restaurant, on that terrace. No need to name it - you've seen the posters. Very chic, very small, very booked solid.  Or, from the residents' terrace, two levels below the restaurant,  Tomorrow, I start the long haul home. I hope it rains in Paris, again. |
|
113)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 111650)
Posted 23 Apr 2023 by Richard Haselgrove Post: I could have given you this:  The girl in full costume bumped into her mate in white jeans and the patterned top in front of me - seemingly by chance. There was full flamenco demo going on on the level below, and the music drifted up to us, so they did a quick impromptu warm-up together in the middle of what was going on around them. Nobody batted an eyelid. |
|
114)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 111646)
Posted 23 Apr 2023 by Richard Haselgrove Post: Well, you didn't have to peek - it would have been even easier if I'd given you the name en clair, even in Spanish. I just thought someone might find it something to think about. Christobal Colon (oh, look it up) gave the Spanish King a Royal monopoly on the tobacco trade, just as the Royal Africa Company gave the English a Royal monopoly on the triangular slave trade (which, again, wasn't taught in school, even though I was schooled in Bristol. I only found out recently). And both families became very rich as a result. Hence, presumably, the moat - though whether it was intended to keep the workers and their contraband inside, or the robbers out, remains unclear. I just hope that in a connected world, both countries would avoid making the same type of mistake again - though the signs aren't great. |
|
115)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 111633)
Posted 22 Apr 2023 by Richard Haselgrove Post: I'm on my travels again. One of the delights of random walks around strange cities is the things you find that aren't mentioned in the tour company's walking notes. Midway between a palace described as "the most beautiful building in [the city]" and the "magnificent plaza of ...", I find this deep, dark moat, surrounding another grand building - locked, silent, guarded by CCTV. No advertisement, except the building itself.  Can you guess what, and where, it is? Fortunately, I have a map, which gives me a name. And you can read the rest on Wikipedia. Just one more thing that is gently being forgotten by history |
|
116)
Message boards :
BOINC client :
"SSL Connect Error" BOINC 7.20.2 for Windows 10 22H2
(Message 111569)
Posted 15 Apr 2023 by Richard Haselgrove Post: The change from ca-bundle.crt to schannel was the subject of intense discussion and testing in early October 2021 - I was heavily involved in that process. The first Windows release for v7.20 (7.20.0) became available in mid-June 2022. Details in https://github.com/BOINC/boinc/pull/4545. My question would be - what is the internet access like for other, more generic, applications on the affected machines? Do web browsers connect to secure sites normally? Is Windows Update working normally? Or is the connection problem limited, specifically, to BOINC and only BOINC? |
|
117)
Message boards :
Questions and problems :
Cannot install/uinstall/run BOINC after disk crush
(Message 111527)
Posted 8 Apr 2023 by Richard Haselgrove Post: There is an FAQ wiki which may be helpful. You have probably hit Error 1714: older version cannot be removed or Error 1327: invalid disk. The easiest route out of that is the one suggested at Step 4 - use regedit to remove the old installation location from the registry. |
|
118)
Message boards :
Questions and problems :
After updating to Darwin 22.4.0 (Macos Ventura 10.3) on intel, lost most live data and all statistics charts.
(Message 111526)
Posted 8 Apr 2023 by Richard Haselgrove Post: No, those are NOT the same person. The regular BOINC developer for the Mac platform is Charlie Fenton, https://github.com/CharlieFenton. |
|
119)
Message boards :
Projects :
News on Project Outages
(Message 111458)
Posted 31 Mar 2023 by Richard Haselgrove Post: Sounds appropriate. |
|
120)
Message boards :
Projects :
News on Project Outages
(Message 111451)
Posted 30 Mar 2023 by Richard Haselgrove Post: Which would make it around 4 AM, local time. I think that's unlikely. |
|
121)
Message boards :
BOINC Manager :
Boinc stuck connecting to localhost
(Message 111406)
Posted 24 Mar 2023 by Richard Haselgrove Post: BOINC doesn't "need" WSL. Some projects concentrate on particular operating system, for some or all of their research programs. If you have WSL installed and active, you can participate in projects that only have native applications for Linux. Without it, you can only be of use to projects which have native Windows applications. The choice is up to you. |
|
122)
Message boards :
Questions and problems :
Too many WUs were downloaded, all settings are ignored
(Message 111341)
Posted 18 Mar 2023 by Richard Haselgrove Post: I have heard that it is a bit tricky to persuade Ubuntu to work together with a newer version. But if it is needed, I would try.You could try https://launchpad.net/~costamagnagianfranco/+archive/ubuntu/boinc I use that on Linux Mint 20.3. It doesn't install on Mint 21.1, which I think is based on Ubuntu 20.4, but that may have different security policies. If you can load it, it's been reliable for me. |
|
123)
Message boards :
Questions and problems :
Too many WUs were downloaded, all settings are ignored
(Message 111336)
Posted 18 Mar 2023 by Richard Haselgrove Post: He says he's running "I run Ubuntu 20.04LTS with BM 7.16.6 which is the latest one for Ubuntu20." By BM, I take it to mean mean the v7.16.6 package of BOINC Manager+client. He's also running app_config.xml files with <max_concurrent> entries: that combination is known to be problematic. Short of running experimental versions, the answer is probably to watch out for v7.20.5 or v7.22.x being added to the Ubuntu repositories, and upgrade as soon as possible. |
|
124)
Message boards :
Questions and problems :
Which phone.
(Message 111314)
Posted 16 Mar 2023 by Richard Haselgrove Post: My phone company (British Telecom, using shared infrastructure provided by Openreach) did the same thing to me 18 months ago. But the BT-supplied router I already had provided a compatible phone socket, ready for use. I did a bespoke modification to my 30 year old house wiring, and plugged in the matching patch cable (about £2), and I have a working fixed phone in every room again. I find that's preferable to carrying a mobile (or even a cordless handset) from room to room. The drawback is that even the fixed phones lose connection if there's a power cut or the router loses connection. So far, I've been able to use a mobile to ring up BT every time it happens, but in a wide-area outage, even that might not work. You might be able to get a VOIP router (that's what they're called), either from the phone company or independently. But check with the phone company that they can transfer your number to a VOIP service provider. |
|
125)
Message boards :
BOINC Manager :
Boinc stuck connecting to localhost
(Message 111265)
Posted 12 Mar 2023 by Richard Haselgrove Post: Next time it happens, the first thing to check is "Which BOINC components are actually running?". Open the Windows Task Manager. You should see three BOINC programs running:
boincmgr.exe is the thing you're looking at; boinc.exe is the thing it's trying to connect to; and boinctray.exe you can ignore (it's a small, invisible, helper application). |
|
126)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 111252)
Posted 10 Mar 2023 by Richard Haselgrove Post: I wouldn't help if I were there either, ...She came back just after I'd started my fourth shift. We worked together for a while, and then she sent me back in, while she continued with the end where we park our cars. Fair dos. (she's an ex-postie - postal delivery worker - so she knows how important it is for her current colleagues, to at least have the paths clear) |
|
127)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 111249)
Posted 10 Mar 2023 by Richard Haselgrove Post: Mine's the same. Fortunately, clearing the yard is very therapeutic: shovel to the left, and scoop it over the wall. Not wise to do too much at once, at my age: 20 minutes will do for a stretch.  One neighbour walked past and simply said 'hi': the other at least said she'd lend a hand when she gets back from walking the dog. |
|
128)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 111247)
Posted 10 Mar 2023 by Richard Haselgrove Post: It's finally arrived. This is the scene from my front door this morning:  The kids are enjoying the recreation ground to the left, but it looks like I've got some work later with the snow shovel. |
|
129)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 111218)
Posted 8 Mar 2023 by Richard Haselgrove Post: Well, I made it to the grocery, so I should be OK over the weekend. And emerged to the first few tentative flakes of the white stuff. I'm at nearly 54° north, so you might think the comparison would be with Canada rather than California. But our Western ocean has a far more benign influence over our local climate than yours does. We get far fewer heavy snow events than you do, and as a result, we can - normally - afford to be complacent about them. We own, and can deploy, far fewer snow clearance vehicles for road/rail/airport use: and most of our drivers have never heard of special winter tyres/tires. In that respect, our behaviour is probably more similar to Texas! |
|
130)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 111207)
Posted 7 Mar 2023 by Richard Haselgrove Post: No, it's spring (according to Met office rules). But it's still below zero here, with a crystal-clear full moon in the night sky. Our date with winter is set for Thursday, with up to 20 cm snow and warnings coming at us with increasing urgency. It's not much, by world standards, but quite enough to paralyse the British road network (and possibly more besides). I feel the need for a pre-emptive visit to the grocery store tomorrow. |
|
131)
Message boards :
Questions and problems :
"Use web preferences" https://mma.xyz/boinc
(Message 111120)
Posted 25 Feb 2023 by Richard Haselgrove Post: The url is supposed to come from the last project where you made a change to your global computing preferences. It's stored in the file 'global_prefs.xml', together with a <mod_time> (unix timestamp) which shows when the last change was made. But I can't find any current BOINC project with the address 'https://mma.xyz/boinc'. Perhaps you joined a short-lived BOINC experiment? I suggest you visit the website of one of your current projects: make an alteration (or create a new set of preferences) there; update your client from that project; check that the live project is now shown in the dialog; and move on from there. |
|
132)
Message boards :
Questions and problems :
CRC checks?
(Message 111056)
Posted 14 Feb 2023 by Richard Haselgrove Post: All my CPDN data files (where this query first arose) have MD5 checksums - either calculated by the server when the file was first loaded to form part of a workunit, or (for an upload file) by the client, when that stage of the computation has completed. Application files are checked for a more sophisticated digital signature. https://github.com/BOINC/boinc/blob/master/client/cs_files.cpp#L127 has this comment: // verify_contents // if true, validate the contents of the file based either on = // the digital signature of the file or its MD5 checksum. // Otherwise just check its existence and size.(and more) That looks good enough for now - we can dig deeper if the errors continue. |
|
133)
Message boards :
Questions and problems :
CRC checks?
(Message 111054)
Posted 14 Feb 2023 by Richard Haselgrove Post: It seems to me that BOINC should ideally do a CRC check on all downloaded files and retry if the check doesn't match.I think it already does, but I'll check the sources. |
|
134)
Message boards :
Questions and problems :
CRC checks?
(Message 111051)
Posted 14 Feb 2023 by Richard Haselgrove Post: There are two sets of files: "project" files and "task/workunit" files.The project files are typically downloaded when you you attach to a project, and often repeated at subsequent contacts. Image files are usually sent in the first group, and are often in a lossy compressible format like JPEG, for eye candy in simple view - intermediate hosts on the internet download path can re-compress them to save bandwidth, which is what the CRC option is designed to allow. In general, task data will be sent in some format like zip, which doesn't introduce lossy changes - the unzipped file should be an exact match to the original. I saw the original posts which possibly triggered this question: I have a nagging fear that BOINC sometimes miscounts when a large download is interrupted part way through and restarted from an intermediate point, but that tends to show up as 'xxxxx expected, yyyyy received' errors - possibly only with debug logging. I'll keep my eye open. |
|
135)
Message boards :
BOINC client :
Authorization required, but no authorization protocol specified
(Message 111037)
Posted 11 Feb 2023 by Richard Haselgrove Post: Where is that keyword 'shuttle' coming from? It's not used in any BOINC program (I've checked the source code). And what is trying to talk to what, without authorisation? |
|
136)
Message boards :
Questions and problems :
BOINC manager crashes with RPC password error, but not when visible
(Message 111022)
Posted 8 Feb 2023 by Richard Haselgrove Post: Does that equate to my unresolved bug report https://github.com/BOINC/boinc/issues/4460? |
|
137)
Message boards :
BOINC client :
BOINC 7.22 Change Log
(Message 110959)
Posted 25 Jan 2023 by Richard Haselgrove Post: The test versions have been withdrawn, because of a major problem with the handling of user preferences. |
|
138)
Message boards :
BOINC client :
BOINC 7.22 Change Log
(Message 110956)
Posted 24 Jan 2023 by Richard Haselgrove Post: TEST VERSION released for evaluation. Windows, Mac and Android versions available via the download all page. A query has been raised over the build procedure for Linux - might be advisable to wait for that one. |
|
139)
Message boards :
Questions and problems :
Installing new BOINC version over old
(Message 110944)
Posted 20 Jan 2023 by Richard Haselgrove Post: No |
|
140)
Message boards :
Questions and problems :
Installing new BOINC version over old
(Message 110940)
Posted 20 Jan 2023 by Richard Haselgrove Post: No |
|
141)
Message boards :
Projects :
If I upgrade my CPU can I keep the existing WU's?
(Message 110923)
Posted 16 Jan 2023 by Richard Haselgrove Post: It's probably not recommended, but I don't see why not - provided you're prepared to accept a small risk of things not working out as planned. I would: 1) Set 'No New Tasks' for all projects. 2) Suspend all tasks, except the ones you're actually running at the moment. 3) Wait until those have finished, and 'update' the project(s) to report those results to the project. 4) Close down BOINC completely - client and manager, not just the manager. Do whatever your operating system requires to prevent BOINC starting when you boot the machine. 5) Install the new hardware. Take good care to ensure that your new i7 is fitted with a good cooling system, and that it is properly in contact with the CPU. 6) Boot the system. Check it thoroughly for error messages, do any software updates that may be advised. Boot it again for luck. 7) Re-enable BOINC, and start it. Run a few tasks - the least important ones first - to check. Ramp up slowly to full production, keeping an eye on the temperatures. And that's about it. Others may chime in with suggestions - good luck. |
|
142)
Message boards :
Questions and problems :
Request new tasks- BOINC
(Message 110873)
Posted 4 Jan 2023 by Richard Haselgrove Post: Are you using an Account Manager (BAM!, Science United)? |
|
143)
Message boards :
Questions and problems :
What is "too many runnable tasks"?
(Message 110868)
Posted 4 Jan 2023 by Richard Haselgrove Post: Presumably, not a horrendous task for someone who rolls their own to increase that?Make sure you read the full context of that change. That makes it even easier! - Make max runnable jobs (1000) into a #define(from https://github.com/BOINC/boinc/commit/40f0cb44f4fcd11eb2789408dfc868de63e42242) https://github.com/BOINC/boinc/blob/master/client/client_state.h#L613 |
|
144)
Message boards :
GPUs :
OpenGL error?! I thought we used OpenCL here?
(Message 110856)
Posted 3 Jan 2023 by Richard Haselgrove Post: OpenGL and OpenCL are different beasts - a Graphics Language and a Computing Language, respectively. OpenGL is commonly used for project graphics apps or screensavers. |
|
145)
Message boards :
Questions and problems :
What is "too many runnable tasks"?
(Message 110848)
Posted 2 Jan 2023 by Richard Haselgrove Post: It'll be a local BOINC client decision. The clue is in the "Not requesting ..." - that's a client decision. The project server isn't involved, so you probably asked for 101 tasks at a moment when you had 999 runnable, or something like that. The question is: why? The purpose would be to stop BOINC wasting computer power on housekeeping, when you could be doing science. And it helps the scientists get their research back in a timely fashion. |
|
146)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110841)
Posted 1 Jan 2023 by Richard Haselgrove Post: What's the point of fireworks in daylight? (no, don't ask them - I doubt the answer would be either polite or intelligible) |
|
147)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 110776)
Posted 19 Dec 2022 by Richard Haselgrove Post: On this side of the pond, a story like that would raise concerns about the package being intercepted between Amazon and purchaser, and the contents replaced. In other words, it's more likely that a rogue van driver and his mates are enjoying the laptop, rather than any Amazon employee. It would still be Amazon's job to sort out the mess, though. |
|
148)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110767)
Posted 18 Dec 2022 by Richard Haselgrove Post: Don't go outside: ice rain has changed our roads into ice surfaces.Managed my UK beer run without problems. It was wet, but not icy. |
|
149)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 110752)
Posted 17 Dec 2022 by Richard Haselgrove Post: How quickly does new technology become ancient history ... Standing in line at the car park Pay & Display machine, the guy in front looked confused and puzzled. His passenger called across: "Should I pay by phone?". He turned to me: "Do you know how this thing works?" "Put coins in the slot, then press the green button?", I suggested. He got the coins into the slot OK, but then pressed the green blob on the instruction panel ... To be fair, that machine has been standing out in all weathers for so many years that most of the green has worn off the 'OK' button, and it's nearly black. |
|
150)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 110750)
Posted 17 Dec 2022 by Richard Haselgrove Post: No, I am not participating.Nor am I. |
|
151)
Message boards :
Projects :
OpenIFS for CPDN
(Message 110709)
Posted 12 Dec 2022 by Richard Haselgrove Post: Uploads update Mon 12th(this affects all CPDN sub-projects, not just IFS) |
|
152)
Message boards :
Questions and problems :
BOINC "High Priority" mode issues!
(Message 110690)
Posted 10 Dec 2022 by Richard Haselgrove Post: Set a shorter cache size. |
|
153)
Message boards :
Questions and problems :
How to edit 'Select computers' list?
(Message 110670)
Posted 7 Dec 2022 by Richard Haselgrove Post: Operating system? If Windows, it's in the Registry - use regedit. Windows Registry Editor Version 5.00 [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager\ComputerMRU]Close BOINC Manager before attempting an edit. |
|
154)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110638)
Posted 5 Dec 2022 by Richard Haselgrove Post: And have done every year since 1972 - so this is the 50th anniversary. The amount has never changed. As an indicator of how much prices have changed, I moved to Yorkshire in 1974 - and was amused to find that there had clearly been an antediluvian unit of Yorkshire currency, namely one-sixth of a pound. Rounded to 17p after decimalisation, one 'Yorkie' would buy you one pint of beer, or one portion of fish'n'chips. Tonight's pint cost me £3.95, and the (good) local chippy is charging £6. |
|
155)
Message boards :
Projects :
OpenIFS for CPDN
(Message 110579)
Posted 28 Nov 2022 by Richard Haselgrove Post: Best restrict these to powerful machines with a good internet connection.They have a high RAM demand per task, and upload a 14 MB result file several times an hour. |
|
156)
Message boards :
Questions and problems :
Downloadpage Boincmanager 7.20.2 for Windows show Windows 64-bit Vista/7/8/8.1/10
(Message 110569)
Posted 27 Nov 2022 by Richard Haselgrove Post: Windows 11 is supported. The page is just a little out of date. |
|
157)
Message boards :
GPUs :
Disable downloading of Tasks that use GPU?
(Message 110552)
Posted 24 Nov 2022 by Richard Haselgrove Post: Easiest way is an option on the project website: https://einsteinathome.org/account/prefs/project Top section, for the venue specified for that computer, has a yes / no switch: Use AMD GPU |
|
158)
Message boards :
The Lounge :
Testing
(Message 110509)
Posted 20 Nov 2022 by Richard Haselgrove Post: 'Copy link address' (right-click in Chrome) renders the last one as http://%22https//boinc.berkeley.edu/%22 |
|
159)
Message boards :
Questions and problems :
Boinc.exe terminates at start on W10
(Message 110466)
Posted 15 Nov 2022 by Richard Haselgrove Post: Just thinking about this. It shouldn't be difficult to get the installer to flag up a warning that to use the repair option you need to be running the same version of BOINC as the installation being repaired. Richard?I always thought the 'repair' option was only offered if the 'installer' version matched the previous 'installed' version. I'll have to look more closely next time. The Windows installer uses commercial (InstallShield) technology, licensed to UCB. I've worked with very early versions of InstallShield, 25 years ago, but by all accounts it's grown far more complicated since then. Rom Walton used to punt all installer bugs into the long grass, pending the arrival of an open source installer which was going to swim over the horizon and save us all. It never arrived. |
|
160)
Message boards :
Questions and problems :
No GPU dectected when started from systemd
(Message 110434)
Posted 13 Nov 2022 by Richard Haselgrove Post: It can sometimes be a timing issue. BOINC detects GPUs by querying the drivers - and if the drivers haven't finished loading properly, they don't give a usable reply. Try stopping the service before you reboot, and disabling it. Then, after a reboot, wait until you have a usable desktop again. Then, enable the service, and start it manually. Works for me. |
|
161)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110417)
Posted 11 Nov 2022 by Richard Haselgrove Post: It gets weirder. Your order has been packedThis is a collection, remember. |
|
162)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110414)
Posted 11 Nov 2022 by Richard Haselgrove Post: It's a chemist, which runs a chain of retail outlets and some franchise outlets in big supermarkets. They have an online link to see if your nearest is affected by the delay, and it brings up a list of 420 outlets, as far south as Lincoln. My local outlet, 5 km away, is in a branch of Sainsbury's which is open 7 days a week. But I don't know what days their van drivers work. |
|
163)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110409)
Posted 10 Nov 2022 by Richard Haselgrove Post: Grumble. For Click & Collect orders, if your chosen pickup point is in the North West, North East, Yorkshire, the Isle of Man, Channel Islands or Scottish Highlands and Islands your order will take an additional working day to our stated delivery times.First time I've seen Yorkshire included in the list of trackless outer fringes of the UK. Send huskies! |
|
164)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 110320)
Posted 4 Nov 2022 by Richard Haselgrove Post: And the download servers are playing hard to get. |
|
165)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110301)
Posted 2 Nov 2022 by Richard Haselgrove Post: Agreed. But since Mint is derived from Ubuntu, it might be worth keeping an eye on the next Ubuntu release, in case action is needed there, |
|
166)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110299)
Posted 2 Nov 2022 by Richard Haselgrove Post: Thanks again. So, we can take it that cvmfs runs OK on Mint 20.3 But it doesn't on Mint 21. I get: richard@Rosa:~$ sudo apt-get install cvmfs [sudo] password for richard: Reading package lists... Done Building dependency tree... Done Reading state information... Done cvmfs is already the newest version (2.9.4~1+ubuntu20.04). The following packages were automatically installed and are no longer required: libflashrom1 libftdi1-2 Use 'sudo apt autoremove' to remove them. 0 to upgrade, 0 to newly install, 0 to remove and 5 not to upgrade. richard@Rosa:~$ sudo cvmfs_config setup Error: failed to load cvmfs library, tried: './libcvmfs_fuse3_stub.so' '/usr/lib/libcvmfs_fuse3_stub.so' '/usr/lib64/libcvmfs_fuse3_stub.so' './libcvmfs_fuse_stub.so' '/usr/lib/libcvmfs_fuse_stub.so' '/usr/lib64/libcvmfs_fuse_stub.so' ./libcvmfs_fuse3_stub.so: cannot open shared object file: No such file or directory libcrypto.so.1.1: cannot open shared object file: No such file or directory /usr/lib64/libcvmfs_fuse3_stub.so: cannot open shared object file: No such file or directory ./libcvmfs_fuse_stub.so: cannot open shared object file: No such file or directory libcrypto.so.1.1: cannot open shared object file: No such file or directory /usr/lib64/libcvmfs_fuse_stub.so: cannot open shared object file: No such file or directory Failed to read CernVM-FS configurationCould you pass that back to the LHC team, please? All those fuse files seem to be present in /usr/lib: 
|
|
167)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110297)
Posted 2 Nov 2022 by Richard Haselgrove Post: And having said that, it bailed out after a further 20 minutes. Uploaded and validated - task 368415294. Looks OK to me. |
|
168)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110296)
Posted 2 Nov 2022 by Richard Haselgrove Post: Thanks for looking into this. I can post the error messages, if it would help? Where would you like them? My second machine (Linux Mint 20.3, very similar hardware) accepted the cvmfs quickstart settings without complaint, and is currently an hour into its first ATLAS native task. It's recorded 4 hours 39 minutes of CPU time in that hour (6 cores, nothing else running, but slow to start the MT phase), so I think this one's OK. |
|
169)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110291)
Posted 1 Nov 2022 by Richard Haselgrove Post: Mint 20.3 seems to take the cvmfs installation. Part 2 tomorrow. |
|
170)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110288)
Posted 1 Nov 2022 by Richard Haselgrove Post: I've been trying - and failing - to run these tasks, but I need cvmfs and it's not installing on Mint 21. I'll take a break, and then try installing it on the older and more stable Mint 20.3 |
|
171)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110286)
Posted 1 Nov 2022 by Richard Haselgrove Post: And I've snagged three. 01/11/2022 11:41:36 | LHC@home | Requesting new tasks for CPU 01/11/2022 11:41:36 | LHC@home | [sched_op] CPU work request: 16034.67 seconds; 0.26 devices 01/11/2022 11:41:37 | LHC@home | [sched_op] estimated total CPU task duration: 23975 secondsThey are designated to run on all six cores, and estimated to run for 2 hours 13 minutes - I make that a total of 143,784 core-seconds. |
|
172)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110285)
Posted 1 Nov 2022 by Richard Haselgrove Post: OK, I've written to Laurence, and attached my Linux boxes. I'm trying to stay native - don't want to add VBox as well - and they may prefer to use their dev site. We'll see what happens. |
|
173)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110283)
Posted 1 Nov 2022 by Richard Haselgrove Post: Or LHC?I was going to say I hadn't seen any MT apps there, but I see they've now got a long_native_mt plan task. No work currently, but I'll ask Laurence. |
|
174)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110278)
Posted 31 Oct 2022 by Richard Haselgrove Post: Looks like we have a taker: https://github.com/BOINC/boinc/pull/4992. Now, how can we persuade MilkyWay or PrimeGrid to load it for testing? |
|
175)
Message boards :
Questions and problems :
Effect of settings on multi-core tasks.
(Message 110265)
Posted 28 Oct 2022 by Richard Haselgrove Post: Already reported at https://github.com/BOINC/boinc/issues/4151 (Jan 2021). This one should be trivial to fix, if anyone can be bothered. |
|
176)
Message boards :
Projects :
News on Project Outages
(Message 110245)
Posted 26 Oct 2022 by Richard Haselgrove Post: One of the researchers at GPUGrid has posted an explanation for their email problem: The issue comes from the fact that the GPUGrid server uses a public IP from the Universitat Pompeu Fabra, so we have to comply with the data protection and security policies of the university. Among other things this implies that we can not send emails from our web server.(message 59003) |
|
177)
Message boards :
Questions and problems :
Error 1325 templates is not a valid short file name
(Message 110224)
Posted 25 Oct 2022 by Richard Haselgrove Post: I've got to go out for a short while, but - in case you need it - I've found where Jord has hidden the sledgehammer. BOINC FAQ for Windows installer Error 1325 I tried to talk you through solution 1: there are six more to try if you're still stuck. |
|
178)
Message boards :
Questions and problems :
Error 1325 templates is not a valid short file name
(Message 110220)
Posted 25 Oct 2022 by Richard Haselgrove Post: Ah - that helps. We know that one. The problem is that installing BOINC is a two-stage process. First, it removes (or tries to remove) the previously installed version: only then does it even attempt to install the new one. Yours is failing at the 'remove' stage. There's almost certainly nothing wrong with the new version you've downloaded. The easiest way to fix this is to try re-installing the exact same version. If you've still got the installer file you used last time, that's great. If not, you'll need to research the version details - look at your account page at one of the projects you run - and find the download in https://boinc.berkeley.edu/dl/?C=M;O=D. When you locate the right version, and run it, it should offer to 'repair' the old installation. Let it do that, and run right through to completion. THEN, try installing the new version. The 'remove' stage should work properly after the repair, and continue on to install the new version. If you can't make it work, come back here and we'll try something else. |
|
179)
Message boards :
The Lounge :
Let's talk politics
(Message 110215)
Posted 25 Oct 2022 by Richard Haselgrove Post: Now, Tory MPs won't know what to think. WhatsApp down for users across UK and world |
|
180)
Message boards :
Questions and problems :
Error 1325 templates is not a valid short file name
(Message 110214)
Posted 25 Oct 2022 by Richard Haselgrove Post: Which version of BOINC? Which operating system? What is the 'installer' file called? Where did you download it from? |
|
181)
Message boards :
The Lounge :
Let's talk politics
(Message 110202)
Posted 24 Oct 2022 by Richard Haselgrove Post: Correction - "the answer they think you want to hear" They aren't always entirely right in interpreting the question. |
|
182)
Message boards :
The Lounge :
Let's talk politics
(Message 110198)
Posted 24 Oct 2022 by Richard Haselgrove Post: He made that (unverified) claim of 102 supporters last night, but there were nowhere that number of MPs prepared to state their backing in public. That means that either Boris made up the number - wouldn't be the first time - or Tory MPs are embarrassed about being seen to back him. Neither would be a good way of starting his second term. |
|
183)
Message boards :
The Lounge :
Let's talk politics
(Message 110196)
Posted 24 Oct 2022 by Richard Haselgrove Post: This is the first time Boris Johnson has been known to pull out of anything. Oh, except the EU. |
|
184)
Message boards :
The Lounge :
Let's talk politics
(Message 110174)
Posted 22 Oct 2022 by Richard Haselgrove Post: He's definitely marmite. Some people are saying he's the only leader that can save them: others say they'll resign or switch to another party if he comes back. And that's just the Tory MPs. Edit - here's one of them. https://www.youtube.com/watch?v=oIFDt4HdLuE |
|
185)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 110158)
Posted 21 Oct 2022 by Richard Haselgrove Post: And there's no sign of any new work today, either - except the occasional one or two stragglers limping in. I hope Krembil are aware - but I fear that they are not - that for a successful scientific research project under BOINC, all the moving parts have to be working at the same time. Work has to be generated, allocated, downloaded, processed, uploaded, validated, and assimilated. The entire pipeline moves at the speed of the slowest. And if one piece breaks, they're all broken. |
|
186)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 110135)
Posted 19 Oct 2022 by Richard Haselgrove Post: And it feels as if they've been turning the up/download servers off and on again, every few seconds. That doesn't bode well. |
|
187)
Message boards :
Questions and problems :
Milkyway@Home - Massive Performance Issue - 1 Job Taking 8 CPUs vs 8 Jobs Taking 1 CPU Each
(Message 110120)
Posted 16 Oct 2022 by Richard Haselgrove Post: The BOINC Statistics display for host average.By which.I assume the graphical display in BOINC Manager. Many years ago, as an undergraduate at one of the major physics labs in the UK, I was taught two lessons which have stayed with me for over fifty years (the actual physics, sadly, has not). Lesson 1: Do every calculation twice. Once, using the best technology and the highest precision available (at the time, that was a sliderule). And again, with a pencil on the back of an envelope, to 'order of magnitude' - nearest power of 10 - precision only. That checks that you put the decimal point at the right place in the first answer. Lesson 2: A number is meaningless unless you state the units it's measured in. In this case, your numbers are 7000, 2000, 4500. Those are measured in "BOINC credits". They should be equivalent to "cobblestones", which is a defined number in terms of the number of calculations performed in reaching the scientific answer. Many of us wish that this was still the case, so that we can do the sort of comparisons that you are attempting. But unfortunately, and disappointingly, the direct link between 'work done' and 'credits awarded' was broken over 10 years ago. Each project is free to choose its own credit reward rate, and as you have found, they don't all keep that consistent, even between the different task types within their own project. As things stand at the moment, you have two options. Either take a deep breath, relax, and stop worrying about it. Or take it up with the research/administration team at Milkyway@Home: they are in control of their own credit rewards, and have the power to change it. But it may be low on their list of priorities. |
|
188)
Message boards :
Questions and problems :
Milkyway@Home - Massive Performance Issue - 1 Job Taking 8 CPUs vs 8 Jobs Taking 1 CPU Each
(Message 110114)
Posted 15 Oct 2022 by Richard Haselgrove Post: The statistics bear this out ...Which statistic, exactly, are you quoting when making that statement? It may be that you've uncovered a blemish in BOINC's record-keeping. My experience is that recorded CPU time (or, more accurately, recorded core time) does usually come out as slightly below 8 times elapsed (wall-clock) time - there's a slight loss of efficiency during synchronisation between threads. But until we know exactly which metric you're referring to, we can't check the others. |
|
189)
Message boards :
Questions and problems :
Running VB tasks for LHC and some others.
(Message 110083)
Posted 12 Oct 2022 by Richard Haselgrove Post: This change was added to the BOINC sources yesterday: https://github.com/BOINC/boinc/pull/4953 But it needs to be rolled out through the distributions and the package management systems before it's installed by default. That won't happen until the next BOINC release, which we expect to be v7.22.x |
|
190)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110074)
Posted 10 Oct 2022 by Richard Haselgrove Post: I don't think anything superficial can be taken at face value with spam messages like this. I've still got two of 'Jacob's' emails: One says it comes from an educational establishment in the Dominican Republic, but the first identifiable IP address is a Microsoft datacentre in Washington DC. The other is again from an educational establishment, this time in South Korea. The IP address is another Microsoft datacentre in Tokyo, Japan. |
|
191)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110061)
Posted 9 Oct 2022 by Richard Haselgrove Post: My email address was on a list stolen from Yahoo years ago. I was warned by my ISP, and got the password changed pronto - I don't think any real damage was done. These spam messages are noticeable because they're using a 'sender name' that's likely to catch my interest. But the actual email account they're sending from bears no relationship to the spoofed name. |
|
192)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110057)
Posted 9 Oct 2022 by Richard Haselgrove Post: I held my nose and risked a peek. It's one of those 'get rich quick' scams, faked up as a newspaper website story. Yawn. |
|
193)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110051)
Posted 8 Oct 2022 by Richard Haselgrove Post: Likewise. I don't dare follow the links that Jacob Klein keeps sending me, even on my best-hardened computer. |
|
194)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 110049)
Posted 8 Oct 2022 by Richard Haselgrove Post: And did my alleged friend find something interesting to show you on Instagram? Or someone? |
|
195)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 110034)
Posted 7 Oct 2022 by Richard Haselgrove Post: It feels as if Krembil have been concentrating on individual components - oiling the squeaking wheels. But they seem not to have an overall, wholistic, comprehensive plan. It can't walk and chew gum at the same time. |
|
196)
Message boards :
Questions and problems :
BOINC Certificate Not Valid (Very Important!!)
(Message 110023)
Posted 5 Oct 2022 by Richard Haselgrove Post: What were you expecting to find at boinc.science? It's not one of ours. |
|
197)
Message boards :
BOINC client :
BOINC 7.22 Change Log
(Message 110019)
Posted 5 Oct 2022 by Richard Haselgrove Post: As the opening post says, there is no version available for download, even for testing. This is just preliminary announcement, so that testers will know what to test for. |
|
198)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 110006)
Posted 5 Oct 2022 by Richard Haselgrove Post: Opened https://github.com/BOINC/boinc/issues/4952 |
|
199)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 110004)
Posted 5 Oct 2022 by Richard Haselgrove Post: That seems to be a Curl thing.I was assuming that, but thanks for confirming and the link. But what's the limit in our clients, and is it of the order of that 45 thousand current ID count? If so, is it required/efficient in a client setting? I'll probably post that question in Git. |
|
200)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 110001)
Posted 5 Oct 2022 by Richard Haselgrove Post: As folks have said. I've got many OPNG transfers waiting from last night, and they're clearing very, very slowly under my OCD. But I caught an interesting new error message when checking it was the same problem as before: 05/10/2022 08:53:34 | World Community Grid | [http] [ID#45505] Received header from server: HTTP/2 503 05/10/2022 08:53:34 | World Community Grid | [http] [ID#45505] Received header from server: <html><body><h1>503 Service Unavailable</h1> 05/10/2022 08:53:34 | World Community Grid | [http] [ID#45505] Received header from server: No server is available to handle this request. 05/10/2022 08:53:34 | World Community Grid | [http] [ID#45505] Info: Connection cache is full, closing the oldest one.I'll look into that one. |
|
201)
Message boards :
Questions and problems :
Unable to launch BOINC CLIENT/MANAGER since upgrade
(Message 110000)
Posted 5 Oct 2022 by Richard Haselgrove Post: Both my machines - currently Mint 21 / BOINC 7.18.1 and Mint 20.3 / BOINC 7.20.2 - run NVidia GPUs just fine. You just have to take time and trouble over choosing and installing the right drivers. |
|
202)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109991)
Posted 4 Oct 2022 by Richard Haselgrove Post: Well, they backed out of that one pretty quickly (temporarily or permanently, time will tell) - I've been back to 'normal' delays for a while. I've given up trying for tonight. |
|
203)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109985)
Posted 4 Oct 2022 by Richard Haselgrove Post: New error mode: 04/10/2022 20:11:33 | World Community Grid | [http] HTTP error: Error in the HTTP2 framing layerdrops the attempted connection immediately. I'm taking a break. |
|
204)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109982)
Posted 4 Oct 2022 by Richard Haselgrove Post: I'm starting to get some OPNG, helped by a 5-minute update script. Some batches very slow to download, but I've just seen one go through at full speed. Depends on which way the wind is blowing, or something. |
|
205)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109975)
Posted 3 Oct 2022 by Richard Haselgrove Post: Yup, somebody's running around with an oilcan, oiling the squeaking wheels. Sometimes, downloads stop, then limp into action again. Then, uploads will pause for a minute or to, and then start again. The forum loads - slowly, but the page with today's news item can't be found. And so on. |
|
206)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109972)
Posted 3 Oct 2022 by Richard Haselgrove Post: ... reduces the need for ocd on the mouse index finger a bit.I've moved the OCD to the enter key - it's easier to find if you're doing something else while waiting. Select all pending transfers. I use a Windows remote Manager even for Linux clients, so I can 'marquee select' - hold the mouse button down in the empty space below the list, drag it to the top, and let it go. Click the 'Retry Now' button once, and then stop molesting the poor wee beastie. Hitting the enter key retries all selected transfers again, and again, and again... |
|
207)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109971)
Posted 3 Oct 2022 by Richard Haselgrove Post: One thing that confuses me is that the small files are the ones that don't download cleanly, and once a few small files are stalled everything falls to bits.....I thought that for a while, but I don't think it can be true. Whenever I've looked into that "transient HTTP error" on downloads, it's been a "503 HTTP service unavailable" - in other words, no connection has been made. And if there's no connection, the rest of the server - in particular, the BOINC bits - has no idea what you were asking for. I think the small ones stand out because the 107 bytes or 0.10 KB show as progress for the small ones. But I think it's been established that those very consistent byte counts are just the HTML version of the error message, not really BOINC data. |
|
208)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109965)
Posted 3 Oct 2022 by Richard Haselgrove Post: Maybe I should do some editing and see if upping the number increases throughput?You could try, but I doubt it would help much. I've been getting quite a lot of new tasks today, and returning them pretty steadily. Uploads are normally going through automatically, though some of them get stuck for a few seconds. The real problem remains with downloads, and the speed and connectability seems to fluctuate every few seconds. I'm only doing a significant amount of work thanks to OCD of the retry finger... |
|
209)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109963)
Posted 3 Oct 2022 by Richard Haselgrove Post: I tried opening the website in an incognito window. It opened i-n-c-r-e-d-i-b-l-y slowly, especially the photographs - but it looked as if it was all there. Then, I clicked for the latest news. And the latest news is ... {"errors":[{"error":"An unexpected error occurred: Could not open Hibernate Session for transaction; nested exception is org.hibernate.exception.GenericJDBCException: Unable to acquire JDBC Connection"}]}
|
|
210)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109961)
Posted 3 Oct 2022 by Richard Haselgrove Post: The battery in yours just runs the digital display stuff.I have a battery one as well, as was the precious one. The battery powers the display, sure - the old one continuously displayed the temperature on LCD, the newer one can display much more on LED, but only on demand - it's normally blank. But the primary purpose of the batteries is to send the temperature reading to the main controller by radio. |
|
211)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109946)
Posted 2 Oct 2022 by Richard Haselgrove Post: I don't think it's a hack. It's just a weekend off for the server team. Oh, for everyone except the one who sent the email encouraging everyone to start demanding tasks again. |
|
212)
Message boards :
Questions and problems :
Unable to launch BOINC CLIENT/MANAGER since upgrade
(Message 109938)
Posted 2 Oct 2022 by Richard Haselgrove Post: Well, that was fun - not. Fished off yesterday's overfetch, without any drama. Then removed v7.18.1, with a view to installing Gianfranco's PPA. That didn't work - there appears to be a new key security policy in place for Mint 21 (I've reported it to him). Then I put v7.18.1 back in place - and it wouldn't work, either. I had to add myself to the BOINC group all over again, before the GUI password could be read. Another reboot, and at last I was back in business. Seems to be behaving itself now. But I still haven't got a clue why other people are having so much trouble. The keys seem to be:
|
|
213)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109933)
Posted 2 Oct 2022 by Richard Haselgrove Post: I've just tried to log on to their forums, and got: Error 500: javax.servlet.ServletException: net.myvietnam.mvncore.exception.DatabaseException: Error executing SQL in MVNForumPermissionWebHelper.getPermissionsForGroupGuest.Edit - still can't get in, but they're still creating and issuing new work far faster than their download servers can handle it. Oh, and I got a copy of that email too. Are we all going to get one? |
|
214)
Message boards :
Questions and problems :
Unable to launch BOINC CLIENT/MANAGER since upgrade
(Message 109924)
Posted 1 Oct 2022 by Richard Haselgrove Post: Yup, that's probably tomorrows job. I've just allowed the second GPUGrid task to run on the second GPU, because they like their work back quickly. That'll probably keep it busy overnight, and then I'll take stock. Luckily, I upgraded the RAM on that machine, ready for the CPDN IFS trials, so it should cope comfortably. But as a diagnosis for why the client/manager won't launch on Mint 21 (see thread title): I'm afraid it was an abject failure. |
|
215)
Message boards :
Questions and problems :
Unable to launch BOINC CLIENT/MANAGER since upgrade
(Message 109922)
Posted 1 Oct 2022 by Richard Haselgrove Post: No, nothing except the 'out of work' message. And it's gone back into the endless fetch'n'fill: Sat 01 Oct 2022 16:53:20 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:54:30 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devicesAnd now it's doing the same for GPUGrid: Sat 01 Oct 2022 16:57:37 BST | GPUGRID | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:57:38 BST | GPUGRID | [sched_op] estimated total NVIDIA GPU task duration: 1440557 seconds Sat 01 Oct 2022 16:58:13 BST | GPUGRID | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:58:14 BST | GPUGRID | [sched_op] estimated total NVIDIA GPU task duration: 1440557 secondsGPUGrid is only allowed to run on one of my GPUs - that remained in place: Sat 01 Oct 2022 16:08:50 BST | GPUGRID | Config: excluded GPU. Type: NVIDIA. App: all. Device: 0 |
|
216)
Message boards :
Questions and problems :
Unable to launch BOINC CLIENT/MANAGER since upgrade
(Message 109921)
Posted 1 Oct 2022 by Richard Haselgrove Post: Well, this is very odd. I finished off all work, set 'no new tasks' on every project, reset every project, stopped and disabled the systemd service, and took a deep breath. Then I updated Mint v20.3 (as advised), and launched the upgrade tool for Mint 21. Many hours, and a good lunch, later, it went through the final grumbles and declared itself complete. Reboot. More grumbles on the pre-launch text screen, but it loaded fully. I still had a BOINC icon on the desktop, so I clicked it. And it was all there, manager and client, all running - including detection of my NVidia GPUs and their drivers. Some oddities: Sat 01 Oct 2022 16:08:50 BST | | Version change (7.16.6 -> 7.18.1)I was running 7.20.2 before, and I haven't run 7.16.6 for years. Then it went into a mode I've complained about before. Sat 01 Oct 2022 16:14:21 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:15:24 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 864.00 seconds; 1.00 devices Sat 01 Oct 2022 16:16:26 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 432.00 seconds; 0.50 devices Sat 01 Oct 2022 16:17:32 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:18:36 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:19:38 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:20:40 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:21:41 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:22:43 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:23:46 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:24:48 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:25:50 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:26:52 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:27:54 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:28:56 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:29:59 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:31:00 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devicesI had an absolutely minimal work request set (0.01 days + 0.00 additional). So why does it ask for (and get - Einstein is a very consistent project) the full amount every 60 seconds - even though it's got it already? It's as if it doesn't register all the previous successful requests. Grumble. It went on asking and getting while I was typing, until Sat 01 Oct 2022 16:43:22 BST | Einstein@Home | [sched_op] NVIDIA GPU work request: 1728.00 seconds; 2.00 devices Sat 01 Oct 2022 16:43:24 BST | Einstein@Home | No work is available for Gamma-ray pulsar binary search #1 on GPUs Sat 01 Oct 2022 16:43:24 BST | Einstein@Home | see scheduler log messages on https://einsteinathome.org/host/12808716/logI'll go and see what that log says. |
|
217)
Message boards :
Questions and problems :
Boinc not working on new Linux Mint 21 edition
(Message 109918)
Posted 30 Sep 2022 by Richard Haselgrove Post: Read several recent threads. I'm planning to upgrade to Mint 21 tomorrow, and try to work out what is going wrong for myself. |
|
218)
Message boards :
Questions and problems :
A question regarding "This project seems to have changed its URL." notices from BOINC.
(Message 109909)
Posted 28 Sep 2022 by Richard Haselgrove Post: That's the 'Yes' question. |
|
219)
Message boards :
Questions and problems :
A question regarding "This project seems to have changed its URL." notices from BOINC.
(Message 109905)
Posted 28 Sep 2022 by Richard Haselgrove Post: The answers to your questions are, respectively, yes and no. These annoying messages may go away over time, or you may just simply have to live with them. BOINC will go on working, either way. |
|
220)
Message boards :
Questions and problems :
BOINC on Ubuntu 22.04.1 LTS
(Message 109900)
Posted 28 Sep 2022 by Richard Haselgrove Post: You can't - unfortunately, v7.18.1 was never intended for Linux release, but the Ubuntu guys picked it up and released it anyway. It's broken. Two options: a) I posted a workround in https://boinc.berkeley.edu/forum_thread.php?id=14792 yesterday. b) Wait until the powers-that-be have actioned my request in https://github.com/BOINC/boinc/issues/4930. That may be slower. |
|
221)
Message boards :
Questions and problems :
Unable to launch BOINC CLIENT/MANAGER since upgrade
(Message 109891)
Posted 27 Sep 2022 by Richard Haselgrove Post: Great. I see that Mint have released an upgrade tool for converting v20.3 installations like mine to v21. I was planning to reduce my computing load from 01 October anyway (when my electricity supplier increases their prices yet again), so I'll take a crunching timeout and try it out. My working hypothesis for the problem you encountered rests on one simple fact: BOINC v7.18.1 was never released for the desktop operating systems Apple, Linux, Windows. It was only ever released for Android. I suspect the Ubuntu repo maintainers saw the Android release notice, and pulled a broken source tree. Now I just need some help in proving it, and examples like this help me down that route. |
|
222)
Message boards :
Questions and problems :
Unable to launch BOINC CLIENT/MANAGER since upgrade
(Message 109888)
Posted 27 Sep 2022 by Richard Haselgrove Post: There's something suspicious about the versions of BOINC v7.18.1 included in those repos - we haven't got got to the bottom of it yet. Could you try removing those versions, and try these instead? sudo add-apt-repository ppa:costamagnagianfranco/boinc sudo apt updateGianfranco Costamagna's PPA is a fast-track link for recent BOINC releases, automatically compiled from Debian sources. The current version, v7.20.2, is running fine on my Mint 20.3 - I'd be interested to hear how it runs on Mint 21 - that would help us distinguish between a problem in Mint 21 / Ubuntu 22 themselves, and a problem in the BOINC packages specific to those OSs. Edit - source reference for the suggested PPA: https://launchpad.net/~costamagnagianfranco/+archive/ubuntu/boinc |
|
223)
Message boards :
Questions and problems :
Unable to launch BOINC CLIENT/MANAGER since upgrade
(Message 109885)
Posted 27 Sep 2022 by Richard Haselgrove Post: Where did you get those various systemd and boinc-... files from? Which repository? |
|
224)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109878)
Posted 27 Sep 2022 by Richard Haselgrove Post: Glory: Northern Rail have approved my fare refund (at 9 pm last night) Grumble: I haven't got the money yet (they're probably sending the cheque by train) |
|
225)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109848)
Posted 18 Sep 2022 by Richard Haselgrove Post: Wait... have I missed EM station twice? Thought I'd seen the whole town...'Local' in this case means down in the valley bottom, beyond the newsagent (which you have seen). If it was any closer to EM, it would be ... ... drumroll ... ... too far from the rail tracks. |
|
226)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109846)
Posted 18 Sep 2022 by Richard Haselgrove Post: Mine was an easy, relaxed, walk in some watery countryside, to ward off the oppression of the summer heatwave. Two trains, both run by the same operator, with a half-hourly service on each line - what could possibly go wrong? The first line is a simple, and recently modernised, commuter shuttle into the nearest city - it normally runs as regular as clockwork. Arrived at my local station, to find that the train I was targetting had been cancelled. Another train on the line had broken down mid-route - they were a train short, and had to work past the blockage. My train took the hit, and the next one was already delayed. I don't know if the breakdown was heat-related, but it wouldn't surprise me. I bypassed the blockage by bus, and got a slightly different route into the city - arrived in time to get the connection half-an-hour later. That service required a second change mid-way: it rattled along just fine until the last station before the change-point, and there it stopped: held by a red light until there was a free platform at the change point. So I missed that as well, and the delay went up to an hour. Ironically, I was on my way to see a piece of railway history:  Bennerley Viaduct - the longest wrought-iron railway viaduct in the world, now preserved. We invented the bloody railways - you'd have thought we would have worked out how to run them by now. But we can't stop tinkering with them - ripping up the tracks and replacing them, replacing the entire management structure every few years as political fashions change. Sigh. |
|
227)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109843)
Posted 18 Sep 2022 by Richard Haselgrove Post: Only on British Railways ... Dear Richard Haselgrove,If only they hadn't sent me that survey form after my day out... If only they hadn't programmed their survey bot to say "From your answers, it appears you may be entitled to a refund on your fare. Click here ..." |
|
228)
Message boards :
Projects :
News on Project Outages
(Message 109834)
Posted 15 Sep 2022 by Richard Haselgrove Post: @Bill, Please ensure that David A receives a copy of that email. He updated all_projects_list yesterday (#4922) ready for the next release, but TACC is still in the list. Edit - it's also still listed at Science United. |
|
229)
Message boards :
Questions and problems :
"chdir: Permission denied" error when running BOINC on kubuntu 22.04
(Message 109823)
Posted 12 Sep 2022 by Richard Haselgrove Post: Referring to Linux Mint Releases, I'm currently at v20.3 'Una', with the 'Mate' desktop management tools (I haven't introduced myself to Vanessa yet). That's supposed to be the most painless introduction for Windows users, which is where I came in. I've paired that with Ginfranco Costamagna's PPA, which is a fasttrack release vehicle for BOINC, currently v7.20.2 |
|
230)
Message boards :
Questions and problems :
"chdir: Permission denied" error when running BOINC on kubuntu 22.04
(Message 109819)
Posted 12 Sep 2022 by Richard Haselgrove Post: In Linux Mint (various versions over the last three years), it looks like this:  (my memorable password is slightly different to yours, but the principle is the same) |
|
231)
Message boards :
Questions and problems :
"chdir: Permission denied" error when running BOINC on kubuntu 22.04
(Message 109816)
Posted 12 Sep 2022 by Richard Haselgrove Post: You can try deleting the password file and hope that boinc will just generate another, but I suspect you are going to have to re-install boinc.DON'T do that. If you don't understand what you're trying to achieve, it's worse than useless. First, make sure the BOINC client is listening for a known password. You've changed the password in gui-rpc-auth.cfg to 123456 - that's fine: it's not a high-security data risk. But now you must re-start the BOINC client, to make sure it's listening for the new password. Second, look at the launcher for the BOINC manager. Examine its properties. If the 'Command' line just says "boincmgr", change it to: boincmgr --password=123456 Now, the client and the manager are on the same page - they're speaking the same language. There's a key for the lock. They should work together. |
|
232)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109790)
Posted 10 Sep 2022 by Richard Haselgrove Post: I had a few in transit - both uploading and downloading - when the certificate failed. They've all now cleared (with a bit of gentle encouragement), so I've started requesting new work again. Got some new allocations, and with the same gentle prodding, they've all downloaded. We seem to have reached status quo ante... |
|
233)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109781)
Posted 9 Sep 2022 by Richard Haselgrove Post: WCG comms broken from Fri, 09 Sep 2022 12:00:00 GMT - SSL certificate expiry on their side. They're still using the old IBM certificate, and it's blown. Anyone got an alternative comms link we can use to notify them? (They're using HSTS - HTTP Strict Transport Security - to enforce HTTPS-only connections) |
|
234)
Message boards :
Questions and problems :
Tasks barely running
(Message 109742)
Posted 31 Aug 2022 by Richard Haselgrove Post: Before you go to bed, 1) Check the Windows 11 power management settings, to make sure the computer isn't going to sleep when you do. (you weren't running either Windows 11 or BOINC 20 years ago). 2) There's a nasty little catch-22 on the BOINC preferences screens these days: Website version: Suspend when no mouse/keyboard input in last --- minutesManager version: (my emphasis) Read that carefully - yes, DON'T do any work when you're not working either. It sounds wrong from the perspective of someone like you, but it's there for a purpose. You should avoid it. |
|
235)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109710)
Posted 27 Aug 2022 by Richard Haselgrove Post: I have certainly seen a vastly increased supply of WUs in the last 30 hours, but the network congestion - especially on task data file downloads - is still an inhibiting factor. |
|
236)
Message boards :
Questions and problems :
Download lots of files or zip into one?
(Message 109697)
Posted 26 Aug 2022 by Richard Haselgrove Post: Most projects try to send all their apps to as many platforms as possible, but come can't manage them all. I think a specialist policy for a complicated project like CPDN would be fair enough. |
|
237)
Message boards :
Questions and problems :
Download lots of files or zip into one?
(Message 109693)
Posted 26 Aug 2022 by Richard Haselgrove Post: We had that type of problem at GPUGrid. Project administrators tend to be more comfortable with Linux than Windows, so they sent out an app with a .gz archive, and tried to unpack it with tar. Apparently Windows 10 has a native tar app, but Windows 7 doesn't. I suggested 7-zip, and they got that working - not as versatile as tar, but it worked. My first copy of WinZip came on a paid-for floppy disk, but Windows added native zip, somewhere down the line. And then took it out again from Windows 11. Standards are tricky things - there are too many of them, and you have to support them all. |
|
238)
Message boards :
Questions and problems :
Blank notices screen
(Message 109685)
Posted 23 Aug 2022 by Richard Haselgrove Post: Thanks. Confirmed on Linux Mint - I'll report it upstream. Edit - done. https://github.com/BOINC/boinc/issues/4902 |
|
239)
Message boards :
Questions and problems :
Broken version of boinc in some repositories??
(Message 109675)
Posted 20 Aug 2022 by Richard Haselgrove Post: The earliest report I can find is message 107883, from 18 Apr 2022. Read on from there. |
|
240)
Message boards :
Questions and problems :
Downloaded tasks showing as Downlaoding only, seem stuck
(Message 109672)
Posted 20 Aug 2022 by Richard Haselgrove Post: WCG is still suffering from severe network congestion issues following their recent move from the US to Canada. They issued more work than recently yesterday, revealing the problems to yet more users. See my thread https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,44256 there for an example of the sort of http_debug messages to watch out for. Resetting the project is not an ideal method of dealing with this problem. There is a 'Retry now' button on the transfers tab which lets you select (click on) a single line or multi-select in the list of pending transfers, and retry it/them. Or there's a 'Retry pending transfers' menu item in the 'Tools' menu. Either may require multiple attempts, but they'll complete in the end. |
|
241)
Message boards :
Questions and problems :
Mint linux 21 fresh install boinc won't start
(Message 109655)
Posted 18 Aug 2022 by Richard Haselgrove Post: The installation instructions don't yet include an in-situ upgrade option - I'm a bit too busy at the moment for a bare-metal reinstall 
|
|
242)
Message boards :
Questions and problems :
Mint linux 21 fresh install boinc won't start
(Message 109653)
Posted 18 Aug 2022 by Richard Haselgrove Post: I'm running Linux Mint 20.3 successfully (both client and manager), but I haven't been offered an upgrade beyond that yet. (May be still in testing?) I'll watch out for it when it comes. |
|
243)
Message boards :
Questions and problems :
Mint linux 21 fresh install boinc won't start
(Message 109646)
Posted 17 Aug 2022 by Richard Haselgrove Post: Or, Change the auto-password to something memorable, and add that to the Manager launcher instead. You'll have to exit and restart the client before that will work. |
|
244)
Message boards :
Questions and problems :
Mint linux 21 fresh install boinc won't start
(Message 109645)
Posted 17 Aug 2022 by Richard Haselgrove Post: Or, Find the auto-generated 32-byte password yourself, in gui_rpc_auth.cfg. Then, add that password to the Manager launcher. |
|
245)
Message boards :
Questions and problems :
v7.20.2 intall issues - Win 11
(Message 109637)
Posted 17 Aug 2022 by Richard Haselgrove Post: The other thing that happened today is that a five-day cricket match started in London. Guess what happened? English batting collapsed, with 6 wickets before lunch. Then the heavens opened, with no play possible in the afternoon or evening sessions. They've just cancelled the second half of the TV highlights show - there were no highlights. |
|
246)
Message boards :
Questions and problems :
v7.20.2 intall issues - Win 11
(Message 109636)
Posted 17 Aug 2022 by Richard Haselgrove Post: Thanks, but we have our own rain-maker for Yorkshire. Some fool declared the whole county to be a drought area yesterday, so of course it started raining almost immediately, and rained for much of the night. |
|
247)
Message boards :
Questions and problems :
Mint linux 21 fresh install boinc won't start
(Message 109634)
Posted 17 Aug 2022 by Richard Haselgrove Post: There have been reports that some recent Ubuntu version have installed a broken version of BOINC in their repositories - it claims to be v7.18.1, but they appear to have packaged the Android sources, rather than the x64 sources. Also, the Linux repository packages are usually deployed as systemd services these days. The client isn't started by the Manager (as it would be under Windows), but in terminal, with: sudo systemctl start boinc-clientYou may have to issue this first: sudo systemctl enable boinc-clientDepends what you're trying to do with BOINC. Are you just running science projects, or do you want to learn about BOINC? Test and improve BOINC? Add new code and features in BOINC? We can help, if you tell us a bit more. |
|
248)
Message boards :
Questions and problems :
v7.20.2 intall issues - Win 11
(Message 109626)
Posted 17 Aug 2022 by Richard Haselgrove Post: I'll do it as soon as it's offered, but I don't want to break BOINC too badly just yet. I've got some accuracy/validation testing to do for Bernd at Einstein, and I need to do it on that particular CPU - just waiting for the cooler weather. I'll fire that up now, and maybe try the 'Insider' pathway at the weekend if we haven't got to the bottom of this by then. |
|
249)
Message boards :
Questions and problems :
v7.20.2 intall issues - Win 11
(Message 109624)
Posted 17 Aug 2022 by Richard Haselgrove Post: I have the Feature Experience Pack, too - but an earlier version, 1000.22000.856.0 I'm not on any Beta or Insider program, and my main version is 21H2. I've just run a manual update, and I've got nothing new offered from that - just an anti-virus definition update for Microsoft Defender. |
|
250)
Message boards :
Questions and problems :
defaultProblem with BOINCstats
(Message 109622)
Posted 17 Aug 2022 by Richard Haselgrove Post: The link between BOINC itself and boincstats is made via a 32-byte hash code called the CPID (cross project identifier). You can see truncated versions of those in your two links - and they are different. It sounds as if you have (or have had) two or more different computers, running two or more different projects, and the CPIDs have not synchronised properly. It's necessary that there's a common pathway, with the same email address, whereby every computer and every project can pass information between them. Computer 1, running project A Computer 2, running project B won't work. Adding computer 3, running projects A+B, should line things up (over a day or two). So would adding Project C to both computers 1 and 2. |
|
251)
Message boards :
Questions and problems :
v7.20.2 intall issues - Win 11
(Message 109609)
Posted 15 Aug 2022 by Richard Haselgrove Post: I've just upgraded from v7.16.20 to v7.20.2 under Windows 11 Pro, with no problems - so this isn't universal. Another question occurs to me: are these machines each your own personal property, free of any restrictions? Or were they supplied by an employer or educational establishment, and possibly subject to group policy restrictions when used on the establishment's network? |
|
252)
Message boards :
BOINC client :
Software update.
(Message 109608)
Posted 15 Aug 2022 by Richard Haselgrove Post: See thread 14749. Sometimes relying on assumptions like that can allow a genuine problem to pass unheeded. That's why I like to get to the real facts behind the report. |
|
253)
Message boards :
BOINC client :
Software update.
(Message 109602)
Posted 14 Aug 2022 by Richard Haselgrove Post: I wasn't (and I'm still not) aware of any problem with Windows - which was why I asked. Unless the Mac developer broke it while fixing his own OS ??? |
|
254)
Message boards :
BOINC client :
Software update.
(Message 109599)
Posted 14 Aug 2022 by Richard Haselgrove Post: Which operating system and hardware? There was a fault with the notification system for Apple macs in the M1 - Apple silicon - range, but that should be fixed now. For mainstream hardware and operating systems, all should have v7.20.2 available by now. |
|
255)
Message boards :
BOINC client :
BOINC 7.22 Change Log
(Message 109595)
Posted 13 Aug 2022 by Richard Haselgrove Post: Changes from the https://boinc.berkeley.edu/wiki/Release_Notes
Please note that this is an advanced planning list only. Final design and detailed implementation has not yet been agreed: there is no version available for download, even for testing. |
|
256)
Message boards :
Questions and problems :
Connect to original account
(Message 109575)
Posted 10 Aug 2022 by Richard Haselgrove Post: Data protection law in UK after Brexit 2020 Here are the overall changes to UK data privacy law after Brexit –
from https://www.cookiebot.com/en/data-protection-act-2018/ |
|
257)
Message boards :
Questions and problems :
BOINC version confusion
(Message 109574)
Posted 10 Aug 2022 by Richard Haselgrove Post: Thanks. Reported upstream: https://github.com/BOINC/boinc/issues/4879 |
|
258)
Message boards :
Questions and problems :
BOINC version confusion
(Message 109571)
Posted 10 Aug 2022 by Richard Haselgrove Post: Try looking directly at the 'Download all' page. There, it lists a v7.20.2 version for "Apple Mac (64-bit Intel and Apple Silicon), MacOS 10.9+". Verify that matches the specifications of your M1 mac mini: if it does, download it and see if it will install. (the process should be the same as for your i5 Intel iMac). Please come back and tell us, either way. The M1 macs are still pretty new: it's possible that the process for notifying you of new versions hasn't been updated properly yet. We can ask them to take another look at it, if needed. (as Keith said in your original thread) |
|
259)
Message boards :
The Lounge :
These are no longer with us, may they have peace.
(Message 109570)
Posted 10 Aug 2022 by Richard Haselgrove Post: Raymond Briggs The Snowman age 88When the Wind Blows |
|
260)
Message boards :
Questions and problems :
Connect to original account
(Message 109566)
Posted 10 Aug 2022 by Richard Haselgrove Post: And different courts. Sigh. |
|
261)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109541)
Posted 8 Aug 2022 by Richard Haselgrove Post: The ice-cream trade has always been a law unto itself - they'll freeze and sell anything if they think that jaded punters can be persuaded to buy them. Those sound like they would go well with the deep-fried mars bars, which come from the same heritage. |
|
262)
Message boards :
Questions and problems :
Can't enter any words when logging in the project manager
(Message 109535)
Posted 8 Aug 2022 by Richard Haselgrove Post: Your profile flag identifies you with the Taiwan locale. What keyboard language are you using? There's a known problem logging in with the Windows default Chinese input method, Microsoft Pinyin. Temporarily switching to a US or UK keyboard will bypass the problem. |
|
263)
Message boards :
Questions and problems :
Probably Stupid Questions
(Message 109531)
Posted 7 Aug 2022 by Richard Haselgrove Post: In the days of Windows 7, it used to be easy to manage the display of a BOINC icon in the Notification Area (system tray), down in the bottom right corner near the clock:  The BOINC installer set it up to display 'Notifications only', but it was easy to change to permanent display:  Windows 10 made that harder to achieve, and the workrounds varied from release to release. I haven't tried with Windows 11 specifically, but searching the web for something like "Windows 11 show notification area icons always" might turn up something useful. |
|
264)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 109512)
Posted 5 Aug 2022 by Richard Haselgrove Post: Also looking at ways to increase still further the 5.2KW of solar we have.Dave, what's the current best back-of-an-envelope benchmark for usable watts per square metre? I live in a traditional terrace house, and all I can use is the roof area of the house itself, about 6m x 7.5m. The ridge line is roughly east-west, so half of that has a good aspect for solar generation. The other wrinkle is that I live in a conservation area, and the roof is clearly visible from a public open space. I think the planners would have a fit if I put up anything clunky: so my feeling up till now has been that putting up enough panels to make a difference would be more costly than average, and not worth the hassle. There would also be problems getting the feed down from the roof through the traditional building materials (Yorkshire stone slabs throughout). Obervations/suggestions? |
|
265)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 109506)
Posted 5 Aug 2022 by Richard Haselgrove Post: I haven't been getting any for GPUs. I'm thinking of mothballing the GPUs at the end of next month, when the next electricity price hike hits in the UK. That'll free a lot of cores for CPU work, so I might have another look then. |
|
266)
Message boards :
Questions and problems :
Issues with BOINC 7.20.2 x64 behind a proxy
(Message 109504)
Posted 5 Aug 2022 by Richard Haselgrove Post: There was a bigger Windows change than just certs - I'll try to dig it out. edit - I was thinking of Issue: https://github.com/BOINC/boinc/issues/4542 Pull request: https://github.com/BOINC/boinc/pull/4545 The prime purpose was indeed to use the Windows certificate store, but the PR goes into considerable discussion about how that was achieved using schannel. I don't see any consideration of proxies, though. It might be best to ping @AenBleidd directly with the problem. |
|
267)
Message boards :
Questions and problems :
Issues with BOINC 7.20.2 x64 behind a proxy
(Message 109500)
Posted 5 Aug 2022 by Richard Haselgrove Post: I think you'll have to raise this as an issue with the development team directly - especially as it seems to be a regression between v7.16 and v7.20. They may need to ask you questions about the proxies you've tried, and the configurations you've used. Issues can be raised at https://github.com/BOINC/boinc |
|
268)
Message boards :
Questions and problems :
Trouble download new Boinc
(Message 109491)
Posted 3 Aug 2022 by Richard Haselgrove Post: There are files at https://boinc.berkeley.edu/dl/boinc_7.20.2_universal-apple-darwin.dmg https://boinc.berkeley.edu/dl/boinc_7.20.2_macOSX_universal.zip though. That follows the naming convention for the most recent previous releases - might be worth a punt, if you're feeling adventurous? |
|
269)
Message boards :
BOINC Manager :
Simple View
(Message 109478)
Posted 1 Aug 2022 by Richard Haselgrove Post: I assume that, when you compile your own Manager under Linux, you get the full five line File menu in Advanced view?  On my PPA-derived copy, I get the Linux corporate hack version, which removes some actions:  That's because Linux repo maintainers assume that their packages will be used in wage-slave conditions, where users don't have full control over their employers' computers: the slaves can't be allowed to shut down the client, because they don't have the authority to restart it. But I digress. When I first encountered and reported this problem, I wondered if the removal of the corporate hack options might be implicated? Simple View wasn't hacked, so possibly the switch back to Advanced View called a function which was no longer available. That might be something to watch out for in your wanderings. |
|
270)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 109459)
Posted 28 Jul 2022 by Richard Haselgrove Post: Enjoy your break. Les dealt with it. |
|
271)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 109451)
Posted 27 Jul 2022 by Richard Haselgrove Post: He's gone and botherated the CPDN guys instead. |
|
272)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109427)
Posted 24 Jul 2022 by Richard Haselgrove Post: All this fuss - allegedly - about a stamp in a passport. I had mine stamped at St. Pancras and Gare du Nord for EuroStar, and I didn't even notice till I got home. It takes no time at all. |
|
273)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109391)
Posted 19 Jul 2022 by Richard Haselgrove Post: How do they cope I wonder?By using diesel traction power. I don't think there are many, if any, real long-distance routes using electric traction. Thanks to the moderator. Count me among the list of people who was disturbed by recent - er, what were they? Certainly not 'conversations'. Statements to which any rational response was greeted by irrational re-statement? I didn't know how to respond either, which is why I mostly kept out of it. |
|
274)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109370)
Posted 19 Jul 2022 by Richard Haselgrove Post: It was very still in Yorkshire this morning - no breeze at all to alleviate the hot, hazy, sunlight - it got you from every side. But there have been brief puffs of wind this afternoon - definitely gearing up for something. |
|
275)
Message boards :
BOINC client :
BOINC 7.20 Change Log
(Message 109329)
Posted 18 Jul 2022 by Richard Haselgrove Post: BOINC 7.20.2 released for testing for Windows.
|
|
276)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109202)
Posted 13 Jul 2022 by Richard Haselgrove Post: Most people just have a couple of days in bed, hopefully she's ok. Just give her first class room service for a bit.One of my regular clients is currently in hospital. She was taken in on the 5th of July, and repeated covid tests have returned positive. Today I was asked to go to her house, and confirm that a laptop was properly set up to access email, for her husband to deliver in his daily visit. She's over 80, and has not been vaccinated. |
|
277)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109188)
Posted 12 Jul 2022 by Richard Haselgrove Post: Just as we shouldn't have silly rules about catering for vegetarians in restaurants.I was impressed when I toured India, in the 1980s. There, restaurants had two entrances of equal status: veg and non-veg. Each led to a dining area of equal size and architectural design, serviced by it's own kitchen. [vegetarianism for religious reasons requires that the cooking utensils aren't contaminated by contact with animal carcasses] "equal but different". |
|
278)
Message boards :
The Lounge :
Let's talk politics
(Message 109160)
Posted 12 Jul 2022 by Richard Haselgrove Post: Greenwich Mean Time itself was only brought into domestic (as opposed to scientific) use to make railway timetables work: before that need arose, each city and town used their own local solar time. The synchronisation of European summer / winter shifts was done for much the same reason: an entire international rail journey could keep to a stable timetable, without blocks of stations shifting an hour back and forth relative to their neighbours. |
|
279)
Message boards :
The Lounge :
Let's talk politics
(Message 109062)
Posted 8 Jul 2022 by Richard Haselgrove Post: Or the freedom to drive at 100mph. Get us into trouble IF we crashGlad to hear that you can guarantee to resurrect the people you crash INTO. |
|
280)
Message boards :
The Lounge :
Let's talk politics
(Message 109053)
Posted 8 Jul 2022 by Richard Haselgrove Post: Why do people accept our freedoms being taken away?Because some other people abuse the freedoms that would otherwise exist. |
|
281)
Message boards :
The Lounge :
Let's talk politics
(Message 109045)
Posted 8 Jul 2022 by Richard Haselgrove Post: It's been suggested that he wants to stay there until he has his wedding in Chequers. The BBC now understands the party will not be held at Chequers.Another case of Boris opening his mouth before thinking. I wonder which donor will pick up the tab for that 'alternative venue'? Anyway, now that excuse has crumbled, what else will he find to hold on for? |
|
282)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109026)
Posted 7 Jul 2022 by Richard Haselgrove Post: And this morning, Johnson has agreed to stand down as Conservative Party leader but will carry on as prime minister till their conference in October.In his dreams. I doubt that will hold through the weekend. |
|
283)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 109001)
Posted 6 Jul 2022 by Richard Haselgrove Post: The BBC News Channel is keeping a running score of resignations on their 'breaking news' strapline. |
|
284)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108832)
Posted 1 Jul 2022 by Richard Haselgrove Post: Fair comment. Largely because of Collatz, other projects went their own way, too, and broke the original plan of cross-project comparability. And also broke any semblance of honesty in the reverse calculation of work done in petaflops across BOINC as a whole. that seems to have vanished from this website at long last, but David used to use a broken global estimate in his applications for money from the NSF. |
|
285)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108815)
Posted 30 Jun 2022 by Richard Haselgrove Post: I've answered David already, given him the more correct line of how everything went re: forum moderation. Good thing I kept all those emails.Good. I thought that bit was very weak - misleading, even. I did a very short stint as a moderator at SETI, at the end of the 'wild west' period, but I burned out within two months. My problem was with the other moderators (especially with those factional fighters who were hanging on from the old days), not with the users - they were lovely. I've helped out the moderators at other projects with technical matters over the years - notably Einstein and CPDN - but I've never had formal message board moderator status, except brief periods of temporary cover here. AFAIK it waned out because a lot of things people wanted to add, David didn't want. If you run into that wall enough times, you just back away and never return.That's my feeling as well, but it's very hand to provide proof. Most of them just faded into the background: they didn't post a (public) resignation letter with reasons. I have some mental timelines of when things went wrong.
But I don't know what to do with that information, or who to direct it at, now Kevin's departed the scene. |
|
286)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108812)
Posted 30 Jun 2022 by Richard Haselgrove Post: Grumble: all that credit Richard got, damn. He wasn't even here until late in the game. ;-)You got this morning's email, then? I'm still trying to process that, and decide how to handle it. To me, that 'history' piece feels very much like a view from the inside, looking out: and it's somewhat different from a similar history, delivered as a talk at the 2014 BOINC Workshop in Budapest. That script has never been published online, unfortunately. I'd love to see a companion piece, written from the outside, looking in. As you say, I wasn't involved in the early stages: I joined SETI early on, but I only volunteered my computers, not my mind and energy. I only got involved in BOINC when SETI Classic closed down, and I was impressed by what I found - that's what drew me in. The trac wiki has a list of over 120 early contributors to the project - plus you and I, in a separate ghetto all of our own. Where did all that energy go? I might ask that question in a reply email - see if we can draw out the other half of the analysis, |
|
287)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108783)
Posted 28 Jun 2022 by Richard Haselgrove Post: The worst thing is one set to a very short time and it goes off while you're in the bathroom.There's worse. If the PIR covers the general entrance and washroom area of the facility, but doesn't cover the W.C. cubicle itself... You can be sitting in pitch darkness, with your trousers around your ankles, but without anything to wave at to restore the light. Remote car-parks at the beginning or end of a country walk are particularly prone to that (and you tend to be in there for longer if you have to manage rucksacks and cagoules, as well as your normal clothing). |
|
288)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108753)
Posted 27 Jun 2022 by Richard Haselgrove Post: Same here, but why monthly, I do it when they ask, every 6 months. Also I just never responded to any smart meter requests. I throw the mail in the bin and block all text and phone numbers calling about it. [checks phone] FIFTY TWO calls blocked from that number now. Why are they so keen? There has to be something dodgy going on for them to be that pushy.My particular band of spivs and hucksters likes to have them monthly. Why the pressure? Because it's a Government scheme, and the Government pay them a bonus for shifting some quota of meters. |
|
289)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108745)
Posted 27 Jun 2022 by Richard Haselgrove Post: This debate isn't helped by poor Government terminology.Why did you guys ever accept a smart meter?'cos it's easier to look at the smart meter in the kitchen than climbing over the bench in the garage to look at a steam meter.. A 'meter' measures things - no more, no less. The thing in your kitchen isn't doing that: it's a short-range radio receiver, a calculator, and a display panel. All of those functions require raw materials to manufacture, and extra energy to run. And in what way are they 'smart'? My energy supplier wanted to push me onto the program. So I rang them up and asked them what degree of smartness was included in the metering systems currently being installed? Could they, for example, communicate with switching systems which would turn individual appliances on and off, perhaps at particular price points at different times of day? No, they can't. Any 'smartness' remains with users, reading and acting on the information displayed. I declined their offer, and I'm reading my own meters monthly and sending the readings to their website. |
|
290)
Message boards :
Questions and problems :
Scheduler Request failed : HTTP Services unavailable
(Message 108692)
Posted 25 Jun 2022 by Richard Haselgrove Post: WCG is still having troubles getting their full multi-server working environment properly configured. It's their problem, not yours - nothing you can do at your end. |
|
291)
Message boards :
BOINC Manager :
Application error on Windows 10 shutdown
(Message 108598)
Posted 20 Jun 2022 by Richard Haselgrove Post: Please add your report to the issue already opened at https://github.com/BOINC/boinc/issues/4793 |
|
292)
Message boards :
Projects :
News on Project Outages
(Message 108596)
Posted 20 Jun 2022 by Richard Haselgrove Post: And some volunteers write scripts that request work every few seconds, so that they're first in line when work does become available. |
|
293)
Message boards :
Projects :
News on Project Outages
(Message 108594)
Posted 20 Jun 2022 by Richard Haselgrove Post: Why do projects often ask us to set no new tasks? Surely they can just turn off the server tap?Because if you don't turn off work fetch, your client will keep pestering their server with "Is there any more yet?" requests. Unlike us, they probably have to pay for their own bandwidth. It's easier, simpler, and more polite if we just shut up for the duration. |
|
294)
Message boards :
Questions and problems :
Ansible installed node not getting credit
(Message 108573)
Posted 19 Jun 2022 by Richard Haselgrove Post: For credit issues, you usually need to go back to the science project you're working for. Look at their website first - specifically, your account. Is this computer shown on your account? Has it been allocated any tasks? Have they been completed? Have the been reported? Were they valid? Have they been granted credit? Don't simply rely on external statistics sites like BOINCstats. They're too far down the food chain. Go to the source, where the credit would be granted. |
|
295)
Message boards :
BOINC Manager :
Question for Linux users of BOINC Manager
(Message 108490)
Posted 16 Jun 2022 by Richard Haselgrove Post: Thanks guys. It turns out that the code for finding and loading the skin files dates back to 2007, and has probably never worked on Linux. If we haven't missed it until now, I don't think we need to start worrying about it now. |
|
296)
Message boards :
Projects :
News on Project Outages
(Message 108484)
Posted 16 Jun 2022 by Richard Haselgrove Post: Strangely, it's not on the list here: https://boinc.berkeley.edu/projects.php and half of my computers list it as a project I can attach to, and the other half don't. Is there a problem with Boinc getting updated project lists?There's a curious, undocumented, but deliberate design choice in BOINC. If you set the configuration flag 'dont_contact_ref_site' - if, for example, you've got better ways of diagnosing the health of your internet connection - then the periodic update of all_projects_list is also supressed. https://github.com/BOINC/boinc/issues/2352 |
|
297)
Message boards :
BOINC Manager :
Question for Linux users of BOINC Manager
(Message 108476)
Posted 15 Jun 2022 by Richard Haselgrove Post: 'Skins' are a central BOINC deployment, not up to individual projects. |
|
298)
Message boards :
BOINC Manager :
Question for Linux users of BOINC Manager
(Message 108471)
Posted 15 Jun 2022 by Richard Haselgrove Post: I saw the Manager crash on return switch from simple to advanced myself - that's one of the bugs I've reported and am investigating. Windows users get a 'skins' folder delivered to their program directory by the Windows installer. I'm trying to work out on Linux, a) the distros don't deliver the skins, or, b) The Linux build doesn't load and display the skins even if present. So far, either or both are possible. |
|
299)
Message boards :
BOINC Manager :
Question for Linux users of BOINC Manager
(Message 108465)
Posted 15 Jun 2022 by Richard Haselgrove Post: I'm trying to help debug some Manager issues in the Linux version of BOINC Manager ready for the forthcoming BOINC v7.20.x release. Question - when you switch the Manager to simple view, can you see a bitmapped background like  in either, or both, v7.20.0 or earlier versions. Or, if you work for a corporate running a skinned version of BOINC, can you see their skin instead? It seems to have gone missing... |
|
300)
Message boards :
BOINC Manager :
make failing
(Message 108440)
Posted 13 Jun 2022 by Richard Haselgrove Post: 'master' is probably not recommended because it's where all the changes are placed first, and sometimes one change has consequences which mean that another change is needed soon after. So timing is critical - you need to catch a master download when things are quiet, not in the middle of a flurry of changes. A release branch, on the other hand, should be the result of a coherent choice to freeze the code at a point where everything is aligned - it should stay stable until the next fixed point. Having said that, I find that master usually builds and runs cleanly, but it's not guaranteed. |
|
301)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108432)
Posted 12 Jun 2022 by Richard Haselgrove Post: ... I backspaced and corrected something in Word without looking at the screen.My fingers know when I've typed something wrong, but I have to look at the screen to see what it was. |
|
302)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108428)
Posted 12 Jun 2022 by Richard Haselgrove Post: Just had to drive half way across West Yorkshire (well, about two miles, really) to deliver a lecture that I thought I'd retired about 25 years ago. When you were taught to type in secretarial college, you were trained not to look at the piece of typing in your machine - after all, it's got nothing on it except what you've already typed. Concentrate instead on your shorthand pad, or whatever you're copying - the stuff you're going to type next.25 years on, it's still true, and still useful. |
|
303)
Message boards :
Questions and problems :
Gotta simple question.
(Message 108423)
Posted 12 Jun 2022 by Richard Haselgrove Post: The question is: which version of the source code did they use? Did they use the master branch, (potential) bugs and all, or did they use the client_release/7/7.20 branch? Pre-release status is actually set in line 27 of version.h Currently: master is at v7.21.0, pre-release branch is at v7.20.0, release |
|
304)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108418)
Posted 12 Jun 2022 by Richard Haselgrove Post: I don't. Some third party has set up an email account with their own credentials, but have attached my plain-text name to the spoof account. You would have to question their motivation.Grumble: "Richard Haselgrove" is sending me spam emails again and I can't even forward them to him as his email server does block them as spam. 😭Why on earth would Richard spam you? |
|
305)
Message boards :
Projects :
News on Project Outages
(Message 108417)
Posted 12 Jun 2022 by Richard Haselgrove Post: PHP code is normally executed on the server. All you should see is the output from running the code, not the source code itself. It sounds as if your copies of Opera and Edge contacted the server in question at a time when the PHP interpreter was not fully active, and had cached the resulting response for that url. Yes, a thorough cache reset is the appropriate remedy. |
|
306)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108388)
Posted 10 Jun 2022 by Richard Haselgrove Post: Grumble 2: No sooner had they finished resurfacing the pavements than the telephone people are digging them up to put in fibre. No idea if that is to give me broad band or for some other provider.If the vans are labelled 'Openreach', then they're a common carrier, and any retail provider can lease the lines and provide the service. The only other major supplier is Virgin, who inherited the Cable TV networks and are repurposing them to provide broadband as well. Everything else is private enterprise - usually for commercial customers only. |
|
307)
Message boards :
BOINC client :
BOINC 7.20 Change Log
(Message 108336)
Posted 7 Jun 2022 by Richard Haselgrove Post: Be aware that it's a pre-release version, not yet fully tested. And no announcement has yet been made concerning real-world testing (though it's pretty thoroughly free of syntax errors). |
|
308)
Message boards :
Projects :
CPDN
(Message 108333)
Posted 7 Jun 2022 by Richard Haselgrove Post: A new batch (931) of hadam4h tasks for Linux is being distributed this morning. I don't know how long they'll last. |
|
309)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108328)
Posted 6 Jun 2022 by Richard Haselgrove Post: I'd rather doubt any were. Boris has been totally eclipsed, publicity-wise, by the royal Jubilee over the last few days - and he was even booed on arrival at one event. I think he was planning some eye-catching announcements of future goodies over the next few days, to remind the punters what they like about him: but now he's only got time for a quick pep-talk to his own MPs in private before the vote. I doubt that'll do much good - the payroll vote will probably see him past this bump in the road anyway. The question is - how long will he survive after that? Margaret Thatcher and Theresa May both survived their initial vote, but neither could hang on for long. |
|
310)
Message boards :
BOINC Manager :
Simple View
(Message 108317)
Posted 4 Jun 2022 by Richard Haselgrove Post: Your simplest option might be to make Advanced the default view, and remove/disable the menu command for switching between views. Simple view would still be in the build, but inaccessible. |
|
311)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108314)
Posted 4 Jun 2022 by Richard Haselgrove Post: Have I ever shared the analogy I use to teach novice users the difference between downloading and installing software? It's like coming home from holiday. First, the taxi driver dumps your suitcase on the garden path. (that's downloading) Then, you unpack and distribute. The souvenirs go on the mantelpiece, the guide books on the bookshelf, the half-eaten British Rail sandwich in the bin, the dirty laundry into the washing machine, and the empty suitcase into the closet under the stairs. (that's installing) |
|
312)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108306)
Posted 2 Jun 2022 by Richard Haselgrove Post: Made it back - just an hour late in the end. And I'm due on the conference call in 45 minutes. Yikes. |
|
313)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108303)
Posted 2 Jun 2022 by Richard Haselgrove Post: ... and we're moving. Latest guess for arrival is 13:57, but it changes every time I look. I'm going to have to haggle with LNER for a replacement for my 13:33 ticket. |
|
314)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108302)
Posted 2 Jun 2022 by Richard Haselgrove Post: Well, it was going so well... Here I am, still at Gare du Nord, an hour and a half after I should have left. "Fatality near Lille" - I thought that was the English excuse. Still, they've bumped me forward to the first train out, and I'm onboard, baggage safely stowed. Just got the warning about the automatic closing doors.... |
|
315)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108300)
Posted 1 Jun 2022 by Richard Haselgrove Post: Last lap. I think I ought to be able to find my way from hotel to station in the morning?  Mind you, it's not a patch on last night...  Back home tomorrow evening, all going well - and it's gone fantastically well so far. I'll fill in the gaps - assuming anyone wants me to - after a few sleeps and a few ales. |
|
316)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 108286)
Posted 31 May 2022 by Richard Haselgrove Post: Well, they didn't yesterday. But on their own website, they simply said "We will provide further details early next week." - the week is yet young, especially in Canada.World Community Grid |
|
317)
Message boards :
Android :
Permission for project to download an app?
(Message 108283)
Posted 31 May 2022 by Richard Haselgrove Post: Some projects use a 'wrapper' app between BOINC and their main executable - but even then, it's BOINC which does the downloading, and should have the correct permissions. It can be easier to work out how projects like YoYo are operating by attaching it to a mainstream Linux or Windows machine, where you can examine the downloads and metadata much more easily. Their Android app is likely to work in a similar way, but with the details hidden. |
|
318)
Message boards :
Projects :
New WuProp app for Android and BOINC 7.18
(Message 108269)
Posted 27 May 2022 by Richard Haselgrove Post: P.S. The above was checked on my Android 11 tablet, which I have with me on my travels. No guarantee that it still works on Android 12. |
|
319)
Message boards :
Projects :
New WuProp app for Android and BOINC 7.18
(Message 108268)
Posted 27 May 2022 by Richard Haselgrove Post: What's remote control? Something on the app? Or is it an app you need from the app store?'Enable remote control' is (a long way down) on the preferences page of the BOINC Manager in v7.18.1 for Android. With it checked, normal GUI RPC management tools can be used on other devices who have TCP/IP access to the Android device. |
|
320)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108245)
Posted 26 May 2022 by Richard Haselgrove Post: Exploring the Douro. This:  close as a pleasant shock to the system after:  So far, I've found the beer fridge (better than port for this time - and heat - of the day. |
|
321)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108242)
Posted 26 May 2022 by Richard Haselgrove Post: I think I found the problem: "The spread of the virus, which originated in the UK, would have been slowed if BT Openworld had installed the latest Microsoft Outlook patches." - Microsoft software always does insane dangerous things. Should have used a proper email client.That was exactly the point I was trying to make to them. I was working solo on a client's site, installing a new server or something, when the beastie dropped. I submitted it to Symantec for investigation from site, then packed up and drove home. No response from Symantec, so I rang BT support line. I was on the phone until well after midnight - we had to drop the call at midnight for a shift change, and they rang me back. THEY were trying to help ME sort out my laptop: I was telling THEM that they had a serious problem at their end, and should be getting the night duty manager out of bed. I forget exactly how (it was 20 years ago, and we were both new to ADSL), but it was clear to me even that night that they were missing a critical service pack for Windows 2000. I failed to get them to budge, which is why I went first to an alternative AV firm to confirm the payload was malicious, and then to the computer press. End of conversation. I have the vineyards of the Douro to explore. |
|
322)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 108236)
Posted 25 May 2022 by Richard Haselgrove Post: I discovered a virus once: https://www.computerweekly.com/news/2240043325/BT-Openworld-sends-virus-to-customers Virus writers are cunning and devious. That one first appeared (and was sent to me by email) on the BT server at about 9 pm on the Friday of US Thanksgiving weekend. Everyone was asleep at the wheel. The virus checkers of the day only looked for matches to known patterns. This pattern was unknown, so it slipped through. Badtrans-B went on to become one of the commonest pests to date. I rang up a UK-based AV firm (the name escapes me at this time of night) on Saturday morning. Kudos to them: they answered the phone immediately, acknowledged that it sounded like a serious report, called one of their specialists into the office, asked me to test the potential fix, and had it published by about 4 pm. The American firms didn't react until Monday. |
|
323)
Message boards :
Projects :
World Community Grid has announced an extended outage from Feb 14 to April 22, 2022
(Message 108216)
Posted 24 May 2022 by Richard Haselgrove Post: The Boinc client is currently set to check for work units only once per day. You might check manually a few times per day to see if you can snag some work. Ni!The WCG server has been set to require a 24 hour delay during the whole transition period. Hopefully, they will remove that requirement when, or soon after, they create new work and verify that everything is running nominally. After that, the contact backoff will go back to seconds or minutes - all you have to do is catch your first task (not always easy at WCG!), and after that everything should run by itself. |
|
324)
Message boards :
Questions and problems :
macOS Monterey
(Message 108213)
Posted 24 May 2022 by Richard Haselgrove Post: Would you happen to know which projects work on the Apple M1?I think Einstein@Home were first off the blocks. |
|
325)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108208)
Posted 23 May 2022 by Richard Haselgrove Post: Would that be like a drive thru car wash experience without the car🚗? 🤔Pretty much, only without the spinning furry rollers. Adding those would have really made things interesting. |
|
326)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108204)
Posted 23 May 2022 by Richard Haselgrove Post: Found out today that the picture I took from the hotel roof (follow reply link) was the fabled Biblioteca Joanina itself. Vistted this morning - it lives up to the billing. No photography, unfortunately - even from the open doorway on the way out. |
|
327)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108203)
Posted 23 May 2022 by Richard Haselgrove Post: Here you go. Welcome to the navel jet wash.  The best I ever had was a tiny hotel in Paris, chosen because it was half way between the RER from the airport, and Gare Montparnasse for my train the next day. Just a stack of rooms, really, with room for a single chair for the porter at the bottom. The best all-over body wash - jets coming at you from all sides and all heights. |
|
328)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108198)
Posted 23 May 2022 by Richard Haselgrove Post: The shower room has both electricity and water - hallelujah. I can fill my travelling kettle and boil it in the same room. Not allowed in Britain. But it's safely separate from the actual shower - that uses hot water from elsewhere. |
|
329)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108195)
Posted 23 May 2022 by Richard Haselgrove Post: The latest one has two settings: "thunderstorm directly overhead" or "handheld wand". You can't use them both together. 'Thunderstorm' I can accept as a choice - not my preferred usage, but OK. The handheld wand has only one parking position: pointing 90 degrees from the wall, at about belly-button height. And it's fierce: I currently have the best power jet-washed belly button ever. |
|
330)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108193)
Posted 23 May 2022 by Richard Haselgrove Post: Why can nobody design a decent hotel shower? |
|
331)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108189)
Posted 22 May 2022 by Richard Haselgrove Post: Moved on, and arrived at the ancient University city of Coimbra:  rather like this  It's graduation day, and apparently the colours represent faculty. I was a student once, but we never did it like this!  
|
|
332)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 108161)
Posted 20 May 2022 by Richard Haselgrove Post: First posting session from Lisbon, Portugal - after three solid days of rail travel. Now, breakfast - a rare luxury! |
|
333)
Message boards :
Questions and problems :
Purpose of placeholder.txt
(Message 108160)
Posted 20 May 2022 by Richard Haselgrove Post: Some operating systems do. |
|
334)
Message boards :
Questions and problems :
Use more threads than available?
(Message 108159)
Posted 20 May 2022 by Richard Haselgrove Post: avg_ncpus will have an effect on how many other BOINC tasks (not threads) are allowed to run alongside this (multi-threaded) application. nthreads will actually control the number of threads spawned by this application (if the application is written to use the calling specification used by BOINC - not all do). They should be used together - the first to clear the right amount of space, the second to use it efficiently. |
|
335)
Message boards :
Questions and problems :
HELP!! Paspberry Pi 4 + BOINC - What am I doing wrong??
(Message 108085)
Posted 14 May 2022 by Richard Haselgrove Post: still attach projects in boinc manager has no project list to attach??You should have a file in var/lib/boinc-client called all_projects_list.xmlIt should be downloaded automatically by the boinc client, now that you've got the permissions sorted out and the client is running. If the automatic download doesn't seem to be working, you can copy that file from another machine. It should also be possible to download it manually from this BOINC web site, but just at the moment I can't find it. |
|
336)
Message boards :
Questions and problems :
Error installing
(Message 108024)
Posted 7 May 2022 by Richard Haselgrove Post: Neither of those. The problem is that Windows thinks that BOINC is still installed, and is trying to find the old installation package - but somebody or something has 'cleaned' that up, so it can't be found. The easiest solution is to re-install the exact same version as you used last time. If that is still in your downloads folder, just double-click on it. That should lead you through a 'repair' process, following which the upgrade should proceed normally. If you can't remember which version you installed last time, look at your account details at any project you've been attached to. Then, download the matching installer file from the archive folder, and try the repair from that. There's a thread somewhere here with umpteen alternative clear-ups, but they're all more complicated. Try the easy way first. |
|
337)
Message boards :
Questions and problems :
BOINC 7.18.1 doesn't work.
(Message 107979)
Posted 28 Apr 2022 by Richard Haselgrove Post: One last thing to try before you give up on the Manager. The client should have picked up the gui_rpc_auth password, and will be listening for it. But the manager doesn't detect the password automatically: you need to help it out. Right-click on the Manager icon on your desktop, and choose Properties. Then,  Use your own gui_rpc_auth, obviously. |
|
338)
Message boards :
Questions and problems :
WUProp not stopping when Boinc auto-suspends
(Message 107976)
Posted 27 Apr 2022 by Richard Haselgrove Post: 'Non compute intensive' projects are deliberately left running. One of the earliest projects in this category was the Quake Catcher Network: earthquakes don't always happen when you're conveniently close to a power outlet. It's untidy, I agree, but I don't think it'll be worth the effort to further sub-divide this category. |
|
339)
Message boards :
Questions and problems :
BOINC 7.18.1 doesn't work.
(Message 107963)
Posted 26 Apr 2022 by Richard Haselgrove Post: If there were a problem there we would have heard about it by now.The biggest problem I have with Gianfranco's distribution is that it's experimental and cutting edge - it's deliberately designed as a test tool for forthcoming developments. That's why I run it. Occasionally, you get offered a surprise update, and sometimes a bug slips into the system. But a bug serious enough to cause problems in general running is rare, and quickly fixed. |
|
340)
Message boards :
Questions and problems :
Unable to download 7.16.21 for Mac
(Message 107957)
Posted 25 Apr 2022 by Richard Haselgrove Post: The link that Dave provided downloads a single file called "boinc_7.16.21_macOSX_universal.zip" - the '.zip' indicates that it's a compressed archive file, containing all the real moving parts. Macs should be able to "extract all..." - that's the Windows terminology, but the Mac should be similar. Inside, there's another folder called "boinc_7.16.21_macOSX_universal" (no extension). And inside that, there's a document titled "ReadMe.rtf". That's your starting point. |
|
341)
Message boards :
Questions and problems :
Stop file deletion on reporting of tasks that error out.
(Message 107934)
Posted 22 Apr 2022 by Richard Haselgrove Post: Isn’t there a problem that affects CPDN due to their large final files or where multiple tasks finish at the same time where Boinc reports that the files have not been processed within the time limit and kills the process.That certainly has been a problem for CPDN in the past (many years ago), but that was on an even slower scale than Dave's bored band. The life expectancy of an upload file was originally 14 days, but that was too short to replace a failed upload server: the limit was extended (IIRC) to 90 days. |
|
342)
Message boards :
Questions and problems :
Stop file deletion on reporting of tasks that error out.
(Message 107930)
Posted 22 Apr 2022 by Richard Haselgrove Post: I am trying to get my head round the sequence of events here. Are you saying that the task crashes because of some programming/data error, and immediately goes into "ready to report" status: but some intermediate upload files, created earlier in the run, are still plodding along and worth saving? Or are you saying that the task finishes normally, with a "success" status, completes its normal end-of-run housekeeping (including uploading final exit files), but transitions to "ready to report" too quickly? The BOINC client should monitor those end-of-run files, and delay changing status and reporting the task until uploads are complete. Anything else would be a bug, and I think we would have heard about it by now. I suspect that the real problem may be that CPDN makes extensive use of intermediate upload files not associated with the final exit from the task. These uploads are associated with, but distinct from, 'trickle' reports of "progress so far". I suspect that these intermediate files are uploaded asynchronously (*) while the tasks are running, and not monitored for status when considering whether the task is "ready to report". If that's the case, I can't think of a way of preventing a report once "ready to report" is reached - it'll be reported an hour later, even if suspended. But if you can suspend the task after the final intermediate ('trickle') file has been created, but before the final wrap-up, that might work. Otherwise, it'll need a bug-fix in the client. * The sequence of events for a current main-project task is 21/04/2022 23:17:18 | climateprediction.net | Sending scheduler request: To send trickle-up message. 21/04/2022 23:17:19 | climateprediction.net | Scheduler request completed 21/04/2022 23:17:22 | climateprediction.net | Started upload of hadam4_a1f0_200010_13_929_012137174_2_r1038383722_7.zipSo the upload is never complete when reported as a trickle. The project might look at that. |
|
343)
Message boards :
Questions and problems :
Next new release?
(Message 107924)
Posted 21 Apr 2022 by Richard Haselgrove Post: Hello there. May I ask how you came to this BOINC Project Call on 14 January 2022. I was not aware of this, though I'm very glad to see you make mention of it.BOINC maintains a number of BOINC email lists, nowadays in the form of Google Groups. The top list on that page, BOINC Projects, is intended for BOINC project scientists and administrators to exchange information and useful tips, and to seek mutual assistance with problems. The somewhat erratic cycle of 'Project calls' is publicised through that list. You should be able to self-subscribe, though I haven't tested the procedure recently. |
|
344)
Message boards :
Questions and problems :
BOINC Tasks Processing - No Credits
(Message 107901)
Posted 20 Apr 2022 by Richard Haselgrove Post: SETI closed down about a year ago.Two years ago (31 March 2020) News item |
|
345)
Message boards :
Questions and problems :
BOINC 7.18.1 doesn't work.
(Message 107894)
Posted 19 Apr 2022 by Richard Haselgrove Post: There is one viable, usable, source for BOINC v7.18.1 for Linux (Debian, Ubuntu and related flavours): Gianfranco Costamagna's PPA. That build is completely separate from the v7.18.1 publicly released for Android - the current version numbering scheme has gone completely haywire. The PPA is automatically generated from Debian development sources whenever a significant change is detected, so it should be treated with caution. If you use it, watch out for, and be prepared to report, temporary bugs introduced during the development process. |
|
346)
Message boards :
Questions and problems :
BOINC 7.18.1 doesn't work.
(Message 107890)
Posted 19 Apr 2022 by Richard Haselgrove Post: The correct BOINC package for Ubuntu 22.04 LTS is definitely version 7.18.1That will be a version chosen by the Ubuntu package maintainers - but they got it wrong. BOINC certified a version called 7.18.1 for use on Android only around 3 August 2021, and that code hasn't been touched since. There is still no sign even of preparation work for any newer release that might become v7.20.x for the major desktop platforms, even though it was promised in 'around two months' - three months ago. |
|
347)
Message boards :
Android :
Android app can't be found in google search
(Message 107871)
Posted 15 Apr 2022 by Richard Haselgrove Post: Have they been submitted to the play store? If not, what is the reason?I think there was some chatter around the time of the release, to the effect that "BOINC violates some Play Store rules, by downloading third-party executables from external, unverified, sites" - i.e. project science applications. That would mean that a Play Store version would have to include every application, from every participating project, so they could all be validated as a single composite package. And re-built, re-packaged, and re-submitted for testing every time even a single project app was changed. The alternative - distributing a relatively stable platform package, and letting it download project apps as needed - was easier, and fitted more neatly with the 'BOINC for desktops' design. |
|
348)
Message boards :
Questions and problems :
Is there a way to get BOINC to give priority to Tasks with short "time to deadline"?
(Message 107827)
Posted 12 Apr 2022 by Richard Haselgrove Post: And now CPDN has no tasks for Linux either. Is that project actually alive?Yes. It's just allocated me four new tasks: three resends and one brand new. |
|
349)
Message boards :
Questions and problems :
boinccmd RPC: "Authorization failure: -155"
(Message 107766)
Posted 8 Apr 2022 by Richard Haselgrove Post: This fails:Typo. The boinccmd switch is --passwdBoinccmd tool |
|
350)
Message boards :
Questions and problems :
boinccmd RPC: "Authorization failure: -155"
(Message 107765)
Posted 8 Apr 2022 by Richard Haselgrove Post: I'd put it slightly differently. The remote machine needs to have a gui_rpc_auth.cfg password set. The BOINC client (not boinccmd) will be listening for that password. If you change that password, or set one for the first time, you need to re-start the BOINC client before it starts listening for the new one. If you want to send instructions from the local machine to the remote machine, you have to send the remote machine's password from the command line on the local machine. You can get general information from the remote machine without a password, but you must send a password to make any changes. |
|
351)
Message boards :
BOINC client :
BOINC 7.20 Change Log
(Message 107756)
Posted 7 Apr 2022 by Richard Haselgrove Post: I don't see it either on the 'download all' page or in the https://boinc.berkeley.edu/dl/ directory. (and there's been no announcement of any testing programme) (and no new or updated 'release branch' of the source code) |
|
352)
Message boards :
Questions and problems :
boinc-client crash and reboot my machine
(Message 107709)
Posted 4 Apr 2022 by Richard Haselgrove Post: BOINC (the program) is highly unlikely to stress "4 CPUs are at more than 90% load", except for 60 seconds or less at startup during benchmarking. That level of CPU activity is more likely to be attributable to one or more science projects, running under the direction of BOINC. We haven't discussed projects yet in this thread. |
|
353)
Message boards :
Questions and problems :
Order files download in.
(Message 107658)
Posted 30 Mar 2022 by Richard Haselgrove Post: No. Our request (on behalf of CPDN) to be able to control file transfers individually was declined. IIRC, that was at a time when one of the several CPDN regional upload servers was offline or full, and nothing could be uploaded from further down the transfer list until the blockage was fixed. That was eventually cured a different way - every transfer is now always attempted at least once when first set up - and the code that Thyme Lawn offered was never used. |
|
354)
Message boards :
Questions and problems :
E_mail "New post....." when post is by someone in your twit list.
(Message 107654)
Posted 30 Mar 2022 by Richard Haselgrove Post: How does this tie up with the thread title? Were you thinking of a different problem? |
|
355)
Message boards :
Android :
No jobs for android?
(Message 107645)
Posted 30 Mar 2022 by Richard Haselgrove Post: 1) Event log - those are just the administrative eye-candy files, not the meat of the business - and yet you still couldn't download all of them at the first attempt. For a comms device, it doesn't communicate very reliably. 2) Venue - the 'details' page for my Android device (12858658) ends:  You can only see that part of the page for a machine on your account, so I don't know what you have - but that's where you'd set it. |
|
356)
Message boards :
Android :
No jobs for android?
(Message 107642)
Posted 30 Mar 2022 by Richard Haselgrove Post: Or perhaps the designation somehow didn't take on the new phone?Venue designations are set and stored on the server. |
|
357)
Message boards :
Android :
No jobs for android?
(Message 107641)
Posted 30 Mar 2022 by Richard Haselgrove Post: That sounds more like a comms problem than a CPU problem. Einstein is quite strong on https security. What does the event log say? |
|
358)
Message boards :
Questions and problems :
boinc-client crash and reboot my machine
(Message 107639)
Posted 30 Mar 2022 by Richard Haselgrove Post: Richard,I don't really know. I'm still very much a newbie on Linux, and I only have personal experience of the Debian / Ubuntu / Mint family of distros. My guess would be that it depends on the distro and the precise version of BOINC in use. There's a note in the sources from February 2018: Both Fedora and Debian have their own systemd units for boinc;There were a number of systemd changes around that time, which roughly equates to v7.10.xx of BOINC (which I had a hand in releasing). It probably depends on how closely the Rasbian maintainers followed those changes. |
|
359)
Message boards :
Questions and problems :
boinc-client crash and reboot my machine
(Message 107635)
Posted 29 Mar 2022 by Richard Haselgrove Post: The log history for Linux systemd installations is kept in the system journal. |
|
360)
Message boards :
Questions and problems :
Wrong URL for Einstein
(Message 107542)
Posted 22 Mar 2022 by Richard Haselgrove Post: When I use an account manager, it creates a folder in c:\program data\boinc\projects called "https://einsteinathome.org", but when I attach manually, it's "http://einstein.phys.uwm.edu/".That's the mess Christian and I were trying to sort out in 2018 - I think Christian was on the Einstein staff team at that point. But it looks like the current team have messed it up again... |
|
361)
Message boards :
Questions and problems :
Wrong URL for Einstein
(Message 107539)
Posted 22 Mar 2022 by Richard Haselgrove Post: If you paste this address into a web browser http://einstein.phys.uwm.edu/get_project_config.phpyou'll be mimicking what BOINC does. On my machine (Windows 7 Chrome) it automatically redirects to the https version of the same address. The response starts <project_config> <name>Einstein@Home</name> <master_url>http://einstein.phys.uwm.edu/</master_url> <web_rpc_url_base>https://einsteinathome.org/</web_rpc_url_base> <web_stopped>0</web_stopped> <min_passwd_length>6</min_passwd_length> <sched_stopped>0</sched_stopped> <min_client_version>50800</min_client_version> <platforms> ...and so on. So maybe the advice on the web page is wrong? |
|
362)
Message boards :
Questions and problems :
Wrong URL for Einstein
(Message 107536)
Posted 22 Mar 2022 by Richard Haselgrove Post: Could you please check a few things for me? In your BOINC data folder, you should find a file called "all_projects_list.xml" Could you please make a note of the datestamp for that file, and then look inside it for the url for Einstein? You'll need to use a programmer's editor like Notepad++ to handle the Linux-style line endings. Mine is dated 3 October 2021, but it should be updated every two weeks - I've surpressed that. And the line I'm interested in says <url>http://einstein.phys.uwm.edu/</url>(not the web_url on the line below) The url is not the master url, but it should be what you see in the 'attach' wizard. And it's the place where BOINC will ask to get the real master url. If that lookup mechanism is broken, we're really screwed. I'll be checking that out while you look at what your machine has. |
|
363)
Message boards :
Questions and problems :
Wrong URL for Einstein
(Message 107530)
Posted 22 Mar 2022 by Richard Haselgrove Post: If you use the BOINC Manager 'attach' wizard, it goes through a extra stage: it calls get_project_config.php to fetch the master_url and attaches to that. If you use boinccmd to attach to the project, you are responsible for verifying that you have the up-to-date address, by following links link the one you've kindly posted. If you are attached to a project by an Account Manager, you are in the hands of the administrator of that AM, who may or may not be keeping alert to changes and applying them in a timely fashion. There's an open request to get this mess sorted out (#2642), but no progress. |
|
364)
Message boards :
Questions and problems :
Next new release?
(Message 107482)
Posted 18 Mar 2022 by Richard Haselgrove Post: Those files were actually compiled on a little low-power Celeron box that cost me less than £200, including a fully-licensed copy of Windows 10 Pro. It was running PrimeGrid multi-threaded tasks throughout. |
|
365)
Message boards :
Questions and problems :
Next new release?
(Message 107476)
Posted 18 Mar 2022 by Richard Haselgrove Post: Of course it does. That's why I could go and get a cup of coffee while it carried on working. You have mail. |
|
366)
Message boards :
Questions and problems :
Next new release?
(Message 107472)
Posted 18 Mar 2022 by Richard Haselgrove Post: Bran flakes? Oh, that kind of fibre - I've got 70 Mbit. Downloading is not a problem - it's the complexity of Microsoft's packaging, uninstalling expired versions and replacing them with new ones that takes the time. |
|
367)
Message boards :
Questions and problems :
Next new release?
(Message 107470)
Posted 18 Mar 2022 by Richard Haselgrove Post: Ugh. Don't hold your breath. I've got the source, but I've also got a huge number of errors when I try to compile it. Seems VS2019 is no longer the version I downloaded, so I'm downloading 1.28 GB of files in 457 packages to update it. I think I'll go and make myself a cup of coffee while it sorts itself out. |
|
368)
Message boards :
Questions and problems :
Android, getting too many wu's to handle in time
(Message 107469)
Posted 18 Mar 2022 by Richard Haselgrove Post: The absolute size of the work cache doesn't seem to matter. Here's an example of the type of log that illustrates what happens: Sat 19 Feb 2022 10:41:27 GMT | Einstein@Home | [sched_op] NVIDIA GPU work request: 21.84 seconds; 0.00 devicesAt the time, my cache settings must have been 0.05 + 0.01 days - that adds up to 5184 seconds. Note that the first request is for something trivial, with no idle devices - and it got one task, as you'd expect. That should have been the end of the matter. But a minute later, it asked again - for the full cache setting, plus an idle device. Each request got the full four tasks - I only saved the first and the last, for clarity. The problem seems to be that the incoming new work allocation, whilst complete - I watched them all download - doesn't get added to the client's knowledge of 'work ready to run'. So it asks again, and again, and again. This particular event was recorded on a Linux machine, so that's a third OS that can be affected. |
|
369)
Message boards :
Questions and problems :
Next new release?
(Message 107467)
Posted 18 Mar 2022 by Richard Haselgrove Post: That zip contains the manager and the tray programs, but not the command or the main executable. Can't you just email me the ones you've compiled?I have the tools to compile, but I don't do it regularly - I prefer to run the 'release' versions, so I can watch and reproduce the bugs that other people report. And report a few of my own... I've just been looking for other downloadable artifacts, but I think the system has changed yet again - I can't find any. So it's not just Jord's memory. I'll try and download a fresh copy of the source 'as is' today, and build from that - it'll take a while. Do I have your email address? |
|
370)
Message boards :
Questions and problems :
Next new release?
(Message 107462)
Posted 18 Mar 2022 by Richard Haselgrove Post: What about those nightly builds? Are they still done?I've just received an email from Vitalii, asking me to test his solution to a bug I reported some time ago. He's given me a download link for the version with this specific (Manager) patch: https://github.com/BOINC/boinc/suites/5591557712/artifacts/181337362 But that turns out to be exactly that - a single-file download. I'll try to find the more general page where you can choose which components you want to download. Be aware that all test builds only comprise the raw component files - boinc.exe / boinccmd.exe / boincmgr.exe. There's no installer - only David can build that (UCB licence and code-signing key required). |
|
371)
Message boards :
Questions and problems :
Next new release?
(Message 107454)
Posted 18 Mar 2022 by Richard Haselgrove Post: I'd prefer to see the scientific community applying collective pressure on David to get off his butt, and stop playing 'the dog in the manger'. We should all be getting the benefit of a new release, not just those of us who can self-compile. |
|
372)
Message boards :
Questions and problems :
Next new release?
(Message 107452)
Posted 18 Mar 2022 by Richard Haselgrove Post: David Anderson was asked exactly that question during the BOINC Projects Call on 14 January 2022. Minutes and a recording of that call can be found at https://drive.google.com/drive/folders/1_YDPZi5OQ10nU91QwICGCKaP7vJTS_N4 David's answer was that a new release would be made in 'around 2 months'. It will not have escaped your notice that more than two months have elapsed since that statement. There is no sign of even any preparatory work having been started. |
|
373)
Message boards :
Questions and problems :
Boinc uses wrong GPU for SRBase only
(Message 107444)
Posted 18 Mar 2022 by Richard Haselgrove Post: That is incorrect, if I don't forbid any GPUs, It runs on both.I'd double-check that theory against reality, if I were you. BOINC will start two tasks, and give one the instructions to run on device 0, and the other instructions to run on device 1. BOINC Manager will show you that the instructions have been issued. But what happens next is entirely dependent on the project's science application. If the application is not capable of reading and acting on the instructions, it'll do its own thing. I've seen that happen in the last few months, while a new application was being developed and tested at GPUGrid. They made a mistake in the code to implement device selection (or left it out entirely), and both tasks ran on device 0. We noticed, pointed it out, and they corrected it. Part of the confusion arises from a change (many years ago) in the way BOINC passes the device instruction. When GPUs were first introduced, it was a very simple parameter on the command line: as GPUs got commoner and more diverse, that was changed to a more complex structure in a file called init_data.xml. Not every project got the memo. |
|
374)
Message boards :
Questions and problems :
Android, getting too many wu's to handle in time
(Message 107439)
Posted 17 Mar 2022 by Richard Haselgrove Post: Richard, the device I checked requested a scheduler update every 70 seconds but this may have been also because I had deactivated new work four days ago to get the work queue down. It was an Einstein device. I installed Universe and it downloaded 67 wu's at once and no new requests in the last 10 minutes. It leaves the device sitting with 167 wu's in the queue total so maybe that is the new target now. Maybe I have to deplete the queue completely before asking for new work.70 seconds exactly fits in with my observations over the years. It's made up of: 61 seconds requested by Einstein (60 set by the admins, plus 1% added by BOINC in case of clock synchronisation errors) 4 seconds typical time for the Einstein server to process a request 5 seconds inserted by the BOINC client between scheduler requests (all subject to miscellaneous +/- seconds depending on how busy everything is) Seeing it at Einstein, but not at Universe, also matches my observations. I've only ever seen it in relation to Einstein tasks. I usually run Einstein on GPUs, and have other projects on CPUs. When this problem strikes (very, very, rarely), Einstein fetches incessantly, the CPU project(s) doesn't. I'd vaguely assumed that the bug was in the GPU part of the client work-fetch algorithm, but the triggering state was caused by something unusual in the Einstein scheduler reply. And there it rests. |
|
375)
Message boards :
Questions and problems :
Android, getting too many wu's to handle in time
(Message 107434)
Posted 17 Mar 2022 by Richard Haselgrove Post: Check the current event log first, if you haven't changed it yet. You should see 'Starting scheduler request' every minute-and-a-bit. If you could set <sched_op_debug> without breaking the behaviour, you should see exactly how many seconds of work you're requesting each time. |
|
376)
Message boards :
Questions and problems :
Android, getting too many wu's to handle in time
(Message 107431)
Posted 17 Mar 2022 by Richard Haselgrove Post: Hey Dave. Yes the additional work buffer is 0. Until february I never had problems with changes in the work buffer. What may have happened is that the last Android update provided by Samsung somehow has let to the BOINC buffer settings not being taken into account anymore. This is a strange behaviour and it happens on ALL of my devices.That sounds like a very unlikely explanation, but odd things have happened in computing ... I have a 4-core Samsung tablet which was running WCG and Einstein 50-50: since the WCG pause, it's been doing exclusively Einstein work. I haven't updated the operating system for a while (it's currently on Android 4.4.177-22626479 (Android 11)), but I'll keep an eye on it when I do next update it. At the moment, it's keeping '4 running and 4 ready to start', as per instructions. More interesting question: how does the excessive workload arrive? Is it a single work request, resulting in a huge spike of work, or a series of separate work requests, again and again and again, spaced 1 minute apart (the Einstein server backoff setting)? I've seen the latter behaviour (endless, repeated, requests) on Windows machines, but never been able to track it down: it stops if you change any of the Event Log settings. [There's a known problem if you use the max_concurrent setting in an app_config file (similar but subtly different). But you won't have installed an app_config under Android, will you?] |
|
377)
Message boards :
Questions and problems :
Suspend on battery power doesn't work
(Message 107348)
Posted 12 Mar 2022 by Richard Haselgrove Post: Make sure your activity preferences are set to 'run based on preferences', not 'run always', before you do. |
|
378)
Message boards :
Questions and problems :
How recent is "users with recent credit" on a project's server status page?
(Message 107308)
Posted 8 Mar 2022 by Richard Haselgrove Post: Dave's experience doesn't tally with what you said.This is the BOINC message board. I gave you the generic BOINC answer. If your question was specific to the CPDN project (Dave's chosen specialism), ask at CPDN. If any other project is of more interest to you - ask there. |
|
379)
Message boards :
Questions and problems :
How recent is "users with recent credit" on a project's server status page?
(Message 107306)
Posted 8 Mar 2022 by Richard Haselgrove Post: How long does it take for RAC to drop to zero? Take an average user with reasonable credit in a project, if they stopped today, when would their RAC be 0?Never. It's like radioactivity - it has a half-life (7 days, in this case). So it gets smaller and smaller and smaller, but never vanishes entirely. For the 'recent' question, you mean 'how long until it reaches 1'? It entirely depends where you start from. |
|
380)
Message boards :
Questions and problems :
How recent is "users with recent credit" on a project's server status page?
(Message 107297)
Posted 7 Mar 2022 by Richard Haselgrove Post: https://github.com/BOINC/boinc/blob/master/html/user/server_status.php#L441: $s->users_with_recent_credit = BoincUser::count("expavg_credit>1"); So: RAC greater than 1. Time doesn't come into it. |
|
381)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 107294)
Posted 7 Mar 2022 by Richard Haselgrove Post: I had to clamber over the remains of some of that yesterday. I decided to repeat the walk around Thruscroft Reservoir, west of Harrogate in the Yorkshire dales. The website (checking it today - I didn't before I set out) says "Our car park and path are temporarily closed": the actual carpark was open and in use, but access to the woodlands had been blocked by a notice, a single wooden bar, and some tatty hazard tape. A quick personal risk assessment later, I decided to do the walk anyway. Most of it consists of broad, well-made and well-maintained footpaths, much of it through mature pine woods. The reservoir dam was completed in 1966 and the valley flooded: I suspect much of the woodland is older than that (*). Right from the entry-point, I could see fallen trees, leaning trees, uprooted trees, damaged trees. Much of the debris had been tidied off the walkways, but one symbolic fallen tree had been left, starkly and pointedly, lying across the path as a warning. Worse was to come. I couldn't get beyond half-way up the western bank. A whole stand of trees had been blown down, completely blocking all progress. I managed to find a place where one of the boundary trees had fallen across and partially flattened the barbed wire fence: I scrambled out and bypassed the blocked section, through the fields. In nearly fifty years of driving and walking through the dales, that's the worst storm damage I've seen. A little early for a 'once in 100 years' event. * Maybe not. My 1958 map has no lake, and only about a third of the area of woodland. |
|
382)
Message boards :
Projects :
Need a new project
(Message 107268)
Posted 3 Mar 2022 by Richard Haselgrove Post: Odd, as they report Linux tasks at 10 TeraFLOPS, almost as much as Windows.Without specifying the version number. |
|
383)
Message boards :
Projects :
Need a new project
(Message 107266)
Posted 3 Mar 2022 by Richard Haselgrove Post: Two separate issues.You also have to have Boinc 7.16.20As a minimum I presume as my self build 7.19.0 works fine with them. BOINC v7.16.20 was a special release for Windows only. There's nothing special about the BOINC code in that version. The vital change is in one of the supporting, ancillary, files distributed in the installer for that version. IF you are running Windows IF you have compiled some sort of v7.19.0 code under Windows, to run under Windows IF you were paying attention to the discussion about that ancillary file, and have an updated version then it should work. If you missed out on any of those steps, the outcome will be indeterminate. |
|
384)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 107150)
Posted 18 Feb 2022 by Richard Haselgrove Post: Y'all still there? Or did Eunice blow you away?Not yet. But the north of England is running some 12 hours behind the southerners |
|
385)
Message boards :
Questions and problems :
Getting GPU work when told not to
(Message 107142)
Posted 18 Feb 2022 by Richard Haselgrove Post: From the log you posted at GitHub: 590 Einstein@Home 17-02-2022 10:24 PM [sched_op] Starting scheduler requestYou asked, you got. That's how Einstein works. Did you follow the advice you've been given in #4624? In the context of that issue, Einstein is running a very old server - the base version reports 611, so they won't have the patch that appears to have been applied between 713 and 717 |
|
386)
Message boards :
Questions and problems :
Now that WCG is down, looking for Android 10/11 projects; Still no asteroids?
(Message 107124)
Posted 17 Feb 2022 by Richard Haselgrove Post: Einstein > message has been sent to project admin asking for someone to fix the Android task issue.And it's working again now. |
|
387)
Message boards :
Questions and problems :
Suspended CPU is busy.
(Message 107079)
Posted 14 Feb 2022 by Richard Haselgrove Post: Perhaps it should have an uninstall option somewhere. It could then ask if you want to complete the uninstall. I see what you are saying, but uninstalling a software package, in my humble opinion, should completely uninstall the package.No, removing the program doesn't usually remove the data. Imagine uninstalling a word processor? Should that delete the novel you've just written with it? |
|
388)
Message boards :
Questions and problems :
Suspended CPU is busy.
(Message 107061)
Posted 12 Feb 2022 by Richard Haselgrove Post: Or stdoutdae.txt in the BOINC data directory. |
|
389)
Message boards :
Questions and problems :
AI WUs possible in the future?
(Message 107051)
Posted 11 Feb 2022 by Richard Haselgrove Post: GPUGrid has been experimenting with such an application (it's the application which will have to change, not the data in the WUs) for some months now. The researcher concerned described their plans some time ago in GPUGrid message 57766. Most of the development stages described in that post have now been solved, and the most recent test tasks have run to completion without error. We await the next step. |
|
390)
Message boards :
GPUs :
Need help getting Intel IGP to work again
(Message 107029)
Posted 9 Feb 2022 by Richard Haselgrove Post: already did.But you didn't show us the startup messages. I agree with Ian - it's likely to be a driver issue. Intel produces separate (distinct) drivers for their CPUs and the GPU component in the CPU package. BOINC would list them separately. If you picked the CPU driver only, that wouldn't make the GPU available. |
|
391)
Message boards :
GPUs :
Need help getting Intel IGP to work again
(Message 107025)
Posted 9 Feb 2022 by Richard Haselgrove Post: Check the first 30-40 lines of the message log from the last time BOINC was restarted, or restart again to get a fresh set. |
|
392)
Message boards :
Questions and problems :
GPU Processing broke: Linux - upgraded to 20.0.4 LTS - Intel and 2 Nvidia 1070 GPUs
(Message 107000)
Posted 4 Feb 2022 by Richard Haselgrove Post: Always nice when things go right! |
|
393)
Message boards :
Questions and problems :
GPU Processing broke: Linux - upgraded to 20.0.4 LTS - Intel and 2 Nvidia 1070 GPUs
(Message 106997)
Posted 4 Feb 2022 by Richard Haselgrove Post: The Amicable Numbers applications page shows that their GPU programs are written using the OpenCL programming language. Your event log shows no sign of an OpenCL driver - only CUDA. Try to add an NVidia OpenCL component to your existing driver package. You'll probably have to reboot for it to be recognised by BOINC. |
|
394)
Message boards :
Questions and problems :
Getting GPU work when told not to
(Message 106986)
Posted 3 Feb 2022 by Richard Haselgrove Post: In your opening post, you say that you are pausing GPU work "to save power". That sounds like a long term decision - not just a five minute tea break while you have a quick game of solitaire. In which case, you could consider excluding the Milky Way project from your GPU(s) in cc_config.xml |
|
395)
Message boards :
Questions and problems :
Getting GPU work when told not to
(Message 106982)
Posted 3 Feb 2022 by Richard Haselgrove Post: Set preference "Don't use GPU" at the project. |
|
396)
Message boards :
Questions and problems :
Getting GPU work when told not to
(Message 106980)
Posted 3 Feb 2022 by Richard Haselgrove Post: Which project? Does the Event Log contain lines commencing "Resent lost task ..." for the download event? |
|
397)
Message boards :
Questions and problems :
Changing BOINC Data Directory on Fedora35
(Message 106971)
Posted 3 Feb 2022 by Richard Haselgrove Post: I have an open request for the BOINC Linux package maintainers to implement this - Linux: support moving data directory to larger drive. After some initial indications of support, it appears to have fallen off the radar. My initial request was based on a suggestion on this message board: message 100356. If you haven't seen that one yet, it might be worth a try. |
|
398)
Message boards :
BOINC client :
Where do we post release notes these days?
(Message 106943)
Posted 1 Feb 2022 by Richard Haselgrove Post: The preparation of release notes is in the documented 'process' section of https://github.com/BOINC/boinc-policy/blob/master/Development_Documents/Client_Release_Process.md. An individual known as the 'Release Manager' is responsible for overseeing the process overall, but doesn't have to do each step themselves. When I was asked to fill this role (briefly - for one time only) for BOINC v7.10, I didn't do the release notes directly myself, but I asked Jord if he could help out, and gave him the best list I could of the significant changes. Unfortunately, the current (self-appointed) Release Manager seems neither to prepare the notes himself, nor to nominate or request somebody else to do them for him. |
|
399)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 106941)
Posted 1 Feb 2022 by Richard Haselgrove Post: Grumble: First appointment to have new boiler installed, engineer didn't turn up, boiler returned. Today second visit, engineer says can't do it the way envisaged. On the plus side, very glad we got a fixed price quotation that the company are I suspect beginning to regret about now as I am guessing, it is going to be at least an extra £750 between labour and materials, possibly more. - serves them right for offering me quote based on pictures they asked for and what I was able to tell them over the phone.They never learn, do they? I had that 25 years ago, when I first replaced an old clunker of a boiler, floor-standing in the cellar, with a modern one that could do hot water as well. That wasn't even a photo job - somebody described as a engineer came to the house, saw exactly what I'd got, and based the fixed quotation on that. The real engineers, when they arrived, were most scathing about salesmen who masquerade as engineers, and only do half the job: fitting the new boiler was easy, but they had to work out their own kludge for connecting it to the existing radiators. |
|
400)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106906)
Posted 28 Jan 2022 by Richard Haselgrove Post: As a driver, pedestrian, and one-time cyclist, I'd like to think that all three of my mirrors are properly adjusted so that I, when driving, can see as clearly as possible behind me. As I approach 70 (age, not mph!), I'm not as flexible as the youngsters when it comes to turning head and torso in the driving seat, so I'm concentrating even harder on the mirrors. I haven't seen the new highway code yet, but I hope it stresses that passengers - sans benefit of mirrors - are in most need of the 'stretch and turn' technique. |
|
401)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106903)
Posted 27 Jan 2022 by Richard Haselgrove Post: Australia scientists find 'spooky' spinning object in Milky WayMy dead-tree reading over dinner yielded a better description: Astronomers discover mysterious pulsing object that may be new class of star |
|
402)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106898)
Posted 27 Jan 2022 by Richard Haselgrove Post: How about Australia scientists find 'spooky' spinning object in Milky Way? |
|
403)
Message boards :
Questions and problems :
how lower the impact BOINC on startup ?
(Message 106839)
Posted 21 Jan 2022 by Richard Haselgrove Post: It's unlikely to be BOINC itself which is delaying startup: the BOINC programs themselves are quite lightweight, and shouldn't take long to load. It's more likely to be the science programs from whichever projects you are attached to that are causing the delay. You could test that by suspending all projects (or all tasks) before closing down one day, and seeing how long the restart takes. But remember to re-enable the projects/tasks when you've finished the test! If it is the project startup which causes the delay, you can postpone it with a configuration option: <start_delay>nseconds</start_delay>Read the configuration section of the user manual for details of how to do that. |
|
404)
Message boards :
The Lounge :
Let's talk politics
(Message 106818)
Posted 19 Jan 2022 by Richard Haselgrove Post: When the answer becomes "nobody told me", the problem becomes "I've lost the ability to work things out for myself". And that means it's time for the men in grey suits to pay him a visit. |
|
405)
Message boards :
Promotion :
New Credit Award System
(Message 106779)
Posted 14 Jan 2022 by Richard Haselgrove Post: That sounds more like inflation to me. |
|
406)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106752)
Posted 12 Jan 2022 by Richard Haselgrove Post: for CPDN it is the floating point benchmark that is significant.I'd say that's true of most scientific projects, except the pure mathematics ones. |
|
407)
Message boards :
Questions and problems :
Infected Project
(Message 106692)
Posted 6 Jan 2022 by Richard Haselgrove Post: Double-check those files with a service like virustotal - that checks them against multiple virus databases. Most BOINC virus 'detections' turn out to be false alarms: they are reported by virus checkers using "heuristics" - looking at the program's behaviour, rather than any actual payload. |
|
408)
Message boards :
Projects :
News on Project Outages
(Message 106686)
Posted 5 Jan 2022 by Richard Haselgrove Post: It was Chrome (under Windows 7) that let me retrieve those certificate details. |
|
409)
Message boards :
Projects :
News on Project Outages
(Message 106684)
Posted 5 Jan 2022 by Richard Haselgrove Post: The BOINCstats website (as seen from here) is using a Cloudflare certificate valid from 11 August 2021 to 10 August 2022. I've sometimes seen problems establishing a secure connection to a busy and active server: I think https requires significantly more server resource to establish the connection than http. If server operators have succumbed to the pressure for 'https for everything', without upgrading their hardware, glitches can happen. They usually clear within a few minutes or hours. |
|
410)
Message boards :
Questions and problems :
Force regular project update?
(Message 106671)
Posted 4 Jan 2022 by Richard Haselgrove Post: An alternative (and the one I use at WCG/OPN) is to open a command window on the boinccmd folder, and run a circular script with an embedded wait command. No need for the task scheduler. No, I didn't write it: one was posted on the project forum, and I tweaked it a bit until I was happy with it. |
|
411)
Message boards :
Questions and problems :
Force regular project update?
(Message 106662)
Posted 4 Jan 2022 by Richard Haselgrove Post: The facility within BOINC to do that is set at the project's end - GPUGrid is one of mine that uses it. If a user wants to mimic the behaviour, they normally write an operating system script that calls the boinccmd utility. |
|
412)
Message boards :
Questions and problems :
Can't communicate with one particular project on one computer only
(Message 106630)
Posted 31 Dec 2021 by Richard Haselgrove Post: Is there a way for Windows to have a file that updates automagically that can be linked to in the same way the Linux BOINC certificate file does?We did have a conversation about that, while we were waiting for what eventually became v7.16.20 to emerge. Read from about https://github.com/BOINC/boinc/issues/4544#issuecomment-937558197 The general consensus seemed to be 1) Windows doesn't make it easy to locate a single 'bundle' file which would work with BOINC's implementation of curl. 2) If there is such a file, it might be in a different place on each release of Windows. 3) If Windows did have such a file, it would be the responsibility of Windows to keep it up-to-date. Many BOINC users run BOINC on old, retired, machines - perhaps with an earlier version of the OS (I'm typing this on a Windows 7 machine). Can we rely on Microsoft to keep the certificates updated on older systems? |
|
413)
Message boards :
Questions and problems :
Can't communicate with one particular project on one computer only
(Message 106628)
Posted 31 Dec 2021 by Richard Haselgrove Post: To be horribly picky, the problem with Windows clients recently was with the included data, rather than the protocol in use. Many of us edited or replaced the faulty file long before the full re-bundled package was assembled. |
|
414)
Message boards :
Questions and problems :
How to tell the BOINC Client to not accept jobs above a certain number of CPUs?
(Message 106623)
Posted 31 Dec 2021 by Richard Haselgrove Post: OK, that looks interesting. However, it doesn't seem to work for the existing jobs. If I look at their properties:One of the quirks of app_config.xml is that it doesn't update the displayed information about already downloaded tasks, but does (if you've got the syntax right, and issued a 'read config files' instruction) act on most of them immediately. The new settings will be displayed for new tasks downloaded after the new file is activated. |
|
415)
Message boards :
Questions and problems :
Can't communicate with one particular project on one computer only
(Message 106615)
Posted 30 Dec 2021 by Richard Haselgrove Post: I have been running BOINC v7.18.1 on Android pretty much since it was released in August. I can connect to it with a standard BOINC Manager, and also with BoincView (a forerunner of BoincTasks, from about 2007. The RPC protocols haven't changed much.) If the Android client can't communicate with one particular project, that isn't the fault of one particular Manager. |
|
416)
Message boards :
News :
Windows communication failure: temporary workaround
(Message 106504)
Posted 24 Dec 2021 by Richard Haselgrove Post: No, it's T.Brada's certificate that expired, on 19 December.  At least, that's assuming they use the same one for their web site. |
|
417)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106503)
Posted 24 Dec 2021 by Richard Haselgrove Post: That's the point that David doesn't seem to appreciate yet. SU relies on [a working] BOINC client. Moving to SU, so users can't whinge, moan, and generally hold his feet to the fire doesn't absolve him from the responsibility of continuing to maintain BOINC. |
|
418)
Message boards :
Questions and problems :
BOINC will only run 1 GPU when set to less than 10 days of work.
(Message 106502)
Posted 24 Dec 2021 by Richard Haselgrove Post: Have you set <use_all_gpus>1</use_all_gpus>in cc_config.xml? Does BOINC list both GPUs in the event log at startup, without saying that the second is unused? |
|
419)
Message boards :
Projects :
News on Project Outages
(Message 106483)
Posted 22 Dec 2021 by Richard Haselgrove Post: Bernd wrote: A few hours ago the Einstein@Home servers experienced a hard shutdown. Apparently several safety measurements failed, for most of these it's still unclear why. We're trying to get the servers back up running and replace the ones with terminal failures and are working on restoring all services before Friday. Until then you may experience some limitations. In particular the O3AS workunit generator isn't running. |
|
420)
Message boards :
Projects :
News on Project Outages
(Message 106481)
Posted 22 Dec 2021 by Richard Haselgrove Post: https://einsteinathome.org/ seems to be down.Seems to be affecting the website and the scheduler only: completed tasks can be uploaded, but not reported. I think the upload servers are in Europe, but the web servers are in America. It may be this afternoon (European time) before they can investigate. Edit: I was wrong - they're back. |
|
421)
Message boards :
Projects :
iThena.Measurements & iThena.Computation are timing out with SSL errors
(Message 106463)
Posted 18 Dec 2021 by Richard Haselgrove Post: Or they are using one of the two SSL certificates in the BOINC client bundle (yes, the emergency fix v7.16.20 version) which expired on 15 December. They are: GlobalSign Cybertrust Global Root Can you reach the iThena website via a secure SSL browser? If so, what does the browser say about the ceritifcate status? Edit - as I warned in https://github.com/BOINC/boinc/issues/4530#issuecomment-933664532 |
|
422)
Message boards :
Questions and problems :
missing project in BM project list
(Message 106460)
Posted 18 Dec 2021 by Richard Haselgrove Post: There is a file "all_projects_list.xml" in your BOINC data directory which defines the entries shown on the 'attach to project' wizard in BOINC Manager. The same list also drives the display at https://boinc.berkeley.edu/projects.php - the modern version of the wiki page you reference. As the Wiki says, "This page is no longer kept up to date". The replacement list is maintained by David Anderson personally. He prefers to be contacted by email. |
|
423)
Message boards :
Questions and problems :
Getting too may WCG tasks on systems that had been working ok
(Message 106444)
Posted 16 Dec 2021 by Richard Haselgrove Post: I suspect there is a dump of WCG tasks. I did not see the 1600 download all at once as a dump as the event log was too big and got truncated.Yes, WCG work (especially, GPU work on Covid, task name prefix OPNG) tends to get released in batches - and the batches are getting bigger: I got 35 in one go at lunchtime 16/12/2021 12:17:49 | World Community Grid | Scheduler request completed: got 35 new tasksBut I keep my requests reasonable, and never get more than I request. 16/12/2021 12:17:48 | World Community Grid | [sched_op] NVIDIA GPU work request: 42684.92 seconds; 2.00 devices 16/12/2021 12:17:49 | World Community Grid | [sched_op] estimated total NVIDIA GPU task duration: 21396 secondsThe Event Log flag <sched_op_debug> is active on all my machines, and can be useful in tracking down issues like this. I find it unlikely that WCG would issue 1600 tasks in response to a single request: most projects set a lower limit in their server's feeder configuration (100 or 200). Even without the current log, you can still track the history. * Under Windows, in files stdoutdae.txt and stdoutdae.old in the data folder. You can configure those to retain any size you like. * In the task list (either in the BOINC Manager, or on the project website), by inspecting the deadlines of the allocated tasks. WCG requests a delay of two minutes between fetches: you would be able to sees a discontinuity of 2+ minutes between batches if multiple fetches were involved. Question: when building the x64 release I get an executable that is 2x as big as the 7.16.20 that Berkeley has. There must be some setting the in my VS2019 that is different from Berkeley's. Usually the debug version is the size hog.Yes, VS2019 files are bigger. Earlier versions relied on library routines delivered as separate, external, .DLL files: with the VS2019 build, the libraries are embedded in the main executable. [edit-2] All three systems used app_config to limit the number of concurrent WCG tasks. I replaced boinc.exe with that version you posted. Maybe this will fix the problem?Yes, that's exactly what the #4592 patch was designed to fix. That's why it's called "client: fix work-fetch logic when max concurrent limits are used". The problem makes itself apparent by causing multiple, repeated, limitless, work fetch requests. Which Is why I keep asking if multiple, repeated, limitless, work fetch requests are visible in your logs. |
|
424)
Message boards :
Questions and problems :
Getting too may WCG tasks on systems that had been working ok
(Message 106440)
Posted 16 Dec 2021 by Richard Haselgrove Post: Oh dear. The resource share you read in a <work_fetch_debug> segment of the event log has NOTHING TO DO with the resource share you set on a project web site. The <work_fetch_debug> usage is an instantaneous snapshot - literally, "what can we do now, this second, in this single instance of work-fetch decision making?". The first question is: "are we allowed, now, this instant, to fetch work from this project?" If no, the value will be zero: if yes, the value will be positive. The routine loops over all attached projects, and counts the positives. If N projects are in the 'can fetch' state, each will show a share of 1/N. The project resource share is a long-term value. That is designed to balance out the project work allocation over days, weeks, months - not second by second. When I asked "what does the event log say about fetching?", I was hoping for a broad-brush overview, at least in the first instance. When you said you "had to abort" a number of tasks, how did they arrive? Did they arrive in one huge dump? Or did they arrive in a trickle, a few at a time, again and again and again? To a first approximation, a dump indicates a server problem: a trickle indicates a client problem. We need to know where to start looking. |
|
425)
Message boards :
Questions and problems :
Getting too may WCG tasks on systems that had been working ok
(Message 106430)
Posted 15 Dec 2021 by Richard Haselgrove Post: WHAT DID THE EVENT LOG SAY ABOUT FETCHING? |
|
426)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 106406)
Posted 13 Dec 2021 by Richard Haselgrove Post: You need to be able to identify yourself, which you cannot do with expired IDs.I don't think it's anything obvious like that. It'll be more to do with the obscure evolution of the British political/civil service mindset. Our passports evolved early, and have carried photographs from the very beginning. Driving licences came much later, and didn't carry photographs until even later - and I don't think they're obligatory even now. I was driving around, perfectly legally, on a plain paper licence until a couple of years ago. (I'd had a new copy printed when I moved to this house thirty years ago, and it would have been valid until my seventieth birthday next year). I changed it when I visited the USA a couple of years ago, in case I felt the urge to hire a car while I was over there. I was able to get a photo-plastic EU-style card by online form-filling: the photograph (a very bad one!) was copied by me giving explicit permission for the driving licence people to access the records held by the passport people. I very much doubt that the passport people destroy the records associated with expired passports: it's more likely that there's an arbitrary rule that says that the photograph on a newly-issued driving licence has to have been taken in the last ten years. I much prefer the system for local officials to issue photo ID for free bus travel (the only other photo ID I hold). Stand still, look into the webcam, thank you. And it's a much better photo! |
|
427)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106382)
Posted 10 Dec 2021 by Richard Haselgrove Post: The optionI think it works the way the developers intended: <fetch_on_update>0|1</fetch_on_update>Setting it to 1 adds extra fetching, but 0 doesn't block normal fetches. That quote comes from the User Manual. |
|
428)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106377)
Posted 10 Dec 2021 by Richard Haselgrove Post: And today, it didn't even attempt to fetch work. Stayed at GPUGRID | [work_fetch] share 0.000 job cache fullthroughout the Einstein refill. As it should. No configuration changes, apart from the log flag selection. Maybe it just doesn't like Thursdays? |
|
429)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106376)
Posted 10 Dec 2021 by Richard Haselgrove Post: Well, I've got a routine log for one of the scheduled GPUGrid updates: 10/12/2021 14:59:17 | GPUGRID | [sched_op] sched RPC pending: Requested by project 10/12/2021 14:59:17 | GPUGRID | piggyback_work_request() 10/12/2021 14:59:17 | | [rr_sim] doing sim: work fetch 10/12/2021 14:59:17 | | [rr_sim] start: work_buf min 21600 additional 4320 total 25920 on_frac 1.000 active_frac 1.000 10/12/2021 14:59:17 | GPUGRID | [rr_sim] 82570.33: e9s627_e1s741p0f526-ADRIA_BanditGPCR_APJ_b0-0-1-RND6065_3 finishes (1.00 CPU + 1.00 NVIDIA GPU) (3715287.81G/47.27G) 10/12/2021 14:59:17 | | [rr_sim] end 10/12/2021 14:59:17 | | [work_fetch] ------- start work fetch state ------- 10/12/2021 14:59:17 | | [work_fetch] target work buffer: 21600.00 + 4320.00 sec 10/12/2021 14:59:17 | | [work_fetch] --- project states --- 10/12/2021 14:59:17 | GPUGRID | [work_fetch] REC 391197.604 prio -1.010 can request work 10/12/2021 14:59:17 | | [work_fetch] --- state for CPU --- 10/12/2021 14:59:17 | | [work_fetch] shortfall 0.00 nidle 0.00 saturated 26946.30 busy 0.00 10/12/2021 14:59:17 | GPUGRID | [work_fetch] share 0.000 blocked by project preferences 10/12/2021 14:59:17 | | [work_fetch] --- state for NVIDIA GPU --- 10/12/2021 14:59:17 | | [work_fetch] shortfall 15302.30 nidle 0.00 saturated 10522.10 busy 0.00 10/12/2021 14:59:17 | GPUGRID | [work_fetch] share 0.000 job cache full 10/12/2021 14:59:17 | | [work_fetch] --- state for Intel GPU --- 10/12/2021 14:59:17 | | [work_fetch] shortfall 0.00 nidle 0.00 saturated 29576.94 busy 0.00 10/12/2021 14:59:17 | GPUGRID | [work_fetch] share 0.000 project is backed off (resource backoff: 116728.86, inc 86400.00) 10/12/2021 14:59:17 | | [work_fetch] ------- end work fetch state ------- 10/12/2021 14:59:17 | GPUGRID | piggyback: resource CPU 10/12/2021 14:59:17 | GPUGRID | piggyback: can't fetch CPU: blocked by project preferences 10/12/2021 14:59:17 | GPUGRID | piggyback: resource NVIDIA GPU 10/12/2021 14:59:17 | GPUGRID | piggyback: can't fetch NVIDIA GPU: job cache full 10/12/2021 14:59:17 | GPUGRID | piggyback: resource Intel GPU 10/12/2021 14:59:17 | GPUGRID | piggyback: don't need Intel GPU 10/12/2021 14:59:17 | GPUGRID | [rr_sim] piggyback: don't need work 10/12/2021 14:59:17 | GPUGRID | [sched_op] Starting scheduler request 10/12/2021 14:59:17 | GPUGRID | [work_fetch] request: CPU (0.00 sec, 0.00 inst) NVIDIA GPU (0.00 sec, 0.00 inst) Intel GPU (0.00 sec, 0.00 inst) 10/12/2021 14:59:17 | GPUGRID | Sending scheduler request: Requested by project. 10/12/2021 14:59:17 | GPUGRID | Not requesting tasks: don't need (CPU: ; NVIDIA GPU: ; Intel GPU: job cache full) 10/12/2021 14:59:17 | GPUGRID | [sched_op] CPU work request: 0.00 seconds; 0.00 devices 10/12/2021 14:59:17 | GPUGRID | [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 10/12/2021 14:59:17 | GPUGRID | [sched_op] Intel GPU work request: 0.00 seconds; 0.00 devices 10/12/2021 14:59:18 | GPUGRID | Scheduler request completedPreserving so we can see what's different if we allow Einstein to fetch as well. |
|
430)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106375)
Posted 10 Dec 2021 by Richard Haselgrove Post: Thanks for the reminder. We had a discussion about that in the projects area in October (that might be what I was remembering, though it doesn't quite match). In that discussion, I suggested I might give it a try, but I've looked - I confirm that option is not active on the machine in question. Also, in the event log extracts I showed you yesterday, there was a work request to GPUGrid an hour earlier, and NVidia work was not requested - NVidia work was only requested when I was trying to fill the Einstein cache. Here are the events on either side of that earlier work request: 09-Dec-2021 13:25:24 [World Community Grid] Sending scheduler request: To fetch work. 09-Dec-2021 13:25:24 [World Community Grid] Requesting new tasks for NVIDIA GPU and Intel GPU 09-Dec-2021 13:25:24 [World Community Grid] [sched_op] NVIDIA GPU work request: 21042.74 seconds; 2.00 devices 09-Dec-2021 13:25:24 [World Community Grid] [sched_op] Intel GPU work request: 25920.00 seconds; 1.00 devices 09-Dec-2021 13:26:30 [GPUGRID] Sending scheduler request: Requested by project. 09-Dec-2021 13:26:30 [GPUGRID] Requesting new tasks for Intel GPU 09-Dec-2021 13:26:30 [GPUGRID] [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 09-Dec-2021 13:26:30 [GPUGRID] [sched_op] Intel GPU work request: 25920.00 seconds; 1.00 devices 09-Dec-2021 13:26:31 [GPUGRID] Scheduler request completed: got 0 new tasks 09-Dec-2021 13:30:21 [World Community Grid] Sending scheduler request: To fetch work. 09-Dec-2021 13:30:21 [World Community Grid] Requesting new tasks for NVIDIA GPU and Intel GPU 09-Dec-2021 13:30:21 [World Community Grid] [sched_op] NVIDIA GPU work request: 21420.77 seconds; 2.00 devices 09-Dec-2021 13:30:21 [World Community Grid] [sched_op] Intel GPU work request: 25920.00 seconds; 1.00 devicesSo the overall cache was definitely low, but the running GPU task and the exclusion of the second GPU meant that it wasn't appropriate for the client to request any from GPUGrid - as intended. |
|
431)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106366)
Posted 9 Dec 2021 by Richard Haselgrove Post: ... work_fetch.cpp tied into rr_simulation is such a kludge now ...Yup, but logic is still logic, even if it doesn't do what David thinks it does. I have a memory gnawing away at the back of my mind. Sometime fairly recently - say last six months, maybe even more recent - I think I saw an issue or PR on GitHub to the effect of 'always request work when contacting a project' - thus over-riding the work fetch priority values, or so it seemed. Ring any bells with anyone here? The Github search tools aren't good enough to find it, and I don't remember the exact wording. It may even have been somewhere else, like these boards, and a request rather than an actual change. I'll keep poking, but any assistance would be welcome. |
|
432)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106362)
Posted 9 Dec 2021 by Richard Haselgrove Post: I don't think much of that will come into play. That machine has a TSI (Task Switch Interval - 'Switch between tasks every ...') of 3,000 minutes - over 2 days. Tasks could still be switched if any of them were reaching a deadline, but my shortest deadline is 1.5 days for WCG resends, and my cache request is 0.25 + 0.05 days - about 7 hours. Nothing should hit any of those triggers in normal running. My biggest risk is fractional GPU running. As the screenshot shows, Einstein is set to use 0.5 GPUs, and so is WCG. GPUGrid is allowed to wallow in a whole GPU to itself, so won't start automatically when there's only half a GPU free. That requires a little gentle nudging (one GPUGrid task will follow another, if only the project would keep up a regular supply). My big worry is simply the work fetch algorithm. Something has unleashed work fetch for GPUGrid when it shouldn't have, and I didn't have enough active in the Event Log to show what it was. I'll turn on some extra flags which I reach the equivalent stage tomorrow, and try again. |
|
433)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106360)
Posted 9 Dec 2021 by Richard Haselgrove Post: Is this a new problem in #4592? I've just had a rogue fetch from GPUGrid - an NVidia GPU fetch, different from the CPU fetches that #4592 was designed to address. I have two NVidia GPUs in the machine running the #4592 artifact - a GTX 1660, and a GTX 1650. I am currently running three projects which can use NVidia GPUs: WCG/Covid, Einstein, and GPUGrid. WCG GPU tasks are short - three or four minutes. They can fit in anywhere, but are rarely available. GPUGrid tasks are looong - around 20 hours on the 1660, much longer on the 1650. And they would like their science back quickly, please - ideally within 24 hours. So the 1650 is no use, and I have 09/12/2021 14:44:10 | GPUGRID | Config: excluded GPU. Type: NVIDIA. App: all. Device: 1Einstein almost always has work, and it's of intermediate size - 15 to 25 minutes, depending on the card. So, my recent work plan has been: Fetch 6 hours of work from Einstein. Suspend the last few, so the machine is constantly ready to download more, but isn't allowed to fetch from Einstein. Script an 'update' from WCG every few minutes, so it has a chance of catching new work whenever available. Allow GPUGrid work to download when needed. Repeat three times per day. That has been working fine. I got a new GPUGrid task this morning, and it's running - about 3 hours into its 20-hour stretch. And I can see this in the Event Log: 09/12/2021 13:26:30 | GPUGRID | Sending scheduler request: Requested by project. 09/12/2021 13:26:30 | GPUGRID | Requesting new tasks for Intel GPU 09/12/2021 13:26:30 | GPUGRID | [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 09/12/2021 13:26:30 | GPUGRID | [sched_op] Intel GPU work request: 25920.00 seconds; 1.00 devices 09/12/2021 12:26:26 | GPUGRID | Scheduler request completed: got 0 new tasks(GPUGrid doesn't support Intel GPU, so that's as it should be) But it was time to refill the Einstein pot, and in the middle of that, I got 09/12/2021 14:16:19 | GPUGRID | Sending scheduler request: To fetch work. 09/12/2021 14:16:19 | GPUGRID | Requesting new tasks for NVIDIA GPU and Intel GPU 09/12/2021 14:16:19 | GPUGRID | [sched_op] NVIDIA GPU work request: 15428.76 seconds; 0.00 devices 09/12/2021 14:16:19 | GPUGRID | [sched_op] Intel GPU work request: 25920.00 seconds; 1.00 devices 09/12/2021 14:16:20 | GPUGRID | Scheduler request completed: got 1 new tasksThat shouldn't have happened (and has never happened before), because the only device GPUGrid is allowed to run on is going to be busy all night.  Any comparable observations? Should I ask David to investigate (and risk delaying the next release for another decade), or just work round it? |
|
434)
Message boards :
Questions and problems :
Suspended CPU is busy.
(Message 106357)
Posted 8 Dec 2021 by Richard Haselgrove Post: It depends on which version of Windows you're using. Windows 7 shows user tasks by default, but shows a clear button - 'Show processes from all users' - to include system and service uses as well. [bottom left of window] Windows 10 and 11 have a toggle in the same place for 'More details' / 'Fewer details'. |
|
435)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106348)
Posted 7 Dec 2021 by Richard Haselgrove Post: why the following code in cs_statefile.cpp?I don't know, but I have traced the history. That code was introduced in https://github.com/BOINC/boinc/commit/86ccb6eed36aec51d4611869a86ce8a1066eb3c4, which is described as 'fix my last checkin'. Fortunately, I've kept a chronological listing, and I can confirm that David's immediately preceding checkin was https://github.com/BOINC/boinc/commit/f716dcf7ae828b64f1bafe9d016ca6c1aeebca4d, which is where the whole concept of a backup project was introduced. I can only presume that the implementation of 'backup project' went west if every project was backing up each other. You'd have to argue pretty strongly to reverse a supposed fix. |
|
436)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106342)
Posted 7 Dec 2021 by Richard Haselgrove Post: I've just been having the same conversation with another user by email. So this is conveniently on my clipboard: https://drive.google.com/drive/folders/14C1sfF9wDbG1U0fPSwkXx3jq_M1HrxwB?usp=sharing You'll need both a .ZIP handler and a 7-zip handler to unpack boinc.exe - so good they compressed it twice. |
|
437)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106339)
Posted 7 Dec 2021 by Richard Haselgrove Post: I've dug out the log: 03-Dec-2021 19:40:10 [NumberFields@home] [sched_op] CPU work request: 5143.59 seconds; 0.00 devices 03-Dec-2021 19:40:12 [NumberFields@home] [sched_op] estimated total CPU task duration: 8445 seconds 03-Dec-2021 19:40:48 [NumberFields@home] [sched_op] CPU work request: 5450.16 seconds; 0.00 devices 03-Dec-2021 19:40:50 [NumberFields@home] [sched_op] estimated total CPU task duration: 8488 seconds 03-Dec-2021 19:41:24 [NumberFields@home] [sched_op] CPU work request: 5414.62 seconds; 0.00 devices 03-Dec-2021 19:41:26 [NumberFields@home] [sched_op] estimated total CPU task duration: 8488 seconds 03-Dec-2021 19:41:58 [NumberFields@home] [sched_op] CPU work request: 5540.89 seconds; 0.00 devices 03-Dec-2021 19:42:00 [NumberFields@home] [sched_op] estimated total CPU task duration: 8488 secondsIf that's how your 'hundreds of tasks' reached your machine, we know the answer and it's been fixed (though not yet released). If they arrived by some other mechanism, please give us the details. |
|
438)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106337)
Posted 7 Dec 2021 by Richard Haselgrove Post: I deliberately put one machine into the state where it was fetching the same quantum of new work every 30 seconds, and getting it, every time - so it was disregarding the new work when calculating what to fetch next time. Is that how your excess tasks arrive? I downloaded and installed the CI test build of #4592: that cured it. |
|
439)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 106293)
Posted 4 Dec 2021 by Richard Haselgrove Post: Spammers... Sigh.David has (belatedly) implemented email verification for posting here. 1) When you reach the page that tells you to look for a verification email, there's no "go back to what you were doing" link (i.e., the thread where you were about to post or reply) 2) I got two emails, with different tokens, about a minute apart. |
|
440)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106269)
Posted 2 Dec 2021 by Richard Haselgrove Post: No probs. Many of us old-timers probably haven't added a new project for years, and forget how it goes. As soon as you do add a project, the client asks for 1 second of work, just to get things started. You have to move very quickly indeed to preempt that initial fetch. But then, the client is likely to ask again, as soon as the project backoff allows (and I've seen delays as low as 7 seconds at some projects). In the worst case scenario, that second fetch may ask for enough work to fill all available cores for your full cache setting time. |
|
441)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106259)
Posted 1 Dec 2021 by Richard Haselgrove Post: There was, and I think still is, a bug the the BOINC client which caused those repeated requests when <max_concurrent> was used in an app_config.xml file: remove optional extensions like that until you have got to know the normal behaviour of your new machines, and have allowed them to settle down into a stable state.I was right - that bug does still exist in the current code. But tonight, David has published #4592, which may correct it. I'll test in the morning. |
|
442)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106248)
Posted 30 Nov 2021 by Richard Haselgrove Post: Added to which: Set "No New Tasks" at the very first moment you can after the 'attach to project' step is complete - within one or two seconds, if possible. Only allow new work again after ensuring that all project optional settings (resource share, venue, devices to use, sub-projects to run) have been successfully downloaded. |
|
443)
Message boards :
Questions and problems :
WCG: new systems download 100s of CPU work units, not possible to work all
(Message 106246)
Posted 30 Nov 2021 by Richard Haselgrove Post: There were a lot of changes, and bugs, round about v7.16.3 - personally, I wouldn't touch it (again) with a bargepole. v7.16.20 is much, much better - though I can't speak to the specific problem you're experiencing. BOINC projects cannot send work arbitrarily - the internet, and your router, makes that impossible. BOINC can only send work as a reply to a request your client has made. So, your first port of call is the event log, perhaps with <sched_op_debug> added to the basic list. What is your client requesting, how often, and what is it getting in return? Did those 400+ arrive in a single batch, all with the same deadline, or did they arrive in repeated batches, every two minutes, with deadlines offset by about 125 seconds? There was, and I think still is, a bug the the BOINC client which caused those repeated requests when <max_concurrent> was used in an app_config.xml file: remove optional extensions like that until you have got to know the normal behaviour of your new machines, and have allowed them to settle down into a stable state. Setting a resource share of zero implies that you intend the new machines to spend most of their time doing something else. What's that? If it's one or more other BOINC projects, have you set any optional parameters to control those, too? |
|
444)
Message boards :
Projects :
News on Project Outages
(Message 106229)
Posted 29 Nov 2021 by Richard Haselgrove Post: GPUGrid has been repaired, with a new certificate valid until 27 February 2022 (watch out for a repeat performance then). But the server will be very busy with recovery today. |
|
445)
Message boards :
Projects :
News on Project Outages
(Message 106217)
Posted 28 Nov 2021 by Richard Haselgrove Post: GPUGrid have let their server https certificate expire. No uploads or downloads, no scheduler contact, website can't be accessed unless you bypass the warnings. |
|
446)
Message boards :
Projects :
Why No Tasks?
(Message 106178)
Posted 25 Nov 2021 by Richard Haselgrove Post: GPUGrid in fact has sent hundreds of work units today. Unfortunately, the datasets are all corrupted or incomplete and the tasks have all errored out in a few seconds on all hosts.And the stragglers are still being flushed through the system. Don't get too excited if one lands in your inbox. |
|
447)
Message boards :
Questions and problems :
Possible to Set BOINC Such that It Will Not Use Page File?
(Message 106161)
Posted 20 Nov 2021 by Richard Haselgrove Post: This is where you have to explain how you're using BOINC. By design, BOINC expects you to be connected to multiple projects - more than your computer can run concurrently. So, it switches between tasks from different projects, depending on the resource shares you've given each project, and your Task Swap Interval. Any task which has started to run, but been swapped out of 'running' status so another project can take a turn before it finishes, counts as 'suspended', and will display status 'waiting to run' in BOINC Manager. Do you have any of those? |
|
448)
Message boards :
Questions and problems :
Hybrid service and user install
(Message 106159)
Posted 20 Nov 2021 by Richard Haselgrove Post: Either solution should be possible. My primary BOINC instance is user-mode with auto log on at power-up. My command lines for starting and running a second user-mode instance alongside the first, for testing, are: Start client #2: D:\BOINC\rh_boinc_test.exe --allow_multiple_clients --allow_remote_gui_rpc --redirectio --detach_console --gui_rpc_port 31418 --dir D:\BOINCdata2Start Manager #2: start D:\BOINC\boincmgr.exe /m /n 127.0.0.1 /g 31418 /p passwordThey should still work if the primary instance is running in service mode. |
|
449)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106146)
Posted 19 Nov 2021 by Richard Haselgrove Post: I recognise those arguments. Back in the 1990s, I was self-employed, and advising small community organisations on their emerging technology needs. My primary concern was with stand-alone computers and local area networking (no internet at that point), but I got drawn into conversations about telephony as well. This was a time when small electronic switchgear was becoming available and affordable. The regulations were that every installation had to include a defined number of 'power fail' sockets, that would remain operational on the old POTS (exchange-powered) network to call the emergency services in the event of, say, a fire or a mains supply failure. Several times, I encountered systems where the sockets were provided but no POTS handsets were available (or nobody knew where they were). The electronic PABX handsets wouldn't work on a POTS line. Bernie - what's the current status of those regulations? |
|
450)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106141)
Posted 18 Nov 2021 by Richard Haselgrove Post: Richard, At least you are back home, of sound mind (?maybe?), your unexpected health issue(s) were addressed and until BT has a new screw-up things are working.Yup, that's a fair summary. It's fun and games like that that keep us young, and out of the funny farm. |
|
451)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106139)
Posted 18 Nov 2021 by Richard Haselgrove Post: Well, my house is certainly a village legend. Have fun with this photo of my 'downstairs' front door, taken several decades before I bought the house. Think British TV soap operas... 
|
|
452)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106136)
Posted 18 Nov 2021 by Richard Haselgrove Post: I've got two front doors :-) Upstairs is a house - two up, two down, end terrace - typical vernacular architecture for the mid nineteenth century in these parts. Downstairs - the void beneath my neighbour's house - was once a shop, selling fine-art wallpaper and decorator's materials. Not exactly what you'd expect to find in a small Yorkshire village. The narrow passageway linking them, with all the steps and head-bumps - ah, that's the stuff of village legend. But for another time. |
|
453)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106134)
Posted 18 Nov 2021 by Richard Haselgrove Post: Piecing together the fragments after the event, although I heard dialtone on the landline on Friday morning, it wouldn't accept a dialed number ("The number you dialled has not been recognised" - standard message on UK phones). My frequent caller had rung me during Friday morning, and - presumably - heard a standard ring tone from my not-yet-connected VOIP adapter. And I didn't answer... Panic set in. She drove (or was driven) to my house, knocked on the front door, and didn't get an answer - probably because I was down in the workroom wrestling with the wires. [Jord has seen my house, and the distance between the front and back doors. There's a story behind that, too.] And being a caring sort of person, she visualised me lying unconscious - or worse - on the floor. And called the ambulance service. They gave me the standard tests, found enough outside normal range to warrant further tests, and carted me off to hospital: it's their job, and hard to argue with when you're surrounded by three of them, all asking questions at the same time. Hospital kept me in for 'observation' overnight - actually, I was probably watching them more than they were watching me - and then let me find my own way home by taxi. I'll keep taking the tablets, and I've got an appointment to see the doc in a couple of weeks to see if they've worked. And I've also registered the salient points of the tale with British Telecom as a complaint. If VOIP is going to be the national solution for the future, they're going to have to put a lot more work into their implementation of the changeover. |
|
454)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106133)
Posted 18 Nov 2021 by Richard Haselgrove Post: Step 3, and email 3 - received at 02:43 on 12 November. You're good to goNaturally, I didn't leap out of bed at three in the morning to re-wire my house. Lift phone handset, hear normal dial tone - so work can wait. Start morning chores, some shopping, a bite of breakfast: then down into my work-room when I felt awake enough to complete the job properly. Remove old face-plate, connect old wiring to newly modified face-plate, plug patch cable between router and wall assembly, test with calls to/from cellfone. All extensions are working - job done by about 1 pm. About 2 pm, my frequent caller rings - first successful outside call. And shortly after that, the ambulance crew arrived at my door... |
|
455)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106131)
Posted 18 Nov 2021 by Richard Haselgrove Post: (in the middle of all this, I had a minor medical episode, which I mentioned in passing to a regular caller) |
|
456)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106130)
Posted 18 Nov 2021 by Richard Haselgrove Post: Step 2 - a second email, timed at 02:23 on 5 November. Your phone service is changing to Digital VoiceWhat sort of an appointment is the bit I've picked out in red? That could be anything from 00:01 Friday to 23:59 Sunday - and I have to make the final change by hand. Grumble. But, time to get down to work. British telephone sockets have a two-part design:  The lower part is user-accessible. You are permitted to remove it, connect extension telephone sockets via your own wiring and a punchdown connection block. I've got half-a-dozen sockets spread over three floors in my house. The connection to the outside world is made via an integral plug/socket when you replace the faceplate on the rest of the master socket (which you are not allowed to modify). With VOIP, the house wiring has to be isolated from the external line. A couple of minutes with a hacksaw sorted that (check you haven't shorted any conductors together by doing that). |
|
457)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106129)
Posted 18 Nov 2021 by Richard Haselgrove Post: Step 1 - I texted for an adapter. Got a reply to say I'd already requested a handset (??? - false), and couldn't have both. Rang them - even the Digital Voice team couldn't see any sign of an order. Confirmed I wanted an adapter, sorted out the delivery details, job done. On arrival, it turned out to be an electric plug-in wall wart, with an IP WPS 'pair' button, and a phone socket. Not what I wanted, but it's a start. Plugged in, pressed the two WPS pairing buttons in the right order and for the right length of time, but indicator did not change to solid green for successful pairing. Stayed flashing, suggesting not paired. Checked router management screen - VOIP not configured. Shrugged shoulders, unplugged device, gave up for now. |
|
458)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106128)
Posted 18 Nov 2021 by Richard Haselgrove Post: OK, let's start the hare running. I'm doing some Beta testing at the same time, so it may come in stages. It starts with an email on 19 October: Your home phone is going digital. |
|
459)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 106125)
Posted 18 Nov 2021 by Richard Haselgrove Post: Anyone interested in the story of how British Telecom's unilateral conversion of landline telephones to VOIP landed me in hospital for 24 hours? (and no, although a punchdown tool does feature in the story, that wasn't the cause of the hospital visit). |
|
460)
Message boards :
Questions and problems :
Can not get Rosetta Python (Vbox) tasks
(Message 106022)
Posted 8 Nov 2021 by Richard Haselgrove Post: I don't think fah is a boinc app so it does not know about its resource needs correct?Correct. FAH and BOINC studiously ignore each other's existence. If both are running on the same machine, each will assume that it has exclusive use of the resources found. Everything will be double-counted, and, roughly speaking, everything will drop back to 50% efficiency or worse. Each platform would need to be manually restrained to work within whatever proportion of the machine resources you are willing to assign it. |
|
461)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 106001)
Posted 6 Nov 2021 by Richard Haselgrove Post: I still think it would be helpful to identify exactly which of the many possible 'transient upload errors' is being encountered in your use case. You referred to my #3778, but that was concerned with the problems which occur with the boinc backoffs when files fail to upload: The cause of that at GPUGrid is well known - a DDoS block on the connection when multiple computers try to connect to the server from the same IP address, in too short a time interval. Your issue #4572 gives an example of a simple 'transient HTTP error', but the extended http_debug log shows an error-free 'Finished upload of ...' log, which takes us no further forward. I also run WCG, and I've seen examples of an error "no server is available to handle your request" - which I don't think I've seen at any other BOINC project. You also mention that you think that the errors are connected with the uploading of 'large' files - can you quantify that? I've recently handled problems caused by attempting to upload files larger than 150 megabytes at CPDN, and 500 megabytes at GPUGrid. Are your files in that sort of size range? Also, in my experience, once a connection has been established and data has started to flow, in general uploads complete: the exception being projects which transfer uploaded files to backing storage, which sometimes accept the whole file and then generate an http gateway error because the backing store is full or offline. Does your error occur at the start of the transfer (in which case the size probably isn't implicated), or does it occur part-way through, or at the end? |
|
462)
Message boards :
Questions and problems :
SLI computing
(Message 105980)
Posted 5 Nov 2021 by Richard Haselgrove Post: ... does Rosetta@Home just only runs that way?Looking at their applications page, I only see CPU apps. A regular Rosetta volunteer could possibly advise? |
|
463)
Message boards :
Questions and problems :
SLI computing
(Message 105978)
Posted 5 Nov 2021 by Richard Haselgrove Post: y=2 if SLI is active or y=1 if SLI is INactive?I haven't been following all the most recent developments in GPU design and project application development, but I've never seen - or read discussion about - multi-GPU applications running under BOINC. BOINC itself is not SLI aware. It will always see and report an SLI pair as two separate devices, and assign a task to run on one or the other. I'll look at the code, but I don't think it's possible (yet) to schedule a task to wait until two GPUs are both available, and then launch a single task to occupy both. In my (only brief) experience of running an SLI pair, my observation was that SLI operation locked both GPUs to the same clock rate. This rather negated the benefit of modern developments like 'Graphics boost clock'. |
|
464)
Message boards :
Questions and problems :
Please Help Me if You Can
(Message 105962)
Posted 3 Nov 2021 by Richard Haselgrove Post: The exported stats are in a compressed XML format. Your favorite database probably has an XML import tool. Make sure you have plenty of disk space available. |
|
465)
Message boards :
BOINC Manager :
how do you set bonic at start up - win11
(Message 105955)
Posted 3 Nov 2021 by Richard Haselgrove Post: It's usually started from the registry: [HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Run] "boincmgr"="\"D:\\BOINC\\boincmgr.exe\" /a /s"That's created by the Windows installer - you don't have to do anything by hand. Having said that, that example was taken from Windows 7, and probably hasn't been changed since Windows XP. It seems to work under Windows 10: Windows 11 hasn't been offered to my test machine yet - I'll watch out for problems when it arrives. |
|
466)
Message boards :
Questions and problems :
Run Boinc as a service on windows 10?
(Message 105926)
Posted 1 Nov 2021 by Richard Haselgrove Post: I wonder if that is a problem with the opencl or cuda implemenations or at the driver level (ATI, NVidia) or at the Microsoft driver handler.It's a feature, rather than a bug, in the Microsoft driver handler. See WDDM Architecture: notice the dividing line between 'User Mode' and 'Kernel Mode'. Kernel mode applications, like services, can't access the bits of the driver above that line. |
|
467)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105924)
Posted 1 Nov 2021 by Richard Haselgrove Post: Completely irrelevant to my problem.Agreed. But don't shoot the messenger. I just wanted to be sure that you knew where the pushback was going to come from, so that you could respond pre-emptively in your proposed issue. |
|
468)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105918)
Posted 1 Nov 2021 by Richard Haselgrove Post: Before you do - 'too many uploads' is a client decision. Irrespective of how many GPUs you may have, the limit will kick in when the number of uploads on the system - whether ready to upload, or backed off following some previous problem - goes above twice the number of CPUs visible to BOINC in the system. It's designed to protect project servers. If there's a problem blocking uploads - we've all seen 'server storage full', and even complete server breakdowns - there's no point in downloading more and more work, which might (for all BOINC knows) never be uploaded or reported. That would be a waste of energy to compute. |
|
469)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105898)
Posted 29 Oct 2021 by Richard Haselgrove Post: I run smaller machines (max 2 GPUs), so I haven't been tempted to change the defaults. Anyone? |
|
470)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105896)
Posted 29 Oct 2021 by Richard Haselgrove Post: Well, that's got me completely confused. :-( But if they're uploading, that's the main thing. I think it's getting a bit late to take this further tonight - I'll try and read up on multiplexing over the weekend. I think WCG uses multiple servers to share the load - maybe the configuration isn't quite identical across the cluster? |
|
471)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105893)
Posted 29 Oct 2021 by Richard Haselgrove Post: Should fetch on update be set???I don't have it set, but it doesn't sound like it would matter. The explataions, such as they are, are here. Actually, <fetch_on_update>0|1</fetch_on_update>sounds good - I might try that myself. |
|
472)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105891)
Posted 29 Oct 2021 by Richard Haselgrove Post: Tried one on a Linux (Mint v20.2) machine. It starts... 29/10/2021 19:24:35 | World Community Grid | Started upload of OPNG_0098436_00093_1_r329524617_0- I won't bore you with the rest. Seems like the multiplexing might be your problem, but I don't know Linux well enough to go much further. |
|
473)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105889)
Posted 29 Oct 2021 by Richard Haselgrove Post: Knock off the debug flag now - they'll all be the same. I'll look through this one. [edit] - ouch - how many files do you have uploading at once? Each one only just seems to get started, but then attention switches to another one. For this sort of job, it's easier to wait until every file is in backoff, then set the http_debug flag, retry just one file, and knock the flag off again. Then you get something like 29/10/2021 18:58:24 | World Community Grid | Computation for task OPNG_0098417_00141_0 finishedYours seems to fail at the one I've marked, but without seeing one the whole way through - from 'started upload' to 'finished upload' - it's hard to be sure. I'll go and look at a Linux machine, and see how it compares. |
|
474)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105887)
Posted 29 Oct 2021 by Richard Haselgrove Post: Set 'http_debug' in the Event Log options, and find out what that 'transient HTTP error' is. |
|
475)
Message boards :
Questions and problems :
Run Boinc as a service on windows 10?
(Message 105883)
Posted 29 Oct 2021 by Richard Haselgrove Post: No reason why not. The restriction (no GPU in service mode) came about because of a Microsoft security restriction in the GPU driver model: I doubt they'll remove that now, even in Windows 11. The big win is that service mode allows BOINC to do science when no user is logged into the console - ideal for servers. |
|
476)
Message boards :
Questions and problems :
Invalid client RPC Password
(Message 105878)
Posted 28 Oct 2021 by Richard Haselgrove Post: I think it's the group membership which is critical. Loads of 'HowTo' on the web, like https://linuxize.com/post/how-to-add-user-to-group-in-linux/ The command is sudo usermod -a -G groupname usernamewhere: groupname is boinc username is your Linux account name. Do the usermod in terminal, and reboot the machine afterwards. Makes it easier to use the BOINC Manager. |
|
477)
Message boards :
Projects :
News posting from GPUGRID Admin
(Message 105858)
Posted 26 Oct 2021 by Richard Haselgrove Post: Another news posting from GPUGrid We have two openings for a PhD and a postdoc. If you know anybody that could be interested, share it please.https://www.gpugrid.net/forum_thread.php?id=5283 |
|
478)
Message boards :
Projects :
WCG OPNG sans OPN1
(Message 105857)
Posted 26 Oct 2021 by Richard Haselgrove Post: Somebody may be mis-remembering the problem we had (and may still have) getting work for intel GPUs without also specifying something else. I'm getting work for both intel GPUs and NVidia GPUs, sometimes in the same fetch, but without any work for the CPU. |
|
479)
Message boards :
News :
Windows communication failure: temporary workaround
(Message 105848)
Posted 24 Oct 2021 by Richard Haselgrove Post: In general, Linux versions of BOINC use the operating system's own versions of the SSL certificates. So, if they fail, the solution is to update Linux, rather than BOINC. But there are so many different flavours of Linux, and so many different implementations of BOINC, that it's hard to give general advice. |
|
480)
Message boards :
Questions and problems :
Want info about the latest client: VS2019? 7.18?
(Message 105847)
Posted 24 Oct 2021 by Richard Haselgrove Post: On an unrelated question, if I found a boinc binary that was 7.17 what build was it taken from?You have to get to grips with the way with the way BOINC uses GitHub and version numbering. In theory: The BOINC master branch - the code you see when you first visit GitHub, or when you update a cloned copy, has the most recent accepted code contributions. It should compile and run, but that's not guaranteed: it's treated as 'untrusted', and given an odd version number - 7.17, 7.19, etc. BOINC creates release branches when it feels that the code has reached a stable, trusted, point. The release branch is given an even release number - 7.16, 7.18 - and in principle is ready for public use. Bugs may be found during testing, and as fixed the minor version number is incremented: 7.16.11, 7.16.16, 7.16.20. There's no evolution of the minor version number during development: it stays at 7.17.0, 7.19.0 or whatever. So it you see a v7.17 binary, it was built from the master branch, but you have no idea exactly what code was used to build it. In practice: BOINC v7.16 has been been a dog's breakfast ever since the release branch was first created. There were multiple bugs in the early code, and they have been 'fixed', not by precision fixing, but by mass importations from the master branch (at least three times, by my count). Each one has copied in a whole new set of bugs. We hope that a new stable v7.18.? release will be made within the next month, to accommodate Windows 11 and the next certificate expiry. It will be built by VS2019, but that's about all we know. That's my personal interpretation of: https://github.com/BOINC/boinc-policy/blob/master/Development_Documents/Client_Release_Process.md https://boinc.berkeley.edu/trac/wiki/AdminReleaseManagement Those documents may disagree, and other interpretations my be available. Best of luck. |
|
481)
Message boards :
Questions and problems :
Want info about the latest client: VS2019? 7.18?
(Message 105842)
Posted 24 Oct 2021 by Richard Haselgrove Post: VS2019 builds OK on my Windows 10 test machine: it downloads and builds the dependency sources as required. Note - that makes it very slow for a first-time build. Make sure it's got a stable internet connection. It gets those sources via a git transfer, so make sure you've got a Git client and a 7-zip decompresser installed before you start. But: v7.16.20 wasn't - and can't be - built using VS2019. I tried - my uploaded error messages led to https://github.com/BOINC/boinc/issues/4544#issuecomment-935787091 |
|
482)
Message boards :
BOINC Manager :
Clear list of hostnames
(Message 105828)
Posted 22 Oct 2021 by Richard Haselgrove Post: Now you're asking ... There's a hidden file .BOINC Managerin your home folder. That contains the section [ComputerMRU]about 20 lines in. |
|
483)
Message boards :
BOINC Manager :
Clear list of hostnames
(Message 105808)
Posted 19 Oct 2021 by Richard Haselgrove Post: Assuming you're asking about BOINC running on Windows, the computer names are stored in the Registry, in a MRU (Most Recently Used) list. The list is held at [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager\ComputerMRU]and can be managed using the Windows utility regedit. Be very careful not to make any unintended changes to other registry keys or values while you're there - indiscriminate editing of the registry can be dangerous. |
|
484)
Message boards :
News :
Windows client 7.16.20 released
(Message 105804)
Posted 19 Oct 2021 by Richard Haselgrove Post: I got this email overnight, with the headline Science United: please install new BOINC version Dear RDH_SU_test:Wouldn't it be a good idea also to let project administrators know that this update is needed, so that they can publicise it on their own websites (where needed)? |
|
485)
Message boards :
Questions and problems :
New boinc manager and AVG
(Message 105791)
Posted 18 Oct 2021 by Richard Haselgrove Post: It's a very common observation after the release of a new computer program. Modern anti-virus programs do far more than scan the files stored on your hard disk for known threats. They commonly also include behavioral analysis, which the boinc client (in particular) can trigger: the boinc client doesn't have a visible screen display, uses your computer at high power, and communicates extensively over the internet. All very suspicious! The other thing that AV programs can do is check their company's database of known programs and their reputation. New programs don't yet have a reputation, and haven't been though a full evaluation in the AV company's labs. So they can panic: the fuss usually dies down in a few days. It's slightly unusual to see such a report on the screensaver module within boinc: that will have changed little, if at all, in the new release, and it doesn't exhibit the same suspicious behaviour as the client. But the general advice with an unexpected warning like this is to check the offending file(s) with a service like virustotal, which can compare the reports on an uploaded file from a range of independent anti-virus products. |
|
486)
Message boards :
Questions and problems :
Collatz Conjecture issue for many users
(Message 105782)
Posted 17 Oct 2021 by Richard Haselgrove Post: There is now an officially prepared v7.16.20 of BOINC, for Windows only. It's still undergoing testing, but it's been confirmed that it solves the certificate problem, and no other problems have been reported so far. Using that would be safer than taking one from a third-party source. Download from: https://boinc.berkeley.edu/dl/boinc_7.16.20_windows_x86_64.exe https://boinc.berkeley.edu/dl/boinc_7.16.20_windows_x86_64_vbox.exe |
|
487)
Message boards :
The Lounge :
Let's talk politics
(Message 105757)
Posted 14 Oct 2021 by Richard Haselgrove Post: I long ago stopped paying attention to "political statistics" - the statistics politicians speak. But I do pay more attention - and award more credibility to - "statistical statistics" - the statistics that professional record-keepers draw from their records. There was a good example of the latter in last Saturday's Guardian: Could the global Covid death toll be millions higher than thought? The article stresses that the figures suggested were drawn from "vital registration systems" - the underlying government records of, inter alia, births and deaths. The researchers put 'the real number [of deaths] closer to 16 million', so far. |
|
488)
Message boards :
The Lounge :
Let's talk politics
(Message 105754)
Posted 14 Oct 2021 by Richard Haselgrove Post: Yes, I'm fully aware that covid-19 transmission is an analogue, rather than a digital, phenomenon - especially concerning the severity of disease following infection. I'm just concerned about the level of misinformation and false statistics being peddled about vaccination. For example, the difference between the odds of death following vaccination (which is 100%, though it could take 100 years or more), compared with the odds of death caused by vaccination (which is tiny, but non-zero). I don't see any reference to fines for non vaccination, in the population at large. I see a statement that a fine may follow if vaccine refusal is combined with continuing employment in a public-facing job. That's a choice: nobody is forcing you into a particular work setting - just saying that some combinations are not viable. |
|
489)
Message boards :
The Lounge :
Let's talk politics
(Message 105750)
Posted 14 Oct 2021 by Richard Haselgrove Post: I really find it difficult to understand the "don't want to" refusal to accept vaccination. My starting point is: 1) It's an infectious disease. It spreads from person to person. 2) To a first approximation, all people are different. They have a mother and a father, and inherit from both. 3) Also to a first approximation, all Covid-19 viruses are the same. They replicate, rather than breed. So, if you pass a virus to another person, the outcome depends on the other person. Whether or not you are suffering from symptoms or a disease outcome from the virus is completely irrelevant. The outcome in the other person depends on their response to the virus, not on your belief that the virus you pass on in symptom-free. So, why would you be prepared to work in a public-facing organisation, and be prepared to infect your customers / service users / whatever? To me, it doesn't compute. |
|
490)
Message boards :
Questions and problems :
64bit BOINC using WINE
(Message 105737)
Posted 13 Oct 2021 by Richard Haselgrove Post: We had a problem a while back (maybe a couple of years) where the client crashed when trying to use OpenSSL in 64-bit mode on certain processors. I think Wine may have triggered the same bug. The solution was to ship the v1.0.1s versions of the SSL libraries libeay32.dll and ssleay32.dll. The ones on this machine are dated Christmas Day 2019 - what was that about? It might jog some memories. Edit - found it. "Confirmed v7.16.4 x64 installs and runs correctly on the Celeron." [J3455]. My final note on https://github.com/BOINC/boinc/issues/2470. |
|
491)
Message boards :
News :
Windows communication failure: temporary workaround
(Message 105710)
Posted 8 Oct 2021 by Richard Haselgrove Post: I think the bundle in the old version would have expired at the end of September too. I think it's just coincidence that you updated at the same time (near enough) as the certificate problem first struck. |
|
492)
Message boards :
Questions and problems :
Newest post first.
(Message 105709)
Posted 8 Oct 2021 by Richard Haselgrove Post: I think it now applies to all threads in the news area. Makes some sort of sense to keep the actual 'news' at the top, but 1) People may miss resolutions and updates posted later in the thread. 2) The 'Sort order' combo is still displayed, and now lies to you. #4549 |
|
493)
Message boards :
The Lounge :
Let's talk politics
(Message 105698)
Posted 7 Oct 2021 by Richard Haselgrove Post: Now I can't get the image of Boris Johnson chasing Carrie round the cabinet table, calling "boinc, boinc, boinc", out of my mind. |
|
494)
Message boards :
The Lounge :
Let's talk politics
(Message 105695)
Posted 7 Oct 2021 by Richard Haselgrove Post: Build BOINC Better 😁LOL |
|
495)
Message boards :
Questions and problems :
Authentication error when suspending/resuming with boinccmd
(Message 105691)
Posted 7 Oct 2021 by Richard Haselgrove Post: It's explained in the link you gave in your first post: https://boinc.berkeley.edu/wiki/Boinccmd_tool or should be. The basic principle is that 'for information' commands (the state queries group) don't need a password, but 'change something' commands (the control operations group) do. |
|
496)
Message boards :
The Lounge :
Let's talk politics
(Message 105685)
Posted 7 Oct 2021 by Richard Haselgrove Post: My fixed-price contract ends this month, and the replacement fixes are eye-watering. Thanks goodness the Tories stole Ed Miliband's idea of a variable price cap (although they rubbished the idea when he first proposed it). I'll risk my luck with that for the next six months. |
|
497)
Message boards :
Questions and problems :
Authentication error when suspending/resuming with boinccmd
(Message 105682)
Posted 7 Oct 2021 by Richard Haselgrove Post: Depends how you (or Debian) have set up your connection between BOINC Manager and the boinc client. It may have a password set. If there is a password, it will be stored in a file called gui_rpc_auth.cfg in the BOINC data directory. The same password will need to be used on the command line with boinccmd - the 'authentication error' happens when a password is expected, but not given. |
|
498)
Message boards :
News :
Windows communication failure: temporary workaround
(Message 105674)
Posted 6 Oct 2021 by Richard Haselgrove Post: Do you have an exact idea or example how it could be made simple and without all these problems?For many years, I maintained the installer which stopped BOINC (if running), installed optimised applications for SETI@Home in the correct place, stitched together an app_info.xml file to service the optional choices made, and restarted BOINC if needed. Replacing a single file - not needing a stop-start cycle - would be a doddle by comparison. But I'm not going to make one because 1) This problem will be fixed before I could complete and test it. 2) It wouldn't work when the support files are locked up inside the main file. 3) We're going to move to an updated version of the support software. 4) I tested a new version that doesn't even need support software or a certificate bundle this morning. 5) I'm too old |
|
499)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105667)
Posted 5 Oct 2021 by Richard Haselgrove Post: Yes, that's the way I tested it was working. But since that's a remote machine I don't usually spend time watching, it isn't important to me personally. The powers-that-be at UCB will not want to be forced into a kludge like that - it will offend their sense that they know how to manage this stuff. But since they self-evidently can't, it may be the least-worst temporary solution. |
|
500)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105665)
Posted 5 Oct 2021 by Richard Haselgrove Post: I've started to prep up a spare Windows 10 partition that I may be able to use for VS2019. The app I built on VS2013 yesterday is still running fine, so there can't be too much wrong with the code: hopefully just the solution file 'boinc_vs2019.sln'. But I want to check Oliver's post about the order of certificates in the bundle not being a solution, first.Ugh - things have changed a bit since I first started programming for Windows 3.1! But I got Windows 10 up-to-date, installed VS 2019, installed some more software it wouldn't work without (Git, TortoiseGit, Notepad++ - I'd use those anyway), downloaded the v7.16.20 branch source, and gave it a try. boinc.exe compiled OK, but with lots of warning messages. It's running now, under Windows 7 - so that's a plus. boinccmd.exe compiled OK, but I haven't tested it. The boinccmd solution compiles those two, but not much else, so it's a useful test. boincmgr.exe failed completely. That's a show-stopper, and may be the reason we saw no progress except the emergency workround over the weekend. I'll stop there for tonight. I've turned down my cache size, so it will be quicker to start experimental work tomorrow, if it's still needed. |
|
501)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105662)
Posted 5 Oct 2021 by Richard Haselgrove Post: I've started to prep up a spare Windows 10 partition that I may be able to use for VS2019. The app I built on VS2013 yesterday is still running fine, so there can't be too much wrong with the code: hopefully just the solution file 'boinc_vs2019.sln'. But I want to check Oliver's post about the order of certificates in the bundle not being a solution, first. |
|
502)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105658)
Posted 5 Oct 2021 by Richard Haselgrove Post: Well, look what the News thread brought in. Edit - I've had a PM overnight from a user trying to build BOINC for the first time under VS2019. He says there are some error messages, perhaps relating to the changes for the ARM build, but doesn't go into details - he's going to try again later. Does anyone here already have a known, working, VS2019 setup? If so, could they give it a try,.please - that might explain why there's no v7.16.20 build from Berkeley yet. |
|
503)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105635)
Posted 4 Oct 2021 by Richard Haselgrove Post: V 7.16.7 should be a sufficiently modern version I believe.If you're as close as that, why not go the whole hog and update? |
|
504)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105634)
Posted 4 Oct 2021 by Richard Haselgrove Post: Ripping a single file out of the installer isn't as easy as it sounds - it isn't a standard archive. Much easier (and quicker) to use the download link from curl, and drop it in from there. It's the same file. |
|
505)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105630)
Posted 4 Oct 2021 by Richard Haselgrove Post: I've got another way of checking those certs, but I've lost the simple bit of code I wrote to help with the process. I'll try to re-write it, and assemble a full reference list of the certificates in the new bundle. But it may not be ready until tomorrow. |
|
506)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105628)
Posted 4 Oct 2021 by Richard Haselgrove Post: Well, I've downloaded the new v7.16.20 branch, and compiled it under VS2013: it's running, and it can connect to the sites which were blocked on Thursday. 04/10/2021 11:55:04 | | Version change (7.16.16 -> 7.16.20) But it's still using the old version of OpenSSL, so we're storing up a third round of this type of error for the future. 04/10/2021 11:55:02 | | Libraries: libcurl/7.42.1 OpenSSL/1.0.2s zlib/1.2.8 Better than nothing, but I'll report back to the developers. |
|
507)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105626)
Posted 4 Oct 2021 by Richard Haselgrove Post: Signs of life - we have a v7.16.20 release branch since 4 hours ago. That should be the beginnings of an emergency Windows release, but the final assembly will have to be done in Berkeley. I doubt it'll be available for download in Europe before tomorrow. @ drghughes - look back through this thread for certificate download links. There are two: my amateur hack, and an official download site for a reliable version. Either will do - you don't have to edit the file yourself, just drop the replacement (renamed in the latter case) into your BOINC program location. |
|
508)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105617)
Posted 3 Oct 2021 by Richard Haselgrove Post: That's what we've been saying from the beginning - and we said it in May last year, too, when the same thing happened for the same reason. BOINC on Windows relies entirely on private files, which are distributed by the Windows BOINC installer. It doesn't leverage the resources of any other software installed on the same instance of Windows. This has two consequences at the moment: 1) certificates stored in BOINC's copy of ca-bundle.crt can expire between releases - as one did last week. 2) BOINC can't communicate securely when there is an expired certificate present, because the old version of the security software being used has a bug in it that prevents the search for a usable certificate proceeding past the roadblock caused by the expired certificate. That software can't be upgraded in the field. To answer the previous question: no, we have no idea when work will even start on a new release, let alone when it will be tested and ready. There has been total silence on this subject from UC Berkeley. The volunteer initiative which seemed promising yesterday morning has petered out with no further progress. |
|
509)
Message boards :
Questions and problems :
"Failed to add project [GPUGrid.Net] Please try again later"
(Message 105610)
Posted 3 Oct 2021 by Richard Haselgrove Post: It's a known problem, since about Thursday afternoon. GPUGrid is one of several projects affected. See the sticky thread HTTP error: Peer certificate cannot be authenticated... at the top of this message board. |
|
510)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105606)
Posted 2 Oct 2021 by Richard Haselgrove Post: Of course. Only mess around with your installation if you have identified that you have the type of problem being described here. |
|
511)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105604)
Posted 2 Oct 2021 by Richard Haselgrove Post: A volunteer developer is working on creating an emergency release (take a bow, Vitalii). The new release will contain a fresh certificate bundle, sourced from a genuine and reliable public source: https://curl.se/docs/caextract.html You will need to rename cacert.pem to ca-bundle.crt, but it works with some of the projects that were having problems on Thursday (I haven't checked them all). If you feel nervous about downloading amateur hacks like mine, feel free to download from there instead. |
|
512)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105598)
Posted 1 Oct 2021 by Richard Haselgrove Post: I have two machines with CA updates waiting, and GPUGrid tasks around 80%. I'll wait till they finish, and see what happens.My Linux machines have been making their hourly checkin to GPUGrid without problems, so the CA updates are irrelevant to this discussion. Linux (Mint) was ahead of the game at the last update. |
|
513)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105588)
Posted 1 Oct 2021 by Richard Haselgrove Post: looks like for my upload issues seen at Universe@home on Linux Ubuntu 20.04.3, simply running the updates seems to have fixed it, one of the updates was the ca certificate. uploads now processing again.That's a good point. So far, we've been treating it as a "Windows only" problem, but if Linux machines haven't been updated for a while, it might affect them too. I have two machines with CA updates waiting, and GPUGrid tasks around 80%. I'll wait till they finish, and see what happens. At least, Linux updates can be done at home, without needing a new client from Berkeley. Unless, that is, you're a corporate Linux user, with a system locked up by central IT admin tighter than a duck's arse. |
|
514)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105587)
Posted 1 Oct 2021 by Richard Haselgrove Post: It went to boinc_admin@googlegroups.com - I'd have thought you could get into that.I have a strongly-worded email in draft, which I intend to post 24 hours after this thread was opened - i.e. in about five minutes from now - unless something else happens first.Too bad it didn't go to any list I seem to follow. Mind sending me a copy? Two replies so far - Vitalii repeating what I just said, and Matt Blumberg saying that projects could sort it out for themselves. Neither answered the main question - "Where's Wally?" I'll send you the thread so far. |
|
515)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105584)
Posted 1 Oct 2021 by Richard Haselgrove Post: The full solution would be a release of an updated client - which is long overdue, anyway. Project server administrators may be able to take alleviating action, but that would be piecemeal, and not every project may have the necessary skills. The BOINC server has no role to play. I have a strongly-worded email in draft, which I intend to post 24 hours after this thread was opened - i.e. in about five minutes from now - unless something else happens first. Sent. |
|
516)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105582)
Posted 1 Oct 2021 by Richard Haselgrove Post: I can load and edit the file, but cannot save it, it says it needs admin priv. I AM the admin.Right-click on your editor, and choose 'Run as administrator'. I'm in the same position as you. I'm an administrator, but not the administrator. Alternative solution: move the file to a non-protected location before attempting to edit it. |
|
517)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105575)
Posted 1 Oct 2021 by Richard Haselgrove Post: Yes, they had an unplanned outage: Greetings,https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,43772 |
|
518)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105573)
Posted 1 Oct 2021 by Richard Haselgrove Post: I see a very high urgency at the development stage is neededSo do we, but unfortunately it's the middle of the night in California. Fixing this one requires tools only available to authorised users - probably employees only - of the University of California in Berkeley. I hope someone with the appropriate contacts can kick the relevant people out of bed when the sun rises. |
|
519)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105570)
Posted 1 Oct 2021 by Richard Haselgrove Post: Done the replacement on my WINE installation but didn't actually confirm it was affected as not currently running.You can do a simple 'project update', from either BOINC Manager or boinccmd, to verify that. The Event Log will show either 'failed' with the expired certificate in the bundle, or 'completed' with it removed. The bad certificate blocks all https communication between client and server, not just uploads. |
|
520)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105564)
Posted 1 Oct 2021 by Richard Haselgrove Post: Do you plan to release a new windows client next week?I no longer have that responsibility - they took it away from me after one trial run with v7.10 in 2018. So far as I can see, the current Release Manager (who is based in California, in the PDT time zone) has not yet responded to the emergence of this problem, on any of the channels I can monitor. |
|
521)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 105555)
Posted 30 Sep 2021 by Richard Haselgrove Post: I'm back. And my pub has WiFi - you didn't mention that. But I actually go to the pub to read some dead trees - it makes a soothing, people-centred, non-technical, end of the day. |
|
522)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 105553)
Posted 30 Sep 2021 by Richard Haselgrove Post: I'm off to the pub - mind the shop till I'm back. |
|
523)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105552)
Posted 30 Sep 2021 by Richard Haselgrove Post: AT YOUR OWN RISK Google drive Make a safe copy of your old one, just in case. |
|
524)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105551)
Posted 30 Sep 2021 by Richard Haselgrove Post: OK, that worked. I removed the one I showed you, between these two scheduler requests: 30/09/2021 18:52:19 | GPUGRID | Scheduler request failed: Peer certificate cannot be authenticated with given CA certificates 30/09/2021 20:38:59 | GPUGRID | Scheduler request completed: got 1 new tasks Unless they've done something to their server, too ;-) |
|
525)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105549)
Posted 30 Sep 2021 by Richard Haselgrove Post: You mean this one?  Searching for 'DST X3' didn't find anything. |
|
526)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105547)
Posted 30 Sep 2021 by Richard Haselgrove Post: Could somebody please post the {full, exact, searchable} name of the certificate that needs to be removed? Once you have that, it's an easy job to make the change. The certificate bundle is just a plain text file - make a copy, work in a safe space - and change the extension to .txt. Even notepad can handle the job. Just make sure you add or remove complete sections. I think I may have modified mine last time - I can't find any of the ones that are being mentioned here. |
|
527)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105544)
Posted 30 Sep 2021 by Richard Haselgrove Post: I think someone made a working bodge last time, by removing the expired X3 certificate (or equivalent). |
|
528)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105539)
Posted 30 Sep 2021 by Richard Haselgrove Post: What OpenSSL version does BOINC for Windows show at the beginning of the log?OpenSSL/1.0.2s |
|
529)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105538)
Posted 30 Sep 2021 by Richard Haselgrove Post: While we wait, it's an easy job to update Windows clients with a new ca-bundle.crt file, when available: you don't even need to stop BOINC (though it might be a good idea to suspend networking while you do it). To be safe: Navigate to the BOINC program folder Rename the old file Drop in the new one. And that's it. If it goes wrong, you can return to the old file for the projects that are still accepting it, while the dust settles. BOINC on Linux and other operating systems uses system security, which is automatically updated when needed. There should be no need to update BOINC separately on those systems. |
|
530)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105536)
Posted 30 Sep 2021 by Richard Haselgrove Post: Writing to the mailing lists as we speak. |
|
531)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105534)
Posted 30 Sep 2021 by Richard Haselgrove Post: Add CPDN to the list. (sorry, Jord - I was busy replying to the CPDN moderators when you posted) |
|
532)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105528)
Posted 30 Sep 2021 by Richard Haselgrove Post: I'll ask. Just had a report, and confirmed, that GPUGrid is affected as well. I seem to remember a previous panic on the last day of some earlier month, as well. Yes, there was an emergency release of v7.16.7 on 31 May 2020, and the final code change was 'Update CA bundle'. |
|
533)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105525)
Posted 30 Sep 2021 by Richard Haselgrove Post: OK, I got this when attempting to attach to iThena. I expected the attach to fail, because they're not accepting new members: but that would be for the server to say, not SSL. 30/09/2021 18:05:06 | | Fetching configuration file from https://root.ithena.net/usr/get_project_config.php 30/09/2021 18:05:06 | | [http] HTTP_OP::init_get(): https://root.ithena.net/usr/get_project_config.php 30/09/2021 18:05:06 | | [http] HTTP_OP::libcurl_exec(): ca-bundle 'D:\BOINC\ca-bundle.crt' 30/09/2021 18:05:06 | | [http] HTTP_OP::libcurl_exec(): ca-bundle set 30/09/2021 18:05:06 | | [http] [ID#2] Info: Connection 2831 seems to be dead! 30/09/2021 18:05:06 | | [http] [ID#2] Info: Closing connection 2831 30/09/2021 18:05:06 | | [http] [ID#2] Info: Connection 2832 seems to be dead! 30/09/2021 18:05:06 | | [http] [ID#2] Info: Closing connection 2832 30/09/2021 18:05:06 | | [http] [ID#2] Info: Trying 85.204.27.80... 30/09/2021 18:05:07 | | [http] [ID#2] Info: Connected to root.ithena.net (85.204.27.80) port 443 (#2833) 30/09/2021 18:05:07 | | [http] [ID#2] Info: ALPN, offering http/1.1 30/09/2021 18:05:07 | | [http] [ID#2] Info: Cipher selection: ALL:!EXPORT:!EXPORT40:!EXPORT56:!aNULL:!LOW:!RC4:@STRENGTH 30/09/2021 18:05:07 | | [http] [ID#2] Info: successfully set certificate verify locations: 30/09/2021 18:05:07 | | [http] [ID#2] Info: CAfile: D:\BOINC\ca-bundle.crt 30/09/2021 18:05:07 | | [http] [ID#2] Info: CApath: none 30/09/2021 18:05:07 | | [http] [ID#2] Info: TLSv1.2 (OUT), TLS header, Certificate Status (22): 30/09/2021 18:05:07 | | [http] [ID#2] Info: TLSv1.2 (OUT), TLS handshake, Client hello (1): 30/09/2021 18:05:07 | | [http] [ID#2] Info: TLSv1.2 (IN), TLS handshake, Server hello (2): 30/09/2021 18:05:07 | | [http] [ID#2] Info: TLSv1.2 (IN), TLS handshake, Certificate (11): 30/09/2021 18:05:07 | | [http] [ID#2] Info: TLSv1.2 (OUT), TLS alert, Server hello (2): 30/09/2021 18:05:07 | | [http] [ID#2] Info: SSL certificate problem: certificate has expired 30/09/2021 18:05:07 | | [http] [ID#2] Info: Closing connection 2833 30/09/2021 18:05:07 | | [http] HTTP error: Peer certificate cannot be authenticated with given CA certificatesNote "CAfile: D:\BOINC\ca-bundle.crt" (yours is in C:). My BOINC client is also v7.16.11, and the ca-bundle.crt file is dated 31 May 2020. Checking that ca-bundle.crt is still serviceable is on the Release Manager's checklist for new versions (I put it there), and there's supposed to be a new version to coincide with the release of Windows 11 on October 5. I'd suggest you wait for that release, see whether that cures the problem, and be ready to raise merry hell if it doesn't. |
|
534)
Message boards :
The Lounge :
Let's talk politics
(Message 105523)
Posted 30 Sep 2021 by Richard Haselgrove Post: Wasn't the Johnson going to get a new cabinet from Ikea?He's probably still looking for the Allen key. |
|
535)
Message boards :
Questions and problems :
HTTP error: Peer certificate cannot be authenticated with given CA certificates (with workaround)
(Message 105522)
Posted 30 Sep 2021 by Richard Haselgrove Post: I'd suspect it's a BOINC problem. Windows versions of BOINC rely on a 'ca-bundle.crt' file stored in your BOINC program folder, and only updated when a new version is installed. The most recent update to the sources held by BOINC was on May 31, 2020, which should have been included on v7.16.11 (released 2020-09-02) - I'll check. |
|
536)
Message boards :
The Lounge :
Let's talk politics
(Message 105519)
Posted 30 Sep 2021 by Richard Haselgrove Post: And so the outgoing cabinet - which fired itself over the child allowance scandal - has just decided after 6 months of talking who was going to be in the new cabinet, that they are the best choice for the new cabinet. Welcome to the banana republic Netherlands.Twinned with banana republic Britain |
|
537)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 105503)
Posted 28 Sep 2021 by Richard Haselgrove Post: No, it was a CoolerMaster G650M. No sign of external damage, but it must have short-circuited to draw more then 5A at 230V under test conditions. The spare I had was 800W, so it'll do for now. The main problem was that it was placed at the bottom (floor level) of a tower case, and the CPU power connector is at the very top. They'd run the cable under the motherboard fixing plate, through a tiny gap at the top and with a sharp 180° turn over the socket. Covering the release latches, and with no spare loose cable to let me move it out of the way. In the end, I cut it. |
|
538)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 105498)
Posted 27 Sep 2021 by Richard Haselgrove Post: Well, here I am again, this time from the comfort of my usual home pages and task display. Phew. It was the PSU, as the spare from downstairs confirms. But who the **** designed the modern computer? Especially those extra CPU power cables right next to the bacon-slicer of a finned CPU cooler. Had to de-mount a couple of fans, and shift the whole motherboard sideways, to get a proper grip on those. More beer, Jeeves. |
|
539)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 105496)
Posted 27 Sep 2021 by Richard Haselgrove Post: After the grumbles, the glories. The cavalry has arrived - four men and a digger. Neighbours should have power today. The WiFi is working after all. The PSU is definitely f**ked - it blows the fuse in the kettle lead every time I switch the tester on. But I've got a spare in stock. Just need to cut the cables ties the professional builder had sprinkled all over the place. (Said builder - CCL - doesn't seem to carry stock any more. Soonest re-stock would be priority delivery tomorrow. If they can get past the electricity company van in the access road). |
|
540)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 105494)
Posted 27 Sep 2021 by Richard Haselgrove Post: It's official - this country is falling apart as we watch. Lost all power this morning. Took them about an hour and a half to send a couple of chaps round to change the fuse in the sub-station. They were all set to pack up and go home, until we pointed out that only 40 of the 42 houses affected had come back on. That'll need a digger, apparently. I'm mostly all right, but it's looking like I've lost a wireless access point and the PSU in my daily driver. Still checking... |
|
541)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 105482)
Posted 25 Sep 2021 by Richard Haselgrove Post: What delivery? You want it - whatever it is? You come get it yourself.No, I mean the heavy stuff - a whole supermarket-full of milk, frozen food, toilet rolls ... |
|
542)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 105480)
Posted 25 Sep 2021 by Richard Haselgrove Post: That'll be fun for the "just-in-time" delivery theory. |
|
543)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 105478)
Posted 25 Sep 2021 by Richard Haselgrove Post: Can you still get around when tank stations are out of petrol?Yes, I had a lovely clear run into town for lunch - the local fuel station was closed, so no queue and no scrum of turning traffic. It had opened again by the time I'd finished my meal, but I didn't bother to turn in. I do hope that drivers from the mainland have the sense to reject the King's Shilling, at least until we grovel for re-admission to the club. Make that 20 years or so. |
|
544)
Message boards :
Projects :
News on Project Outages
(Message 105476)
Posted 25 Sep 2021 by Richard Haselgrove Post: How come the Oooops is always on a weekend?Because the person fiddling with it goes home at 4 pm on a Friday, finished or not? |
|
545)
Message boards :
GPUs :
"Exclusive Application" option problem
(Message 105474)
Posted 25 Sep 2021 by Richard Haselgrove Post: Ticket 41 was actually opened in July 2007, 14 years ago - well before the mass adoption of multi-core CPUs, and before GPUs had even been considered as a compute resource. David Anderson closed it with a 'wontfix', but it was reopened and variants of it have been requested many times since. “The moving finger writes; and, having writ, |
|
546)
Message boards :
Projects :
News on Project Outages
(Message 105467)
Posted 23 Sep 2021 by Richard Haselgrove Post: It appears to be a network issue at the department level.Not the Department of Engineering IT Support again? |
|
547)
Message boards :
GPUs :
Intel GPU
(Message 105410)
Posted 13 Sep 2021 by Richard Haselgrove Post: Fair enough. |
|
548)
Message boards :
GPUs :
Intel GPU
(Message 105406)
Posted 13 Sep 2021 by Richard Haselgrove Post: Several projects do already have iGPU applications - SETI, Einstein, WCG/Covid-19. Most discussion has focused on efficiency: there is a balance to be struck between high-performance applications, which draw so much power and generate so much heat that the rest of the CPU has to slow down, and low-performance applications which - well, don't add much science. But I'm more concerned about scientific accuracy. Most iGPU applications are built on the back of OpenCL applications originally developed for discrete GPUs - these applications require little or no extra programming effort. But most applications that I've seen use a compiler optimisation called 'cl-mad-enable'. The documentation for mad says: mad approximates a * b + c. Whether or how the product of a * b is rounded and how supernormal or subnormal intermediate products are handled is not defined. mad is intended to be used where speed is preferred over accuracy.From what I've seen over the years, the divergence caused by mad has increased with each new iteration of the silicon and each new generation of runtime support in the drivers. My advice would be to 'proceed at caution' until these effects are more widely understood. |
|
549)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 105375)
Posted 11 Sep 2021 by Richard Haselgrove Post: Seems like they're all watching the WiFi - https://systemstatus.berkeley.edu/event/179274/ makes for entertaining, bur not reassuring, reading. |
|
550)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 105371)
Posted 11 Sep 2021 by Richard Haselgrove Post: There's a server at the end of the line, but ET isn't picking up..... |
|
551)
Message boards :
News :
Certificate of computation
(Message 105355)
Posted 8 Sep 2021 by Richard Haselgrove Post: Well, nothing new has been added to either the BOINC or the boinc-site repositories since this morning (UTC). |
|
552)
Message boards :
Projects :
News on Project Outages
(Message 105347)
Posted 8 Sep 2021 by Richard Haselgrove Post: Andy Bowery has noticed the problem: The project has had to be taken offline. New RAM was installed on one of the key servers. This new RAM has unfortunately failed. It is anticipated that this downtime will last until tomorrow.I've replied to say Could you please check that all essential servers are working and accessible from the internet before re-starting the project? A half-way restart is very wasteful of unissued tasks. |
|
553)
Message boards :
Projects :
News on Project Outages
(Message 105346)
Posted 8 Sep 2021 by Richard Haselgrove Post: CPDN downloads are failing again. |
|
554)
Message boards :
News :
Certificate of computation
(Message 105345)
Posted 8 Sep 2021 by Richard Haselgrove Post: Nah, the CPID isn't error-checked. If you make a slight typo in that field, you get the errors on line 112 and 145. |
|
555)
Message boards :
News :
Certificate of computation
(Message 105344)
Posted 8 Sep 2021 by Richard Haselgrove Post: CPID 16a4a513fbbb77b756173c6d1cf16b44 does not have zero projects.That's funny - I can see 47 projects and 4.35 sextillion floating-point operations OK. Maybe David is still fixing it as we speak. |
|
556)
Message boards :
News :
Certificate of computation
(Message 105343)
Posted 8 Sep 2021 by Richard Haselgrove Post: For me, the new page shows 14 projects. But BOINCstats shows 23 ... It looks like the numbers are pulled from BOINC Combined Statistics, which has a different way of handling retired projects. More seriously, I'm wondering whether the mathematics leading to my statement of "3.77 sextillion floating-point operations" has been audited: BOINC doesn't count or record flops directly. 68aa4b6077c3fe48975e530f6ad94ca5 |
|
557)
Message boards :
News :
Certificate of computation
(Message 105341)
Posted 8 Sep 2021 by Richard Haselgrove Post: David has forgotten that he fixed this problem in another place last year: web: in "all projects" certificate, show message if no projects |
|
558)
Message boards :
Questions and problems :
Install Windows 11 on an "Old PC"
(Message 105310)
Posted 3 Sep 2021 by Richard Haselgrove Post: https://boinc.berkeley.edu/dl/?C=M;O=DAnd your point is? The newest Windows version shown via that link - v7.16.16, dated 27 October 2020 - is to be avoided. It has a nasty bug which prevents re-opening the Manager after closing it, when run on Windows 8.1, 10, and (presumably) 11. I don't know whether the currently recommended v7.16.11 has been rigorously tested under Windows 11 yet, but I suspect not - unless you're in a position to do that? The reported fix in #4468 is purely cosmetic (correct reporting of the Windows version details), and is too new to be included in either the v7.16.11 or v7.16.16 downloads. |
|
559)
Message boards :
Questions and problems :
Cannot open the Boinc manager
(Message 105299)
Posted 1 Sep 2021 by Richard Haselgrove Post: Once I had time to sit down and properly think, turned out to be easy to find. See if #4072 "7.16.16 when doing BOINC Manager Close Window, how does one open BOINC Manager again?" matches what you're seeing? That one was noticed when a stray test copy of BOINC v7.16.16 sort-of 'escaped' on 27 October last year. That one was fixed within two days (#4075), but there has been NO official BOINC release for Windows since then. The notes for #4075 say: This issue affects only Windows 8.1/10 x64 built with VS2013 and VS2019.That might imply that installing the current official v7.16.11 release, dated 2 September 2020 (happy birthday tomorrow!) might be a workround - that one was built using VS2010. But I wouldn't be too optimistic - there have been many changes to those system tray icons as Windows 10, ahem, "matured". And I hope the Windows Release Manager has booked some lab time for the official release of Windows 11 on 5 October 2021... |
|
560)
Message boards :
Questions and problems :
Cannot open the Boinc manager
(Message 105297)
Posted 1 Sep 2021 by Richard Haselgrove Post: I agree. Probably even the Manager is running too, but minimised to the notification area and invisible. We've had all this before with the various flavours of Windows 10: I need to do some heavy searching to find the latest solution. Be back later. |
|
561)
Message boards :
Questions and problems :
Amount of CPU for AMD Radeon vs NVidia GTX on Einstein
(Message 105268)
Posted 26 Aug 2021 by Richard Haselgrove Post: My experience with Nvidia GPUs at Einstein under windows, is that you should always reserve one CPU core for each GPU task you are running plus 2 CPU cores to OS if you plan to use the computer for your daily routines.The same applies to most projects using the OpenCL programming language for NVidia GPUs. It's not so important for projects that use the proprietary CUDA programming language. |
|
562)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 105261)
Posted 26 Aug 2021 by Richard Haselgrove Post: Fresh milk is distributed cheaply and at scale in every British city and town - it's a staple of the British diet. A perennial complaint of British dairy farmers is that the 'economy of scale' purchasing power of British supermarkets forces farm-gate prices close to, or below, the cost of production. |
|
563)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 105255)
Posted 26 Aug 2021 by Richard Haselgrove Post: Finally got around to reading the "milkshake" article.It's probably more a case of obsessive business central control. Why aren't the "restaurants" (as they like to call them) supplied with milk from local cows via local dairies? What's the food-mile cost of supplying every ingredient from a centralisd distribution depot? |
|
564)
Message boards :
Questions and problems :
BOINC stopped working- memory hog? Running Linux
(Message 105200)
Posted 20 Aug 2021 by Richard Haselgrove Post: Four quick points: 1) The messages refer to disk space. We usually try to keep 'disk space' and 'memory' problems distinct, because they have different causes and different solutions. 2) Les has a good point - you may have several failed tasks. Check the CPDN website (your account) to see if that's happened. 3) Try to find out why they failed. The commonest cause is the absence of the required 32-bit compatibility libraries. There are guides for installing them on major Linux distros, again on the CPDN website. 4) Once you've found and eliminated the cause, the quickest way to recover the wasted disk space is through BOINC Manager, advanced view: on the Projects tab, select CPDN and click 'Reset project'. Similar considerations will apply to Rosetta, but I don't know that project so well. |
|
565)
Message boards :
Projects :
News on Project Outages
(Message 105194)
Posted 19 Aug 2021 by Richard Haselgrove Post: Further news from CPDN: Andy Bowery wrote: Just to let you know the affected RAM board of the dev site has now been removed and both the dev and main projects are now back online.I have been able to report overdue trickles. |
|
566)
Message boards :
Questions and problems :
Memory usage by Boinc
(Message 105176)
Posted 18 Aug 2021 by Richard Haselgrove Post: Perhaps selecting to not keep GPU projects in memory, might alleviate some of the issue?BOINC GPU tasks are never kept in memory in the first place. GPUs don't have any access to the swap file or virtual memory that would normally hold a CPU task when idle. |
|
567)
Message boards :
Questions and problems :
BOINC pauses all tasks for no apparent reason
(Message 105175)
Posted 18 Aug 2021 by Richard Haselgrove Post: Low memory won't affect cached tasks, but will reduce the number of tasks you can run concurrently. |
|
568)
Message boards :
Questions and problems :
Cannot input anything in Boinc account manager
(Message 105174)
Posted 18 Aug 2021 by Richard Haselgrove Post: There is a known problem with the versions you mention: it cannot accept input in certain Asian languages, including Chinese. You may be using one of the languages that's affected. If you can switch to using English, or any language using a variant of the English alphabet, you should be able to proceed. |
|
569)
Message boards :
Projects :
News on Project Outages
(Message 105137)
Posted 15 Aug 2021 by Richard Haselgrove Post: Andy has confirmed that both the main CPDN project and the test project have been deliberately stopped as a result of the suspected hardware failure. Further news will be posted as it becomes available. |
|
570)
Message boards :
Projects :
News on Project Outages
(Message 105126)
Posted 14 Aug 2021 by Richard Haselgrove Post: Andy reports that a new problem - tentatively identified as a hardware failure - has been observed on the CPDN 'dev' (test) servers. He has shut down "the project" to minimise data loss, and fears that this closure may last for several days. The main, production, CPDN server is also reporting 'shut down for maintenance'. I have asked Andy to clarify whether both versions of the project need to be shut down because of the single hardware failure, and am awaiting his reply. |
|
571)
Message boards :
Projects :
News on Project Outages
(Message 105053)
Posted 12 Aug 2021 by Richard Haselgrove Post: Staff, or outside contractors? |
|
572)
Message boards :
Projects :
News on Project Outages
(Message 105048)
Posted 12 Aug 2021 by Richard Haselgrove Post: At 15:24 UTC on 12 Aug 2021 Andy Bowery wrote: All services have been restored now to climateprediction.net infrastructure. The Department of Engineering IT Support decided to roll back the changes they made to the networking. This has allowed us to restore all the CPDN services.Edit: Yes, I can confirm that all files for new tasks are being downloaded cleanly. |
|
573)
Message boards :
News :
Flatpak installer available
(Message 105018)
Posted 11 Aug 2021 by Richard Haselgrove Post: Has v7.18.1 been officially released for platforms other than Android, then? It hasn't shown up in Gianfranco's PPA yet - that's usually first. |
|
574)
Message boards :
Projects :
News on Project Outages
(Message 105008)
Posted 10 Aug 2021 by Richard Haselgrove Post: Edit3: Does suspending the project stop the uploads/downloads?I think not. |
|
575)
Message boards :
Projects :
News on Project Outages
(Message 105002)
Posted 10 Aug 2021 by Richard Haselgrove Post: CPDN needs to be aware that BOINC is designed to manage multiple projects in parallel, and that many of us use it that way. There was once a proposal by, I think, user 'Thyme Lawn' to allow/suspend transfers by project: he coded it for precisely this scenario, but it was rejected by the gatekeepers. For that reason, I can't follow your example: all my recent tasks have declared their download errors to be permanent and have reported their task status as 'download failed'. I've set 'no new tasks' until I receive positive confirmation that the network is operating properly again. |
|
576)
Message boards :
Questions and problems :
BOINC pauses all tasks for no apparent reason
(Message 104999)
Posted 10 Aug 2021 by Richard Haselgrove Post: Also, be aware that if you are using BOINC's own internal controls to keep CPU usage "well under 10%", they may not be working in the way you expected. BOINC can't keep your CPU running continuously at a low speed. Instead, it operates by stopping and starting the CPU at 1-second intervals. A setting of 10% will result in the CPU being stopped for 9 seconds and allowed to run for 1 second. The system was designed in the days of single-core CPUs: I don't know how it operates on modern multi-core CPUs. I doubt there is any mechanism for evening out the thermal cycle by running separate tasks at different times in the 10-second window: you may find that all tasks are paused at the same time, and all restart and run for their single second at the same time. |
|
577)
Message boards :
Projects :
News on Project Outages
(Message 104998)
Posted 10 Aug 2021 by Richard Haselgrove Post: So, does anyone know whether the CPDN download servers have been re-connected to the internet yet? I'm on Andy Bowery's email distribution list, and I haven't seen anything yet - and I've completed upgrading my machines to Linux Mint v20.2 Memo to project staff: the project shouldn't be restarted after maintenance until all components are tested and working. |
|
578)
Message boards :
Projects :
News on Project Outages
(Message 104979)
Posted 7 Aug 2021 by Richard Haselgrove Post: I'm not very impressed by the Oxford University Engineering IT Support team. They will have scheduled this work for the summer vacation, when the undergraduate demand is low: but university postgrad and faculty research continues 52 weeks of the year. This is also a very busy time of year for university administration, dealing with applications from next year's intake of new students. Letting a planned infrastructure upgrade over-run by a week is bad management, to say the least. |
|
579)
Message boards :
Projects :
News on Project Outages
(Message 104975)
Posted 6 Aug 2021 by Richard Haselgrove Post: Two tasks reported, four tasks allocated, four sets of downloads failed. You'd have thought they knew about that one by now :-( |
|
580)
Message boards :
News :
Android client version 7.18.1 released.
(Message 104967)
Posted 6 Aug 2021 by Richard Haselgrove Post: WRONG/BAD LINK ON DOWNLOAD PAGE: https://boinc.berkeley.edu/dl/boinc_7.18.0.apkIt can be downloaded manually from https://boinc.berkeley.edu/dl/boinc_7.18.1.apk |
|
581)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 104962)
Posted 5 Aug 2021 by Richard Haselgrove Post: And in winter, when the drivers will need lights, wipers, and heating during the outbound dispersal? |
|
582)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 104960)
Posted 5 Aug 2021 by Richard Haselgrove Post: Low-rate electric charging at a park-and-ride is definitely one of the ways forward. It would encourage both EV use for the broad rural dispersal at the end of the working day, and also public transport use in the congested city centre. PaR sites are also low density and readily accessible, which would make the infrastructure installation phase much easier. Just don't expect to power the whole lot by on-site solar power in the UK - the density's too great for that. |
|
583)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 104950)
Posted 5 Aug 2021 by Richard Haselgrove Post: Unfortunately, we have a phenomenon called 'winter' here. And it rains... |
|
584)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 104945)
Posted 4 Aug 2021 by Richard Haselgrove Post: That's my concern as well. In the UK, 240V (-ish) AC is universal, but the supply up my street looked very ancient when they had to dig it up to fix a fault. It was so old it wasn't even shown on their modern electronic map: they had to resort to detectors and pure guesswork to know where to dig. Sticking an extra 50A (per house) onto that sounds decidedly iffy - not to mention that I would have to get it out to the far end of my private parking lot. My temporary stop-gap is a modern, standards-compliant, pure diesel. range ~600 miles between tanking, and it counts as 'ultra low emission' for city use. I took it for an 1,800 mile run last year, only filling twice away from home, but I don't think I saw a single EV charging point at any of my overnight stops during the run. |
|
585)
Message boards :
Android :
BOINC Android Client 7.18.1 available
(Message 104873)
Posted 25 Jul 2021 by Richard Haselgrove Post: Confirmed, with screenshots. I'll call it in. Done: #4463 |
|
586)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 104847)
Posted 19 Jul 2021 by Richard Haselgrove Post: Fortunately, my night clubbing days are long gone. A pint or two of a nice locally brewed beer in a congenial pub is much more my style!Same here, and the same optimal solution. To which can be added: if you are a de facto regular, whether known by that name or not, you only have to supply contact details etc. once per 'rule set lifetime' (with occasional refresher courses for new staff - I had four old hands explaining to a newbie last week "it's just Richard, he knows what to do"). But that causes a problem for these so-called passports. A vaccination certificate should have a serious lifetime and an expiry date - I'd guess a year, in this case. But a test certificate will be chip-wrappings after three days: you can't combine them. |
|
587)
Message boards :
Questions and problems :
crucial value for resource management
(Message 104846)
Posted 19 Jul 2021 by Richard Haselgrove Post: ... field in <proxy_info> section ...No, it's an option in it's own right, alphabetically next after 'proxy'. Always refer new users back to the User Manual, in this case the Client configuration / Options section. |
|
588)
Message boards :
Questions and problems :
running windows and linux clients at the same time.
(Message 104829)
Posted 18 Jul 2021 by Richard Haselgrove Post: If not using the default, you have to set the control port manually, and separately, for both client and manager. It's usually done on the command line. And make sure they match... |
|
589)
Message boards :
BOINC Manager :
What should I do if a task goes past the deadline?
(Message 104823)
Posted 17 Jul 2021 by Richard Haselgrove Post: The first, and most important change, is to reduce the amount of work you are requesting - so it doesn't happen again. Then, consider your project settings. Your screenshot says you are running 10 'mt' (multi-threaded) tasks at the same time. Normally, an mt task - by itself - uses all CPU cores available in the machine, so that would be extremely inefficient. However, PrimeGrid is a special case, and you can set separate core limits for each search type. Count up your core demand and allocation carefully, and make sure you aren't - literally - overcooking your CPU. There is a particular problem when mt tasks are allocated by the server, which I've written up at GitHub #4151 - using PrimeGrid as one of the examples. When calculating how much work to send in response to your work request, the server doesn't take any account of the number of cores the task(s) will occupy - so it tends to send 4, 8, 12 ... times as much work as you were asking for, depending on the value of your 'cores per task' setting. So - back to the first point - ask for a small amount of work overall, and small amounts at each subsequent request. |
|
590)
Message boards :
BOINC Manager :
Suspend tasks after checkpoint
(Message 104818)
Posted 17 Jul 2021 by Richard Haselgrove Post: On the other hand, BOINC is aware when a science app has checkpointed - the app notifies BOINC - so the idea is feasible. And it would have a minor energy-saving effect. Two things would need to be considered carefully: 1) Timings. Some projects write very large data files to disk as part of the checkpointing process. I don't think BOINC has any awareness of the physical write speed of the device in use, or of the effect of any 'lazy write' caching in the operating system. It would be counter-productive if BOINC acted on the checkpoint notification before all write operations had securely completed. 2) Some projects haven't implemented checkpointing at all. There would need to be a clear policy, communicated to users, for how BOINC would incorporate that scenario into the proposal. |
|
591)
Message boards :
Questions and problems :
Loss of statistical data when machine crashes
(Message 104813)
Posted 16 Jul 2021 by Richard Haselgrove Post: That particular sequence of events started with the unexpected and unexplained appearance of a v7.16.16 installer in the download directory at 2020-10-27 19:45. That turned out to be a VS2019 experiment, built from whatever happened to be in the release branch at the time. That's not a flippant remark: the release branch is supposed to be a tightly controlled, reference version that satisfies the needs of the LGPL 'source code' condition. The v7.16 branch - hasn't, to put it mildly, I won't look further into the details at this time on a Friday night (UK time), but I'm pretty sure no changes have been made to the statistical record keeping in recent years. The statistics for each project are kept in a separate file. I think (subject to checking) that each file is re-written from scratch - using the data held in RAM - each time that project's data changes, usually when a project task completes. That's the only time the data in the file is vulnerable to errors. The fundamental issue would be - your computer shouldn't crash. External factors, like a local thunderstorm affecting the electricity supply, can't easily be avoided, but apart from that, modern computers should run for months without errors. If a particular project should regularly crash the machine - then that's a bad project. Avoid running it on that machine. Even saying that, a bad project should only mess up its own statistical record, because only that single file should be being re-written at the moment the crash occurs. We can explore this further - after I've had a night's sleep! - but we would need further information about which projects are being run, and the nature of the crashes that trigger the problem. |
|
592)
Message boards :
Questions and problems :
Older Boinc version to Ubuntu 20.04.2 LTS
(Message 104754)
Posted 10 Jul 2021 by Richard Haselgrove Post: boinc is correctly spelled at steps 2 and 3. The 'bionic' main universe in step 1 is, I suspect, a different animal altogether. |
|
593)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 104745)
Posted 9 Jul 2021 by Richard Haselgrove Post: I'm sorry to intrude into private grief, but I have to state my case. I regard myself as a liberal, and sometimes even as a socialist. I'm not a communist, and I have never been a formal member of a political party - although I've voted whenever possible. I do find the relentless repetition of the McCarthyite position: "Are you now, or have you ever been a member of the Communist party?" distressing and unhelpful. I prefer the formulation "From each according to their ability, to each according to their needs" - it's much closer to my own position, and in close accordance with the teachings of the New Testament. We've worked together in the past: I'm sorry it's come to this. But I'll leave it there. |
|
594)
Message boards :
Questions and problems :
Older Boinc version to Ubuntu 20.04.2 LTS
(Message 104729)
Posted 8 Jul 2021 by Richard Haselgrove Post: You'd probably have to build your own binaries from sources, using the build tools and libraries native to your own machine. You can download that version's source code from https://github.com/BOINC/boinc/tree/client_release/7%2F7.4 |
|
595)
Message boards :
Questions and problems :
Your settings do not allow fetching tasks for Intel GPU. To fix this, you can change Project Preferences on the project's web site.
(Message 104727)
Posted 8 Jul 2021 by Richard Haselgrove Post: They exist, but all WCG GPU tasks are rare. Some versions of the iGPU are problematic under Windows. |
|
596)
Message boards :
Projects :
News on Project Outages
(Message 104709)
Posted 5 Jul 2021 by Richard Haselgrove Post: Now fixed and tasks downloading.I'll be ready for another one around lunchtime tomorrow - we'll see what happens then |
|
597)
Message boards :
Questions and problems :
Resource share questions
(Message 104690)
Posted 3 Jul 2021 by Richard Haselgrove Post: You can't set the percentages directly. All you can set are the absolute numbers - 100 (default), 15 and 8 for your lower priority projects. Then, BOINC adds up the numbers and converts each individual share to a percentage of the total. Your PCs have three projects still at a share of 100, and a total of 362. On the Mac, you only have one project at 100, and a total of 162. The percentages work out different, because the totals are different. |
|
598)
Message boards :
BOINC Manager :
Can't start client or reinstall on Windows
(Message 104677)
Posted 1 Jul 2021 by Richard Haselgrove Post: It's a Windows registry problem. Use regedit, and navigate to [HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup]You'll see two keys listed INSTALLDIR DATADIR - either or both may refer to drive D. Change them to a valid drive letter, and the installer should run. If you managed to salvage the BOINC data folder from the failed drive, put it at the new DATADIR location before installing BOINC - you might be able to pick up where you left off. Otherwise, just start over again. |
|
599)
Message boards :
Projects :
News on Project Outages
(Message 104673)
Posted 30 Jun 2021 by Richard Haselgrove Post: The retry got allocated three tasks: two downloaded OK, but the third failed. I've posted a request for a sanity check on the project message board - looks like wingmates have been having problems too. |
|
600)
Message boards :
Projects :
News on Project Outages
(Message 104669)
Posted 30 Jun 2021 by Richard Haselgrove Post: It may be time dependant. My other Linux box got some around 08:30 UTC this morning, and one of those is running now. I'll try and see if the failing one can get more work before the current task finishes. They both need a reboot, too, but it's hard to get all the tasks synchronised to finish together when both types are being send out, |
|
601)
Message boards :
Projects :
News on Project Outages
(Message 104666)
Posted 30 Jun 2021 by Richard Haselgrove Post: And it seems like Wednesday is the new Tuesday. Got two new tasks allocated, but: 30/06/2021 16:11:36 | climateprediction.net | Started download of ic_N144_200309_000045.nc.gz 30/06/2021 16:11:36 | climateprediction.net | [http] [ID#24101] Info: Trying 129.67.193.131:80... 30/06/2021 16:11:36 | climateprediction.net | [http] [ID#24101] Info: TCP_NODELAY set 30/06/2021 16:13:37 | climateprediction.net | [http] [ID#24101] Info: Connection timed out after 120664 milliseconds 30/06/2021 16:13:37 | climateprediction.net | [http] [ID#24101] Info: Closing connection 21319 30/06/2021 16:13:37 | climateprediction.net | [http] HTTP error: Timeout was reached 30/06/2021 16:13:37 | climateprediction.net | Temporarily failed download of ic_N144_200309_000045.nc.gz: transient HTTP errorCouldn't connect to cpdn.org, either - but that's come back up. I'll retry the downloads. Edit: Something fishy here. I had nine files failing to download, each with their own two minute timeout (default for Linux). When I retried a single download (all individual file delays had cleared by then), four files disappeared from the list without appearing in the event log, and the other five went through the two-minute timeout routine. Both tasks - one N144, one N216 - are marked as download failures. |
|
602)
Message boards :
BOINC Manager :
Suggestion: automatically free RAM on the use schedule
(Message 104647)
Posted 27 Jun 2021 by Richard Haselgrove Post: I always leave that ticked: mind you, I also run a huge 'Switch between tasks every ...', so I very rarely see a suspended task on my systems - they run 24/7. We each have our own preferred options: this one seemed to be appropriate for the OP's particular needs. |
|
603)
Message boards :
BOINC Manager :
Suggestion: automatically free RAM on the use schedule
(Message 104643)
Posted 27 Jun 2021 by Richard Haselgrove Post: There's already a standard global option Leave non-GPU tasks in memory while suspendedUncheck that, and all project tasks will be removed from memory within seconds when your schedule suspends computation. Don't worry about the "non-GPU" bit: all GPU tasks are already removed from memory when suspended, without any choice in the matter. |
|
604)
Message boards :
Questions and problems :
running windows and linux clients at the same time.
(Message 104631)
Posted 26 Jun 2021 by Richard Haselgrove Post: Yes, it's well known that - specifically - VirtualBox doesn't support 'pci passthrough'. That's the formal name for the ability even to see, let alone use, peripherals like GPUs in a virtualised environment. Other virtualisation solutions do support passthrough, but BOINC doesn't support them - so we're stuck. |
|
605)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 104548)
Posted 9 Jun 2021 by Richard Haselgrove Post: One Fastly customer triggered internet meltdown A customer quite legitimately changing their settings had exposed a bug in a software update issued to customers in mid-May, causing "85% of our network to return errors", it said.Let that be a lesson to us all! |
|
606)
Message boards :
Documentation :
32bit apps on 64bit Linux distros.
(Message 104491)
Posted 27 May 2021 by Richard Haselgrove Post: Could the wiki be updated to include lib32stdc++6 or lib32stdc++7. (I discovered that 7 is needed for the latest Ubuntu versions.) A user went to the wiki and followed the instructions but found everything crashing due to this missing. This with CPDN but I know at least one other project uses this.Following discussion with Dave, I've re-arranged this section to assume that current users are installing/running 64-bit versions of Linux by default, and thus the more urgent need is to install extra 32-bit libraries where needed. Previously (from about 2009), the presumption had been that the base system was 32-bit, and that 64-bit libraries were needed for compatibility. Please check, and let me know if any further tweaks are necessary. |
|
607)
Message boards :
Questions and problems :
BOINC 7.16.11 / Intel GPU Jobs - stuck or running indefinitely
(Message 104480)
Posted 26 May 2021 by Richard Haselgrove Post: Received and replied. Sorry, I got distracted by the real world. |
|
608)
Message boards :
Questions and problems :
BOINC 7.16.11 / Intel GPU Jobs - stuck or running indefinitely
(Message 104477)
Posted 26 May 2021 by Richard Haselgrove Post: Ir's a known issue. I ran into it during the Beta phase, observed something which gave me a clue what was going wrong, and then tested/confirmed my suspicions. I've reported the problem and the likely cause to the project staff. It's difficult. I've deliberately written that last sentence very vaguely, because the solution which I've adopted (and which is working fine on my personal machines to this day) involves disabling a safety feature which might be needed in other circumstances: the manufacturer's advice is very strongly that this should not be done except by software developers for testing purposes. I don't come into that category, but I can accept responsibility for my own machines only. I'm not writing details of the cause and solution in public, because I can't take responsibility for rendering another volunteer's machine inoperable - that's the worst-case scenario. In the end, WCG will have to ask Scripps to re-write part of the application code so it can run, in particular, on the slower variants of the Intel GPU range. Until then, the only safe advice is to avoid using those iGPUs for the WCG covid-19 application. Having said that, if you want to PM me and confirm that you understand the risks, I can point you to the official documentation with the warnings and the (simple) change which might be enough to get you running again. |
|
609)
Message boards :
Android :
BOINC Android client 7.16.16 released
(Message 104467)
Posted 25 May 2021 by Richard Haselgrove Post: There has continued to be massive activity on GitHub, but it's hard to get a comprehensive overview. Most of the activity seems to have been computer science 'process' activity (especially in the area of automated testing), rather than scientific research 'purpose' activity. I am particularly concerned by the stress on automated testing. From my observations, this can lead to an excess of complacency: "The code compiles, so it must be right". I have seen a number of logic errors making their way into the code: the computer science is right, but the end result is meaningless. I have a private saying for that sort of code: "Legal but stupid". |
|
610)
Message boards :
Documentation :
Format of job log files
(Message 104449)
Posted 23 May 2021 by Richard Haselgrove Post: He's doing more for Science United, these days. But he doesn't seem to have realised that Science United is entirely reliant on a working, bug-free BOINC infrastructure to hold it up. DA.If the intention was to come back and add something new, like log entries for failed tasks with the reason for failure included, it's taking a mighty long time to get round to it.Is the person who made the change still doing any work on BOINC? |
|
611)
Message boards :
Documentation :
Format of job log files
(Message 104447)
Posted 23 May 2021 by Richard Haselgrove Post: I did some more poking around to try and work out when, and why, [es] was added to the job log format. 'When' is easy: Jul 14, 2013. 'Why' is unfathomable. Nothing is said in the commit that added it. 2013 is the first year for which no separate file of checkin_notes was maintained. So history is silent. And then, as now, it had absolutely zero purpose. Starting at https://github.com/BOINC/boinc/blob/master/client/cs_apps.cpp#L181, the code path reads if (rp->exit_status != 0) {
had_error = true;
}
if (had_error) {
...
[do some housekeeping]
...
} else {
...
rp->append_log_record();
...
}And that's the only place where the code to add to the job log file is called.So, we write to the job log if, and only if, exit status is zero. Adding those five bytes to every line tells us absolutely nothing we didn't know already. If the intention was to come back and add something new, like log entries for failed tasks with the reason for failure included, it's taking a mighty long time to get round to it. |
|
612)
Message boards :
Documentation :
Format of job log files
(Message 104437)
Posted 22 May 2021 by Richard Haselgrove Post: I haven't found any documentation either, but this is what I've worked out for myself. It's text format, with Linux line endings, with fields in the format [identifier, value]. I read them as: Datestamp Unix timestamp at completion [ue] Estimated runtime BOINC Client estimate (seconds) [ct] CPU time Measured CPU runtime at completion (seconds) [fe] Estimated FLOPs count From project (integer) [nm] Task name From project [et] Elapsed time Wallclock runtime at completion (seconds)There's an additional field [es] after those, but I haven't identified it yet - the value is always zero. Edit - [es] stands for 'exit status'. But since job log entries are only written for tasks which finished without error, that's not much use. |
|
613)
Message boards :
Questions and problems :
Cannot accept friend requests
(Message 104360)
Posted 11 May 2021 by Richard Haselgrove Post: If the message came from Science United, you should be able to access it on https://scienceunited.org/su_home.php |
|
614)
Message boards :
Projects :
Open-source projects
(Message 104358)
Posted 11 May 2021 by Richard Haselgrove Post: What about World Community Grid?The main working component for the covid-19 project (Autodock for GPU) is at https://github.com/ccsb-scripps/AutoDock-GPU But WCG themselves add a lot of packaging to make it run under BOINC - I'm not sure if that is open-source. And WCG runs multiple science projects: whether each working component is open-source will depend on each separate research group/institution. |
|
615)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104320)
Posted 7 May 2021 by Richard Haselgrove Post: Hi Richard, I was wondering what you thought about my post earlier about the possible scheduler bug that I found(unrelated to this issue)? It looks like there are two orders for calling sched_work_old and sched_work_locality, but 1 of the times the result is saved before the scheduler is actually called.Well, I saw it and had a quick look... Sat down to have another read-through (starting my head-spin again), and this is what I think is happening: Send_work goes through some basic checks, and then - starting at the line you link - goes through five possible configurations 1) locality_scheduler_fraction - mixed, locality first. 2) debug_locality - mixed, old scheduler first 3) locality_scheduling 4) sched_old 5) send_work_score It's only the 'mixed' cases where two separate schedulers are called, and some flags are saved and reused. So only the 'locality' and 'old' schedulers are in play. With the 'locality' scheduler, the 'insufficient' flags are set during the scheduler run, and need to be preserved. But I can't find any sign of the flags being set in any of the other schedulers, including 'old'. So, although in the 'debug' case, the 'save and reset' precaution is in the wrong place, (send_work_old(); should be around line 1685), in reality there aren't any volatile flags that need to be preserved. Further, there's only one project that we know of that uses locality scheduling (Einstein) - this scheduler was written, by Bruce Allen, specifically for Einstein. Their main server is using config (1), which has the 'save and reset' in the right order. They may get a bit of a shock if they ever try 'debug' mode in the future, but probably not. Phew. |
|
616)
Message boards :
News :
2021 BOINC workshop
(Message 104308)
Posted 5 May 2021 by Richard Haselgrove Post: It was spoken in the Zoom workshop - so you'll be able to hear it, rather than read it. In the recordings we can't access yet... |
|
617)
Message boards :
Questions and problems :
boincが使えなくなりました
(Message 104305)
Posted 5 May 2021 by Richard Haselgrove Post: There is a known problem with BOINC Manager not accepting input in some far-Eastern languages, and it's likely that Japanese falls into that category. If it's possible for you to use an English language setting temporarily while you re-attach to your projects, you should be able to return to Japanese afterwards. |
|
618)
Message boards :
Questions and problems :
Display initial information without restarting client?
(Message 104302)
Posted 4 May 2021 by Richard Haselgrove Post: OK - https://github.com/BOINC/boinc/blob/master/client/gpu_amd.cpp#L131 callib is 'CAL library'. CAL is the first proprietary language used for programming ATI cards, when they were ATI and not AMD. It must be legacy support, in case you want to install a driver and run programs still using that old language - it's been discontinued for several years now. No harm from the legacy check, but no benefit, either. |
|
619)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104301)
Posted 4 May 2021 by Richard Haselgrove Post: At long last, a smoking gun. This looks very suggestive: 04/05/2021 17:17:33 | World Community Grid | Host location: noneIt does look like you don't only have to enable CPU work, you have to request it as well. Edit - ouch! That poor little celeron is going to have it's work cut out. Of those 8 tasks, 7 were for the iGPU. Estimated at 41 minutes, but actually running for nearly five hours. And they were all _2, _3, _4 resends, with a 36-hour 'hurry up' deadline. I've cut down the CPU workload. |
|
620)
Message boards :
Questions and problems :
Display initial information without restarting client?
(Message 104298)
Posted 4 May 2021 by Richard Haselgrove Post: Should I be concerned about this at the end of that file? "<warning>No ATI library found.</warning>"Only if you have an ATI card and no driver for it. |
|
621)
Message boards :
News :
2021 BOINC workshop
(Message 104293)
Posted 4 May 2021 by Richard Haselgrove Post: Still nothing online...Last week, Matt said 'Wednesday' - but he's said things like that before. |
|
622)
Message boards :
Questions and problems :
Display initial information without restarting client?
(Message 104291)
Posted 4 May 2021 by Richard Haselgrove Post: On Windows, learn where stdoutdae.txt is in the data folder. On Linux, you'll need journalctl. |
|
623)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104279)
Posted 2 May 2021 by Richard Haselgrove Post: Yes, I think we can pretty well conclude that the effect is real - there's definitely a causal link between 'no CPU work' and 'no Intel GPU work'. But the question we asked ourselves at the beginning was - is that WCG's fault, or BOINC's fault? (I think we can rule out the client, by now). Neither of us has found a smoking gun in the BOINC code, so are we inclining towards the WCG modifications? That's pretty much me for the evening - the UK has a watercooler appointment with the TV in about 10 minutes. |
|
624)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104277)
Posted 2 May 2021 by Richard Haselgrove Post: Since you mention SETI, I posted the instructions for offline testing in SETI message 2072928Also, I converted one of the early Betas to run offline at a command prompt, which removed the dependency on new work.Cool! I was thinking about looking how to do that to run the tests in the link grumpy_swede posted about how much running CPU tasks can slow down the iGPU. I ended up just running them with the SETI AP WUs, but the slowdown was not representative of what was shown with my data collection during the beta. There was a lot more slowdown with AP than with OPN. |
|
625)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104274)
Posted 2 May 2021 by Richard Haselgrove Post: Just saw reports that work was flowing more freely, bumped the cache, and got 14 new tasks for iGPU. So that rules out a limit, unless they've been changing settings while we experiment. I've turned off CPU tasks for that profile, so we'll see how it holds up, maybe with another cache boost later (when the rush has died back down again - downloads are busy ATM). |
|
626)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104271)
Posted 2 May 2021 by Richard Haselgrove Post: This seems to be behaving as expected - CPU and Intel GPU computing is enabled and you get both CPU and Intel GPU. Since I set my test machine back to stock boinc this morning(almost 13 hours ago) it has not gotten Intel GPU tasks.Are you running any sort of 'retry' automation? Otherwise, the backoffs will cut you down to very few requests. |
|
627)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104269)
Posted 2 May 2021 by Richard Haselgrove Post: Now I'm up to five iGPU tasks: 02/05/2021 19:06:44 | World Community Grid | Sending scheduler request: To fetch work.Though the first is due to finish in about five minutes. Edit - iGPU completed and reported, and I got my third CPU task in return. No configuration changes in the last two hours. 02/05/2021 19:26:05 | World Community Grid | Reporting 1 completed tasks |
|
628)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104268)
Posted 2 May 2021 by Richard Haselgrove Post: Having work constantly available enabled me to run tests and replicate the issue on command repeatedly.We have scripts for that now!. Also, I converted one of the early Betas to run offline at a command prompt, which removed the dependency on new work. |
|
629)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104266)
Posted 2 May 2021 by Richard Haselgrove Post: If the end of work is that soon, is it time to post about the intel_gpu issue in the WCG forums? Or did you want to do some more testing first?It's not the end of work, just the end of the stress test. Then back to a trickle of 2,000 every half hour, or whatever it was. I'd imagine they'd want to process the resulting server load issues first: I'd imagine it'll be better to wait until we have a constructive diagnosis to pass on. |
|
630)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104265)
Posted 2 May 2021 by Richard Haselgrove Post: This can be from the config.xml limits or the user's project preferences.Indeed. I was caught out initially (I don't usually use project preferences for that), but I'd corrected it by then - and the limit is still being applied. First iGPU task still has 45 minutes to run - we'll see what happens then. |
|
631)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104260)
Posted 2 May 2021 by Richard Haselgrove Post: The 'school' venue is one I usually reserve for an Android tablet to run CPU tasks. It was set to maximum 2 tasks in WCG device profiles, but by 17:21 I'd realised that and removed the restriction. The only limit I can think of after that would be 'four per (intel) GPU', which has never been mentioned, and I think I've seen exceeded on 'big' machines. |
|
632)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104257)
Posted 2 May 2021 by Richard Haselgrove Post: Kevin just said: There are between 40-48 hours of work left to be run at the current pace of the stress test (so around Tuesday 12:00 UTC +/- 3-4 hours).Good thing tomorrow is a public holiday in the UK - and forecast to be very wet. I can stay indoors and keep trying. |
|
633)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104256)
Posted 2 May 2021 by Richard Haselgrove Post: I would take this as the case you mentioned before. Meaning where the scheduler happens to not have work(for the CPU or intel GPU with whichever wcg projects are selected) at the precise time when you requested it.The problem with that one is that it's so badly implemented (at all projects, not just WCG) that it chucks out every possible excuse. That last 17:21 reply, in full, was: 02/05/2021 17:21:45 | World Community Grid | No tasks sentTake your choice: the final one is usually the most reliable. |
|
634)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104252)
Posted 2 May 2021 by Richard Haselgrove Post: But I can't get beyond here. Computer is a quad-core plus iGPU: I wanted to run 3xCPU + iGPU, but instead I've got 2xCPU (both running) and 4xiGPU (one running). And 02/05/2021 17:21:05 | World Community Grid | Computer location: schoolDo we know that limit? |
|
635)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104251)
Posted 2 May 2021 by Richard Haselgrove Post: And we have lift-off: 02/05/2021 16:46:08 | World Community Grid | Computer location: schoolOr maybe not: 02/05/2021 16:50:16 | World Community Grid | Sending scheduler request: To fetch work. |
|
636)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104250)
Posted 2 May 2021 by Richard Haselgrove Post: Changing to a new location and back again 'sort-of' works: 02/05/2021 16:35:35 | World Community Grid | New computer location: |
|
637)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104249)
Posted 2 May 2021 by Richard Haselgrove Post: First oddity: the reply after initialisation says "Computer location: home". But the WCG website says it's on 'default'. I think default is the default on my account, but I'll check... |
|
638)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104248)
Posted 2 May 2021 by Richard Haselgrove Post: OK, now we're sucking diesel - or diesel fumes, at any rate. 02/05/2021 16:12:11 | World Community Grid | Sending scheduler request: Project initialization.So far, so bad. I'll keep trying a few different combinations. |
|
639)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104246)
Posted 2 May 2021 by Richard Haselgrove Post: It may for the machines with multiple gpus that we discussed earlier. I think that it does not fit for machines with only 1 gpu which is an intel_gpu.Yup, setting one of the other devices to allow work seems to liberate the intel_gpu - allowing CPU in your case, allowing NV in mine. And yup, I've spent a lot of time looking at BOINC - it's my retirement hobby, to keep the little grey cells supple. But I've spent more time looking at the clients, rather than the server. You're moving at least as fast as me on this one. I do have a suitable 'CPU plus intel_gpu only' machine to explore with. It's just running a few PrimeGrid MT tasks to keep warm at the moment, but I'll flush those out (2-3 hours), and set it up on a separate WCG profile so I can fiddle with different settings. |
|
640)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104243)
Posted 2 May 2021 by Richard Haselgrove Post: I've never used the simulator, except in its web-interface form. It often lags behind the master code: I had a problem involving an app_config.xml file, but I couldn't include it in the simulation run. That was added to the input page, but my next problem seemed to involve the use of two app_config.xml files. And so on. But if you can overcome those constraints by building locally, it may give you some insights. NB - it can't simulate the server response "... reached a limit on tasks in progress". On the server side, your description is fine. But be aware it leaves out the dimension of time. For your req_seconds, the scheduler picks one candidate task, locks it, and performs a number of checks to see if it's right for your machine. If it is, it's assigned to you, added to a buffer, and the whole process starts again. Adding multiple tasks to the buffer - sequentially - takes time, and is optimised as far as possible. The main optimisation is the use of a 'feeder'. The feeder preforms all the database lookups to populate a small pool of available tasks for allocation. On small projects, that pool contains maybe 100 or 200 tasks at a time, and is held in the fastest-access memory available. I shudder to think how many may be needed for WCG in this mode! That single pool has to be used by every scheduler instance currently running (multiple requests are processed in parallel). So, while your scheduler is plodding through your request, other schedulers are fishing in the same pool. For a non-trivial request, it's entirely possible for the feeder pool to have been emptied before your scheduler has found enough compatible tasks to add up to your request quantum. You'll get a reply comprising less work than you asked for, and you ask again a couple of minutes later. By which time, the feeder pool will have been emptied and refilled many times over (I think the usual cycle time is 2 seconds). That may even account for the missing intel_gpu tasks on its own. As we saw, your request is processed device by device - NV, AMD, Intel (in that order). The intel_gpu is most likely to hit 'empty feeder syndrome'. Look back at my summary log (message 104231) - does that explanation fit? |
|
641)
Message boards :
GPUs :
AMD GPU Windows 10 21H1 Computation Error
(Message 104240)
Posted 2 May 2021 by Richard Haselgrove Post: "Output file ... absent" is a symptom, not a cause. Try looking at the failed task on the project's website: there may be additional clues in the 'stderr.txt' reported back to the server. |
|
642)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104236)
Posted 1 May 2021 by Richard Haselgrove Post: It's not necessarily as bad as that. I found a problem with another project earlier this year, and their response when I pointed it out to them was along the lines of "we maintain our own separate version of the server code, except for the scheduler, where we've brought in the latest version as a complete replacement" (or words to that effect - I'm relying on memory). All the issues we've discussed have been related to the scheduler, and It's entirely possible that WCG will have done something similar. It's unlikely that they will have written an entirely separate replacement: as you and I have found, it's fiendishly complicated, and I think even experienced coders like Bruce have refrained from meddling as much as possible. The BOINC code is the best guide we've got, and it's likely to be pretty close - it has to be compatible with the requests from the newest clients, at the very least. |
|
643)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104234)
Posted 1 May 2021 by Richard Haselgrove Post: Thereby hangs a very nasty tale. This machine actually reports <coproc_cuda> <count>2</count> <name>GeForce GTX 1660 SUPER</name>but in practice it's got one 1660 and one 1650. Left to its own devices, the 1650 would be idle, but you can wake it up with <use_all_gpus>1</use_all_gpus> in cc_config.xml Having two different cards, but only reporting the characteristics of the 'better' one, causes real problems. If the 'smaller' one has less memory, or can't do double precision, or is pre-OpenCL v1.2, the server will still send impossible tasks, and the client will still try to run them (unless you exclude that project on that GPU, also in cc_config.xml). I went to the 2014 BOINC workshop in Budapest (my local low-cost airline happened to fly direct to Hungary from the airport just up the road, on the right day of the week both out and back. I could hardly pass that one up!), and I heard the 'historical summary' talk David Anderson gave that year. I heard him say that the decision not to separately identify GPUs was a mistake, in hindsight. I don't think either the text or a recording of that talk is available in the public domain, but the slides can be downloaded from https://boinc.berkeley.edu/trac/attachment/wiki/WorkShop14/workshop_14.pdf. Slide 52 was on screen as he said that: "Reflections on software: things we need to change - Coprocessor model". But nobody has. |
|
644)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104231)
Posted 1 May 2021 by Richard Haselgrove Post: Summary log from the second machine: 01-May-2021 16:31:50 [World Community Grid] [sched_op] Starting scheduler request 01-May-2021 16:31:50 [World Community Grid] [sched_op] NVIDIA GPU work request: 41413.98 seconds; 0.00 devices 01-May-2021 16:31:50 [World Community Grid] [sched_op] Intel GPU work request: 36455.24 seconds; 0.00 devices 01-May-2021 16:31:51 [World Community Grid] [sched_op] estimated total NVIDIA GPU task duration: 976 seconds 01-May-2021 16:31:51 [World Community Grid] [sched_op] estimated total Intel GPU task duration: 0 seconds 01-May-2021 16:33:58 [World Community Grid] [sched_op] Starting scheduler request 01-May-2021 16:33:59 [World Community Grid] [sched_op] NVIDIA GPU work request: 40652.56 seconds; 0.00 devices 01-May-2021 16:33:59 [World Community Grid] [sched_op] Intel GPU work request: 36604.44 seconds; 0.00 devices 01-May-2021 16:34:01 [World Community Grid] [sched_op] estimated total NVIDIA GPU task duration: 0 seconds 01-May-2021 16:34:01 [World Community Grid] [sched_op] estimated total Intel GPU task duration: 5818 seconds 01-May-2021 16:36:51 [World Community Grid] [sched_op] Starting scheduler request 01-May-2021 16:36:51 [World Community Grid] [sched_op] NVIDIA GPU work request: 40845.92 seconds; 0.00 devices 01-May-2021 16:36:51 [World Community Grid] [sched_op] Intel GPU work request: 30935.17 seconds; 0.00 devices 01-May-2021 16:36:52 [World Community Grid] [sched_op] estimated total NVIDIA GPU task duration: 0 seconds 01-May-2021 16:36:52 [World Community Grid] [sched_op] estimated total Intel GPU task duration: 5818 seconds 01-May-2021 16:38:54 [World Community Grid] [sched_op] Starting scheduler request 01-May-2021 16:38:54 [World Community Grid] [sched_op] NVIDIA GPU work request: 41158.52 seconds; 0.00 devices 01-May-2021 16:38:54 [World Community Grid] [sched_op] Intel GPU work request: 25191.31 seconds; 0.00 devices 01-May-2021 16:38:57 [World Community Grid] [sched_op] estimated total NVIDIA GPU task duration: 0 seconds 01-May-2021 16:38:57 [World Community Grid] [sched_op] estimated total Intel GPU task duration: 8728 seconds 01-May-2021 16:41:51 [World Community Grid] [sched_op] Starting scheduler request 01-May-2021 16:41:52 [World Community Grid] [sched_op] NVIDIA GPU work request: 41495.04 seconds; 0.00 devices 01-May-2021 16:41:52 [World Community Grid] [sched_op] Intel GPU work request: 16539.24 seconds; 0.00 devices 01-May-2021 16:41:53 [World Community Grid] [sched_op] estimated total NVIDIA GPU task duration: 0 seconds 01-May-2021 16:41:53 [World Community Grid] [sched_op] estimated total Intel GPU task duration: 8728 seconds 01-May-2021 16:43:56 [World Community Grid] [sched_op] Starting scheduler request 01-May-2021 16:43:56 [World Community Grid] [sched_op] NVIDIA GPU work request: 41752.40 seconds; 0.00 devices 01-May-2021 16:43:56 [World Community Grid] [sched_op] Intel GPU work request: 7959.73 seconds; 0.00 devices 01-May-2021 16:43:57 [World Community Grid] [sched_op] estimated total NVIDIA GPU task duration: 967 seconds 01-May-2021 16:43:57 [World Community Grid] [sched_op] estimated total Intel GPU task duration: 5818 seconds 01-May-2021 16:46:52 [World Community Grid] [sched_op] Starting scheduler request 01-May-2021 16:46:53 [World Community Grid] [sched_op] NVIDIA GPU work request: 42095.51 seconds; 0.00 devices 01-May-2021 16:46:53 [World Community Grid] [sched_op] Intel GPU work request: 2216.89 seconds; 0.00 devices 01-May-2021 16:46:54 [World Community Grid] [sched_op] estimated total NVIDIA GPU task duration: 1934 seconds 01-May-2021 16:46:54 [World Community Grid] [sched_op] estimated total Intel GPU task duration: 2909 seconds 01-May-2021 16:48:57 [World Community Grid] [sched_op] Starting scheduler request 01-May-2021 16:48:58 [World Community Grid] [sched_op] NVIDIA GPU work request: 41290.60 seconds; 0.00 devices 01-May-2021 16:48:58 [World Community Grid] [sched_op] Intel GPU work request: 0.00 seconds; 0.00 devices 01-May-2021 16:48:59 [World Community Grid] [sched_op] estimated total NVIDIA GPU task duration: 1918 seconds 01-May-2021 16:48:59 [World Community Grid] [sched_op] estimated total Intel GPU task duration: 0 secondsSo, it eventually filled the Intel queue, but it took several goes, and started topping up NV along the way. Final thought, before I take a break - my head's spinning. In pulling out that summary, I deleted all the references to "Server version 701". In theory, that dates back to 2013 or even earlier - almost as old as Einstein's! - but I suspect that in reality, they've forked off the mainline and done their own tweaks. |
|
645)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104228)
Posted 1 May 2021 by Richard Haselgrove Post: I'm getting twitchy about https://github.com/BOINC/boinc/blob/master/sched/sched_send.cpp#L739: // return true if additional work is needed,That leads on to https://github.com/BOINC/boinc/blob/master/sched/sched_send.cpp#L796: // check config.xml limits on in-progress jobsThat should loop through #define PROC_TYPE_CPU 0and when the maximum is exceeded for one proc_type, zero out the work request for that type, and proceed to check the others. But I've got a log on a different machine that appears to say different. Hold on, and I'll edit it in. 01/05/2021 16:29:47 | World Community Grid | Requesting new tasks for NVIDIA GPU and Intel GPU 01/05/2021 16:29:47 | World Community Grid | [sched_op] NVIDIA GPU work request: 41076.75 seconds; 0.00 devices 01/05/2021 16:29:47 | World Community Grid | [sched_op] Intel GPU work request: 36306.05 seconds; 0.00 devices 01/05/2021 16:29:48 | World Community Grid | Scheduler request completed: got 0 new tasks 01/05/2021 16:29:48 | World Community Grid | This computer has reached a limit on tasks in progressThat machine is at the same 100 limit for NV, which accounts for the 'limit' message, but why didn't it go on to check the intel request? I've only got three of those, so nothing to stop it. But now I've reported a few NV tasks, it's gone into a different mode: 01/05/2021 16:38:54 | World Community Grid | Sending scheduler request: To fetch work. 01/05/2021 16:38:54 | World Community Grid | Reporting 1 completed tasks 01/05/2021 16:38:54 | World Community Grid | Requesting new tasks for NVIDIA GPU and Intel GPU 01/05/2021 16:38:54 | World Community Grid | [sched_op] NVIDIA GPU work request: 41158.52 seconds; 0.00 devices 01/05/2021 16:38:54 | World Community Grid | [sched_op] Intel GPU work request: 25191.31 seconds; 0.00 devices 01/05/2021 16:38:57 | World Community Grid | Scheduler request completed: got 3 new tasks 01/05/2021 16:38:57 | World Community Grid | [sched_op] estimated total NVIDIA GPU task duration: 0 seconds 01/05/2021 16:38:57 | World Community Grid | [sched_op] estimated total Intel GPU task duration: 8728 secondsLater, it started fetching a litle NV, and even later - drum roll! - both types together. I'll try and extract a summary of the log. |
|
646)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104226)
Posted 1 May 2021 by Richard Haselgrove Post: Ah ok. Yes, I misread that. It would be interesting to know if the client is only requesting nvidia since presumably they have their cache set high enough for 150 NV or if the client is requesting all of them and the WCG server sends for the NV GPU until: the 50 per GPU or the req_secs is reached.Mine is behaving a little differently. I'm working at a Windows machine, with WCG enabled for NV (x2) and Intel_gpu, but not for CPU. It's also got work for other projects, but I'm not fetching from them at the moment. I'm stable at 120 tasks: 8 WCG for intel, 12 from other projects, and thus 100 for NV - and it says "This computer has reached a limit on tasks in progress". Which is what the book says, but not what's been reported. Meanwhile, see what you think of https://boinc.berkeley.edu/trac/wiki/ProjectOptions#Joblimits. That's the instruction manual for server operators. It's not bang up to date, but four of the last five changes have been made by Kevin Reed - another of the WCG admins - so I guess they're paying attention. Neither that section ('Job limits'), not the next section down ('Job limits (advanced)') seems to allow them to set different limits for different classes of GPU - so, "All GPUs are created equal". ??? So, "don't save the settings if nothing's changed" instead of "don't send the settings if nothing's changed"Yup, I'll go with that - looked inside the reply file this time. |
|
647)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104224)
Posted 1 May 2021 by Richard Haselgrove Post: If I am interpreting the code properly, the settings are sent with every sched_reply. The account_[WCG] file is overwritten if "project->gui_urls != old_gui_urls || update_project_prefs". update_projects_prefs is set if the venue has changed or if the sent project settings are different from the current project settings. See cs_scheduler.cppThe alternative way of reading that is "don't send the settings if nothing's changed". I must have changed mine on Wednesday, but not since. They probably haven't considered the case of "user modified client record, so now it's different from what the server remembered" - maybe that's the mis-match that prompts your send. |
|
648)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104223)
Posted 1 May 2021 by Richard Haselgrove Post: IIRC, the limit is 50 per GPU with a max of 200 per machine. People with multiple GPUs have pointed that out before.Yes, we've got that one sorted. But I was drawing attention to "gets up to 150 NV wu's" in the post I quoted. He shouldn't have space for 150 tasks for a single NV GPU if they were allocated strictly '50 for the NV, 50 for the ATI, 50 for the APU'. It seems to '150 in total - first come, first served' - that's the effect we're chasing ("Why are there none left for tail-end Charlie?"). |
|
649)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104219)
Posted 1 May 2021 by Richard Haselgrove Post: The interesting question is "why do we have separate lines for <work_req_seconds> and <cpu_req_seconds>?".The 'historical legacy' answer seems to hold up. David Jan 10 2009 My 'account_[WCG]' file still has a datestamp of 28 April - so it's not re-written every time. Probably only when the setting on the server has changed. |
|
650)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104218)
Posted 1 May 2021 by Richard Haselgrove Post: The interesting question is "why do we have separate lines for <work_req_seconds> and <cpu_req_seconds>?". For the project where I get my CPU tasks from, both are populated, and identical to the microsecond. I suspect the answer may be historical: when BOINC was first written, there was only one possible sort of work (GPUs couldn't compute), and maybe the wording was changed in newer servers? The alternative may (and probably should) have been retained for backward compatibility. I'll add that to my code reading list ('blame' is useful in these situations). The other thing I'll look at is the timing evidence from the modification datestamp on the various files - notably, request/reply and the account file. Edit - interesting post at https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,43386_offset,570#657527: Just mentioned that my GPU cruncher with 1660 Super, 280X and A12-9800 APU gets up to 150 NV wu's and almost non for the ATI/AMD. That's no real problem because I can handle this with a second Boinc instance.With a 50 per GPU limit, is the limit being applied globally? Does the NVidia card get all 150, and leave no quota for the other two? That makes some sense, because NVidia was the first GPU type to be handled by BOINC, followed by AMD and finally Intel. The requests are almost certainly handled in that order in the server code, but I'll check. |
|
651)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104207)
Posted 30 Apr 2021 by Richard Haselgrove Post: Would it be correct to say that if I get it to use the locally modified account_ file and it starts getting the intel gpu tasks that it may be an issue with the client?I think it would be fair to say that if you got it to work - we don't know what the heck will happen! Your mission, should you choose to accept it, is to ... I'm not getting the feeling that this is a client problem. You're getting the right sort of figures for 'work request' in the event log, and I think we'd have noticed by now if those figures didn't also appear in sched_request (though that's something you could check). The problem seems to happen after you send the request, and before you get the reply - and the only thing between those two events is the server. |
|
652)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104202)
Posted 30 Apr 2021 by Richard Haselgrove Post: Is there a way for me to edit this file and have BOINC use it to test having OPN1(I think 92) set to 1 and OPNG(I think 94) still set to 25? Just for testing to be able give more info to the WCG people. That way it could have CPU enabled but only get a max of 1 task at a time. Unless something (maybe "-1"?) could be put in to mean no tasks. Or probably better to remove 92 from:Yes, that should be possible - for testing, at least. But I think it might only work once: it can only have reached your machine via a scheduler request/reply, and it might be repeated in all subsequent replies. Have a look in sched_reply_[WCG].xml (root of BOINC data folder), and see what you can find in there. It might be safest to stop the client before making that change, and see what happens on restart. The value to use for 'no restriction' in BOINC is undefined. Sometimes, David has used 0, and sometimes he's used -1. There's even one file where he's used both within 15 lines of code. [https://github.com/BOINC/boinc/blob/master/lib/common_defs.h#L86]. I think he flipped somewhere around 2015, so 'recent' changes like this one (2016) probably use -1, if it's defined at all. I'll look. |
|
653)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104198)
Posted 30 Apr 2021 by Richard Haselgrove Post: If it was not clear earlier in the thread. There is only one thing I change in the project preferences to switch between getting the intel GPU tasks and not getting the intel GPU tasks:I think that is possibly a specific WCG problem with the implementation of those controls: their way of transferring those choices from the web site to the scheduler will be unique to them (because their website is unique). But I still intend to look through the generic BOINC server code when I get a chance, but it'll be slow - I have open diagnosis calls to both Einstein and CPDN on the go at the moment. Edit - ah, I'd forgotten that WCG also has a user selection for 'max tasks in progress'. That's another complication: I was going to say 'unique to WCG', but I've recently seen it on 'traditional' BOINC projects too. It'll be a thread on this board - can anyone remind me while I finish breakfast? OK, you can stop looking: message 103817 |
|
654)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104188)
Posted 30 Apr 2021 by Richard Haselgrove Post: Do you mean on the server side or the client side? The log made me think that the client was requesting the intel gpu tasks.I mean on the server side. The client makes the requests as expected, but the server seems to skip that part of the processing. |
|
655)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104184)
Posted 30 Apr 2021 by Richard Haselgrove Post: That sounds like a related class of problem: if something blocks one part of the request, it doesn't fall through and process the Intel part. Either your 'CPU disabled'. or my 'NV at task limit' is enough to skip processing the intel part of the request. I think the 'must enable CPU to get intel' variant was reported fairly early in the Beta test, and from memory Uplinger fixed it? Might be worth looking through those Beta messages. I'll try to look through the relevant bit of the server code, but this weekend is beginning to get busy. |
|
656)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104181)
Posted 30 Apr 2021 by Richard Haselgrove Post: Confirmed - while I was typing, 30/04/2021 09:32:23 | World Community Grid | [sched_op] NVIDIA GPU work request: 8525.33 seconds; 0.00 devicesNothing for Intel while NVidia is at the limit. But if you report one, you can get a replacement: 30/04/2021 09:27:23 | World Community Grid | Reporting 1 completed tasks |
|
657)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104180)
Posted 30 Apr 2021 by Richard Haselgrove Post: Two more logs - I updated the cache to 0.3 days. 30/04/2021 09:17:27 | World Community Grid | Sending scheduler request: To fetch work. 30/04/2021 09:19:31 | World Community Grid | Sending scheduler request: To fetch work.I now have 107 tasks, and 7 of them are for the intel. We have a limit of 50 tasks per GPU, so my NV cards are maxxed out. Reporting 2 tasks gave me space for 2 replacements, but at that point, NV processing would have been stopped. None for Intel, although work was requested? I suspect server processing stops when one part hits a limit, and doesn't move on to the next part of the request. Best of luck in getting David Anderson to fix that one. |
|
658)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104179)
Posted 30 Apr 2021 by Richard Haselgrove Post: I guess what I am asking since I know that you a lot more about the inner workings of BOINC than I do is:Hi there. I've been working for a while on the generic BOINC server code, but I'm very new to the specifics of WCG - I only joined at the start of the Covid-19 GPU beta test. And we have many, many, variables flying around at once, and being changed at short notice. But about 30 minutes ago, I logged: 30/04/2021 08:33:28 | World Community Grid | Sending scheduler request: To fetch work.So, at that much-reduced level, it's working as designed. My cache settings, during this test, are 0.1 days + 0.01 days: I have 63 tasks on that machine, running two-up on each of two NVidia GPUs (GTX 1660 and GTX 1650). I also have a 'update every five minutes', script running, not that it's really needed during the stress test. So, I suspect, the 'little and often' approach works best under current circumstances. That means your experience is perhaps best put down to strange things happening when a BOINC server is stressed beyond its normal limits. Similar things were seen at SETI@Home at busy times. |
|
659)
Message boards :
Questions and problems :
GPU not receiving tasks when CPU computing disabled
(Message 104176)
Posted 30 Apr 2021 by Richard Haselgrove Post: Actually, the 'devices' field is only non-zero if you have absolutely no work left left and the device - Intel GPU - is idle. 108243.50 seconds (over 30 hours) is a lot of work to ask for in a single gulp, and for a project under stress (deliberately - they're running a stress test!), it will make life even harder for the project server. Under normal circumstances - unless you have an intermittent internet connection or something - your machine will be contacting the server at least once an hour to report results, and it's better to ask for a top-up then. |
|
660)
Message boards :
Questions and problems :
Boinc sees 2 graphics cards when only 1 is installed.
(Message 104167)
Posted 29 Apr 2021 by Richard Haselgrove Post: It sounds like you have two separate drivers installed for your single graphics card. I'd suggest you use something like Display Driver Uninstaller to tidy things up, and then install (or allow Windows 10 to install) a single driver for the new card. |
|
661)
Message boards :
Questions and problems :
BOINC Manager won't let me connect to a client
(Message 104146)
Posted 27 Apr 2021 by Richard Haselgrove Post: OK, three things: 1) SETI@Home stopped sending out new work over a year ago. There is no point in spending time trying to connect to them. At the least, set 'No New Tasks' for that project when you get a chance. (Details are on the project web site) 2) Event Log: your optional choices generate a huge volume of traffic, but provide very little useful information at this stage. [gui_rpc], in particular, is a resource hog. At this stage, you're probably stuck. Shut down BOINC completely: make sure both boinc.exe and boincmgr.exe have disappeared from Task Manager. You may have to force them to quit. Then, you need to find your BOINC data directory. If you chose your own location when installing BOINC, and know where it is - congratulations. Otherwise, the default location is C:\ProgramData\BOINC\That location is a hidden folder: you'll either need to tell Windows Explorer to display hidden folders, or paste that address into Windows Explorer's address bar. The file we're looking for is called cc_config.xmlIt's an optional file, but it's where your Event Log choices are stored. Rename that file so BOINC can't see it, but we can examine it later if we need to. Then, you can restart BOINC again. It'll use the default settings for the log, but your projects and tasks should be where you left them: you should have control via BOINC Manager as normal. 3) WCG: they are running a stress test of their server, and it's failed - communications are horribly difficult. That's their problem, not yours. Concentrate on getting your own machine working normally first, and post again if you need help with any specific problems. |
|
662)
Message boards :
GPUs :
Hardware accelerated GPU scheduling
(Message 104103)
Posted 25 Apr 2021 by Richard Haselgrove Post: Sorry - that rules me out, probably on all three counts.Turns out I spoke too soon. I woke up the Windows 10 boot drive on a perfectly ordinary desktop machine, did a year's worth of updates (the machine runs Windows 7 day-to-day), and now that hardware acceleration setting is active. The machine is an Intel i5, with HD 530 iGPU and NVidia GTX 1050 Ti - both drivers from Microsoft, so I get OpenCL support from Intel. At least five years old, and not the sharpest knife in the box even then. It all seems to run, but I don't think I can run a useful performance test. |
|
663)
Message boards :
Questions and problems :
World Community Grid gets stuck
(Message 104100)
Posted 25 Apr 2021 by Richard Haselgrove Post: Richard Haselgrove:From your original description of the problem, it does sound as if the AMD APU is triggering the same Windows protection mechanism as the Intel iGPU. My own iGPUs have performed perfectly since I traced the cause of the problem: but unfortunately, I don't have any APUs here, so - although it seems likely - I can't confirm whether the same tweak will work on the equivalent AMD platform. If you - or anyone else with a similar APU, and experiencing the same problem - would be prepared to test my suspicion, I could send you the details by PM or email (initial dot surname at btinternet dot com - be careful with spelling). It's a simple one-line tweak, and completely reversible - but it would require a Windows restart. |
|
664)
Message boards :
GPUs :
Hardware accelerated GPU scheduling
(Message 104084)
Posted 24 Apr 2021 by Richard Haselgrove Post: https://devblogs.microsoft.com/directx/hardware-accelerated-gpu-scheduling/Thanks for that. So, we need a late-revision Windows 10 version (May 2020 update or later); a 'recent GPU that has the necessary hardware'; and 'a WDDMv2.7 driver that exposes this support to Windows'. Sorry - that rules me out, probably on all three counts. And there's nothing about whether this will enhance the computational use of GPUs. The only technical commentary concentrates on frame buffering, user input, and latency: in other words, gaming. Unless we have any Windows 'Insider' participants reading here (and not under NDA), we'll just have to wait until the promised The goal of the first phase of hardware accelerated GPU scheduling is to modernize a fundamental pillar of the graphics subsystem and to set the stage for things to come… but that’s going to be a story for a another time 😊. |
|
665)
Message boards :
Questions and problems :
World Community Grid gets stuck
(Message 104083)
Posted 24 Apr 2021 by Richard Haselgrove Post: Sadly, this is a known problem which the WCG team are aware of, but haven't yet addressed. I'm guessing that you're running an Intel GPU (iGPU) under Windows. What follows only applies to that combination: if your situation is different, please stop reading here and post back with details of your GPU. Not every iGPU is affected, which is why they haven't turned off the supply entirely. But for those affected - it's mainly the i5 range of processors and below - the WCG iGPU is simply too slow, and triggers a safeguard/watchdog in Windows. The app continues to run, but isn't actually using the GPU any longer. There are several possible courses of action, but most of them would need to be carried out by the WCG project staff working with the Scripps Institute science team. I think they're all too busy to work on it yet. It is possible to disable the Windows watchdog, but it's dangerous: depending on what else you use the computer for, you could encounter other problems far worse than this one. So I'm not going to post details in public: if anyone else understands Windows internals enough to deduce what I'm alluding to, feel free to use your own judgement as to whether it's safe in your environment. Otherwise, the Scripps programmers are going to have to redesign their application pretty substantially. I think it's probably best for you to withdraw your computer from the iGPU part of the project for the time being, and watch out for any new version announcements on the WCG project forums. |
|
666)
Message boards :
GPUs :
Hardware accelerated GPU scheduling
(Message 104073)
Posted 23 Apr 2021 by Richard Haselgrove Post: What Microsoft has released is fairly new 2020 and it enables GPU cards to do their own scheduling rather than having the System CPU do the Scheduling.Was that Microsoft release aimed at developers/programmers, or at end users? Do you have a link? |
|
667)
Message boards :
Questions and problems :
Performance drop after Win 10 April Update?
(Message 104025)
Posted 20 Apr 2021 by Richard Haselgrove Post: First question: Does your RAC fall come from 'less credit per task' or 'fewer tasks completed'? Without details like that, we don't have a clue where to start looking. |
|
668)
Message boards :
Questions and problems :
Finding Bottlenecks
(Message 103987)
Posted 17 Apr 2021 by Richard Haselgrove Post: Another possible explanation is that your 'better CPU' - the one with the time differential - is using a technology called 'hyperthreading'. This enables one physical CPU 'core' to work on two different computer programs at once. When the two programs are different, they might use different resources, or have pauses when, for example, they're waiting for you to press the next key on the keyboard. Under these circumstances, a hyperthreaded core can feel almost as fast as two completely separate cores. But when the two running programs are similar - say, two Rosetta tasks - and both use the same parts of the CPU hardware, they are competing rather than collaborating, and get in each other's way. There aren't so many gaps available, and overall progress doesn't increase so much. In general, hyperthreading (or HT) isn't as much use on BOINC projects as it is in other areas of computing. |
|
669)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 103908)
Posted 9 Apr 2021 by Richard Haselgrove Post: It was in Northern Ireland. |
|
670)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 103906)
Posted 9 Apr 2021 by Richard Haselgrove Post: Re: this post Are we all (in UK) going to get a day off work to not go to the funeral?No - I don't think I'm allowed to take a day off from my pension. But I might raise a glass on Monday. |
|
671)
Message boards :
BOINC Manager :
Manager closes (crashes?) on various keyboard / mouse actions
(Message 103888)
Posted 7 Apr 2021 by Richard Haselgrove Post: The specific fix was intended to handle the alt-space key combination. I don't know if that's defined for Linux - I'll try it later. Edit - my current Linux Manager (built 4 February 2021) reacts to alt-space the same way as native Linux tools: it displays a menu analogous to the Windows menu, and doesn't crash. |
|
672)
Message boards :
Questions and problems :
How do I place a remark in app_config?
(Message 103884)
Posted 7 Apr 2021 by Richard Haselgrove Post: The full name of that file is "app_config.xml" The xml bit is important: it tells you what type of file it is, and therefore where to look for the answer to your question. A search for 'xml comment format' took me to https://www.tutorialspoint.com/xml/xml_comments.htm Warning - BOINC doesn't always implement every feature of a file format - it only handles the parts that are commonly needed. Experiment with a simple comment to start with, and check it with the 'unparsed_xml' event log flag. |
|
673)
Message boards :
BOINC Manager :
Manager closes (crashes?) on various keyboard / mouse actions
(Message 103883)
Posted 7 Apr 2021 by Richard Haselgrove Post: Yup, that's a thing. It's been reported and fixed (#4186, #4198). Unfortunately, that was only a couple of months ago, and there hasn't been a public release since then containing the fix. It's on its way, but as yet without a timetable or ETA. |
|
674)
Message boards :
Questions and problems :
Changing drive letter after install
(Message 103836)
Posted 3 Apr 2021 by Richard Haselgrove Post: No, not quite as easy as that. To save me typing it all over again, try message 98168 (and the other posts in that thread if you need them). |
|
675)
Message boards :
Questions and problems :
Virtualbox hogs machine
(Message 103823)
Posted 3 Apr 2021 by Richard Haselgrove Post: It's not secret - I'm sure I've posted it here before. Here goes. First client and first manager are absolutely standard installations. My default locations are D:/BOINC D:/BOINCdata - YMMV, adjust accordingly. In the cc_config.xml file for that instance, set <allow_multiple_clients>1</allow_multiple_clients>in the <options> section. May need a restart, certainly at least 'read config files' - I can't remember. To start the second client, make a one-line batch file with D:\BOINC\rh_boinc_test --allow_multiple_clients --allow_remote_gui_rpc --redirectio --detach_console --gui_rpc_port 31418 --dir D:\BOINCdata2and to attach a Manager to it start D:\BOINC\boincmgr.exe /m /n 127.0.0.1 /g 31418 /p passwordThose live on my test machine desktop. Notes: 1) I have a second copy of boinc.exe in my program directory, called rh_boinc_test.exe. That lets me run two different versions at once - you could just re-use the same one. 2) I have a gui_rpc_auth.cfg password set for the first instance. 3) The gui_rpc_port has to be not 31416, but the same in both batch files. 4) The second data directory has to be kept separate from the first. Note D:\BOINCdata2 5) Remember you have two separate clients to shut down in an orderly fashion, if you need to. |
|
676)
Message boards :
Questions and problems :
Virtualbox hogs machine
(Message 103820)
Posted 3 Apr 2021 by Richard Haselgrove Post: I have a tried-and-tested configuration for running two clients and two separate managers to control them, if you want to borrow it. I don't use it in normal running (barely worth it on a quad core), but it's quite tricky to set up, so it's nice to have "one I prepared earlier". Edit - Windows only, before you ask. But the principle would still apply. |
|
677)
Message boards :
Questions and problems :
Virtualbox hogs machine
(Message 103817)
Posted 3 Apr 2021 by Richard Haselgrove Post: Thanks for that. For the second time in a week, that's something that's crept into the BOINC code, and I have no memory of it. In this case, it's been in the code-base since Jul 31, 2016 - the 'dark ages' of BOINC, after Rom Walton left and nobody was left to watch over the store. I hope this one works - I had to fix the last one. It's a slight sadness that this has been implemented at the Project level. You couldn't use it, say in the example in this thread, to limit LHC to four VBox tasks, but run sixtrack on the 20 remaining cores. You'll still have to use the app_config.xml file for that - or ask LHC to split itself into two distinct projects. Edit - or run two separate clients on the same machine, with two different preference sets: any VBox app but 4 CPUs max, and sixtrack with 20 CPUs max. |
|
678)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 103803)
Posted 2 Apr 2021 by Richard Haselgrove Post: Yup, I hit that one from the other side. I'd kept my paper driving licence for many years - the one I got in about 1976 said it was valid until 2022, so I took them at their word! (Swapped it a couple of times when I changed address, but stuck with paper). Visited the USA in 2019, and thought I might want to hire a car, so getting with the 21st. century seemed like a good idea. Found that I could convert online, precisely because I had a passport. I authorised them to reach across from the DVLA database to the Passport Office database, and grab that photo. Strange thing is, that my passport photo is horribly insecure. It was taken in an automated booth in a Post Office (£5.00 - ker-ching), and sent off by post. No verification needed for a renewal: it could have been anyone, though I suppose they checked it against the one I'd sent them 10 years earlier. Later that year I signed up for my free 'seniors' bus pass, expecting to use another of the same batch of photos from the machine. Went to the bus station just opposite the Post Office, filled out the form, showed them where I lived and when I was born. Photo? Just stand there, she said. Fiddled with the webcam perched on top of her computer monitor. and the job was done. No charge, it's definitely verified as the same person who provided the documents - and it's a much better photo! |
|
679)
Message boards :
Questions and problems :
Error 2718. Missing package name for product code '{B3B5C017-61CD-4587-9FA8-7E5E2D94810E}
(Message 103766)
Posted 31 Mar 2021 by Richard Haselgrove Post: OK, here goes. 1) 'Missing package'. This is a package of instructions and data used for installing your old version. In short, it's called an 'msi' file, and is normally squirrelled away by Windows so that it can be used again when the time comes to UNinstall the old version, in preparation for installing the new version. 2) HDD failure - 'missing drive D:'. There is one registry key where you need to fix this. [HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup] "UpgradingTo"="7.6.22" "INSTALLDIR"="D:\\BOINC\\" "DATADIR"="D:\\BOINCdata\\"Looks like I used the same version as you did! First, you need to change (or delete) any reference to a D: drive in that 'BOINC Setup' registry branch. The setup program doesn't like imaginary drives. Second, if you still get the error about the missing package, try installing the old version again - exactly the same old version. If you haven't still got the old setup file, you can download it from https://boinc.berkeley.edu/dl/?C=M;O=D. Scroll down to end 2015 / early 2016, and download the matching version - 32-bit or 64-bit, with or without VBox. Hopefully, this will offer to "repair" your old version, copying the missing files to C:. Let it complete: this will replace the missing msi package as well. Then, your planned upgrade to v7.16.11 should go smoothly. If not, post again and describe the new state of affairs. |
|
680)
Message boards :
The Lounge :
Looking for a proportional font with unique characters
(Message 103742)
Posted 29 Mar 2021 by Richard Haselgrove Post: Why does the code container have a different font than the rest of the forums?The code container uses a fixed-width font - every character is giver the same width - which makes it easier to line up indents / columns / numeric tables. Proportional fonts, as the OP requested, make it much harder to do that. iiiiiiiiii wwwwwwwwwwiiiiiiiiii wwwwwwwwww |
|
681)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 103733)
Posted 29 Mar 2021 by Richard Haselgrove Post: Having at one time totally denounced the idea, Alexander Boris de Pfeffel Johnson is now floating the idea of vaccine passports. While I have no objection to them myself, I do see the main plus side as being no rowdy teenagers in pubs.What a fuss about nothing. In my experience, pubs have regulars, and drinkers have locals. Any passport will be checked once, and won't need to have an expiry date. Then it's back to normal. |
|
682)
Message boards :
Questions and problems :
How in the hell do you stop the BOINC client from putting a password in gui_rpc_auth.cfg?
(Message 103705)
Posted 26 Mar 2021 by Richard Haselgrove Post: The requirement to use a password has already been removed from the source code. client: allow an empty GUI RPC password, but generate alert message All you have to do is persuade the distribution release manager of your choice to build a new version. Warning: you will probably find them even more resistant to threats and bluster than we are. Alternatively, you could download the source code and build your very own copy - it's all publicly available. You could even remove the generated alert message. |
|
683)
Message boards :
Questions and problems :
BOINC unable to honor project shares (at all, not only in short run)
(Message 103680)
Posted 24 Mar 2021 by Richard Haselgrove Post: If GPU app really need full core for support it just provides 1.But if you, dear reader, decide that the GPU app you've been sent needs a full CPU core in support - you will have to configure that yourself. Whatever the science, whatever the programming language, whatever the observed behaviour - it's still down to you. BOINC automation won't set it for you. Project administrators won't set it for you. Project programmers won't change their programming language for you. |
|
684)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103673)
Posted 22 Mar 2021 by Richard Haselgrove Post: ... no real re-write with re-design ...Guess who was bemoaning that last week. https://setiathome.berkeley.edu/forum_thread.php?id=85717&postid=2070907#2070907 |
|
685)
Message boards :
Questions and problems :
BOINC unable to honor project shares (at all, not only in short run)
(Message 103670)
Posted 22 Mar 2021 by Richard Haselgrove Post: 0.9 = 0 (not 0.9)That's exactly correct. BOINC 'overcommits' the CPU by 'not more than one complete core'. |
|
686)
Message boards :
Questions and problems :
BOINC unable to honor project shares (at all, not only in short run)
(Message 103668)
Posted 22 Mar 2021 by Richard Haselgrove Post: "0.9CPU"Exactly. As I tried to express in https://github.com/BOINC/boinc/issues/2949 As I said in that report, the code in sched_customize.cpp (line 505 ff.) "still relies on assumptions about the relative speeds of CPU and GPU devices, and the proportion of the work to be done on each device. As speeds have diverged, these assumptions have become less and less realistic." |
|
687)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103667)
Posted 22 Mar 2021 by Richard Haselgrove Post: Fair enough. I think we've established what I set out to achieve - that there is a bug (several bugs!), and that a simple hack eliminates the massive overfetch that's given in the thread title. To go further, and eliminate the extra bugs that pertain to your setup, would require re-writing the whole of rr_sim to keep track of max_concurrent at every step of the way. I don't think I'm skilled enough to do that. We'll have to stop at the proof-of-concept. |
|
688)
Message boards :
Questions and problems :
BOINC unable to honor project shares (at all, not only in short run)
(Message 103650)
Posted 22 Mar 2021 by Richard Haselgrove Post: He said in the previous thread that he wanted Einstein to run on one of the {CPU|NV} devices - I forget which - but not the other. That rules out RS zero. The previous thread also wrote: 1 GPU MW task that doesn't need full CPU core!What makes you think that MW has cracked the busy-wait OpenCL problem on NVidia? |
|
689)
Message boards :
Questions and problems :
Use of BOINC and WSL: some confusion
(Message 103618)
Posted 20 Mar 2021 by Richard Haselgrove Post: Okay. I seem to have had success with the command:Yes, that makes sense. That command starts a BOINC client running, and tells it to listen for commands on the non-standard port. Later on, you'll want to command it to do something, or manage what's happening already. You'll use boinccmd or boincmgr to do that, and you'll have to tell those tools to send their commands on the same non-standard port. |
|
690)
Message boards :
BOINC client :
BOINC interface freeze
(Message 103600)
Posted 19 Mar 2021 by Richard Haselgrove Post: A user in another place has posted this: BOINC conflicts with Windows' default Chinese input method, Microsoft Pinyin. We can switch the keyboard to US and then everything is fine.Could you test whether this works for you, please, and let us know? The same user has added: I was using Google Pinyin on old devices. I haven't tested if Google Pinyin works.If you have access to that tool, it may help - but I'm aware that Google services are not accessible in all territories where Pinyin is used. |
|
691)
Message boards :
Questions and problems :
gui_rpc_auth.cfg exists but can't be read. Check the File permissions.
(Message 103599)
Posted 19 Mar 2021 by Richard Haselgrove Post: I've found a new (?) page, which may assist readers and save them wading through this whole thread. Accessing BOINC on Linux The final suggestion on that page is exec su $USERThat only works if you want to launch BOINC Manager from the command line in terminal. If you want to use a graphical launch icon on your desktop, or a menu item created by your distribution, you'll need to reboot the machine. |
|
692)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103594)
Posted 18 Mar 2021 by Richard Haselgrove Post: You probably pushed up the priority of the other projects by being locked into working on TN-Grid for so long by deadline pressure. It will return to normal gradually, but over a period of several days. See the Configuration Options page of the User Manual. Try setting the line <rec_half_life_days>X</rec_half_life_days>to something much smaller: one day, instead of the default 10, would sort things out quicker. |
|
693)
Message boards :
Questions and problems :
BOINC Menager keeps connecting and disconnecting to localhost
(Message 103588)
Posted 18 Mar 2021 by Richard Haselgrove Post: 'Remove button inactive' usually results when a project has been added by an 'Account Manager'. Do you now use, or have you ever used, an account manager on that machine? |
|
694)
Message boards :
Questions and problems :
Old style SETI@Home Screensaver
(Message 103556)
Posted 15 Mar 2021 by Richard Haselgrove Post: Unfortunately, not at the moment. The SETI screensavers (of any age) show details of the actual data being crunched. And for the time being, SETI is in hibernation. No SETI data is being crunched by anyone. |
|
695)
Message boards :
The Lounge :
Ubuntu 20.10
(Message 103551)
Posted 15 Mar 2021 by Richard Haselgrove Post: Linux is a steep learning curve. I'm a Windows user (developer, even), but decided to dip a toe in the Linux waters a couple of years ago, to see what all the fuss was about. I found the best way was to plonk the new, soon-to-be, Linux box on the bench next to a fully-working Windows computer - and keep the search engine of choice open at all times. |
|
696)
Message boards :
The Lounge :
Ubuntu 20.10
(Message 103547)
Posted 15 Mar 2021 by Richard Haselgrove Post: That will be the old file downloaded from the berkeley.edu domain. These days, you are recommended to download and install BOINC from the Linux repository for your distribution. |
|
697)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 103525)
Posted 14 Mar 2021 by Richard Haselgrove Post: Found that quote about "lowest possible". I forgot where I'd seen it for a while! 
|
|
698)
Message boards :
The Lounge :
Ubuntu 20.10
(Message 103524)
Posted 14 Mar 2021 by Richard Haselgrove Post: In Linux Mint 20.1 (based on Ubuntu), I see it here: 
|
|
699)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 103522)
Posted 14 Mar 2021 by Richard Haselgrove Post: Just discovered some unexpected information. I wanted to record it, but not to start a competition between projects. This is posted in the hope of encouraging collaboration and cooperation. I recently offered to help Sandman192 work out why his BOINC tasks weren't finishing in time. We still haven't fully got to the bottom of it, but the first thing that struck me was that his CPU (i7, Windows) was running extraordinarily slowly - one task was showing elapsed time 338,484.76 seconds for 15,419.56 CPU time. His wingmates were finishing with an et only slightly above ct. I pointed this out, and he told me that he had been running Folding@Home at the same time as BOINC. He's stopped folding, and BOINC elapsed times have dropped dramatically. So I decided to conduct my own experiment. These two screenshots were taken from the same computer, running similar tasks (the BOINC task was multithreaded). Folding:  BOINC:  Folding FAQs say this: How do I manually adjust the priority of the Folding@home core?The Folding core is shown as running at base priority 4, which they describe elsewhere as 'the lowest possible', but BOINC has undercut it with a base priority of 1. BOINC, on the other hand, is using all four cores, whereas Folding leave one free. Horses for courses, but this shows it's better, for the time being, to pick just one platform and stick with it. |
|
700)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103507)
Posted 12 Mar 2021 by Richard Haselgrove Post: @Sandman192 - you have replies to your PMs. |
|
701)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103506)
Posted 12 Mar 2021 by Richard Haselgrove Post: Sorry, I was away from the screen for a while. I've got a couple of PMs in my inbox too. I'll get on to them. |
|
702)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 103493)
Posted 12 Mar 2021 by Richard Haselgrove Post: I suspect it would be very difficult to sell a new house without a decent internet connection - or at least, there'd be a huge price penalty. Imagine trying to sell a house with no electricity supply. The developer will pay - grumbling, over a barrel, but they'll pay. There are stories from a decade or more ago where the developer laid the telecoms infrastructure with aluminium, instead of copper, wiring - to save money. It turned out to be incompatible with ADSL broadband. I don't think they'll make that mistake again, with so many people working from home. |
|
703)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103484)
Posted 11 Mar 2021 by Richard Haselgrove Post: I finished tidying up the re-worked hack yesterday, and it's been running for about 18 hours without problems. Here's a slightly artificial log to show the effect - I manually boosted the number of CPU tasks cached, to increase the numbers. 11/03/2021 11:46:03 | | [rr_sim] doing sim: CPU sched 11/03/2021 11:46:03 | | [rr_sim] start: work_buf min 864 additional 864 total 1728 on_frac 0.997 active_frac 1.000 11/03/2021 11:46:03 | Einstein@Home | [rr_sim] 9.23: h1_0676.05_O2C02Cl4In0__O2MDFS3a_Spotlight_676.85Hz_929_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (2585.35G/280.08G) 11/03/2021 11:46:03 | Einstein@Home | [rr_sim] 378.46: h1_0676.05_O2C02Cl4In0__O2MDFS3a_Spotlight_676.85Hz_925_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (105999.42G/280.08G) 11/03/2021 11:46:03 | Einstein@Home | [rr_sim] 932.30: h1_0676.05_O2C02Cl4In0__O2MDFS3a_Spotlight_676.85Hz_926_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (258535.16G/280.08G) 11/03/2021 11:46:03 | Einstein@Home | [rr_sim] 1301.53: h1_0648.65_O2C02Cl4In0__O2MDFS3a_Spotlight_649.20Hz_511_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (258535.16G/280.08G) 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] 1758.54: wu_sf3_DS-16x271-2_Grp197929of1000000_1 finishes (1.00 CPU) (11087.93G/6.31G) 11/03/2021 11:46:03 | Einstein@Home | [rr_sim] 1855.37: h1_0648.65_O2C02Cl4In0__O2MDFS3a_Spotlight_649.20Hz_510_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (258535.16G/280.08G) 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | Einstein@Home | [rr_sim] 1922.87: p2030.20170627.G31.90+02.51.S.b0s0g0.00000_2911_3 finishes (0.50 CPU + 1.00 Intel GPU) (31419.20G/16.34G) 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | Einstein@Home | [rr_sim] 2224.60: h1_0648.05_O2C02Cl4In0__O2MDFS3a_Spotlight_648.60Hz_532_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (258535.16G/280.08G) 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] at app max concurrent for GetDecics 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] 2424.21: wu_sf3_DS-16x271-2_Grp618165of1000000_0 finishes (1.00 CPU) (15285.09G/6.31G) 11/03/2021 11:46:03 | NumberFields@home | [rr_sim] 6516.53: wu_sf3_DS-16x271-2_Grp617510of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | | choose_project(): 1615463165.647471 11/03/2021 11:46:05 | | [rr_sim] doing sim: work fetch 11/03/2021 11:46:05 | | [rr_sim] start: work_buf min 864 additional 864 total 1728 on_frac 0.997 active_frac 1.000 11/03/2021 11:46:05 | Einstein@Home | [rr_sim] 9.23: h1_0676.05_O2C02Cl4In0__O2MDFS3a_Spotlight_676.85Hz_929_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (2585.35G/280.08G) 11/03/2021 11:46:05 | Einstein@Home | [rr_sim] 377.46: h1_0676.05_O2C02Cl4In0__O2MDFS3a_Spotlight_676.85Hz_925_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (105718.49G/280.08G) 11/03/2021 11:46:05 | Einstein@Home | [rr_sim] 932.30: h1_0676.05_O2C02Cl4In0__O2MDFS3a_Spotlight_676.85Hz_926_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (258535.16G/280.08G) 11/03/2021 11:46:05 | Einstein@Home | [rr_sim] 1300.53: h1_0648.65_O2C02Cl4In0__O2MDFS3a_Spotlight_649.20Hz_511_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (258535.16G/280.08G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 1758.54: wu_sf3_DS-16x271-2_Grp197929of1000000_1 finishes (1.00 CPU) (11087.90G/6.31G) 11/03/2021 11:46:05 | Einstein@Home | [rr_sim] 1855.37: h1_0648.65_O2C02Cl4In0__O2MDFS3a_Spotlight_649.20Hz_510_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (258535.16G/280.08G) 11/03/2021 11:46:05 | Einstein@Home | [rr_sim] 1921.76: p2030.20170627.G31.90+02.51.S.b0s0g0.00000_2911_3 finishes (0.50 CPU + 1.00 Intel GPU) (31400.97G/16.34G) 11/03/2021 11:46:05 | Einstein@Home | [rr_sim] 2223.60: h1_0648.05_O2C02Cl4In0__O2MDFS3a_Spotlight_648.60Hz_532_2 finishes (1.00 CPU + 1.00 NVIDIA GPU) (258535.16G/280.08G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 2423.21: wu_sf3_DS-16x271-2_Grp618165of1000000_0 finishes (1.00 CPU) (15278.75G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 6516.53: wu_sf3_DS-16x271-2_Grp617510of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 6613.36: wu_sf3_DS-16x271-2_Grp618075of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 6981.59: wu_sf3_DS-16x271-2_Grp616684of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 7181.20: wu_sf3_DS-16x271-2_Grp619492of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 11274.52: wu_sf3_DS-16x271-2_Grp616806of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 11371.35: wu_sf3_DS-16x271-2_Grp618515of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 11739.58: wu_sf3_DS-16x271-2_Grp617586of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 11939.19: wu_sf3_DS-16x271-2_Grp619493of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 16032.51: wu_sf3_DS-16x271-2_Grp619476of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 16129.34: wu_sf3_DS-16x271-2_Grp619477of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 16497.57: wu_sf3_DS-16x271-2_Grp618817of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 16697.18: wu_sf3_DS-16x271-2_Grp618644of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 20790.50: wu_sf3_DS-16x271-2_Grp619016of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 20887.33: wu_sf3_DS-16x271-2_Grp619479of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 21255.56: wu_sf3_DS-16x271-2_Grp619478of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | NumberFields@home | [rr_sim] 21455.17: wu_sf3_DS-16x271-2_Grp619017of1000000_0 finishes (1.00 CPU) (30000.00G/6.31G) 11/03/2021 11:46:05 | | [work_fetch] ------- start work fetch state ------- 11/03/2021 11:46:05 | | [work_fetch] target work buffer: 864.00 + 864.00 sec 11/03/2021 11:46:05 | | [work_fetch] --- project states --- 11/03/2021 11:46:05 | Einstein@Home | [work_fetch] REC 824078.357 prio -2.159 can request work 11/03/2021 11:46:05 | NumberFields@home | [work_fetch] REC 1860.297 prio -0.075 can request work 11/03/2021 11:46:05 | | [work_fetch] --- state for CPU --- 11/03/2021 11:46:05 | | [work_fetch] shortfall 0.00 nidle 0.00 saturated 20790.50 busy 0.00 11/03/2021 11:46:05 | Einstein@Home | [work_fetch] share 0.000 blocked by project preferences 11/03/2021 11:46:05 | NumberFields@home | [work_fetch] share 1.000 11/03/2021 11:46:05 | | [work_fetch] --- state for NVIDIA GPU --- 11/03/2021 11:46:05 | | [work_fetch] shortfall 0.00 nidle 0.00 saturated 1855.37 busy 0.00 11/03/2021 11:46:05 | Einstein@Home | [work_fetch] share 1.000 11/03/2021 11:46:05 | NumberFields@home | [work_fetch] share 0.000 blocked by project preferences 11/03/2021 11:46:05 | | [work_fetch] --- state for Intel GPU --- 11/03/2021 11:46:05 | | [work_fetch] shortfall 0.00 nidle 0.00 saturated 1921.76 busy 0.00 11/03/2021 11:46:05 | Einstein@Home | [work_fetch] share 1.000 11/03/2021 11:46:05 | NumberFields@home | [work_fetch] share 0.000 no applications 11/03/2021 11:46:05 | | [work_fetch] ------- end work fetch state -------David only ran a single type of [rr_sim], the first one shown here (which I've left untouched). You can see that it concludes with the last NumberFields task - my CPU app - finishing after 6516 seconds. My hack runs a separate version of [rr_sim] for work fetch, giving a more realistic buffer size of 21,455 seconds, which is reflected in the 'saturated' work fetch figure for the CPU. I haven't done anything special for GPUs - they're normally controlled by the number of GPUs in the system, rather than by app_config.xml. If anyone wants to test it (and has experience of the CPU overfetch we've been discussing), let me know. |
|
704)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103474)
Posted 9 Mar 2021 by Richard Haselgrove Post: I chose to use an integer - allows for future expansion of the rr_sim space ;-) It builds, without warnings and - after some effort - errors. I've re-created the problem case, but I'm going to leave it running on the old app overnight, so I can test properly with fresh eyes in the morning. |
|
705)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103470)
Posted 9 Mar 2021 by Richard Haselgrove Post: OK, plan of action, after yet another report of the same problem (thread 14182, from someone who wants to gripe about the problem, but not to be part of the solution). I'm going to try and work up my hack into something presentable, and let a few trusted beta testers take it for a spin. Windows only, at this stage.
With any luck, and provided real life doesn't intrude, I may have something for you by the weekend. |
|
706)
Message boards :
Questions and problems :
Replacement of Cord for Mac VirtualBox applications
(Message 103466)
Posted 8 Mar 2021 by Richard Haselgrove Post: Confirmed that CoRD has been hard-coded as the only available RDP client since 2012. The downloaded files from the CoRD page that the OP linked also have a 2012 datestamp. The CoRD development project has been terminated. Source: https://github.com/BOINC/boinc/blob/master/clientgui/MainDocument.cpp#L1833 There is an open, but deferred, development issue requesting that this be re-visited: Allow any remote desktop client OSX #2231. I'll add a reminder pointing back here. |
|
707)
Message boards :
Questions and problems :
Problem with "max concurrent" in app config.
(Message 103465)
Posted 8 Mar 2021 by Richard Haselgrove Post: In theory, the current effective cache is enumerated by cycling through all known cached tasks and adding the estimated runtimes to reach a 'saturated' figure for core-seconds, or separately for device-seconds in the case of GPU apps.In general, 'target work buffer' is measured in wall time, and 'shortfall' is measured in core-time. If cache is empty, shortfall=ncpus*buffer.When shortfall variable get increase? Shortfall is the difference between target and saturated. At the moment, tasks from projects with a max_concurrent set are sometimes excluded from saturated, so saturated is low, and shortall is high. You can see the exclusions in the output of rr_simulation. That's a very crude over-simplification from memory. There are numerous ifs, buts, maybes, and edge-cases, but that's the principle so far as I've been able to derive it. |
|
708)
Message boards :
Questions and problems :
Problem with "max concurrent" in app config.
(Message 103453)
Posted 7 Mar 2021 by Richard Haselgrove Post: In general, 'target work buffer' is measured in wall time, and 'shortfall' is measured in core-time. If cache is empty, shortfall=ncpus*buffer. As I did in issue 4117, comment 738194152, the best way for this problem is to use the GUI event log configurator in BOINC Manager. Check both rr_simulation and work_fetch_debug. Click apply: uncheck both, count to five, and save. You should then have one, or at most two, cycles of each. Drop the complete segment (no edits) in a dropbucket somewhere, give me a link, and I'll start looking in the morning. It'll take some time. |
|
709)
Message boards :
Questions and problems :
Problem with "max concurrent" in app config.
(Message 103448)
Posted 7 Mar 2021 by Richard Haselgrove Post: CPU work request: 38618.23 seconds; 0.00 devicesIs not, of itself, a significant over-allocation of work. If the problem is that the human eye sees no need for extra work to be requested, but the BOINC eye does, we need evidence of that. At first glance, your comments suggest something similar to the issue I've been discussing most recently with Raistmer in GPU tasks skipped after scheduler overcommits CPU cores, and before that in Client: loses track of new work requirements. Those are complex problems, and they need forensic teamwork to address. Who's in? |
|
710)
Message boards :
Questions and problems :
Problem with "max concurrent" in app config.
(Message 103440)
Posted 6 Mar 2021 by Richard Haselgrove Post: I can only assume the server is saying "er.... no, you have loads to do, you ain't getting any more".The actual wording in the log would be "This computer has reached a limit of tasks in progress", after an attempted work fetch. If the project has set such a thing. |
|
711)
Message boards :
Questions and problems :
Problem with "max concurrent" in app config.
(Message 103436)
Posted 6 Mar 2021 by Richard Haselgrove Post: Well, on those figures, Kryptos is always going to be the highest priority for work fetch - it'll always be under-represented compared to its resource share (and that won't take any notice of cc_config.xml, either). Tweak your resource shares to more closely match what you actually want to run. |
|
712)
Message boards :
Questions and problems :
Problem with "max concurrent" in app config.
(Message 103435)
Posted 6 Mar 2021 by Richard Haselgrove Post: And going for a walk on the internet is a bit wobbly.You can get newspapers on t'internet nowadays lad!There are some things one might find rather uncomfortable or unpleasant to do with an i-pad or the like......Wrapping up one's fish and chips being but one of them ;-) |
|
713)
Message boards :
Questions and problems :
Problem with "max concurrent" in app config.
(Message 103432)
Posted 6 Mar 2021 by Richard Haselgrove Post: Sorry, I was out for a walk, fetching the newspaper. The key 'starter for ten' is <sched_op_debug> - that's quiet enough to leave running all the time. The key thing to post here is the number of seconds of work requested, and the number of (estimated) seconds returned. Depending on the answer, we might start asking questions about DCF, and even for a single (PLEASE - only one) cycle of <work_fetch_debug>. DON'T leave that one running! I don't know the project, so I can only go on what you report. |
|
714)
Message boards :
Questions and problems :
Client asks for work but get 1 fictive task in return
(Message 103419)
Posted 4 Mar 2021 by Richard Haselgrove Post: Not saying it happened on this occasion, but it is entirely possible to receive a new task from Einstein with no visible downloads. It only happens with the Gravity Wave search, which uses locality scheduling. Even for your first task, you need to download something like 48 data files, up to 200MB. But once you've got them, the server tries, and - after a bug was fixed last week - often succeeds in sending you a new 'task' which is just a set of parameters to specify how to analyse the data which you've already got. |
|
715)
Message boards :
BOINC Manager :
Boinc 100% disk activity, locks up
(Message 103398)
Posted 3 Mar 2021 by Richard Haselgrove Post: Edit: actually 3 units running, 2 on the single core and 1 on the GPU.Ah. You hadn't mentioned a GPU. That brings O2MDF and FGRPB1G into play as well, with a wider range of dates. My understanding is that for the tasks using the large data files - the Gravity Wave tasks - the first action is to read and unpack the raw data. Possibly, the portions related to the parameters of the current task have to be extracted. But only after the data pre-processing is complete, can the mathematical processing begin. Only the Einstein team would be able to guide you through all that. |
|
716)
Message boards :
BOINC Manager :
Boinc 100% disk activity, locks up
(Message 103395)
Posted 3 Mar 2021 by Richard Haselgrove Post: The point I was trying to make was that Einstein has more than one type of work, and they change over time. Looking at their applications page today, they have two applications for CPUs at the moment: Gravitational Wave search O2 Multi-Directional (O2MD1) - started 7 Dec 2020 Gamma-ray pulsar search #5 (FGRP5) - started 1 Aug 2017 Given the dates, it's possible that there's a particular problem with the O2MD1 app on your hardware. Have you discussed this on the Einstein message board yet? (BTW, you'll see that the Windows version of O2MD1 was revised on 1 Feb 2021. You can read about that here - it won't affect your problem). |
|
717)
Message boards :
BOINC Manager :
Boinc 100% disk activity, locks up
(Message 103387)
Posted 3 Mar 2021 by Richard Haselgrove Post: In that case, it would also help to consider which of the many sectors of scientific research that Einstein comprises were active at the time. Their Gravity Wave search, in particular, relies on having something like 48 downloaded data files, some 150MB - 200MB of raw data, present while the task is running. I don't have figures for the effect this has on memory usage by the application, but I suspect it can't be small. |
|
718)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103363)
Posted 2 Mar 2021 by Richard Haselgrove Post: Just curious:No. 02/03/2021 12:33:51 | Einstein@Home | [sched_op] CPU work request: 0.00 seconds; 0.00 devices 02/03/2021 12:33:51 | Einstein@Home | [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 02/03/2021 12:33:51 | Einstein@Home | [sched_op] Intel GPU work request: 1059.49 seconds; 0.00 devicesThe client will report all app_versions it has available (read sched_request.xml!), but the server will decide what to send. Edit: at the moment, I'm only choosing to run one of the two available intel_gpu apps, but that's a preference set on the project, not in the client. |
|
719)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103359)
Posted 2 Mar 2021 by Richard Haselgrove Post: But first of all, why simulation should be skipped if concurrent limit in place?https://github.com/BOINC/boinc/commit/40f0cb44f4fcd11eb2789408dfc868de63e42242 Buried in all that mucking about is: - work fetch: if a project is at a max concurrent limit,
don't fetch work from it.
The jobs we get (possibly) wouldn't be runnable.
NOTE: we currently provide max concurrent limits
at both project and app level.
The problem with app level is that apps can have versions that
use different resources.
It would be better to have limits at the resource level instead.We still don't have limits at the resource level: Project-level configurationmax_concurrent is available at the <app> level. plan_class (which implies resource) is available at the <app_version> level. Never the twain shall meet. These days, David is operating as a coder, but he's alienated all the skilled program designers. Coding without designing first is a recipe for failure. |
|
720)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103334)
Posted 1 Mar 2021 by Richard Haselgrove Post: Part 2: In the "work_fetch" case (only, I think), we have to stop it ignoring max_concurrent. The current test and action are at #L340 (test) and #L299 (action, both in client/rr_sim.cpp). The action is } else if (p->pwf.at_max_concurrent_limit) {
rsc_pwf.pending_iter = rsc_pwf.pending.erase(
rsc_pwf.pending_iter
);That 'erase' has to go, in the work fetch case.The tricky bits are: 1) Which mode we're in - sched or fetch - are passed in a const char* variable called "why". Why, indeed? I'd define them as symbolic integer constants, I think. 2) There's an optimisation in rr_sim.cpp, #L662 that skips all the heavy lifting if we've done all the heavy lifting already this second. But that has to come out if we're doing it for the other reason this time. I think the only 'functional' code that remains after all my botched hacks are the 'erase' action in L299, and the removed optimisation in L662. But that damned 'why' variable has to be passed through several subroutine calls. |
|
721)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103333)
Posted 1 Mar 2021 by Richard Haselgrove Post: EDIT: but I don't quite understand, in title post there you say "patched version". So patch exists already? Where is the problem then?Well, it exists on one test machine here. Did I say I hate this language? I hacked it: it's a horrible hack, and I'm not going to try to display it (much of the horribleness is because David passed a text string as a debug message: I needed to use it as a test for action choices). And it's three months ago, and I've forgotten most of what I hacked. But here goes... The client does a round-robin simulation on the current cache, to check what's about to happen. It's called in two distinct places: cpu_sched.cpp, #L882 ("rr_simulation("CPU sched");") work_fetch.cpp, #L659 ("rr_simulation("work fetch");") (end of part 1, "too many links" for the anti-spam watchdog) |
|
722)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103330)
Posted 1 Mar 2021 by Richard Haselgrove Post: Yes, that's the conclusion I reached in early December, and wrote up in the later stages of #4117 - specifically, https://github.com/BOINC/boinc/issues/4117#issuecomment-744042005. The patched version I mention there is still running fine. The calculation of current cache size is done in rr_sim, which is used for two distinct purposes: 1) CPU sched - checking for missed deadlines etc. 2) Work fetch - seeing if there's a shortfall. If a task isn't going to run at all (well, not yet) because the project is at max_concurrent, it's excluded from rr_sim. At the time that code was added, we weren't going to fetch at all from that project (see comment above linked comment). So it didn't matter. But the code was changed again. Now, projects at max_concurrent can fetch work. But rr_sim wasn't updated. The exclusion of max_concurrent tasks from the work fetch version of rr_sim is at least in part (and I suggest the major part) responsible for the over-calculation of shortfall that you're seeing. What happens next depends on what you're trying to achieve. If you want your personal copy of BOINC to run Einstein smoothly - reduce your cache (especially 'additional' days), and be careful (minimal) in your use of max_concurrent If you want BOINC to work better for everyone - join BOINC's coding team, and help persuade David to solve the problem he created. I find the C++ programming language, and David's telegraphic coding style, intensely difficult to follow - I was raised on Algol, 60 and W. And he's stopped talking to me. See if you can find the magic button to make him listen. |
|
723)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103318)
Posted 28 Feb 2021 by Richard Haselgrove Post: And another observation: time to complete of CPU tasks is changing on GPU task completion. Why? Different devices... Which project? If Einstein: they boycotted Credit New. They're still using the original DCF - and it's single-valued. All tasks, of whatever types, belonging to the project are adjusted. |
|
724)
Message boards :
Questions and problems :
BOINC unable to access internet from Windows 10
(Message 103278)
Posted 27 Feb 2021 by Richard Haselgrove Post: Sorry for not responding sooner as I did not receive notifications of prior postings.You can request notifications by clicking the 'subscribe' button at the top of this message list. But it's not automatic. |
|
725)
Message boards :
Questions and problems :
BOINC unable to access internet from Windows 10
(Message 103277)
Posted 27 Feb 2021 by Richard Haselgrove Post: Only 32-bit versions of BOINC working, and 64-bit versions failing, on low power intel CPUs is a known problem, and we know how to work round it. Hold on while I look up the earlier references. Noting for future reference: https://github.com/BOINC/boinc/issues/2470 To decode the rather obscure language in that reference: we found that 64-bit BOINC crashed when it tried to access 'secure' (https) websites. This was due to BOINC being distributed with old versions of some security files. BUT: that was in 2018. A change was made at the beginning of 2020, and all BOINC versions from v7.16.4 onwards should be shipped with the updated files, and should work properly. If you're having a problem with (one only?) copy of v7.16.11, we may have to look further. |
|
726)
Message boards :
Questions and problems :
BOINC Manager not showing tasks
(Message 103244)
Posted 26 Feb 2021 by Richard Haselgrove Post: Ouch. That's why I also copy and paste, then amend. Not totally foolproof but it saves me relying on my memory.Trouble is, copy'n'paste with incomplete edit can be even worse... #3249 fixes #3247. |
|
727)
Message boards :
Questions and problems :
BOINC Manager keeps attaching to a different account
(Message 103239)
Posted 26 Feb 2021 by Richard Haselgrove Post: AFAIK not IP address as this may change.I was thinking of the case where the server was trying to identify a machine from garbled information. https://github.com/BOINC/boinc/blob/master/sched/handle_request.cpp#L78 // find the user's most recently-created host with given various characteristicsI was forgetting that the computer name is the most fundamental of all... :-( |
|
728)
Message boards :
Questions and problems :
BOINC Manager keeps attaching to a different account
(Message 103236)
Posted 26 Feb 2021 by Richard Haselgrove Post: I'm finding it hard, from my knowledge of how BOINC is designed to work, to see how this solution could relate back to the original problem. I've had a look through my own registry, and in all the - multitude of - entries, I haven't seen a single one which relates to a user account key. Those are all stored in the files in the BOINC data folder. The data folder - by design - does not get deleted when the BOINC programs are uninstalled. This is to allow for version upgrades: an upgrade requires the removal of the old version, and its replacement by the files for the new version. The plan is for the user's data to remain unchanged during this process, even to the extent of keeping part-processed work intact, and restarting it at the exact point the previous version left it at. All of which is a long way of saying that this may be a temporary fix - don't be too disheartened if the problem returns in the future. One thing that hasn't been discussed so far is the potential role of project servers in all this. The BOINC server code contains tools which try to re-unite a lost computer with its owner's account. From memory, it considers things like hardware configuration and IP address - there may be others. Might it be a possibility that one of the projects you're attached to is using a buggy version of that code? It would be a hard one to track down, but the first step would be to make a note of the computer's 'HostID' or 'Computer ID' at each of the projects it is attached to - it will be different in each case. Check it locally first ('Properties' button on the 'Projects' tab, BOINC Manager, Advanced view). Then check that the same number is used on your own account on each project's web site. Then, if it changes again to the old account again in the future, check the list of HostIDs again. I would expect that at least one - hopefully just one - will also have changed. Go to the project web page, and see what account the new HostID is attached to. That would give us some clues to work with. |
|
729)
Message boards :
Projects :
bad news everyone, results of the BOINC Collatz project are invalid
(Message 103233)
Posted 25 Feb 2021 by Richard Haselgrove Post: My concern is that, whatever the credit scale that the project - any project - has chosen to pay credits at, David converts them back into GFlops at the defined, Cobblestone, rate. And then claims that the resulting figure is a true measure of the contribution that volunteer computing collectively makes to science. In the case of the project which started this discussion, that claim is quite possibly fraudulent. But It's David's reverse calculation from credit to flops which is fraudulent, not the project's way of allocating credits in the first place. I wrote it up once at SETI@Home. |
|
730)
Message boards :
Questions and problems :
BOINC Manager not showing tasks
(Message 103185)
Posted 24 Feb 2021 by Richard Haselgrove Post: The other, more permanent, way of fixing this is to add the owner to the boinc security group: sudo usermod -a -G boinc ownernameAnd reboot |
|
731)
Message boards :
BOINC client :
Stuck on "BOINC screensaver loading"
(Message 103138)
Posted 23 Feb 2021 by Richard Haselgrove Post: I think it's highly unlikely to be "the task" which causes this, but more likely to be "the type of task". Try to gather information about which type of task is running when this happens. |
|
732)
Message boards :
The Lounge :
Apple M1 & BOINC
(Message 103126)
Posted 22 Feb 2021 by Richard Haselgrove Post: Bernd Machenschalk (administrator of Einstein @ Home) has announced that they have started beta testing the first (to my knowledge) native Apple silicon science application for BOINC: We do have the first native app version for Apple Silicon (M1) available. It is a FGRP5 app. So with a new M1 Mac and the latest "universal" BOINC Client you should get native work for your machine. Note that this build is pretty hackish, so far I don't even know whether it runs on any other machine than it was built on. So far the app completed a handful of tasks, all pending validation. That's also something I awaiting eagerly, whether the results of this app really validate. Oh, and yes, this is a "Beta" App, so make sure in your preferences that you want to receive "Beta test work".Feel free to test, and report feedback direct to Einstein. |
|
733)
Message boards :
BOINC client :
Move Client to another drive
(Message 103102)
Posted 19 Feb 2021 by Richard Haselgrove Post: To save me typing it all over again, try message 98168 |
|
734)
Message boards :
Android :
BOINC Android client 7.16.16 released
(Message 103097)
Posted 19 Feb 2021 by Richard Haselgrove Post: Not quite ;-) |
|
735)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 103082)
Posted 18 Feb 2021 by Richard Haselgrove Post: People who are invited to be vaccinated, make an appointment, go to get vaccinated at the selected location and time. THEN refuse to be vaccinated for no good reason at all.Probably the same people who go to a polling station to spoil their ballot papers. It's called 'abstaining in person'. I did that once. A particularly pointless election, where nobody knew any of the candidates. I wrote a long version of "none of the above", complaining about the waste of money. Spoke with the person in charge of the election, later (in the pub) after he was off duty. "You spent a long time writing in that voting booth", he said. And a few moments later: "A lot of people did that". |
|
736)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 103063)
Posted 18 Feb 2021 by Richard Haselgrove Post: And that has something to do with the message board timing out?Well, if it involved loading new software, and then reinstating a code database onto a replacement server that's trying to handle too much already, it might. Just a suggestion. |
|
737)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 103061)
Posted 18 Feb 2021 by Richard Haselgrove Post: Raistmer emailed Eric K yesterday, to ask him to re-instate the SVN repository holding all the SETI application code...Who is DoS'ing the SETI messageboard?No idea, I'm seeing the same. https://einsteinathome.org/content/igpu-beta-app-validation-issue#comment-183576 |
|
738)
Message boards :
Promotion :
Is it ok if I put the BOINC logo on my car?
(Message 103060)
Posted 18 Feb 2021 by Richard Haselgrove Post: I have stickers for organisations that I'm a member of on my car - they get me free parking at some venues. Other people have bumper stickers - either jokes, or showing support for environmental or political causes. I doubt something like that would cause any problems, and might spark some interesting conversations. But I'd steer clear of a giant-sized paint job that might imply sponsorship or endorsement by the wider BOINC community. Might not look good if you found yourself in an embarrassing position while driving it! |
|
739)
Message boards :
Android :
BOINC Android client 7.16.16 released
(Message 103052)
Posted 18 Feb 2021 by Richard Haselgrove Post: You could look at https://github.com/BOINC/boinc/milestone/24?closed=1, but I don't think it'll be very helpful. |
|
740)
Message boards :
Android :
BOINC Android client 7.16.16 released
(Message 103032)
Posted 16 Feb 2021 by Richard Haselgrove Post: Saw that. He probably wouldn't have bothered if I'd reported it myself... One down, still 11 to go. |
|
741)
Message boards :
Android :
BOINC Android client 7.16.16 released
(Message 103025)
Posted 16 Feb 2021 by Richard Haselgrove Post: Very wise. Time to hand the shift over to the upper-left USA. See if they can sort out some fixes by the morning! |
|
742)
Message boards :
Android :
BOINC Android client 7.16.16 released
(Message 103022)
Posted 16 Feb 2021 by Richard Haselgrove Post: Another one for you. When downloading the new version from the BOINC website (as advised in the announcement), the BOINC download page advises you to go to the Google play store instead. Left hand, right hand? |
|
743)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103005)
Posted 15 Feb 2021 by Richard Haselgrove Post: It's probably easiest on Einstein if you work out a reasonable target for what you can work off before you start hitting deadlines (given your various resource constraints), and then abort the excess sooner rather than later. No shame in that, if it's going to happen anyway, and better for the project if it happens sooner, so the resends can go out quickly. If the website says you have more in progress than you can see locally, brace yourself for resent lost tasks - Einstein has that setting switched on permanently. |
|
744)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 103002)
Posted 15 Feb 2021 by Richard Haselgrove Post: Well, Einstein's 384 limit is implemented as a single global value. So it's a pretty blunt weapon, more of a long stop than something you should expect to ride on day by day. Ease off any app_config.xml files that contain a <max_concurrent> value, and I'm not sure how Einstein implemented the back stop: if it's saying "daily limit", you'll probably have to wait until midnight (German time) to start filling the cache with GPU tasks. "Max tasks in progress" might have been better here, but I suspect their rathed old server doesn't have that option. |
|
745)
Message boards :
Questions and problems :
x86 or x64 - what is faster?
(Message 102998)
Posted 15 Feb 2021 by Richard Haselgrove Post: Also the better addressing of large memory spaces can help when pulling data from memory rather than diskThus encouraging bloatware, instead of mean and clean programming - and slowing things down again. |
|
746)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 102995)
Posted 15 Feb 2021 by Richard Haselgrove Post: Is it possible to fix this bug ?Well, you're a programmer, so you know that it's always possible to fix a bug - but it's not always easy. There are two prerequisites: 1) That somebody carefully analyses the cause and location of the bug, so that it can be expressed in programming terms. 2) That a programmer familiar with the existing code chooses to devote enough time and effort to re-writing that code, with sufficient care to avoid introducing a new set of bugs. I've tried to address (1) by writing up https://github.com/BOINC/boinc/issues/4117, and supplying as much information (simulation run, extensive logs, etc.) as I can. The first thing to get clear in your head is that this is a client problem - your client repetitively requests new work, in spite of receiving everything it asks for (Einstein is good at that). If you haven't already, check your local Event Log (or history in stdoutdae.txt), and see the repeated requests/replies. I have to confess to a total failure to move David on to prerequisite (2). |
|
747)
Message boards :
GPUs :
How to increase GPU performance?
(Message 102957)
Posted 10 Feb 2021 by Richard Haselgrove Post: By all means try it, but be prepared for disappointment. Have a look at the User manual, especially the section on Project-level configuration. You can create a configuration file that specifies how much of the GPU will be scheduled to run the app. Note that this doesn't control the app: it just means that BOINC allows extra copies to run on the same hardware. |
|
748)
Message boards :
GPUs :
How to increase GPU performance?
(Message 102955)
Posted 10 Feb 2021 by Richard Haselgrove Post: The efficiency of the application, in this situation, is very much determined by the way it is coded. Eric Driver, the administrator of NumberFields, is new to GPU programming, and this is his first attempt at it. I also tried this app, and noticed a different problem: if the GPU is also providing the video signal for the monitor, screen refresh becomes noticeably sluggish. Both of these are symptoms of an immature application. They can't be cured externally, although sometimes workrounds can mitigate the effects. I would expect them to be improved in later releases. Eric has posted about releasing the source code on Github once he's sure it's working properly. If you are an experienced GPU programmer, perhaps you could help him out? |
|
749)
Message boards :
Projects :
How to assign profile to project on a specific machine
(Message 102856)
Posted 7 Feb 2021 by Richard Haselgrove Post: MilkyWay certainly has the full set of locations available: Preferences for this project MilkyWay@home preferences. CPDN is going through a minimalist phase, but the framework is there, too. |
|
750)
Message boards :
Projects :
How to assign profile to project on a specific machine
(Message 102850)
Posted 7 Feb 2021 by Richard Haselgrove Post: Through project web sites. Your 'account' page at any given project website has an entry for "Computers on this account", with a link to 'view' them. For each computer linked to the project, you can see the host name, and a link for that computer's details. [Other users can't see the name - just an ID number - and a limited amount of detail] At the foot of the page of details, you can see the 'Location' that computer is currently assigned to: this is a drop-down list, and you can change it to another profile. Click the 'Update location' to confirm a change. |
|
751)
Message boards :
Questions and problems :
x86 or x64 - what is faster?
(Message 102830)
Posted 6 Feb 2021 by Richard Haselgrove Post: Depends who's writing the program. But in principle, and provided everything fits within the 4 GB address space limit, and the same hardware can handle either mode, x86 should be quicker because the memory transfers will be smaller. |
|
752)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 102808)
Posted 4 Feb 2021 by Richard Haselgrove Post: I've always thought it pretty daft that a tiny little part of that island is "ours". When I were a lad, my friend and his family left Ireland to avoid the conflicts at the time. His words (and he's Irish obviously) - "Only the Irish would be stupid enough to start a war against themselves". More likely it's religion, Christians fighting er... Christians.Read about the 'Plantations of Ireland'. It was a deliberate act of colonisation, and that's never ended well. |
|
753)
Message boards :
BOINC client :
Stuck on "BOINC screensaver loading"
(Message 102775)
Posted 1 Feb 2021 by Richard Haselgrove Post: I've had a positive response from Oliver this morning, and he confirms the problem. They've re-issued the Gravity Wave application as v2.09 (for CPUs only, for Windows only) and the graphics should work properly now. |
|
754)
Message boards :
BOINC client :
Stuck on "BOINC screensaver loading"
(Message 102748)
Posted 30 Jan 2021 by Richard Haselgrove Post: I think I've found a configuration error at Einstein. <app_version>
<app_name>hsgamma_FGRP5</app_name>
<version_num>108</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>1.000000</avg_ncpus>
<flops>4502389234.875296</flops>
<plan_class>FGRPSSE</plan_class>
<api_version>7.3.0</api_version>
<file_ref>
<file_name>hsgamma_FGRP5_1.08_windows_intelx86__FGRPSSE.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>einstein_S5R6_3.01_graphics_windows_intelx86.exe</file_name>
<open_name>graphics_app</open_name>
</file_ref>
<file_ref>
<file_name>fftwf-wisdom_FGRP5_1.08_windows_intelx86.exe</file_name>
<open_name>fftwf-wisdom_FGRP5_1.08_windows_intelx86.exe</open_name>
</file_ref>
</app_version><app_version>
<app_name>einstein_O2MD1</app_name>
<version_num>208</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>1.000000</avg_ncpus>
<flops>4502389234.875296</flops>
<plan_class>GWnew</plan_class>
<api_version>7.3.0</api_version>
<file_ref>
<file_name>einstein_O2MD1_2.08_windows_x86_64__GWnew.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>einstein_O2MD1_2.08_graphics_windows_intelx86_64.exe</file_name>
<open_name>einstein_O2MD1_2.08_graphics_windows_intelx86_64.exe</open_name>
</file_ref>
</app_version>The second one lists a 'graphics' application, but doesn't declare it as a "graphics_app". I think that means that the 'Show graphics' will work while a LAT task is running, but not while an h1 task is running. Time will tell.I've sent a PM to Oliver, but probably not before he went home for the weekend. Today, they've got a 504 Gateway Timeout error to deal with on uploads - that will get priority. |
|
755)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 102745)
Posted 29 Jan 2021 by Richard Haselgrove Post: OK, two issues arise from that. 1) You mention "limiting the number of concurrent jobs" - I assume by setting one or other of the possible <max_concurrent> flags in one or more app_config.xml files. I have reported a possible bug in the boinc client - the component that lives on, and controls the work done on, your computer. The effect of the bug makes your computer ask, repeatedly, for more and more work. If you use app_config files, and especially if you use max_concurrent, keep a close eye on the Event Log and be prepared to step in if you see repeated requests for the same amount of work over a short period of time. By 'short', I mean really short: I think Rosetta would request work every twelve seconds. 2) "Cache size" - on these, and other BOINC message boards - is a shorthand for the configuration parameters Store at least 0.01 days of work Store up to an additional --- days of workYou can set those either on your account page at a project web site, or directly through BOINC Manager on your computer. You can't use both techniques at the same time. If you use both, the ones on your machine will take precedence, and the ones from a project website will be ignored. IIRC, Rosetta has a deadline of 3 days from the time new tasks are downloaded. Your cache (sum of 'at least' and 'additional') should always be less than half that, and while you're working out what's going wrong, keep it even smaller. I personally use 0.25 + 0.05 days: that means 'report the work you've completed, and get a new batch, about once an hour', and the whole lot should turn over in about 6 - 8 hours. Adjust those figures if you have internet constraints, if you don't let the computer run 24 hours a day, or suchlike. |
|
756)
Message boards :
BOINC client :
Stuck on "BOINC screensaver loading"
(Message 102742)
Posted 29 Jan 2021 by Richard Haselgrove Post: Some do. <file_ref>
<file_name>einsteinbinary_BRP4_1.34_windows_x86_64__opencl-intel_gpu-Beta.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>einsteinbinary_BRP4_1.00_graphics_windows_intelx86.exe</file_name>
<open_name>graphics_app</open_name>
</file_ref>
<coproc>
<type>intel_gpu</type>
<count>1.000000</count>
</coproc>But I don't think the ones for the 'big' GPUs do. I could look them up, but I don't use them myself.
|
|
757)
Message boards :
Questions and problems :
Definition of tasks as CPU or GPU
(Message 102733)
Posted 28 Jan 2021 by Richard Haselgrove Post: GPUs can do very fast maths, but are not good at the management and control - decision-making - parts of a scientific application. Different applications, different hardware, different operating system - and different programmers - have different ways of coping with this. If a GPU application keeps a CPU running at 100%, the CPU is (in most cases - not every time: check the small print) very lightly loaded. It's in what's called a busy-wait state, where it asks - millions of times a second - "Do you need me for anything?". Usually, the answer is no, but every few milliseconds, the GPU will want to have the old results taken away, and to be given something new to do. For efficiency, it's important that the CPU is alert to respond immediately. To answer your direct question: yes, BOINC will count that as using a whole CPU, and it'll prevent a pure CPU task from running as a result. But it won't be using power-hungry components like the floating point unit hard, if at all. If you reduced the CPU loading for thermal purposes, you might be able - careful! - to get away with increasing the permitted %age a bit. |
|
758)
Message boards :
Questions and problems :
GPU tasks skipped after scheduler overcommits CPU cores
(Message 102728)
Posted 28 Jan 2021 by Richard Haselgrove Post: Yes, that's a thing: client/cpu_sched.cpp, line 1250. It's designed to happen, but only when CPU tasks have been given unusual precedence. We need to work out why that's happened. There are two basic cases when CPU tasks are promoted before GPU tasks. 1) When a CPU task needs to run in 'high priority' or 'Earliest Deadline First' (EDF) mode, to avoid missing its deadline. 2) When a CPU task is running 'multi-threaded' (MT). GPU tasks can still run, but only to the extent that all GPU tasks between them are assessed to need less than one complete CPU between them for support purposes. A GPU task which takes the total CPU requirement beyond 1 would be blocked. Would either of those cases apply here? In your simulation, six Rosetta tasks are running in EDF, and the first CPU sched debug ends all CPUs used (12.84 >= 12), skipping OPN1_0032301_05289_0But I've sometimes found that the simulator doesn't exactly reproduce the decisions of the real-world client. The Einstein NVidia task is allocated a whole CPU, and the Collatz intel_gpu task is allocated 0.83 CPU, so something has to be freed up to allow both to run. Rosetta tasks have short deadlines, so will need EDF more often. Reducing your cache size to well below the Rosetta deadline may well help. |
|
759)
Message boards :
BOINC client :
Request: Set Vbox to low piority
(Message 102700)
Posted 25 Jan 2021 by Richard Haselgrove Post: This issue was on the BOINC Roadmap 2020 - and yes, I do know it's 2021 now, and it hasn't happened. There are two parts to this. First, are you thinking about the priority at which the VBox engine itself runs, or the priority at which the science apps inside the container run? It used to be possible to set the science apps to run at a lower priority than the container (as happens with BOINC itself), but the VBox developers removed this feature from their code. Without their support, it's not easy to see how that would be possible now. |
|
760)
Message boards :
Projects :
News on Project Outages
(Message 102694)
Posted 25 Jan 2021 by Richard Haselgrove Post: Projects which have closed down usually - eventually - make that visible publicly. Either they post a polite notice of thanks on their website, or their domain registration lapses and gets taken over by a spammer.If the closed projects don't tell that they have closed (there is a projects email list they can email that to), nor the new projects say that they are new and in production, it's no wonder the developers don't know about these things. New projects that are still in alpha/beta stage will generally not be added to the projects list.So, create a thread in which volunteers (that are more "ready" to projects than developers) can report the closed projects is a bad idea? If you come across one of those, it's worth posting the information to the mailing list Jord mentioned. But a message board thread is an inefficient tool in those circumstances - the people you most want to be aware of and act on the information, are the least likely to read it here. |
|
761)
Message boards :
Projects :
News on Project Outages
(Message 102690)
Posted 25 Jan 2021 by Richard Haselgrove Post: Is it possible to update only this xml file??A fresh copy of 'all_projects_list.xml' is fetched by your client every 14 days [*]. Whatever is contained in the most recent copy will be displayed by the 'attach to project' wizard in BOINC Manager. But most of the time, the new copy will be identical to the previous one. The file is updated by a human being, and it's their judgement and knowledge which determines whether a new project is deemed 'ready for the big time' after testing, or has finally died after a period of unresponsiveness. That information flow is often sparse. The same file [**] is used to drive the display at https://boinc.berkeley.edu/projects.php, so in theory that should be as up-to-date (or not) as the BOINC Manager display. * Unless you have an obscure and completely unrelated configuration flag set. Grumble. ** Strictly, a different version of the same file, and they should be updated together. But that process is also liable to human error. |
|
762)
Message boards :
Questions and problems :
gui_rpc_auth.cfg exists but can't be read. Check the File permissions.
(Message 102684)
Posted 24 Jan 2021 by Richard Haselgrove Post: Add your user name to the boinc user group. It's a false message. |
|
763)
Message boards :
Questions and problems :
boinctui on pi zero - calculations fail
(Message 102672)
Posted 23 Jan 2021 by Richard Haselgrove Post: That's exactly what was being requested by way of a log file - you've found the right place. These lines help us to get further information from the project web sites. Jan 23 16:38:40 pizero boinc[563]: 23-Jan-2021 16:38:40 [Einstein@Home] URL http://einstein.phys.uwm.edu/; Computer ID 12868528; resource share 100Just taking the first, it gets us to: TASKS FOR COMPUTER 12868528 (sorry - their caps, not mine!) As you surmised, all the tasks finished early with an error - that was the reason for them failing to produce an output file. The few I've checked all say process exited with code 127 (0x7f, -129)</message> <stderr_txt> ../../projects/einstein.phys.uwm.edu/einsteinbinary_BRP4_1.06_arm-unknown-linux-gnueabihf: relocation error: ../../projects/einstein.phys.uwm.edu/einsteinbinary_BRP4_1.06_arm-unknown-linux-gnueabihf: symbol h_errno version GLIBC_PRIVATE not defined in file libc.so.6 with link time referenceI don't know enough about Linux library linkage to take you any further, but it's a starting point. |
|
764)
Message boards :
BOINC Manager :
BOINC for M1 Macs
(Message 102637)
Posted 21 Jan 2021 by Richard Haselgrove Post: Slight OT: BOINC can run natively on M1 Macs...?There is a version of BOINC itself available, but that's not the end of the story. BOINC itself does nothing: it's just an enabler for the science applications written and provided by the projects. Until the projects have developed new applications for the changed environment, and utilising the features of the new Apple silicon, there's not must point in using a new version of BOINC. We had a discussion in the lounge area towards the end of last year: Apple M1 & BOINC. That seemed to suggest that even the new version of BOINC isn't yet able to recognise that the Apple emulation layer (Rosetta 2) can run old science apps written for efficient running on Intel CPUs, so there's still some way to go. The Einstein project has now taken delivery of a test machine, and is starting to work out how best to deploy it. |
|
765)
Message boards :
Questions and problems :
Numberfields@home is down?
(Message 102608)
Posted 17 Jan 2021 by Richard Haselgrove Post: Not your problem - it's the project. See Eric Driver's post at Message 102569. |
|
766)
Message boards :
Questions and problems :
Can't get boinc-client to start computing
(Message 102598)
Posted 16 Jan 2021 by Richard Haselgrove Post: Sorry to be a noob - and I have searched the internet for 30 mins to try and find more info - also looked in /var/lib/boinc-clinet and /etc/boinc-client but can only find GPU detect logs and not stdoutdae.txt which I understand is the event log file? In fact I searched my entire machine and couldn't find it... must be doing something wrong.How did you install BOINC? Recent repository versions have transferred what used to be stdoutdae.txt into the system journal. Edit - try variations on journalctl --boot=-1 --unit=boinc-client |
|
767)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102590)
Posted 15 Jan 2021 by Richard Haselgrove Post: You mean I was right?!?It took some time to turn your general remarks in message 101928 into a reportable bug report - the question 'why?' is important when considering reports like that. But yes, you were right ;-) |
|
768)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102576)
Posted 15 Jan 2021 by Richard Haselgrove Post: #4151 |
|
769)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102573)
Posted 14 Jan 2021 by Richard Haselgrove Post: Yup. MilkyWay confirms my suspicions - it's a generic bug in BOINC, or at least a 'design flaw': the sort of thing that has to be picked up by thinking, not by automated syntax checkers. I ran the test machine down to empty - no tasks remaining. Then allowed MilkyWay to ask for work. 14/01/2021 19:30:21 | | [work_fetch] Request work fetch: project work fetch resumed by user 14/01/2021 19:30:22 | | [work_fetch] target work buffer: 8640.00 + 864.00 sec 14/01/2021 19:30:22 | | [work_fetch] shortfall 38016.00 nidle 4.00 saturated 0.00 busy 0.00 14/01/2021 19:30:22 | Milkyway@Home | [sched_op] CPU work request: 38016.00 seconds; 4.00 devices 14/01/2021 19:30:24 | Milkyway@Home | Scheduler request completed: got 23 new tasks 14/01/2021 19:30:24 | Milkyway@Home | [sched_op] estimated total CPU task duration: 39233 secondsSo I got what I was asking for - or did I? My request was for 9,504 wall-seconds, or fullCPU-seconds. On a quad core with four idle cores, that translated into 38,016 core-seconds. But the tasks I received were MT, using all four cores. So I got 39,233 fullCPU-seconds. 10.6 hours of work, not the 2.64 hours target. A four-fold over supply. BUG: The server estimates the duration at https://github.com/BOINC/boinc/blob/master/sched/sched_send.cpp#L429 (estimate_duration_unscaled) and https://github.com/BOINC/boinc/blob/master/sched/sched_send.cpp#L477 (estimate_duration) The comment says // estimate the amount of real time to complete this WU,but it's not true: there's no 'ncpus' term in either calculation. That comment was written in the days of single-core CPUs, and the answer is in core-seconds. For MT tasks, that's wrong by a factor of ncpus. The more cores your CPU has, the more wrong it is. |
|
770)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102572)
Posted 14 Jan 2021 by Richard Haselgrove Post: Or look at https://github.com/BOINC/boinc/blob/server_release/1.0/1.0.4/version.h. That's where the scheduler picks up the 713 from #define BOINC_MAJOR_VERSION 7 #define BOINC_MINOR_VERSION 13 |
|
771)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102568)
Posted 14 Jan 2021 by Richard Haselgrove Post: See edit! |
|
772)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102566)
Posted 14 Jan 2021 by Richard Haselgrove Post: Milkyway@Home maybe?It's worth a thought. Their N-body tasks (the only MT ones) were in total chaos a few years ago, but I remember someone saying they'd cleaned up their act. All my other projects have given up on me (GPUGrid - end of research run; NumberFields - crashed hard disk; SETI - went to meet its maker; ...) Oh, ********. Server status page says "Upstream server release: 1.0.4", scheduler reply says "[sched_op] Server version 713". So which, and when, is it? I thought I'd asked CERN to sort that one out. |
|
773)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102561)
Posted 14 Jan 2021 by Richard Haselgrove Post: Now that everyone has returned refreshed after the holidays (??!!), I've opened communications with PrimeGrid and had some preliminary discussion with Rytis. They've made some small changes to their private code which allows people to specify a lower level of participation in the MT tasks than "100% of available CPUs". I'm pretty sure that there is a generic flaw in the basic server code used by every BOINC project, but PrimeGrid's is so customised that it's hard to provide convincing proof. Does anyone know of a project which (1) uses 'close to standard' server code, and (2) provides a reliable supply of consistent MT tasks? |
|
774)
Message boards :
Questions and problems :
invalid rpc client password in windows 10
(Message 102523)
Posted 11 Jan 2021 by Richard Haselgrove Post: Sadly, the error message 'invalid rpc client password' is notoriously misleading - it is shown by default whenever an RPC fails, whatever the real reason for the failure. I'm sorry your time has been wasted by this message, but I'm pleased you're out of the woods now. |
|
775)
Message boards :
Questions and problems :
Only 1 out of 2 GPUs working
(Message 102484)
Posted 8 Jan 2021 by Richard Haselgrove Post: reinstalled using 'sudo apt install --reinstall boinc-client, but no luck :(And reinstalling BOINC wasn't what Keith suggested, either. He said reinstall the drivers - and that means the entire driver suite, not just the fan controller. |
|
776)
Message boards :
Questions and problems :
Only 1 out of 2 GPUs working
(Message 102476)
Posted 7 Jan 2021 by Richard Haselgrove Post: My CC config even states to use all gpus:Did you verify that had been pickup up and acknowledged in the event log? 30-Dec-2020 08:56:30 [---] Config: use all coprocessors |
|
777)
Message boards :
Questions and problems :
Per-device project level configuration?
(Message 102467)
Posted 7 Jan 2021 by Richard Haselgrove Post: It was in my mind when I wrote #2992 |
|
778)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 102427)
Posted 5 Jan 2021 by Richard Haselgrove Post: You mean SETI has a cleaner? Doesn't match my memory of Eric's office. |
|
779)
Message boards :
The Lounge :
Apple M1 & BOINC
(Message 102425)
Posted 5 Jan 2021 by Richard Haselgrove Post: The issue at GitHub has started quite a discussion, including input from two Einstein staff. They have 'new Mac' hardware on order, and will be starting to explore it when it arrives. I suspect they might be first... @ JohnD, I notice that your machine hasn't returned any more Einstein work since I asked you to get server logs from the new universal BOINC. It's also still reporting "CPU [' '] lacks feature ' sse '" today. Unless you have another reason for persevering with that version, feel free to revert to the pre-universal BOINC - I think this one has given us enough information now. I'd also quite like to see how the CPU was described under the old version. |
|
780)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 102424)
Posted 5 Jan 2021 by Richard Haselgrove Post: Just gone down.They don't usually run backups at two in the morning ??? |
|
781)
Message boards :
BOINC Manager :
Does 7.16.16 for Linux have a snooze feature?
(Message 102418)
Posted 4 Jan 2021 by Richard Haselgrove Post: It was a decision of the linux repository maintainers to remove the tray icon completely from the Linux builds: apparently it didn't work consistently across the multitude of different Linux desktops. The snooze function, alas, vanished with it. |
|
782)
Message boards :
Questions and problems :
Any way to exempt one project from suspending tasks?
(Message 102405)
Posted 4 Jan 2021 by Richard Haselgrove Post: The 24 core tasks often don't fully utilise all the cores anyway.Precisely. The more cores you have working in parallel, the longer you have to wait at the synchronisation points for the laggards to catch up. I'd even suggest two PG tasks at 10 threads each, plus 4 spare cores for anything else. Keep away from multiples of 4 or 6, else PG will fill the beast and you're back to square 1. |
|
783)
Message boards :
Questions and problems :
Any way to exempt one project from suspending tasks?
(Message 102401)
Posted 4 Jan 2021 by Richard Haselgrove Post: I think I'd constrain the Primegrid task to say 20 threads using app_config.xml, and let CPDN have continuous use of one (or more) of the remaining cores. |
|
784)
Message boards :
The Lounge :
Apple M1 & BOINC
(Message 102395)
Posted 4 Jan 2021 by Richard Haselgrove Post: Yup, that's the sort of gotcha that happens when a manufacturer makes a big design change and everybody else has to scrabble to catch up. Streaming SIMD Extensions (SSE) is a single instruction, multiple data (SIMD) instruction set extension to the x86 architecture, designed by Intel and introduced in 1999 in their Pentium III series of Central processing units (CPUs)And in the twenty years since then, SSE (and many subsequent evolutions later, through AVX) have become the bedrock of high performance computing. Now it's back to the drawing board. Edit: reported upstream as #4137. Many thanks for the report. |
|
785)
Message boards :
News :
Server outage
(Message 102389)
Posted 2 Jan 2021 by Richard Haselgrove Post: David doesn't give a reason, but it's likely to be this: https://technology.berkeley.edu/dc-shutdown Data Center Power Shutdown Systems will likely be down from 8 a.m. on Saturday, Jan. 2 through 6 p.m. on Sunday, Jan. 3 (times are estimated).That's 16:00 UTC Saturday until 02:00 UTC Monday. |
|
786)
Message boards :
The Lounge :
Apple M1 & BOINC
(Message 102386)
Posted 1 Jan 2021 by Richard Haselgrove Post: The Einstein project allows you to see the server's decision-making processes, in the form of the server log for the most recent contact. If you could arrange to make one more request for work with the universal build, and immediately capture the server log (click on the 'time of last contact' link for the host, on the Einstein web site), I'm sure it would help the developers work out what's going wrong. At this stage, very early in this transition from Intel to Apple silicon, it isn't clear whether the bug lies in the new version of BOINC or at the Einstein server. |
|
787)
Message boards :
BOINC Manager :
Boinc Manager won't start on Ubuntu 20.04.1 LTS (Focal Fossa)
(Message 102331)
Posted 25 Dec 2020 by Richard Haselgrove Post: Or, if it doesn't start at all, look for posts about a lock file in your home directory. |
|
788)
Message boards :
Projects :
MLC@home output file absent
(Message 102317)
Posted 24 Dec 2020 by Richard Haselgrove Post: But here we are remote from the scene of the action. "Output file absent" - is merely a symptom. The file is absent because the application never got far enough to write it - it crashed. You have to look, at least, at the task lists on the project website, and then drill down to the logged details of an individual failed task. That may reveal an obvious clue, or it may need to be deciphered by the project's programming team. You won't know until you look. |
|
789)
Message boards :
Projects :
GPUGrid
(Message 102306)
Posted 22 Dec 2020 by Richard Haselgrove Post: For the record: the 'disk full' situation was resolved yesterday (21 December), and the GPUGrid project is now operating normally. |
|
790)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 102270)
Posted 20 Dec 2020 by Richard Haselgrove Post: Ireland and Scotland are thinking about it - Air and ferries only, the roads north are open but forbidden. |
|
791)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 102265)
Posted 20 Dec 2020 by Richard Haselgrove Post: And now Italy ... |
|
792)
Message boards :
Projects :
GPUGrid
(Message 102254)
Posted 19 Dec 2020 by Richard Haselgrove Post: Two messages from Toni, the project admin: Disk full Disks are full. I'm suspending the project. Sorry for the missed credits. |
|
793)
Message boards :
Questions and problems :
Task transfer
(Message 102232)
Posted 18 Dec 2020 by Richard Haselgrove Post: DON'T try to copy separate individual components. The 'slots' are the individual work-spaces for tasks in progress. They are wiped clean at the end of each task, and created / deleted as needed. If you did a clean and orderly shutdown, all the slots would have been empty - well done, that's the best way. You can put the BOINC data anywhere - it doesn't have to be in ProgramData, or even on the system disk - you have a free choice. But that's where it will be if you didn't make an explicit decision to be different. My regular instructions, tried and tested over the years, are: 1) Copy the entire BOINC data structure, subfolders and all, from your demised drive / system image, and put it wherever is most convenient. Personally, I've standardised on D:\BOINCdata 2) Download (or recover from the old machine) a full BOINC installer. That's for the program, not the data. Pick any version you're comfortable with. 3) Run the installer. Pause on the second screen, which tells you where everything will be put. Does that match what you just did? 4a) Yes - the locations are right. Continue as usual. 4b) No - I've chose somewhere new. Click the 'Advanced' button at bottom-right. Using the buttons provided, ensure that the second line - the data line - matches where you've already put the data. You can change the program location as well, but it doesn't matter. Free choice. 5) Finish installation as normal. 6) Launch BOINC Manager. All your old projects, tasks, statistics, options, ID numbers, and anything else you care to think of, should be be in their places and ready to rumble. Start her up! 7) Sit down and enjoy a well earned cup of tea. There's nothing else to do. |
|
794)
Message boards :
BOINC client :
Request: better GPU labelling
(Message 102186)
Posted 15 Dec 2020 by Richard Haselgrove Post: If both cards have been programmed properly, and if both cards are running within their operational envelope, then both cards should compute the same mathematical answer, within tolerance. No problem at all swapping between them midway. |
|
795)
Message boards :
BOINC client :
Request: Ability to change number of cores used when computer in use
(Message 102152)
Posted 14 Dec 2020 by Richard Haselgrove Post: Join the queue on #41. [ I don't even have to look that one up any more :-( ] |
|
796)
Message boards :
BOINC client :
Request: better GPU labelling
(Message 102147)
Posted 14 Dec 2020 by Richard Haselgrove Post: All BOINC GPU detection is done at BOINC startup. I have no evidence of BOINC changing device number mapping while running - although a crash and reboot (or an automated OS update) can change things while no-one's looking. Again, BOINC does not re-assign tasks from one device to another arbitrarily. But there are multiple reasons why a task running on a GPU may stop: end of task-switch interval; pre-emption so a more urgent task can run; project application crash; OS/driver crash; user operation; and probably many more. Some of these may cause the entire task to fail, but often it will become 'waiting to run'. When a device of the same type becomes available, the waiting task will be started on that device. The decision is 'first come, first served': whether it's the same device as last time is irrelevant and ignored. |
|
797)
Message boards :
Questions and problems :
How to install Boinc on Fedora with GPU and SELinux?
(Message 102143)
Posted 14 Dec 2020 by Richard Haselgrove Post: That particular line was added by a user called 'RustyBSD' in 2012. I don't recognise the handle, and no explanation is given. It's also likely that the CUDA driver situation has changed, even since the page as a whole was last updated (2017). I'd just give it a try - ignore that line, and report back with your experiences. Please post any error messages your receive. |
|
798)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 102122)
Posted 13 Dec 2020 by Richard Haselgrove Post: And each extra worker, each attachment point hardware, each welding kit - all adds to the load being supported by the existing cables. And with the likely failure mode being a runaway cascade of individual strands breaking, there wouldn't be time to evacuate workers when the final collapse began. As the video confirmed. |
|
799)
Message boards :
GPUs :
Can I run 1 WU across multiple GPUs?
(Message 102108)
Posted 12 Dec 2020 by Richard Haselgrove Post: I'd suggest running 24 threads one week, and 4 tasks x 6 threads each another week. See which gets most work done overall. [Suggesting you stick to divisors of 12, because a cache memory miss on the 'wrong' CPU is likely to be costly. Might even be worth trying 2 tasks x 12 threads each] |
|
800)
Message boards :
GPUs :
Can I run 1 WU across multiple GPUs?
(Message 102086)
Posted 10 Dec 2020 by Richard Haselgrove Post: ... you can play with the project_max_concurrent settings to prevent a project from gobbling up all your threads, allowing some number of spare threads always available to another project.You can, more particularly, play with the full app_config.xml kit of parts to constrain generic MT jobs to use less than the full CPU. MT tasks don't become more efficient by turning the parallelism volume control up to 11. In fact, I believe that efficiency generally decreases as the number of threads increases. If the threading is managed properly (both avg_ncpus and --nthreads have to be set), MT jobs can co-exist perfectly happily with single-threaded jobs from other projects. |
|
801)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 102079)
Posted 10 Dec 2020 by Richard Haselgrove Post:  I can understand why the 1 and 5 could be on the "line before", and the 4 and 8 on the "line after". But how do they get the 3s to the end? |
|
802)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 102070)
Posted 10 Dec 2020 by Richard Haselgrove Post: It's a similar problem to those described in https://www.theforthbridges.org/forth-road-bridge/maintenance/major-projects/main-cable-inspection/ https://www.theforthbridges.org/forth-road-bridge/maintenance/major-projects/main-cable-acoustic-monitoring/ |
|
803)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 102063)
Posted 10 Dec 2020 by Richard Haselgrove Post: To support this much heavier platform, using the same three towers, additions to the number of cables were made: 6 auxiliary mainstay cables (each of 413 inch diameter) were added from the tower tops (2 per tower) to the platform and 6 auxiliary backstay cables (each of 853 inch diameter) were added from the tower tops to the ground.Those would be some cables! The original has 'three and a quarter inches' and 'three and five eighths inches' respectively (88.5 mm : 92 mm). Another triumph for the US adherence to imperial measurements in a technical setting. NASA would be proud. And another triumph for Word-PDF conversion - it transposes the digits when I try to copy-paste them, too. |
|
804)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 102053)
Posted 9 Dec 2020 by Richard Haselgrove Post: The reports also refer to the fallen instrument carrier weighing 900 tons. I haven't been able to find a corresponding weight for the platform as originally installed, but I'm pretty sure that a substantial amount of extra equipment was added over the years - like the 7-beam ALFA receiver that came into use while SETI@Home was active. |
|
805)
Message boards :
GPUs :
Can I run 1 WU across multiple GPUs?
(Message 102048)
Posted 9 Dec 2020 by Richard Haselgrove Post: BOINC manages my dual 12 core CPUs on one motherboard as one....No, it manages them as 24 - cores, that is, although the terminology does describe cores as CPUs. |
|
806)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102036)
Posted 8 Dec 2020 by Richard Haselgrove Post: As an aside, and keeping it separate from the problems at PrimeGrid, the old server code at Einstein is useful for showing how it should be done. 2020-12-08 14:35:42.7720 [PID=4585 ] [send] Intel GPU: req 4338.97 sec, 0.00 instances; est delay 0.00That's an extract from the server log visible at Einstein. Some terms need explaining: 'req' is the request from my machine - how much it wanted. 'DCF' we've come across before. It's calculated by the client, and reported to the server. 'delay bound' is how long the deadline is going to be - 14 days, for these tasks. 'duration' is the server's calculation of how long it will take. 'unscaled' is the raw estimate from speed, 'scaled' takes account of all the fractions, including DCF. The 'unscaled' figure is less than my request, so on that basis it would have gone on to find another task. But the 'scaled' figure is more than enough, so it didn't need to. The modern version of that second line is at https://github.com/BOINC/boinc/blob/master/sched/sched_send.cpp#L1498 log_messages.printf(MSG_NORMAL,
"[send] on_frac %f active_frac %f gpu_active_frac %f\n",
g_reply->host.on_frac,
g_reply->host.active_frac,
g_reply->host.gpu_active_fracNo mention of DCF. Likewise, the 'scaled' runtime estimate ignores DCF.
|
|
807)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102034)
Posted 8 Dec 2020 by Richard Haselgrove Post: And last night's work finished here, too. I asked for a smaller amount, but I've switched to MT work, using all four cores at once - so only one task can run at a time. The replacement fetch was: 08/12/2020 14:34:50 | PrimeGrid | [sched_op] CPU work request: 38016.00 seconds; 4.00 devicesThe work fetch request was for 0.11 days (9,504, times 4 cores, equals 38,016 - correct). The reply was: (speed) <flops>8339247311 (size) <rsc_fpops_est>6525083998674 =782.45 seconds per task times 49 new tasks = 38,340 seconds (first task to go above request - correct) times 1.5503 client DCF = 59,439 (close enough to client estimate - correct) Each task is estimated to run for 1,213 seconds, or just over 20 minutes - that's what the manager shows, and that's what's happening in reality. Correct. So far, so good. BUT: The client request was expressed in cache per core - so four times the requested cache in wall-time. The reply is also expressed as estimated per core, and would be right if each separate core was working on a separate task. The 49 tasks would finish in 12 cycles of 20 minutes, or four hours. But the estimated speed is four times greater than the speed for a single core - it's the aggregated speed of the whole CPU. And the tasks are running singly, using all four cores. So they'll run for 49 cycles of 20 minutes, or over 16 hours. I call ***bug*** in the MT part of the server calculation. That old 'size / speed' is in mixed units - '(size per core) / (speed per CPU)' Now to find it in the code... (or perhaps run it past Rytis first) |
|
808)
Message boards :
Questions and problems :
BOINC ceases to compute
(Message 102030)
Posted 8 Dec 2020 by Richard Haselgrove Post: The answer is contained in your second screen-shot. Status: Suspended - com... Connected to localhost (7.16.6) I think if you expand the column width, the first will expand to 'Suspended - computer is in use'. This is a known bug with BOINC versions 7.16.5/6/7 - try to switch to a different version. |
|
809)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102025)
Posted 7 Dec 2020 by Richard Haselgrove Post: Well, my plucky little Celeron finished its four tasks tonight, very close to the target time. Racked up a DCF of 8.8362. I've now got a batch of 49 Sophie Germain LLR MTs, reckoned to use all four cores for 13 minutes each. Of course, they're showing up at 8.8 times that, or 01:55:13 each - nearly four days, for a work fetch of 0.01 + 0.1 days. But it looks like the first one will be done in about 20 minutes total. I'll see how many I have left in the morning. |
|
810)
Message boards :
GPUs :
Can I run 1 WU across multiple GPUs?
(Message 102018)
Posted 7 Dec 2020 by Richard Haselgrove Post: I think they'll appear twice in the Event Log as 'Device 0', Device 1'. I think it will be because of the computer languages and techniques used for scientific programming, which differ from screen-painting. |
|
811)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102016)
Posted 7 Dec 2020 by Richard Haselgrove Post: Rytis is a familiar name from BOINC mailing lists, and is definitely a good person to approach. And yes, by "adjusted the sizes of their tasks" I was referring to the estimated size conveyed in <rsc_fpops_est>, but I was trying to be less technical in that post. |
|
812)
Message boards :
GPUs :
Can I run 1 WU across multiple GPUs?
(Message 102012)
Posted 7 Dec 2020 by Richard Haselgrove Post: No |
|
813)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 102008)
Posted 6 Dec 2020 by Richard Haselgrove Post: I've been trying to make more sense of it, but I haven't got completely to the bottom of it. It's clear that the jobs issued by PrimeGrid are 'too small' - their server thinks they will take less time than they really will. Those 170 MT jobs that started all this running - the Primegrid server sends them out with an estimate of 92 seconds, and Peter's computer 'corrects' that to 11:55. The tools for doing that - DCF, or 'duration correction factor' - are old. They're kept in the client as a legacy, for use when talking to an old server. Old servers know how to handle them. But PrimeGrid is using a relatively modern server. It reports 'server version 713', which dates it to about the middle of 2018. The old 'DCF' legacy code has been stripped out of the current code, and I suspect that happened longer ago than 2018 - possibly as long ago as 2012. I'll keep looking. Since I doubt we're ever going to get that old code back, it would be better if the PrimeGrid staff adjusted the sizes of their tasks to make them more realistic. |
|
814)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101988)
Posted 5 Dec 2020 by Richard Haselgrove Post: Interesting.....Peter got <app_version>
<app_name>llrTPS</app_name>
<version_num>804</version_num>
<api_version>7.11.0</api_version>
<file_ref>
<file_name>cllr64.3.8.23.exe</file_name>
<open_name>primegrid_cllr.exe</open_name>
<copy_file/>
</file_ref>
<file_ref>
<file_name>llr.ini.6.07</file_name>
<open_name>llr.ini</open_name>
<copy_file/>
</file_ref>
<file_ref>
<file_name>llr_wrapper_8.04_windows_x86_64.exe</file_name>
<main_program/>
</file_ref>
<is_wrapper/>
<platform>windows_x86_64</platform>
<plan_class>mt</plan_class>
<avg_ncpus>24.000000</avg_ncpus>
<flops>70852542799.840393</flops>
<cmdline> --nthreads 24</cmdline>
</app_version>I don't think the server always plays it back if it thinks you have the app already, but Peter sent this in the request:<app_version>
<app_name>llrTPS</app_name>
<version_num>804</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>24.000000</avg_ncpus>
<flops>70852542799.840393</flops>
<plan_class>mt</plan_class>
<api_version>7.11.0</api_version>
<cmdline>--nthreads 24</cmdline>
<is_wrapper/>
</app_version>
|
|
815)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101987)
Posted 5 Dec 2020 by Richard Haselgrove Post: Now, doing a similar calculation for Peter's 24-core MT tasks, using the figures from sched_request and sched_reply. The basic calculation is "size / speed", or " size is <rsc_fpops_est>6525083980042.000000 speed is <flops>70852542799.840393 Which is bloody fast for a single core (over 70 Giga-Flops), so we'll assume it's for the whole CPU, 24 cores in parallel. So duration is 92.093857498882860479551119183157 cpu-secs or 2,210.2525799731886515092268603958 core-secs The client applies a DCF of 7.767842, so the task is estimated at 715.37053422183723671319732473793 cpu-secs, or 00:11:55. QED The total 170 job work-fetch turns out to be 170 tasks * 92ish sec/task * 7ish DCF = 121,612.99081771233024124354520545 cpu-sec The log says 121823 seconds, a little more. That's to take account of <on_frac>0.998370 <active_frac>0.999910 - the client takes account of those little breaks in service. |
|
816)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101985)
Posted 5 Dec 2020 by Richard Haselgrove Post: Checking... 1) Benchmark (from client_state / host_info) 2) Task speed (from sched_reply / app_version) <p_fpops>2084811827.956989</p_fpops>
<flops>2084811827.956989</flops>
|
|
817)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101983)
Posted 5 Dec 2020 by Richard Haselgrove Post: I would posit that Prime Grid are using some rather strange figures in getting to their flops guess.I don't think it's that. For my first group of four, the flops estimate in the <workunit> sent by the server exactly matched (down to the units place, and I think on into the micro-flop) the Benchmark figure calculated by the computer. It's a straight copy. The dodgy one is probably the fpops_est for the <result>..It matters, but projects have a habit of paying very little attention, especially when it should be changing. PrimeGrid will be searching for ever-larger primes, and naively I'd suggest that the search will take ever longer? fpops_est should be growing to keep pace, but they may not have bothered. |
|
818)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101979)
Posted 5 Dec 2020 by Richard Haselgrove Post: Yes, I've got all that now. I was a bit pre-occupied with getting my test started, fetching the paper, and writing my reply to David. And after all that, I think I need a bit of a lie-down... We also have to factor in <duration_correction_factor>7.767842 - most projects have stopped using that. But at least they don't use APR on top of it. I'll do the maths after I've had a break - at least, it's all in the sched files, for which many thanks. |
|
819)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101974)
Posted 5 Dec 2020 by Richard Haselgrove Post: No, I changed it manually. It was set to 1 when I first attached. For reference, here are the first two work requests for a new computer on a new project: 05/12/2020 12:16:25 | | [work_fetch] target work buffer: 864.00 + 8640.00 secNothing wrong with that. The estimated duration matches the speed and size of those first few tasks. But after three hours, we're at <checkpoint_elapsed_time>10853.239532</checkpoint_elapsed_time>
<fraction_done>0.050201</fraction_done>I knew this machine was slow, but not that slow. At this rate, it'll finish at about midnight Monday!
|
|
820)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101972)
Posted 5 Dec 2020 by Richard Haselgrove Post: Ah - you'd answered my question (edit) already. I'll switch to that. But now I see  I'll try LLR (SGS) anyway.
|
|
821)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101970)
Posted 5 Dec 2020 by Richard Haselgrove Post: Need to check one more thing. I've joined up, and my initial state is  So, all four are running at once. That's because the project has a preference I'll switch it to 4 for the next fetch, and try again. Looks like I'll have time to go and fetch my newspaper before we get there...Edit - this is the one I got. Proth Prime Search LLR (PPS)Which one were you running? |
|
822)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101968)
Posted 5 Dec 2020 by Richard Haselgrove Post: Thanks. I've got a little 4-core Celeron (bought for testing a low-power CPU bug), which should be dry in the next half-hour or so - I drain them, rather than abort them. I'll repro that, and grab the screen-shots etc. Sounds like another issue for GitHub, after I've answered David's silly question overnight. Going to be a busy weekend. |
|
823)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101959)
Posted 4 Dec 2020 by Richard Haselgrove Post: If the machine was nearly dry, it would ask for 11232s x 4 (44,928 seconds), to keep all four cores busy for the whole length of the cache. So that looks OK. It got 20 tasks - 3,103.95 seconds each. How do they display on the Tasks tab in BOINC Manager? The average task runtime would be about 52 core-minutes, or 13 CPU-minutes. The Manager should display in wall-time, so the 'correct' figure would be 13 minutes. 20 tasks would take about 4.3 hours - a bit long, but not disastrous. The server translates your work request into a number of tasks, but at most projects mere mortals can't see the working. The honourable exception is Einstein, but they don't have any MT apps. The 'estimated total CPU task duration' is worked out by the client, and we can look into that. If you see a work request which seems to produce the wrong answer, it would be helpful to gather all the evidence (the debug log, like this one: a screenshot of BOINC Manager: and if you get there before the next work request, the 'sched_request_...' and 'sched_reply_...' files for the event). [Those are created new each time, but remain in the data directory until over-written by the next request] |
|
824)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101957)
Posted 4 Dec 2020 by Richard Haselgrove Post: Yes I know, but I assumed the client was saying "I want 1000000 seconds of work" - a long time ago you could see that in the event log. Perhaps they moved it into a debug flag. If I ask for 1 million seconds and get 5 million, it's the server's fault.On the very first attach, the clients should ask for precisely 1 second of work. But the subsequent request should asking for something approximating to your cache setting. Yes, the simple figures are revealed by a debug log flag - the <sched_op_debug> I mentioned in my question. That produces very few, but useful, extra lines in the output - small enough for me to set it as routine on all my rigs. |
|
825)
Message boards :
Projects :
News on Project Outages
(Message 101943)
Posted 4 Dec 2020 by Richard Haselgrove Post: Both website and downloads are OK again. |
|
826)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101935)
Posted 3 Dec 2020 by Richard Haselgrove Post: ... giving the same dose twice but giving one "big" dose and one "little" dose (I can't remember the sequence) ...Actually, it was the other way round - little first, then the big one. Seems as if the human immune system likes to receive an amuse-bouche before the main course ... |
|
827)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101934)
Posted 3 Dec 2020 by Richard Haselgrove Post: ... Primegrid send out far too many tasks ...Can we take a technically precise reading on that statement, please? I'm currently working on what seems to be a real-life programming bug (#4117) which looks like that, but in reality the cause of the problem is that the client asks (repeatedly) for too much work - way beyond what my cache settings require. Your Event Log will tell the full story - <sched_op_debug> will add useful figures for how much you request and how much you receive. Note that the fundamentals of the internet prevent servers from sending work except in response to a request - firewalls and routers would block such traffic. |
|
828)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101932)
Posted 3 Dec 2020 by Richard Haselgrove Post: What I'm trying to do is shorten the deadline for my CPDN tasks ...Yes, I know - but this is a public forum, and other readers might have used my instructions for a different purpose. It's important that they, too, know the full implications so they can make an informed judgement. |
|
829)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101923)
Posted 3 Dec 2020 by Richard Haselgrove Post: As an afterthought - it is possible to edit the locally-displayed deadline, and if you extend it, the BOINC client will let it start after the new date. But the real, effective, deadline is stored in the project's server, and your local edit has no effect there. If you award yourself a period of extra time, it's at your own risk - the project may not accept your scientific result, and may not award you any credit for it. |
|
830)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101920)
Posted 3 Dec 2020 by Richard Haselgrove Post: That's odd. We're being told: Our first vaccine is commonly known as Pfizer, has been approved, is made in Belgium, and has all the problems of low-temperature transport and storage that you list. Our second vaccine is expected to be Oxford. That hasn't been approved yet, but has been submitted. It had the dosage error in testing leading to the 70% - 90% coverage variation that you describe. But it does not have the low-temperature storage requirement, and our GPs/care homes are eagerly awaiting the simpler roll-out. Looks like your GPs are being given a mish-mash of those two stories. Maybe our Brexit misinformation team are trying to put you off buying any, so we can keep it for ourselves? |
|
831)
Message boards :
Questions and problems :
Any way to manually change the deadline of a task?
(Message 101913)
Posted 3 Dec 2020 by Richard Haselgrove Post: You can do it, but I'd strongly urge you not to do so. Everything to do with tasks is stored in a single file - client_state.xml. It's huge, it's vital, it's complicated, and it's fragile. To change anything, you have to stop BOINC, edit the particular value you're interested in, re-save the file, and start BOINC again. A single slip-up in either values or formatting, and the whole pack of cards collapses in a heap. Having said that, deadlines are easy: <report_deadline>1608156032.000000</report_deadline>That's a "Unix timestamp" - the number of seconds since the start of the current Unix epoch, and can be converted via, say, EpochConverter to Wednesday, December 16, 2020 10:00:32 PM And, again, having said that, I wouldn't do it that way. I also run Climate Prediction, and I manage that by Not running a large cache Not running too many projects at once Increasing the time between task switches to a day or longer BOINC takes very little notice of deadlines unless there's a real danger that they might be missed. The key factor in your problem is probably to make sure that no other project grabs the CPU because its own deadline is too close. |
|
832)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101890)
Posted 2 Dec 2020 by Richard Haselgrove Post: The first challenge will be getting it across the channel from Belgium, during Brexit - while keeping the temperature low enough during delays. |
|
833)
Message boards :
Questions and problems :
Reporting Tasks completed
(Message 101889)
Posted 2 Dec 2020 by Richard Haselgrove Post: I'm afraid you'll have to wait until they've fixed their server. More details in https://boinc.berkeley.edu/forum_thread.php?id=14053 |
|
834)
Message boards :
Questions and problems :
Any way to force NOT reporting results immediately?
(Message 101861)
Posted 30 Nov 2020 by Richard Haselgrove Post: It's likely to be in the task specification, sent to you in the sched_reply_... file the last time you received new work. GPUGrid does the same: <result>
<name>e23s14_e4s5p0f177-ADRIA_PBBandit100ns-0-1-RND9533_0</name>
<final_cpu_time>0.000000</final_cpu_time>
<final_elapsed_time>0.000000</final_elapsed_time>
<exit_status>0</exit_status>
<state>2</state>
<platform>windows_x86_64</platform>
<version_num>211</version_num>
<plan_class>cuda101</plan_class>
<report_immediately/>
...It's the way the project manages its workflow. GPUGrid is doing fast-moving research, and wants its science back pronto. I suspect Primegrid may be trying to keep the size of their database of 'results out in the field' as small, and hence as cheap, as possible.It's their science. You are volunteering to help them, not the other way round. Please respect their wishes, as far as possible. |
|
835)
Message boards :
Questions and problems :
names of windows 10 service (to delay start)
(Message 101838)
Posted 28 Nov 2020 by Richard Haselgrove Post: <start_delay> doesn't delay BOINC itself - BOINC itself starts immediately, but that option delays the start of the science applications after BOINC is running. I think the service is simply called BOINC. Win 10 has the "Automatic (delayed start)" service launch option from Win 7, but I don't think you have much control over the delay. I have a machine I can test that on if you need help. |
|
836)
Message boards :
Android :
BOINC for BoincTasks on Android
(Message 101820)
Posted 27 Nov 2020 by Richard Haselgrove Post: Several fruitless tests, and much hard work on Fred's part later, my tablet has a working client and it's crunching science for Einstein - and I can manager it from a perfectly normal Windows BOINC Manager. Still some rough edges to knock off, but it's getting there. Look out for Beta 7 or above on Fred's site. |
|
837)
Message boards :
Questions and problems :
Install Error on MUBUNTU 20.10
(Message 101810)
Posted 27 Nov 2020 by Richard Haselgrove Post: On mine, it adds itself to the 'System Tools' section of the menu. |
|
838)
Message boards :
Questions and problems :
boinc 7.16.12 won't install (on macos 10.7)
(Message 101787)
Posted 26 Nov 2020 by Richard Haselgrove Post: Charlie has provided some more background privately. It seems there was a subtle change in the digital signature system round about version 10.9 - older versions of the Mac OS can't read the newer signatures. Owners of older Macs can go on running BOINC version 7.16.11 - Charlie has made that visible again on the download all page, in case you don't have a copy to hand. He's also replaced 7.16.12 with 7.16.13 - no change to the actual program, just an extra safeguard to stop people trying to install it on the older Macs. There's no added benefit in upgrading to 7.16.12/13 anyway - those versions were only released to support the brand new 'Apple Silicon' range launched this month. |
|
839)
Message boards :
Questions and problems :
Updating BOINC on Mac failed when running downloaded app file with "digital signature is invalid"
(Message 101786)
Posted 26 Nov 2020 by Richard Haselgrove Post: Charlie has provided some more background privately. It seems there was a subtle change in the digital signature system round about version 10.9 - older versions of the Mac OS can't read the newer signatures. Owners of older Macs can go on running BOINC version 7.16.11 - Charlie has made that visible again on the download all page, in case you don't have a copy to hand. He's also replaced 7.16.12 with 7.16.13 - no change to the actual program, just an extra safeguard to stop people trying to install it on the older Macs. There's no added benefit in upgrading to 7.16.12/13 anyway - those versions were only released to support the brand new 'Apple Silicon' range launched this month. |
|
840)
Message boards :
Questions and problems :
Boinc will drain battery on plugged in Mac Book
(Message 101777)
Posted 25 Nov 2020 by Richard Haselgrove Post: Relying on wikipedia, "All USB-C cables must be able to carry a minimum of 3 A current (at 20 V, 60 W) but can also carry high-power 5 A current (at 20 V, 100 W)." My current (non-Apple) laptop, purchased in early 2019, can run BOINC continuously from a 45W charger, well within that USB-C spec. It uses a more traditional power connector, though it also has a USB-C socket as well, for expansion peripherals. You could explore the possibility of getting an after-market 100W charger and high-power cable? |
|
841)
Message boards :
Questions and problems :
Updating BOINC on Mac failed when running downloaded app file with "digital signature is invalid"
(Message 101762)
Posted 25 Nov 2020 by Richard Haselgrove Post: There's nothing much we can do at this end - digital signatures are issued to the developer, and tightly guarded. But I've sent an email to the developer concerned, and drawn attention to this report. There may be a delay in responding, because of time-zones. |
|
842)
Message boards :
Questions and problems :
boinc 7.16.12 won't install (on macos 10.7)
(Message 101761)
Posted 25 Nov 2020 by Richard Haselgrove Post: There's nothing much we can do at this end - digital signatures are issued to the developer, and tightly guarded. But I've sent an email to the developer concerned, and drawn attention to this report. There may be a delay in responding, because of time-zones. |
|
843)
Message boards :
Android :
BOINC for BoincTasks on Android
(Message 101703)
Posted 22 Nov 2020 by Richard Haselgrove Post: Interim report - haven't got this working on the Samsung tablet yet. Fred's app has detected the IP address correctly, but doesn't recognise it as WiFi, and I can't get a normal manager to connect.. We're in contact on the BoincTasks forum. |
|
844)
Message boards :
Android :
BOINC for BoincTasks on Android
(Message 101673)
Posted 20 Nov 2020 by Richard Haselgrove Post: Interesting. I'll run down my current v7.16.3 APK's cache, and take it for a spin. See how it gets on with my current rpc clients: BOINC Manager for Windows and BoincView (an oldie, but a goodie!). Edit - that was convenient - it had lost WiFi overnight and finished all its work anyway! Reported all now, starting on updates... Edit 2 - for the record, it's Samsung Galaxy Tab A 2019 10.1 Inch 32GB |
|
845)
Message boards :
The Lounge :
Apple M1 & BOINC
(Message 101633)
Posted 18 Nov 2020 by Richard Haselgrove Post: This is a three-part problem. 1) Very few people have developed or fully tested new software to get the best out of the M1 chip yet. There is supposed to be a transitional emulation package to enable the new Macs to run older software until the transition is complete, but speeds and efficiency are likely to be lower than theoretically achievable. 2) The developer of BOINC for the Apple platform, Charlie Fenton, has made sterling efforts to ensure that a version of the BOINC package suitable for the new environment will be available as soon as possible after launch. I think it's nearly there, but he wants to wait for test users to report their initial findings before making a public release. 3) Once the machines and BOINC are available, individual science projects will have to update their applications as well. This is likely to be the longest and slowest stage of all, and also the one where the temporary use of the emulation layer has the most (bad) effect. Overall, it's a big job. Early reports will be interesting, but don't interpret them as the final answer. Edit - if you're feeling adventurous, look at the 17 November entries at the top of https://boinc.berkeley.edu/dl/?C=M;O=D |
|
846)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101630)
Posted 18 Nov 2020 by Richard Haselgrove Post: You are clearly going to the wrong URL if visiting their massage boards.LOL. I nearly fell into that one, until I saw what I'd done. But it does massage my ego to be #33 in the world there, until the other SETI orphans catch up. |
|
847)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101622)
Posted 18 Nov 2020 by Richard Haselgrove Post: FacePlant spamming adverts.....Is it my imagination, or are adverts getting more invasive? One of my BOINC projects puts up a full page advertising splat every time I visit their massage boards, another site puts up video covering part of the content. Both are recent developments, unrelated to the shit-shower that infected all tabloid newspaper sites decades ago. |
|
848)
Message boards :
News :
Web site makeover
(Message 101608)
Posted 17 Nov 2020 by Richard Haselgrove Post: My main issue with Science United is that I can't control on disabling GPU workloads being sent to me. I mainly have a laptop that I use for crunching and if I get tasks that are both GPU/CPU then my laptop overhears or Tthrottle pushed down the processing to the point of it not doing anything.I don't like the lack of control either, but that's one you can do something about. Look in the 'Options' section of Client configuration, and you'll find the three possible types of GPUs listed: <ignore_ati_dev>N</ignore_ati_dev>Check your existing logs to see which type of GPU you have. Make a cc_config,xml file to ignore device 0 (zero) of that make. That should prevent the overheating. |
|
849)
Message boards :
Documentation :
Moving BOINC directory to /home under Linux and systemd
(Message 101606)
Posted 17 Nov 2020 by Richard Haselgrove Post: Relayed upstream to https://github.com/BOINC/boinc/issues/4104, and drawn the attention of the main distro maintainers. |
|
850)
Message boards :
Questions and problems :
Boinc won't stop running
(Message 101597)
Posted 16 Nov 2020 by Richard Haselgrove Post: Problem 2: BOINC won't stop (doesn't react to mouse/keyboard use) - BOINC can't read the X-server names. Needs a Linux guru.We've found our Linux guru! Simple bug in creating the systemd file. Gianfranco's PPA has a new test version out - it works. Other users can add it from terminal. Run sudo systemctl edit boinc-client.serviceand add these two lines into the empty override file which appears [Service] PrivateTmp=falseThen reload and restart the service. |
|
851)
Message boards :
Documentation :
Moving BOINC directory to /home under Linux and systemd
(Message 101591)
Posted 16 Nov 2020 by Richard Haselgrove Post: Thanks, Graeme. I missed this the first time round, but my eye was caught by the new post. That certainly sounds like a worthwhile contribution to the documentation and BOINC's arsenal of tools. I have two slight worries, though. 1) The instructions are to 'copy' the data directory. That's wise while you're in the process of upgrading, but omce the new version is settled and running, you'll still have the old copy lying around, using space and waiting to trip up the unwary. BOINC data is ever-changing, and not amenable to incremental or transactional backup. A backup data directory is very little use unless very recent - so it might be better to delete it and recover the space soon after performing this move, after verifying that it's working, and before you forget what you did. 2) What happens when a new version is released through your distro? Does systemd preserve local edits like this, or will an upgrade force a hard reset to /var/lib/boinc-client? That would be confusing, to say the least. As it happens, we're having a very hard look at various systemd issues in https://github.com/BOINC/boinc/pull/4071, with several of the main Linux players participating. I'll cross reference this thread for them, and ask the same questions: if we can get confirmation from the package maintainers, I'll see if we can get official documentation for this. |
|
852)
Message boards :
Questions and problems :
Einstein takes over
(Message 101569)
Posted 13 Nov 2020 by Richard Haselgrove Post: Also, the term 'Credit' is a bit of a misnomer. Most projects issue multiple 'credits' per task (sometimes a great number), but the number awarded per unit time varies greatly. BOINC was originally intended to even out credits over time (in accordance with your work share settings), but that was never implemented. Instead, the balance was made over REC, or recent estimated credit - not what you see on project web pages at all. And finally, I think WCG uses a different method altogether - see their web pages. Edit - see https://www.worldcommunitygrid.org/help/viewTopic.do?shortName=overview#505 BOINC provides a reference about credit and its relation to FLOPS here. However, you should know that seven (7) World Community Grid points are equal to one (1) BOINC credit.I'm slightly doubtful about that FAQ (first time I've seen it) - I though you would get fewer points than credits (and the BOINC calculation has had a factor of 2 correction, which WCG doesn't seem to have noticed). But they should know their own schema! |
|
853)
Message boards :
Questions and problems :
Boinc won't stop running
(Message 101563)
Posted 13 Nov 2020 by Richard Haselgrove Post: tl;dr Problem 1: BOINC won't start (permanently 'in use') - bug in versions 7.16.5/6/7. Avoid those versions. Problem 2: BOINC won't stop (doesn't react to mouse/keyboard use) - BOINC can't read the X-server names. Needs a Linux guru. |
|
854)
Message boards :
BOINC client :
Panic mode needs to start earlier or something
(Message 101541)
Posted 10 Nov 2020 by Richard Haselgrove Post: ... won't BOINC learn how long things actually take if you give it its head and ignore it for long enough?It depends. In principle yes, But the definition of 'long enough' is problematic. In the early years of BOINC (up until 2010), BOINC used a value called DCF ('duration correction factor'). That was carefully constructed to protect deadlines. If even a single task over-ran (and finished normally, to prove it wasn't stuck in a loop), then the estimates were updated instantly - on the client, so even tasks already downloaded and waiting to run were affected. One task might be wasted, but BOINC would 'know' that all the rest should go into panic mode earlier. But that method has problems. It's very difficult for projects to keep track of varying applications with varying speeds through a single value. So with multiplying projects, multiplying applications, and multiplying devices (GPUs of differing speeds), things needed to change. David, in his infinite wisdom (ahem!), decided to take all the calculations back onto the server, and work with slowly-changing averages instead of sudden jumps. The easiest way of tracking these is in the 'application details' shown for each of your computers at each project's website - look for APR ('average processing rate'). There are multiple problems with the APR approach. 1) The initial values for a new application, or a new computer, are badly designed and can be seriously wrong. 2) The 'averaging' approach makes it slow to adjust, even once it's got started. 3) Only runtimes from validated tasks are considered, so if an over-running task is disqualified for lateness, it's difficult to get back in range. So, "How long is enough?" depends on each separate project, and how it manages its estimates. Some mainstream BOINC projects (Einstein, GPUGrid) still use DCF, and will react instantly. Most other mainstream BOINC projects will use APR. And WCG isn't a mainstream BOINC project - they run their own server code, and do things their own way. I haven't studied it: you'll have to ask the folks over there. |
|
855)
Message boards :
News :
Web site makeover
(Message 101530)
Posted 9 Nov 2020 by Richard Haselgrove Post: David Anderson writes the news, but doesn't read it.But please, reintroduce the button for download boinc client.No news??? |
|
856)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101526)
Posted 9 Nov 2020 by Richard Haselgrove Post: Yes, it showed just fine. The question mark and all that follows it are tracking and authorisation data for the hosting site.For images to work within image tags, they need to end on .bmp, .jpg, or .png, so next time remove all the crap, after the jpg, from the question mark to everything after that.Will do. just checked the preview which showed the picture so I assumed it worked. Interestingly, if I put the url up to .jpg into a web browser (but no more), it simply replies "Bad URL timestamp". This particular photo service - I'm not familiar with it - actually needs all that crap before it releases the image. |
|
857)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101494)
Posted 4 Nov 2020 by Richard Haselgrove Post: There was a very interesting small-scale study reported a few days ago, but I can't find the reference quickly. (I'll try to find it) It is now possible to examine the genetic sequence of virus samples taken via single test swabs. Testers collected multiple samples (~100) from single sites on a single day, so they were directly comparable. One sample-set was taken at a food processing factory. Another sample-set was taken from a general hospital. The first sample-set turned out to be closely self-similar: all the individual samples had the same genetic sequence. Conclusion - this outbreak was caused by a virus circulation within the workforce. They didn't have to worry about the wider community. The second sample-set turned out to have wide variation - different patients had picked up widely varying mutations of the virus. Conclusion - the outbreak was not the result of cross-contamination within the hospital, but the patients had been infected in the community before admission. Once that kind of research can be validated and rolled out more widely, it should - I sincerely hope - be possible to trace and contains local cluster spreaders, without resorting to regional or national lockdowns. |
|
858)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101478)
Posted 3 Nov 2020 by Richard Haselgrove Post: That means it's far easier for us to keep up with any changes it makes.That means also that there is a better chance of any future vaccine being rolled out broadly to the population at large. And working, if any of us live that long. Chill, everyone, and have a listen to https://www.youtube.com/watch?v=U6f734MDffs. It's chilling, but beautiful. The guy singing is a former Professor of Sociology at Sheffield University: it's a modern retelling, but the history is accurate. Enjoy, and learn. |
|
859)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101464)
Posted 3 Nov 2020 by Richard Haselgrove Post: Strange, I read something somewhere that coronavirus was unusual in that it did not mutate, it had a parity check or something.'Something, somewhere' is a very helpful source reference. But I've read the same thing: Coronaviruses in general (not just this particular one) are more stable, and have a lower - but non-zero - mutation rate than, for example, the rhinoviruses that cause the majority of common colds. |
|
860)
Message boards :
Questions and problems :
no WSL detected
(Message 101438)
Posted 2 Nov 2020 by Richard Haselgrove Post: Have you, yourself, added WSL to your Windows 10 system? That's a pre-requisite, not installed by docker, whether it needs it or not. |
|
861)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101415)
Posted 1 Nov 2020 by Richard Haselgrove Post: #1 We do have hills.Lol. Yes, I know you do - I cycled up several of them during my years as a student there. |
|
862)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101413)
Posted 1 Nov 2020 by Richard Haselgrove Post: I second that - I dropped in on Malvern on my way back from Cornwall in September. I wasn't so impressed by the town - it was only a fly-by, so I didn't explore any of the eateries or hostelries, and parking was horrendous - but I escaped round the back of the hills and found plenty of open space for both parking and walking. The views from the top will make a tremendous comparison with Cambridgeshire! |
|
863)
Message boards :
Questions and problems :
Primegrid.com - Malwarebytes blocking it
(Message 101396)
Posted 30 Oct 2020 by Richard Haselgrove Post: The report, as with many of these, was about the website - not about any downloaded executables. Please have a look at my message 100040 (July 2020, a similar event concerning a different project). If you have a premium subscription, you can raise a ticket - give them as much detail as possible - and they really will look into it and let you know the outcome. |
|
864)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101390)
Posted 30 Oct 2020 by Richard Haselgrove Post: When I need to test a specific BOINC release version, I: Go to the main GitHub landing page - the one with a full folder listing Use the dropdown list from the 'Master' button - either the release branch, or, for finer control, the tag list Once the correct version is shown, use the dropdown on the green 'Code button' to "Download ZIP". It's surprisingly quick. No, there aren't nightly builds, but each PR gets its own build, for testing that specific change (the rest of the code comes from master) For 4076, you can get the win-manager from https://ci.appveyor.com/project/BOINC/boinc/builds/36041052/job/n3q6kfh8e66vhk2j/artifacts That's the VS2013 x64 build, but I think it was a component (wxWidgets) change, not a compiler change. I've prodded to get VS2019 to save its artifacts, if we're going to be using that, but no news yet. |
|
865)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101387)
Posted 30 Oct 2020 by Richard Haselgrove Post: Test it, or lob hand grenades into the discussion - your choice! |
|
866)
Message boards :
BOINC client :
7.16.11 available for testing for Windows and Macintosh
(Message 101385)
Posted 30 Oct 2020 by Richard Haselgrove Post: I've been searching my inbox for an official announcement, like the one David made on 8 April for the previous version - but I can't find one. The date on the all_versions download page is 2 September, but I think that might be the beta test date (it's also the datestamp on the final assembly files). If we can agree on a public release date, Jord or I will try to edit it in. Edit - oh, David did it in the News thread here this time, not by email. 8 September it is. |
|
867)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101382)
Posted 29 Oct 2020 by Richard Haselgrove Post: I'll leave you to deal with #4076. :-) hic! |
|
868)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101381)
Posted 29 Oct 2020 by Richard Haselgrove Post: I still don't know what to do about it thoughThat's the problem. I don't know either, but I'm not going down silently. At least I got a thumbs-up from Vitalii - I do genuinely think he's on the right side, although walking a tightrope between David and everyone else. |
|
869)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101379)
Posted 29 Oct 2020 by Richard Haselgrove Post: That's interesting. Merge and close, with no evidence of a successful test. |
|
870)
Message boards :
Questions and problems :
Invisible BOINC manager
(Message 101377)
Posted 29 Oct 2020 by Richard Haselgrove Post: My recommendation, open only 1 instance.Without specific manual configuration, only one instance can run at once. Any additional instance will find the first, and close itself. One instance can have multiple viewports, connected to different clients. |
|
871)
Message boards :
Questions and problems :
Boinc wiki needs updating
(Message 101376)
Posted 29 Oct 2020 by Richard Haselgrove Post: https://boinc.berkeley.edu/projects.php Look for 
|
|
872)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101370)
Posted 29 Oct 2020 by Richard Haselgrove Post: I can think of a few candidates. I'm sure they would make good fertiliser. |
|
873)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101368)
Posted 29 Oct 2020 by Richard Haselgrove Post: Well, last week there were two flowerbeds outside my window ... 
|
|
874)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101366)
Posted 29 Oct 2020 by Richard Haselgrove Post: I have the magic word for showing the Windows 7 icon control tool in Windows 10. 
|
|
875)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101365)
Posted 29 Oct 2020 by Richard Haselgrove Post: Continuing a conversation from elsewhere last night. Curious behaviour of this strange v7.16.16 under Windows 10 - seeking confirmation before I move on to Windows 7 and blow the doors off in public. On machine restart: I get BOINC running before I'm fully logged in - testing on a machine that requires manual entry of a password, but BOINC can be viewed over a remote connection before I do that. When I do access the desktop after entering the password, there's no trace of BOINC - no icon in the system tray, even hidden. BOINC Manager won't display on screen from the Windows menu. But it is visible from the Task Manager - crashing out there enables it to restart normally from the Windows menu. I'll try a few extra tricks before I make the formal report. Edit - won't restore from a pinned taskbar icon, either. Windows 10 pro 2004, build 19041.572 |
|
876)
Message boards :
Questions and problems :
Ubuntu 20.10 but getting error on BOINC
(Message 101361)
Posted 28 Oct 2020 by Richard Haselgrove Post: David Anderson has proposed making passwords optional again: https://github.com/BOINC/boinc/pull/4071. It's not in the current alpha build - that's a separate conversation - but it's a move in the right direction. However, there's no obvious way back for the people who are now stuck with a password they didn't ask for and can't (easily) access. I would have preferred that he fixed the real problem (inadequate collaboration with the people who make the distro installers), rather than leave everybody in limbo from his failed experiment. |
|
877)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101359)
Posted 28 Oct 2020 by Richard Haselgrove Post: Well, between us, that's given the pot a good stir. I think that calls for a beer :-) Back later... |
|
878)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101354)
Posted 28 Oct 2020 by Richard Haselgrove Post: Each separate .exe file: v7.16.11
Directory of C:\Test\boinc_inst_11\Program Files 64\BOINC
02/09/2020 15:05 1,306,952 boinc.exe
02/09/2020 15:05 399,176 boinccmd.exe
02/09/2020 15:06 8,281,416 boincmgr.exe
02/09/2020 03:31 1,928,192 boincscr.exe
02/09/2020 15:05 16,712 boincsvcctrl.exe
02/09/2020 15:05 70,472 boinctray.exe
6 File(s) 12,002,920 bytesv7.16.16
Directory of C:\Test\boinc_inst_16\Program Files 64\BOINC
26/10/2020 18:33 4,325,200 boinc.exe
26/10/2020 18:33 684,880 boinccmd.exe
26/10/2020 18:33 10,739,536 boincmgr.exe
26/10/2020 18:33 3,661,648 boincscr.exe
26/10/2020 18:33 129,360 boincsvcctrl.exe
26/10/2020 18:33 345,936 boinctray.exe
6 File(s) 19,886,560 bytesHe probably left the debug flag on...Edit - I'm not risking it running live - those were admin install images. And the registry bug is still there: https://github.com/BOINC/boinc/issues/824 |
|
879)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101352)
Posted 28 Oct 2020 by Richard Haselgrove Post: Well, it looks genuine:  https://github.com/BOINC/boinc/commit/a5ecb07a0b54c9d142c365bcdc1e0ebf4559a4df Might combine VS2019 and a fix for when he typed 7.16.15 for 7.16.5 Edit - it still has the VS2010 DLLs |
|
880)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 101350)
Posted 28 Oct 2020 by Richard Haselgrove Post: Hang on everyone - what's this doing in the BOINC binary download folder? boinc_7.16.16_windows_x86_64.exe 2020-10-27 19:45 13Mfive minor versions beyond anything we've seen before, and 50% bigger. |
|
881)
Message boards :
Questions and problems :
Ubuntu 20.10 but getting error on BOINC
(Message 101349)
Posted 28 Oct 2020 by Richard Haselgrove Post: Can you imagine the outcry if it were to be made that interesting (difficult) for Windows users?Or for Science United users with no access to the normal help channels? |
|
882)
Message boards :
Questions and problems :
Ubuntu 20.10 but getting error on BOINC
(Message 101346)
Posted 28 Oct 2020 by Richard Haselgrove Post: Well, he's responded, with a sort-of climbdown. But I still see problems ahead.I was tempted to add "as I predicted in the very first comment on that PR". But I decided that - because David seems to switch off whenever he sees my name - it would probably be counterproductive.The latest version which requires a password in the gui_rpc_auth.cfg file sure is causing lots of troubles it seems from the plethora of posts for help lately.I've commented on the PR that did this, #3709, so let's see if there's any other comments on this than "yes, but it's more secure this way". There are three cases: 1) Empty password. Manager will work, and you get a warning. That's fine. 2) You know what you're doing, and why. You've set a password, and set the Manager to run with it. That's fine too. 3) You didn't have a password, but now you have one of David's specials. The Manager doesn't work. What's your way out of (3)? We can't just erase all passwords, because that would mightily annoy group (2), and they are probably people David thinks are important. Once again, David has shown that he thinks in the present. But he doesn't think about edge cases, or about transitions from old to new. He's a coder - barely more than a hacker - but not a scientist or an analyst. |
|
883)
Message boards :
Questions and problems :
Invisible BOINC manager
(Message 101345)
Posted 27 Oct 2020 by Richard Haselgrove Post: You could use the <start_delay>nseconds</start_delay> in the cc_config.xml to pause running the applications to allow sufficient time for the video drivers to load and settle.Won't help. 'Start delay' applies to science apps - but these problems are in BOINC itself. Mine are in the client - if GPUs aren't detected in the initial startup, they are never available in the succeeding run. Dave's problems are with the manager - he's uncovered a similar problem there. |
|
884)
Message boards :
Questions and problems :
Invisible BOINC manager
(Message 101335)
Posted 27 Oct 2020 by Richard Haselgrove Post: Stopping the client will leave the manager visible, but empty of content. Having a stray lockfile will prevent the manager running, so shouldn't be visible in top. Neither matches Dave's description. |
|
885)
Message boards :
Questions and problems :
Invisible BOINC manager
(Message 101322)
Posted 26 Oct 2020 by Richard Haselgrove Post: It might also help to explain the context in which the self-compiled programs run. For example, I'm happy to compile my own apps, or swap in test versions compiled by others. But the initial context remains the same: Gianfranco's PPA. So, after a restart (any restart) the current client runs as a service, and the current manager is idle - waiting for me to start it manually when I need it. Mostly that works fine, but after an update the client may start before the graphics drivers are ready - no GPUs. How similar it that to your planned start-up context? |
|
886)
Message boards :
Questions and problems :
Ubuntu 20.10 but getting error on BOINC
(Message 101298)
Posted 25 Oct 2020 by Richard Haselgrove Post: So, if all else fails, use the Index Finger and see what happens.Congratulations - though I must confess I'm sometimes tempted to raise the middle finger instead... For reference next time, this is my monthly (or thereabouts) maintenance routine. 1. sudo systemctl stop boinc-client 2. sudo systemctl disable boinc-client 3. Install all the pending Linux updates that have accumulated 4. Perform a full cold reboot of the machine 5. Wait until everything has stopped moving, and I can see my normal stable desktop 6. sudo systemctl enable boinc-client 7. sudo systemctl start boinc-client Notes: 2,6: The disable/enable commands are to delay the BOINC auto-start. I run GPUs, and the system updates normally include a new kernel and a new video driver. Unless these are given time to sort themselves out after the reboot, BOINC can't find the GPUs, and can't run GPU tasks. 4: I normally perform a software restart from the start menu, but your finger may come in useful here. |
|
887)
Message boards :
Questions and problems :
Ubuntu 20.10 but getting error on BOINC
(Message 101295)
Posted 25 Oct 2020 by Richard Haselgrove Post: I was tempted to add "as I predicted in the very first comment on that PR". But I decided that - because David seems to switch off whenever he sees my name - it would probably be counterproductive.The latest version which requires a password in the gui_rpc_auth.cfg file sure is causing lots of troubles it seems from the plethora of posts for help lately.I've commented on the PR that did this, #3709, so let's see if there's any other comments on this than "yes, but it's more secure this way". |
|
888)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101280)
Posted 24 Oct 2020 by Richard Haselgrove Post: Some things are designed to be left permanently switched on (or should be) - clock, fax, answerphone, fridge. MW clock radios should be like that, but my current DAB radio needs to be rebooted every couple of months, or else the buttons stop working... |
|
889)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101276)
Posted 24 Oct 2020 by Richard Haselgrove Post: Presumably the Rosetta notice invoked a video player app already present on your computer, and passed it a url to use as the video data source? In that case, security is the responsibility of the video player app, not of BOINC. The video may very well be irritating, but it shouldn't be dangerous (in the malware sense). |
|
890)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101274)
Posted 24 Oct 2020 by Richard Haselgrove Post: ... but if BOINC or someone else can write at all they can write anything.The client (and only the client) is designed to download external files, write them to disk, and as you say, anything is possible. High security is needed. The manager doesn't download anything, and only writes its own configuration settings. For that to become a malicious executable, somebody would have to persuade you to download a spoof version of BOINC. That's probably outside the scope of this conversation. The installer is probably the most dangerous of all, but you have to explicitly authorise it. That's why the BOINC developers have always resisted the siren voices that call for an auto-update mechanism. |
|
891)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101269)
Posted 24 Oct 2020 by Richard Haselgrove Post: By that wording, I was meaning to refer primarily to the BOINC data folder structure, where most writing takes place. BOINC writes in other places too - offhand, the sticky GUI configurations in the user's home folder (hidden), the GUI lock file, and maybe more. But these are not areas where an executable file could end up.but should most definitely be blocked from writing outside its defined areaIs there a defined area or are we seeing a corner case of a possible definition? Edit for clarity: the boinc client writes in the data folder, and so far as I know, nowhere else. The boinc manager writes in the user folders - because the user launches and runs it. The boinc installer will need to write in the system areas, and will require administrative elevation to do so. |
|
892)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101250)
Posted 23 Oct 2020 by Richard Haselgrove Post: If I'm right BOINC is blocked out from the rest of the world here.The point is that BOINC is allowed to download and execute binaries from the projects. I don't think it's ever happened (yet!), but a project could be hacked or go rogue, and start distributing malware. That shouldn't be allowed to escape into the host operating environment. BOINC should be able to read system files, but should most definitely be blocked from writing outside its defined area. |
|
893)
Message boards :
Questions and problems :
Ubuntu 20.10 but getting error on BOINC
(Message 101247)
Posted 23 Oct 2020 by Richard Haselgrove Post: You possibly need to restart the computer for those last changes to take effect. |
|
894)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101240)
Posted 23 Oct 2020 by Richard Haselgrove Post: Well, mine wasn't intended to be a fix, just a 'proof of concept' to show that I had localised that 'error(2)' and suggesting that it was the last remaining blockage. As to exactly why BOINC can't access that file (name) - I don't know. It may be somewhere in the systemd implementation: the original manual that David describes reading sounds like it ought to be world-readable, and the permissions I could read were 777. But isn't BOINC supposed to be sandboxed under systemd. Could that block out the rest of the world? |
|
895)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101237)
Posted 23 Oct 2020 by Richard Haselgrove Post: ... managed to wipe out my LAN connections - but left 2.4GHz wifi and internet intact.I had exactly that with an all-singing, all-dancing, modem/router/dual wifi branded "BT Home Hub 5A" - if the hub restarted for any reason, the LAN wouldn't reconnect, though the internet and the wifi would. And BT makes that hub reboot once a fortnight to check it can still negotiate the same ADSL speed. At 3 a.m. After multiple phone calls, a factory replacement router, and home engineer visits - I re-contracted with BT, and got a free Home Hub 2 (which is newer and better than 5A, of course). That's worked flawlessly since April. |
|
896)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101234)
Posted 23 Oct 2020 by Richard Haselgrove Post: The first error message is simply a failure to open a directory (line 1825). Errnum 2 may elucidate.So, after letting my head spin round it a bit, I thought "Can we find the list of X servers any other way - like a static list?" Er, no - that's such a convoluted bit of code. What a stupid way of assembling a list of strings. So then I thought, why don't we put the same list in a place where we know BOINC can read it? I tried to copy the existing folder into boinc_client, but Mint refused - copied the folder, but not the contents. "Can't copy special file" - which seems like a bit of a clue in itself. So I made a new file - empty, two-character name, just like the old one - says it's a text file instead of a socket file, but who's counting by this stage? Built a new client with the new location - AND IT WORKS! Now to tell git... |
|
897)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101218)
Posted 22 Oct 2020 by Richard Haselgrove Post: It seems that we have to put our thinking caps on, and start reading at https://github.com/BOINC/boinc/blob/master/client/hostinfo_unix.cpp#L1806 Observations so far: It seems (and it's a pity) that this is is simply theoretical - just RTFM The first error message is simply a failure to open a directory (line 1825). Errnum 2 may elucidate. The second error message reports "none of the Xservers were queryable" (line 1948) Apart from that, we're looking for a needle in a haystack - or an XScreenSaver running on an Xserver. Send out the search parties. I'm getting "#define ENOENT 2 /* No such file or directory */" - but I can see it on my machine. ?? |
|
898)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101216)
Posted 22 Oct 2020 by Richard Haselgrove Post: What are the 'idle_detection_debug' messages in that scenario? Does that GPU have a monitor attached - i.e., is it known to the xhost? I'll try answering those questions on one of my Mint systems, too. Edit - sadly, doesn't work here. Still Thu 22 Oct 2020 15:11:33 BST | | [idle_detection] Error (2) opening /tmp/.X11-unix/. Thu 22 Oct 2020 15:11:33 BST | | [idle_detection] Could not connect to any DISPLAYs. XSS idle determination impossible. |
|
899)
Message boards :
Questions and problems :
Logging in to Project websites with BOINC manager
(Message 101202)
Posted 21 Oct 2020 by Richard Haselgrove Post: Yes, that's the way David Anderson has designed Science United. We don't know of any way to access those private messages. You have to ask him / SU. |
|
900)
Message boards :
Questions and problems :
gui_rpc_auth.cfg exists but can't be read. Check the File permissions.
(Message 101199)
Posted 21 Oct 2020 by Richard Haselgrove Post: That's a very fair debating point. May I point you to my analysis in https://github.com/BOINC/boinc/pull/3709#issuecomment-627906655 - the same PR that I mentioned earlier? The BOINC developers considered two cases: BOINC running on the local machine (localhost), and BOINC running on a remote machine (Controlling BOINC remotely) They took the decision - pragmatically, but as I said, debateably - that the Mac and Windows GUIs would be able to read and use the password on the local machine, without user action. Mac and Windows users wishing to control a remote machine have to jump through additional security processes. The Linux GUI was never given this ability, and so most Linux installations started with an empty password file - until the May update. Hence the current problems. My personal view is that the original developers got it about right. Giving local users (effectively) password-free access to view and manage the BOINC client running on their own local machine is not a great security risk - the client itself is effectively firewalled away from the rest of the operating system and local storage outside the data folder tree. Additional security, as you suggest, is warranted for a client running on a remote machine, possibly headless and without human oversight. If you feel that the Linux user should be required to deliberately provide a password for local control too - that's the debate - then logically the same requirement should be applied to Mac and Windows users too. |
|
901)
Message boards :
Questions and problems :
gui_rpc_auth.cfg exists but can't be read. Check the File permissions.
(Message 101197)
Posted 21 Oct 2020 by Richard Haselgrove Post: Could a Linux expert please proof-read this analysis? Starting from https://docs.oracle.com/cd/E86824_01/html/E54763/chmod-1.html The Owner of gui_rpc_auth.cfg has to be able to both read and write it - but it's not executable. So 6 The Group of BOINC users has to be able to read it. So 4 Other users - it depends whether you're a member of the boinc group. If yes, you're covered above. If no, you're an 'other user' - so you need a 4. So, 640 for members of the boinc group: 644 if you haven't done that yet. |
|
902)
Message boards :
Questions and problems :
gui_rpc_auth.cfg exists but can't be read. Check the File permissions.
(Message 101196)
Posted 21 Oct 2020 by Richard Haselgrove Post: The critical things to note are: The client service has to be able to both READ and WRITE the file. It seems it already has this ability. The user running the Manager has to be able to READ the file. They do not need to write to it, and for security, should not have permission to do so - else another user on the same system would be able to muck you about. This is the only part you need to preserve from your original changes. I'm not sufficiently Linux-fluent to express that in chmod numerics, but that's the principle. The other solution to the permissions conundrum would be for the principal operator to join a group - boinc - which already has the requisite read permission. |
|
903)
Message boards :
Questions and problems :
gui_rpc_auth.cfg exists but can't be read. Check the File permissions.
(Message 101193)
Posted 21 Oct 2020 by Richard Haselgrove Post: Last time I looked at a fresh password file it was empty. You may need to set a password manually.You need to look again, specifically at https://github.com/BOINC/boinc/pull/3709. Since May this year, a password has been required, and will be created (random 32-byte string) if not present. The Linux tools for ensuring that the Manager can read - and thus supply - the requisite password from gui_rpc_auth.cfg are woefully inadequate. |
|
904)
Message boards :
Questions and problems :
gui_rpc_auth.cfg exists but can't be read. Check the File permissions.
(Message 101191)
Posted 21 Oct 2020 by Richard Haselgrove Post: This is a very well known, very common, error message - and it's completely false. There is a password in use to protect the communications between the client and manager, and you have to ensure that both components are using the same, current, password. First - ensure that the client has been restarted since the contents of gui_rpc_auth.cfg were last altered or made readable. After that restart, the client will be expecting those contents to be sent as the password. Try the manager again (but don't hold your breath). It depends on your distribution and installation method whether the manager can find and read the gui_rpc_auth.cfg file. That won't be made obvious in any messages. The most reliable way of fixing this problem is to provide the new password yourself. Open gui_rpc_auth.cfg as a text file (may need sudo). Copy the contents. Then, find the launcher for your BOINC manager (terminal or icon - whichever you use). Add the password to the command line, thus boincmgr --password=passwordAn alternative technique is to add your user account name (since you are the one running the manager) to the 'boinc' user group, so that your copy of the manager can read boinc's files. This may require a Linux restart - I'm not an expert on Linux. |
|
905)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101143)
Posted 18 Oct 2020 by Richard Haselgrove Post: Sorry, no. Still "XSS idle determination impossible." Next? |
|
906)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101142)
Posted 18 Oct 2020 by Richard Haselgrove Post: I'll try that xhost command later, and report back.Tried it (with sudo). Got the response localuser:boinc being added to access control listwhich looks hopeful - but still the same error message in idle_detection_debug. I'll try a reboot, as seemed necessary when adding my user account to the boinc group. |
|
907)
Message boards :
Questions and problems :
Linux Suspend when computer is in use bug.
(Message 101139)
Posted 18 Oct 2020 by Richard Haselgrove Post: In https://github.com/BOINC/boinc/issues/3715, I've recently posted the output of the 'idle_detection_debug' log under Mint. Sun 11 Oct 2020 15:43:28 BST | | [idle_detection] Error (2) opening /tmp/.X11-unix/.and noted There is a 'socket file' called X0 in /tmp/.X11-unix/I'll try that xhost command later, and report back. |
|
908)
Message boards :
Questions and problems :
Logging in to Project websites with BOINC manager
(Message 101120)
Posted 16 Oct 2020 by Richard Haselgrove Post: Which is how, and why, you can see your credit scores, your preferences, and your computer details directly from a BOINC native message board. |
|
909)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101111)
Posted 15 Oct 2020 by Richard Haselgrove Post: In the UK, with the second iteration of our app (the first didn't work), apparently you get an alert from Android that you've been close to an infected person, followed by a second alert from the app to say you weren't close enough for long enough to catch anything, and to ignore the first alert. I haven't installed it either. |
|
910)
Message boards :
Questions and problems :
NVIDIA RTX3090
(Message 101107)
Posted 15 Oct 2020 by Richard Haselgrove Post: It is known that GPUGrid tasks fail, but Einstein tasks run normally. We await reports from other projects. |
|
911)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 101102)
Posted 14 Oct 2020 by Richard Haselgrove Post: Another night, another sacrifice. Venue NOT accepting orders, the app says. Completely untrue, as the barmaid taking my order confirms - it's only the app which isn't taking orders, because it would only charge the lunchtime price, and even then salami-slice their cut off the top. The landlady is not impressed by the new technology when it meets the real world ... |
|
912)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 101092)
Posted 13 Oct 2020 by Richard Haselgrove Post: It's been a while since we celebrated the Tuesday evening ritual. But I still went down to the place of sacrifice as usual ... On Sunday, for the first time, they had joined up with an 'app' to enable socially-distant ordering. I gave it a whirl. Yesterday, it didn't know it was Monday. (We have an all-evening 'happy hour' on the quietest night of the week) Today, I went right through. The app didn't import all my payment details from Google Pay (it got the expiry month for my debit card, but not the year). I asked the barmaid how much they receive from an app purchase - expecting both the bank, and the app developer, to take a slice. She assured me they get the full amount. Hmmm. We'll keep an eye on that. Paid for my next pint with cash. Heard the next table asking what the policy was on people from different households meeting for a quiet drink. Advised the barmaid that the traditional response was "Don't ask, don't tell." Apart from that, it was a good night. |
|
913)
Message boards :
Questions and problems :
constant writes to disk by boinc
(Message 101060)
Posted 11 Oct 2020 by Richard Haselgrove Post: There's still a boinc_task_state.xml file in every slot directory, to keep track of how far it's got. That's management data for BOINC, not for the project. |
|
914)
Message boards :
Questions and problems :
constant writes to disk by boinc
(Message 101049)
Posted 10 Oct 2020 by Richard Haselgrove Post: Could be the client writing its own internal state into client_state.xml? That file isn't used in normal running, but is read at start-up - so it needs to be current in case there's a sudden power outage. You could check the time-stamp on that file. |
|
915)
Message boards :
Questions and problems :
elapsed time is misleading
(Message 101037)
Posted 10 Oct 2020 by Richard Haselgrove Post: There are two ways of setting this value: either globally, through the website of one of the projects your computer is attached to, or locally via BOINC Manager. They can't be mixed: once you set any local value, local values take precedence and the website values are disregarded. It sounds like you may be using local, BOINC Manager, settings. To see the alternative setting, you need to be in Advanced View, not Simple View. Switch from the View menu, or use the keyboard shortcuts: Ctrl+Shift+A for Advanced, or Ctrl+Shift+V for Simple. |
|
916)
Message boards :
News :
Web site makeover
(Message 101035)
Posted 10 Oct 2020 by Richard Haselgrove Post: But, if i'm not wrong, i remember that SU receives funds from NSA....I think that was presented as NSF, but you may be right. |
|
917)
Message boards :
Projects :
News on Project Outages
(Message 101027)
Posted 9 Oct 2020 by Richard Haselgrove Post: GPUGrid has deployed replacement apps. Work has started flowing again, but is in short supply - it'll take a while to re-fill everyone. |
|
918)
Message boards :
Questions and problems :
Unexpected exit
(Message 101024)
Posted 9 Oct 2020 by Richard Haselgrove Post: Things usually break a) when you change something b) when they get old and overheat When did you last upgrade either Linux or BOINC? Did you hoover out the dust bunnies while you were investigating? |
|
919)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 101021)
Posted 9 Oct 2020 by Richard Haselgrove Post: Welcome to the magic roundabout. |
|
920)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 100991)
Posted 8 Oct 2020 by Richard Haselgrove Post: Sounds like it is time to shut the rigs off.I don't think we should punish the projects' science for David's loss of direction. But maybe we could ask the projects for their collective opinion? |
|
921)
Message boards :
News :
Web site makeover
(Message 100988)
Posted 7 Oct 2020 by Richard Haselgrove Post: Have you made it possible for Science United volunteers to interact with projects - preferences, message boards, friends, private messages - yet? |
|
922)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 100987)
Posted 7 Oct 2020 by Richard Haselgrove Post: And lo - like Gaul, BOINC was divided into three parts. Scientists: Developers: Computing power: No mention of the volunteers who make it all happen. |
|
923)
Message boards :
Projects :
News on Project Outages
(Message 100981)
Posted 7 Oct 2020 by Richard Haselgrove Post: I've only just got through to their forums for the first time since I posted (it helps a lot not to request an https connection). No response from the administration yet, but I've sent a PM off to Toni. Edit - message from Toni in the news forum. Suspending project - Until we update the app. Sorry. |
|
924)
Message boards :
Projects :
News on Project Outages
(Message 100977)
Posted 7 Oct 2020 by Richard Haselgrove Post: GPUGrid task failures All my Linux tasks issued after about 12:30 UTC today (Weds 7 October) started to fail instantly. All Windows tasks issued about an hour later failed the same way. The GPUGrid server is very difficult to reach currently, probably because of the large number of tasks failing and being reported/replaced, but we suspect a license expiry on their software (as has happened before). Please set 'No New Tasks' on this project for the time being, to reduce the load while we investigate further. |
|
925)
Message boards :
BOINC Manager :
When I Click Add Project, There are No Projects
(Message 100959)
Posted 4 Oct 2020 by Richard Haselgrove Post: It's version 7.16.10.v7.16.10 was only ever a short-lived test version for Mac. Try upgrading to the recommended final release version 7.16.11. |
|
926)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100955)
Posted 4 Oct 2020 by Richard Haselgrove Post: Ah - mystery solved. It was requested earlier this year (#3463), and added in June (#3800) - just in time for the much-delayed v7.16.11 release. You'll have to upgrade - if we're talking Linux, Gianfranco Costamagna has a PPA with v7.16.11 in it already. |
|
927)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100954)
Posted 4 Oct 2020 by Richard Haselgrove Post: All of my tn-grid WUs have been v1.10 regardless of which level of optimisation they’re at. That’s running Ubuntu or Mint.Yes, I saw that from the project's Applications page after I posted - v1.10 could be either Linux or Mac, but v1.11 can only be Windows. Here's one from Mint, also at v7.16.11: name: 172467_Hs_T001760-INPP5F_wu-172_1601265762688_2So I think the problem can only be the version of BOINC that he's using. I'll check when that line was added, next time I'm back at a Windows machine. |
|
928)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100936)
Posted 3 Oct 2020 by Richard Haselgrove Post: I wondered if there might be something odd about your project (failure to checkpoint, or something). So I ran a test task (same BOINC version as last time): name: 172770_Hs_T137591-GBE1_wu-121_1601737922481_1So that wasn't it. Where did you get your BOINC version from? (and why did we each get a different app_version number?) |
|
929)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100935)
Posted 3 Oct 2020 by Richard Haselgrove Post: I don't have an "elapsed task time" row in the result of --get_tasks. There is estimated CPU time and there is slot, with nothing in-between for me.In a word, no. Mine was taken from BOINC v7.16.11 (released last month), running under Windows 7/x64. Having looked at the code, I suspect that boinccmd uses exactly the same function to parse the client reply as the GUI manager does - and I don't think that's changed much in 10 years. I'll take a proper look tomorrow. |
|
930)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100932)
Posted 3 Oct 2020 by Richard Haselgrove Post: Actually, looking at the output from --get_tasks, I think it's been done already. For a completed task. name: wu_sf7_DS-16x8_Grp265895of655360_0For a running task, name: wu_sf7_DS-16x8_Grp266373of655360_0For an unstarted task. name: wu_sf7_DS-16x8_Grp268382of655360_0So, what do you still need, apart from the ones I've highlighted? Be aware that your computer also records a permanent list of all completed tasks by project, in the file "job_log_[project].txt". Lines are written on completion, and the one for the completed task above reads 1601725393 ue 3812.124815 ct 2694.590000 fe 10000000000000 nm wu_sf7_DS-16x8_Grp265895of655360_0 et 2766.865900 es 0We can break that down for readability: 1601725393 ue 3812.124815 ct 2694.590000 final CPU time fe 10000000000000 nm wu_sf7_DS-16x8_Grp265895of655360_0 et 2766.865900 final elapsed time es 0 |
|
931)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100930)
Posted 3 Oct 2020 by Richard Haselgrove Post: The boinccmd code is pretty straightforward. I'll see if I can do one of my 'copy and tweak' pull requests over the weekend. |
|
932)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100927)
Posted 3 Oct 2020 by Richard Haselgrove Post: However the `boinccmd` value for "current CPU time" does not correspond to the GUI's task column for "Elapsed".They're two different measurements. 'Elapsed' time means the time the task has been running, as shown by a wall-clock. 'CPU time' means the time the CPU - specifically - has been active during that time. There are many reasons why the CPU may be active for less than 100% of the run time. The commonest is if the task does most of its scientific work on a GPU: in that case, the CPU is only needed for management and to supply data. That needn't take long. Other reasons might include thermal throttling (the CPU is paused periodically to cool down), or an over-committed computer (the task is swapped out of the CPU for a few milliseconds so something more urgent can run instead). BOINC originally worked on CPUs only, and CPU time was the only value measured. Elapsed time became more important when GPUs were introduced, about 10 years ago. The client still records both values, but CPU time was removed from the Manager display, and elapsed time was never added to the boinccmd tool. We should tidy that up some day. |
|
933)
Message boards :
Projects :
News on Project Outages
(Message 100922)
Posted 2 Oct 2020 by Richard Haselgrove Post: Work fetch debug is a bit heavyweight for the initial investigation. I would start with sched op debug - that's lightweight, and can be left running all the time. |
|
934)
Message boards :
Projects :
projects not listed
(Message 100906)
Posted 30 Sep 2020 by Richard Haselgrove Post: YAFU: http://yafu.myfirewall.org/yafu/YAFU describes itself as an 'alpha' (early stage test) project. The invitation code is on the front page, below the 'Join YAFU' button. TN-GRID: http://gene.disi.unitn.it/test/The invitation code is also on the front page, below the 'Join TN-GRID' line. |
|
935)
Message boards :
Projects :
projects not listed
(Message 100892)
Posted 29 Sep 2020 by Richard Haselgrove Post: Use a web browser and search tool to find the web site of any such project. Read the project's own description: many will be experimental, in test mode, or have finished their project and not be sending any new work. If you decide to work for them, read the attachment instructions on their website. Some will require you to register an account on their website first, and some may require you to use an invitation code. There will normally be a public statement of the url to be used when attaching to the project: if that isn't stated, you can try using the website address that you found at the beginning. When everything is ready, open the 'Add project ...' tool in BOINC Manager. Ignore the list of existing projects, and simply paste the url of your new project into the 'Project URL:' box at the foot of the page. Click 'Next', and continue to follow the instructions. |
|
936)
Message boards :
BOINC Manager :
closing boinc manager now kills client
(Message 100890)
Posted 29 Sep 2020 by Richard Haselgrove Post: There is a registry key in Windows [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager] "AutomaticallyShutdownClient"=dword:00000000The equivalent in Linux is a hidden file in the current user's home folder called '.BOINC Manager', containing the line AutomaticallyShutdownClient=0 |
|
937)
Message boards :
Questions and problems :
possible to use boinccmd --quit when multiple clients are running?
(Message 100879)
Posted 28 Sep 2020 by Richard Haselgrove Post: Read the User Manual for the Boinccmd tool. The GUI rpc port can be specified as part of the hostname argument: hostname can be a domain name, an IPv4 address, or an IPv6 address. If the client uses a non-default GUI RPC port, you can specify it as hostname:port, IPv4_addr:port, or [IPv6_addr]:port. |
|
938)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100867)
Posted 26 Sep 2020 by Richard Haselgrove Post: Agreed. They've used a funny sort of CUDA app development which requires explicit pre-knowledge of the card characteristics. The exact error message (on an A100, cc8.0 datacentre GPU) is # Engine failed: Error compiling program: nvrtc: error: invalid value for --gpu-architecture (-arch) |
|
939)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100864)
Posted 26 Sep 2020 by Richard Haselgrove Post: Found the thread, and saw the error message in the results. Yup, that's a show-stopper, even though the A100 card is only cc8.0 Meanwhile, I've submitted #4031 to deal with the flops display. |
|
940)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 100859)
Posted 25 Sep 2020 by Richard Haselgrove Post: Yorkshire 'supercell' storm covers region in hail That one missed me by about five miles to the East. I got spectacular reverberations through the local hills, and a few direct bangs from 'line of sound' strikes, but no hail and only modest amounts of rain. It's suddenly turned brrrrrrr cold, though. |
|
941)
Message boards :
BOINC client :
7.16.11 available for testing for Windows and Macintosh
(Message 100856)
Posted 25 Sep 2020 by Richard Haselgrove Post: Yes - initial dot surname at btinternet dot com will reach me. |
|
942)
Message boards :
BOINC client :
7.16.11 available for testing for Windows and Macintosh
(Message 100854)
Posted 25 Sep 2020 by Richard Haselgrove Post: Could you post the raw log data, please, and I'll look at it. Sometimes you need to look at the whole picture in the round - something completely different might be coming into play. |
|
943)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 100851)
Posted 24 Sep 2020 by Richard Haselgrove Post: Grab 'em while you can. My local supplier has dropped prices on the 20 ranges, but is currently showing: RTX 2060 SUPER: £354.97 - £486.90 RTX 2070 SUPER: £446.39 - £655.96 RTX 2080 SUPER: £765.99 - £827.32 They're not currently showing prices for the 30 range, which are still out of stock. |
|
944)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 100848)
Posted 24 Sep 2020 by Richard Haselgrove Post: So much so for the price of the 3080 being on par with that of a 2080 Super, as those only cost €499-599 here. 2070 Super goes from €549-699I'm not sure you're reading the right column there - in this country, 2080s are still more expensive than 2070s. Those look more like 2060 prices. |
|
945)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100843)
Posted 24 Sep 2020 by Richard Haselgrove Post: Might be wise to throw in a <cpu_usage> line for completeness. It's not marked as optional in the manual. |
|
946)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100839)
Posted 23 Sep 2020 by Richard Haselgrove Post: How do you want to set that experiment up? What parameters are you looking for?This is just for baseline BOINC users, not fancy optimisers. Ideally a single 30x0 card, in a host with plenty of power and cooling (so nothing gets throttled). Run a known - preferably CUDA - app for long enough to get a good idea of performance. Slap in an app_config.xml file with <gpu_usage>.5</gpu_usage>, and record what happens. |
|
947)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100834)
Posted 23 Sep 2020 by Richard Haselgrove Post: I've received reliable reports that BOINC shows 14,884 GFLOPS peak for the RTX 3080, and SIV shows 29,768 - exactly double. Since we use different API calls for getting the shader count, that'll be the difference - SIV will be right, and us wrong. That leaves the question of the doubled FP32 pipeline unresolved. That may require direct experimentation on the hardware - I'm thinking possibly including running two tasks in parallel. |
|
948)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100827)
Posted 22 Sep 2020 by Richard Haselgrove Post: Twice as fast at Primegrid compared to a RTX 2080 Ti running a CUDA app.That's more surprising. I'd have expected the speed increase to be less, because the RTX 2080 can use its INT32 pathway, which it (probably - I'm not fully knowledgeable on Einstein's maths) couldn't use at Einstein. |
|
949)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100824)
Posted 22 Sep 2020 by Richard Haselgrove Post: Thanks for that - I've looked through some of the logs, and it all checks out (like the 10 GB VRAM shown in stderr.txt, against the 4 GB recognised by BOINC. That's for another day.) Remember that this is an OpenCL app, and relies of the efficiency of the OpenCL translation layer in using the new CUDA functions. And it wastes something like 14 seconds at the end on pure CPU work - so we can't put too much reliance on the speed ratio. Also, the host only ran for one day - looks like a burn-in test. Now, if we could just find a CUDA example at GPUGrid... |
|
950)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100820)
Posted 22 Sep 2020 by Richard Haselgrove Post: The Collatz comparison doesn't really help to answer my question. Collatz will be especially well served by the previous Volta and Turing chipsets, because of their additional, independent, pathway for INT32 calculations. The Ampere chipset makes that extra pathway available for FP32 calculations too, which makes it more widely suitable for the type of research that BOINC is designed to support. |
|
951)
Message boards :
Questions and problems :
BOINC install as a service: noob
(Message 100810)
Posted 21 Sep 2020 by Richard Haselgrove Post: Which network machine are you trying to install on? The Domain Controller is likely to be problematic, even though the absolute prohibition has been lifted. If a domain member, it would be as well to ensure that your Domain account is also registered as having local admin rights on that machine. |
|
952)
Message boards :
Projects :
Mind Modeling Issue
(Message 100799)
Posted 21 Sep 2020 by Richard Haselgrove Post: I am probably being a bit picky here but you can also type or paste a url directly into the manager to attach to a project. - For me the easiest way with testing branch of a project which isn't listed in the drop down menu for instance.I think (but I'm not certain) that if you paste a url into the GUI 'Add project' wizard', it'll still get checked through the 'get_project_config.php' mechanism. |
|
953)
Message boards :
GPUs :
Specifications for NVidia RTX 30x0 range?
(Message 100797)
Posted 21 Sep 2020 by Richard Haselgrove Post: May I ask you hardware enthusiasts to double-check my thoughts on running BOINC on the new RTX 3070/3080/3090 range? I've been studying NVIDIA A100 Tensor Core GPU Architecture and NVIDIA Ampere GA102 GPU Architecture BOINC uses the number of CUDA cores per SM, and a flops multiplier, to estimate the GPU's peak speed. I'm getting that the GA102 (and above, but not the A100) benefit from both an increase from 64 to 128 cores per SM, and the ability to process two FP32 streams concurrently. So I think that the current v7.16.11 BOINC client will rate the new cards at one-quarter of the flops reported by other tools. Can anybody confirm that? If it's true, I'll code a patch for the next release of BOINC. |
|
954)
Message boards :
Projects :
Mind Modeling Issue
(Message 100796)
Posted 21 Sep 2020 by Richard Haselgrove Post: That is controlled by the project rather than Boinc.It's even more complicated than that. There are three different ways of connecting to a project. 1) via BOINC Manager. The Manager takes the value from the all_projects_list, but validates it with the project via a call to get_project_config.php The project will be attached using the returned value - which may even be a completely different domain, let alone swapping between https / http. 2) via boinccmd. The project will be attached using the url you supply - unchecked. 3) via an Account Manager. The project will be attached using a url specified by the AM administrator - unchecked. In the current case (MindModelling), somebody suggested using http://mindmodeling.org/. If you paste that into a browser, you are redirected to https://mindmodeling.org/ But https://mindmodeling.org/get_project_config.php returns <master_url>mindmodeling.org/</master_url>without a protocol specifier - which is a project mistake. The results are probably undefined. The only way you can get attached twice is if you use two different methods - usually, both an Account Manager and a native BOINC route. It's best not to do that. |
|
955)
Message boards :
Questions and problems :
BOINC install as a service: noob
(Message 100783)
Posted 18 Sep 2020 by Richard Haselgrove Post: Please tell us your operating system (Apple, Linux, Windows, other) - even that much makes a big difference. |
|
956)
Message boards :
The Lounge :
CPUcoin scam?
(Message 100771)
Posted 17 Sep 2020 by Richard Haselgrove Post: David Anderson wrote:The two are not mutually exclusive - the 'talk' doesn't sound like a full security audit.I did talk with them ...So not a scam. The website - while full of impressive jargon - is actually rather short on practical, working, detail. [But with much better spelling and grammar than most scams] I'd assume that, since you have to download a 'miner' app, some of your CPU power will be diverted in their direction. |
|
957)
Message boards :
The Lounge :
CPUcoin scam?
(Message 100764)
Posted 16 Sep 2020 by Richard Haselgrove Post: this?Well, I've had a look at 'this', and I'm not convinced. 1) It is TRUE that WCG and Rosetta are assisting in covid-19 research using the BOINC platform. They are doing this anyway, irrespective of any coin tie-in. 2) There has been no explicit endorsement of, or collaborative development with, CPUcoin by the BOINC community at large. [There may have been contact with David Anderson individually - his name is mentioned - but that's a different question]. 3) The CPUcoin team seem to be willing to watch you run the specific named BOINC projects, and award you coin (no mention of how much). There's no mention of any technical link between BOINC and CPUcoin in the 'Security & Data Flow Document'. Personally, I'd simply run BOINC as normal, and leave all the self-promotion out of it. But explore if you wish. |
|
958)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 100722)
Posted 12 Sep 2020 by Richard Haselgrove Post: I'm still waiting on the GPU stuff....You'll be waiting a long time before they bring out a GPU capable of holding the whole earth's climate in memory at the same time. CPUs can virtualise memory, but GPUs can't as yet. And when they can, it'll be limited to PCIe bus speeds. |
|
959)
Message boards :
Questions and problems :
Controlling Remote Machines
(Message 100697)
Posted 11 Sep 2020 by Richard Haselgrove Post: Routers and firewalls would normally let your requests out of Crete without a problem, though the firewall may need port 31416 opening. The problem is getting back into the UK (isn't it always). If you didn't set up the inward ports before you left, and don't have full remote systems management access already, security should prevent you changing anything at this late stage. Just relax on the beach until you get back. BoincTasks uses exactly the same protocol and access rules as BOINC Manager, so that would be blocked too. The only gains for your situation would be configurable refresh intervals, and the ability to see both main machines on the same GUI. |
|
960)
Message boards :
Questions and problems :
Controlling Remote Machines
(Message 100689)
Posted 10 Sep 2020 by Richard Haselgrove Post: How 'remote' is your laptop from the main machines? There should be at least a router, and possibly a firewall, that you have to negotiate to access them from 'the internet'. You (or your IT manager, if different) would have to open the two IP addresses and port 31436 on each. DNS won't work inbound across a router. My preferred approach in this scenario was to install a main-site router with VPN termination in the router - log in to that via VPN, and then you're effectively a part of the internal LAN. Everything, including DNS, should work. [My experience was with Windows servers, running as a Domain under Active Directory, so DNS and DHCP were local. YMMV.] Note: you need a very, very, good remote-to-main internet service for this to work - preferably gigabit. BOINC's own manager requests an update every second (non-configurable), and gets a flood of data in return. And has to receive and process that within the second. BoincTasks would be a better bet for slower lines. |
|
961)
Message boards :
Questions and problems :
Boinc is running, but I can't see anything in Boinc Manager (Ubuntu 18.04.5)
(Message 100658)
Posted 9 Sep 2020 by Richard Haselgrove Post: I think that, when I hit this same snag on my first Linux machine, the group membership listing didn't update immediately. But it was correct after the next restart. |
|
962)
Message boards :
BOINC client :
BOINC is not working
(Message 100650)
Posted 8 Sep 2020 by Richard Haselgrove Post: In June, David Anderson announced that a BOINC client called SUNET had been launched. From what I could see then, that client appeared to be based in China, but I can't now see it from outside China. You could try using that client instead. Please let us know how you get on. |
|
963)
Message boards :
BOINC client :
7.16.11 available for testing for Windows and Macintosh
(Message 100608)
Posted 5 Sep 2020 by Richard Haselgrove Post: Yes, it was odd that after Vitalii specifically asked me to do it, and I managed to turn it round quicker than I expected, it remained unmerged for several days. Maybe Vitalii took a holiday, as I intend to do next week. But not making the final mass-transfer for release was nothing to do with that - the timing seems to be completely disconnected from any concept of which changes are important or form a connected whole. As you know, I was much more concerned to see that #3988 joined its ancestors before release. #3999 is another that should have been in the same group, but it's less important. |
|
964)
Message boards :
BOINC client :
7.16.11 available for testing for Windows and Macintosh
(Message 100606)
Posted 5 Sep 2020 by Richard Haselgrove Post: I see your https://github.com/BOINC/boinc/pull/3982 was moved to 7.18, so someone decided that in the meantime people can look in the registry to disable that setting when they set it.Quite a lot of things which should have waited for the next development cycle (and were marked accordingly) were mass-migrated into 7.16 because David doesn't know how to manage the "bugfix and cherrypick" nimble release process. |
|
965)
Message boards :
BOINC client :
7.16.11 available for testing for Windows and Macintosh
(Message 100599)
Posted 4 Sep 2020 by Richard Haselgrove Post: https://boinc.berkeley.edu/alpha/.I think you have to ask David Anderson personally to be admitted to the Holy of Holies. I don't know for certain, because I've never tried. The Alpha testing regime, so called, involves deterministically filling in a chart to display that none of the obvious major functions of BOINC have been clearly broken. It's a bit late for that. Over the last year, we've repeatedly shown that several of the non-obvious (but vital) parts of BOINC have been broken through typos and other flaws in unrelated tweaks. I don't think many of those would have been found through ritualistic testing, so I prefer to spend my time actually watching BOINC as it runs and reporting the anomalies I see in real life. |
|
966)
Message boards :
Questions and problems :
Possible to schedule downloads / uploads
(Message 100585)
Posted 4 Sep 2020 by Richard Haselgrove Post: Yes. Open BOINC Manager, and switch to 'Advanced View' (if you aren't using it already). From the menu, choose 'Options', 'Computing preferences...'. Switch to the fourth tab, 'Daily schedules'. In the lower part of the screen, fill in your preferred schedule, and save it. Also look at the 'Activity' menu. Make sure that 'Network activity according to preferences' is selected. That should do it. |
|
967)
Message boards :
BOINC client :
Feature request - boinccmd - need to control amount of cores used during a specific time (--set_run_mode)
(Message 100565)
Posted 1 Sep 2020 by Richard Haselgrove Post: See, for example, https://github.com/BOINC/boinc/issues/41 (That one was transferred from https://boinc.berkeley.edu/trac/ticket/41 - a mere 13 years old. It'll probably get the right to vote and drink before it's implemented) |
|
968)
Message boards :
Questions and problems :
Is it possible to disable applications in the app_config file?
(Message 100555)
Posted 1 Sep 2020 by Richard Haselgrove Post: I think we're back to app_info, then. We know that works, because of many years' experience at SETI. |
|
969)
Message boards :
Questions and problems :
Is it possible to disable applications in the app_config file?
(Message 100553)
Posted 1 Sep 2020 by Richard Haselgrove Post: I'm not seeing any sign in either log to say that max_concurrent is intervening to say Oi! You can't run that! Edit - until possibly that last log, while I was typing. |
|
970)
Message boards :
Questions and problems :
Is it possible to disable applications in the app_config file?
(Message 100547)
Posted 1 Sep 2020 by Richard Haselgrove Post: What does <cpu_sched_debug> or <rr_simulation> have to say? (looking in code, max_concurrent seeks to be tested in rr_sim) |
|
971)
Message boards :
Questions and problems :
Is it possible to disable applications in the app_config file?
(Message 100538)
Posted 1 Sep 2020 by Richard Haselgrove Post: Surely the scheduler is smarter than that :-)I shouldn't bet on it. Einstein doesn't use a standard scheduler, with all the multitude of twists and tweaks which have been added (individually, not holistically) over the years. And any (server) scheduler can only act on the information it's been given. In this case, the information will be: I have an app, and I want n seconds of work. Max_concurrent isn't included. |
|
972)
Message boards :
Questions and problems :
Is it possible to disable applications in the app_config file?
(Message 100536)
Posted 1 Sep 2020 by Richard Haselgrove Post: Hmm. Hadn't thought of that :-)No, not that way. app_config.xml doesn't pass enough information back to the server about what work you can handle..As a matter of interest, why wouldn’t max_concurrent = 0 work? My worry would be that BOINC downloads a task, but never runs it. Bill would be fetching tasks from Einstein as normal, with no control over what comes down the line - rather different from, say, 'resource share 0' for a backup project. One of us should try it sometime ... |
|
973)
Message boards :
Questions and problems :
Is it possible to disable applications in the app_config file?
(Message 100534)
Posted 1 Sep 2020 by Richard Haselgrove Post: No, not that way. app_config.xml doesn't pass enough information back to the server about what work you can handle. You could do it by going the whole way to Anonymous platform - using an app_info.xml file instead. If you have a compatible machine running the application you want, you'll already have all the files you need, and a suitable <app_version> segment in client_state.xml. You just have to connect them all together... Be warned: writing app_info.xml files is a specialist skill, and you have to get everything exactly right - if you get anything wrong, BOINC will reject the file and delete everything that goes with it (tasks, files, ...). Assemble everything you need in a clean directory: when you think you're ready to copy it across the BOINC, let your cache drain to nothing and only allow minimal work fetch until you're sure it's working. |
|
974)
Message boards :
Projects :
Does anyone know the status of ClimatePrediction.net?
(Message 100502)
Posted 27 Aug 2020 by Richard Haselgrove Post: See notes in News on Project Outages (always at the top of this board). They suffered a major power outage about a week ago - probably related to bad weather at the time, ironically - and their university hosts seem to be struggling to get everything back up. |
|
975)
Message boards :
Questions and problems :
wasted calculation time and cpu power (?)
(Message 100490)
Posted 27 Aug 2020 by Richard Haselgrove Post: BOINC does have a setting to control checkpoints (1 minute from memory)BOINC doesn't 'control' checkpoints, it 'permits' them. A checkpoint is a collection of all the data needed for a scientific calculation to resume from the point reached so far. Sometimes that might be simple (last candidate prime checked, last candidate Collatz dis-proof): sometimes it's megabytes of data (global atmospheric state, folded protein energy levels). And it has to be consistent / coherent, ready for the algorithm to restart. Science apps can only checkpoint when the internal data is in a state of readiness. Yes, we have to educate scientists about the need: no, we can't force them to heed our advice. The BOINC control cited merely allows a user to say "This machine isn't usually switched off, so a checkpoint is rarely needed: you can save time and energy by not writing a checkpoint so often." |
|
976)
Message boards :
Projects :
News on Project Outages
(Message 100480)
Posted 26 Aug 2020 by Richard Haselgrove Post: I've just finished one from the short dev site batch. Final log: 26/08/2020 15:02:03 | cpdnboinc_dev | [sched_op] Fetching master fileSo, we've run so long under this blackout that the local client has moved to the maximum backoff between attempts to report - 24 hours. Once you hear that it's back up, a manual update may help to get things going again. |
|
977)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 100463)
Posted 25 Aug 2020 by Richard Haselgrove Post: For Want of a Nail the battle was lost ... |
|
978)
Message boards :
Questions and problems :
Downloading from weight 0 project instead of weight 1000
(Message 100457)
Posted 24 Aug 2020 by Richard Haselgrove Post: This forum is fulfilling its purpose. Question: How do I change minor parameters in BOINC's operations? Answer: you will be guided to the appropriate point in the user interface, user manual, third-party tools, or wherever. Question: How do I make a fundamental change in BOINC's core programming? Answer: you will be guided to GitHuib, where you can submit your own improved code. This forum is a triage station, not the all-seeing eye. |
|
979)
Message boards :
The Lounge :
Happy Birthday BOINC
(Message 100450)
Posted 24 Aug 2020 by Richard Haselgrove Post: Old enough to get wasted! |
|
980)
Message boards :
Projects :
News on Project Outages
(Message 100437)
Posted 24 Aug 2020 by Richard Haselgrove Post: Judging by the statistics shown at https://www.boincstats.com/stats/188/project/detail/lastDays, BOINC @ TACC has been down since 18 July. |
|
981)
Message boards :
Questions and problems :
Server cancel for an already running task?
(Message 100418)
Posted 23 Aug 2020 by Richard Haselgrove Post: If a task has not even started by the time the deadline is reached, the local client (not the server) will abort it. If the task has been started, the client will let it run to completion. As Dave said, server instructions will vary according to the policy of the individual project concerned. Most projects will simply let nature take its course. The only common exception is when a project realises that an entire batch of tasks has been prepared from faulty data, and has no scientific value. Under these circumstances, server operators have the power to abort a task at whatever stage it has reached. Most projects are extremely reluctant to use this power, and will only use it as a last resort. Project decisions like that are taken on the basis of scientific need alone, and pay no attention to credit. The purpose of aborting a batch of tasks is to free up the machines from wasting time on useless work: it enables your machine to start working on replacement (and hopefully corrected) tasks as quickly as possible. |
|
982)
Message boards :
GPUs :
Remote access stops GPU calculations
(Message 100417)
Posted 23 Aug 2020 by Richard Haselgrove Post: This is a known problem, and is caused by security measures introduced by Microsoft into the GPU driver model from Windows Vista onwards. There is no known way of combining Windows RDP with GPU crunching. The workrounds available would depend on what exactly you wish to do. If you wish simply to observe and manage BOINC remotely, follow the steps in Controlling BOINC remotely. You can either use a standard copy of BOINC Manager to access the remote machine, or you can use a third-party utility like BoincTasks. BOINC Manager can only display a single remote machine at a time: BoincTasks can combine the data from multiple remote machines. If you want to go further and retain remote access to the entire Windows environment on the remote machine, you would need to switch to a different remote access solution, like one of the many flavours of VNC. |
|
983)
Message boards :
Projects :
News on Project Outages
(Message 100375)
Posted 21 Aug 2020 by Richard Haselgrove Post: Zip files are still being uploaded at least for the most recent Linux models as they go to a server somewhere else in the world. However everything at Oxford still appears to be down.I've just uploaded a zip to upload11.cpdn.org (192.171.139.103). Out of curiosity, I tried to find the location of that IP address. The site I chose offered four different database answers: Swindon, London, Nottingham (all in England), and Currie in Scotland. |
|
984)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 100330)
Posted 17 Aug 2020 by Richard Haselgrove Post: Went out to perform the weekly ritual sacrifice. Because it's been Monday all day (don't ask), stayed for an extra one. Emerged to the most intense monsoon I've experienced since Bombay '84 (unless I'm blanking Kenya '86). Listen to the drips... |
|
985)
Message boards :
Questions and problems :
priority of tasks
(Message 100283)
Posted 11 Aug 2020 by Richard Haselgrove Post: If we move on a stage. If you have units that are not completed within 3.5 days, your machine will not be considered as 'reliable' and so will not get re-sends which have only a 3.5 day deadline. If that is the case, then reduce the number of that project that you hold in cache in your Device Profiles.That precise statement is true for at least one project - Numberfields@Home. As others have said, other projects have widely different policies and practices. When speaking of BOINC, you have to distinguish between BOINC generalities, and project specifics. |
|
986)
Message boards :
Questions and problems :
Errors/Timed out
(Message 100261)
Posted 9 Aug 2020 by Richard Haselgrove Post: Turn on <sched_op_debug> logging, and it'll you the size of the work request in seconds. The very first request (when you first join) is always 1 second. If you are entirely empty, the first 'real' request will be 25,920 seconds per device (core or GPU). How many tasks that turns into depends on the running average estimate of the speed of your device. If you have some work already, the request will be reduced accordingly. |
|
987)
Message boards :
Questions and problems :
Errors/Timed out
(Message 100257)
Posted 9 Aug 2020 by Richard Haselgrove Post: Doesn't affect me. I'm just finishing some CPDN tasks, well into their 11th day of running. They were downloaded immediately I released the 'No New Tasks' for the project (it was a special request run - not my usual fare). I think you're confusing runtimes with deadlines. BOINC won't download a task if it thinks you can't complete it before deadline, but that's not the same thing as the cache setting. |
|
988)
Message boards :
Questions and problems :
SRBase suddenly not requesting CPU tasks
(Message 100255)
Posted 9 Aug 2020 by Richard Haselgrove Post: "computer id not assigned" means that the host has never, ever, managed to contact the scheduler for the project in question. There are five projects in this state: 09/08/2020 08:35:45 | http://burp.renderfarming.net/ | URL http://burp.renderfarming.net/; Computer ID not assigned yet; resource share 100 09/08/2020 08:35:45 | http://casathome.ihep.ac.cn/ | URL http://casathome.ihep.ac.cn/; Computer ID not assigned yet; resource share 100 09/08/2020 08:35:45 | http://setiathome.berkeley.edu/ | URL http://setiathome.berkeley.edu/; Computer ID not assigned yet; resource share 100 09/08/2020 08:35:45 | http://setiweb.ssl.berkeley.edu/beta/ | URL http://setiweb.ssl.berkeley.edu/beta/; Computer ID not assigned yet; resource share 100 09/08/2020 08:35:45 | BOINC@TACC | URL https://boinc.tacc.utexas.edu/; Computer ID not assigned yet; resource share 100 Some have been tested this time: 09/08/2020 08:35:45 | http://casathome.ihep.ac.cn/ | Fetching scheduler list 09/08/2020 08:36:16 | http://setiweb.ssl.berkeley.edu/beta/ | Fetching scheduler list 09/08/2020 08:36:25 | BOINC@TACC | Fetching scheduler list Each in turn returned an error. BURP and CAS are old projects which have possibly stopped functioning: they are no longer in the official list (but may be kept by BAM! 'just in case'). SETI has been moving servers around while hibernating, and forget to install a new scheduler on the machine now assigned to that role. TACC is 'down for maintenance'. |
|
989)
Message boards :
Questions and problems :
I just don't understand how BOINC schedules tasks
(Message 100251)
Posted 9 Aug 2020 by Richard Haselgrove Post: Probably because the multithreaded / virtual machine rules ('preempt other tasks so that the MT task can have everything it needs') were added piecemeal after the 'run earliest deadline first if a task is in danger of missing deadline' rule. The interaction between these different rules is very subtle, and if they're not all added together as a coherent, thought-through, set, there can be conflicts - as you have found. |
|
990)
Message boards :
Questions and problems :
Errors/Timed out
(Message 100248)
Posted 8 Aug 2020 by Richard Haselgrove Post: Well, it's kind not to keep hitting anybody else's servers, too - especially when they're taking a mighty hitting from the influx of SETI exiles. I actually set mine to 0.25 (that's the six hours) plus 0.05 (about an hour). If you finish any work, you're going to hit the server anyway within the hour, to report it. So you might as well grab a top-up while you're at it - saves making a double hit. |
|
991)
Message boards :
Questions and problems :
Errors/Timed out
(Message 100246)
Posted 8 Aug 2020 by Richard Haselgrove Post: Unless that machine is away from the internet for extended periods, you really don't need a multi-day work cache. All my machines are on less than a day, and most of them are on 6 hours. |
|
992)
Message boards :
Questions and problems :
SUNET - first impressions
(Message 100240)
Posted 7 Aug 2020 by Richard Haselgrove Post: Nothing strange there - compare Western sites hosted by Amazon! Edit - I still get a 504 error when trying to reach the home page. But I did get verification codes sent to a gmail address. Maybe Google encrypted them? |
|
993)
Message boards :
BOINC Manager :
BSOD - Memory Management
(Message 100237)
Posted 7 Aug 2020 by Richard Haselgrove Post: BOINC will run (as a simplified generality) one task per 'device' - a device being a single core of a (nowadays) muilti-core CPU, a discrete GPU, or an integrated GPU within a CPU package. It doesn't matter how many projects you have available to pick from - simply having more projects won't increase the number of active tasks. |
|
994)
Message boards :
BOINC Manager :
BSOD - Memory Management
(Message 100224)
Posted 6 Aug 2020 by Richard Haselgrove Post: I don't know of any, but this post could be the start of a useful exchange of knowledge and investigation. Despite being attached via Science United, you will be running a normal, unchanged, BOINC client and BOINC Manager. You have full access to all the science project data that you had before. Could you please post here the names of all the projects that Science United has currently attached you to? You could either gather that from BOINC Manager (advanced view, projects tab), or from the Event Log immediately after startup. By comparing the 'before' list that you have already posted, with the 'after' list attached by Science United, we can ask the community to start the process of elimination. |
|
995)
Message boards :
Questions and problems :
Science United & BOINC | Questions about BOINC behavior
(Message 100219)
Posted 5 Aug 2020 by Richard Haselgrove Post: Yes, that's an odd one - I haven't seen it before, either. (But I thought we ought to sort out the account manager stuff first). It comes from https://github.com/BOINC/boinc/blob/master/client/cs_prefs.cpp#L239, so it's a global preference. Looking at general preferences on a standard BOINC project, I see: Suspend when no mouse/keyboard input in last --- minutesin the list of 'When to suspend' - that must be the one. I'd guess the intention might be to allow a laptop to sleep when not in use, to save the battery. I think we've got better ways to handle that now. My own global_prefs.xml file contains <suspend_if_no_recent_input>0</suspend_if_no_recent_input>and works properly. You should be able to correct that either via the website of one of your attached projects, or by correcting the file manually. Check global_prefs_override.xml too, but I don't see the entry there. |
|
996)
Message boards :
Questions and problems :
Science United & BOINC | Questions about BOINC behavior
(Message 100214)
Posted 4 Aug 2020 by Richard Haselgrove Post: A couple of quick points, to help you re-orient yourself to the new environment. First, you didn't use BOINC to run SETI@Home in 1999. SETI did indeed start then, but as a single, integrated, volunteer project. BOINC evolved out of the original SETI@Home, and came on the scene five or six years later. That changed things a lot. Second, there are two basic ways of running BOINC. Either you manage everything yourself, or you hand over the management tasks to a third-party utility called an 'Account Manager'. Science United is one of those account managers. Both approaches draw on the same pool of scientific projects, so there's nothing to be gained by trying to run both techniques at once. You should either leave everything to Science United (and detach from the projects you joined manually) - or manage BOINC yourself (and detach from Science United). Pick one, and leave the other alone. |
|
997)
Message boards :
Questions and problems :
SUNET - first impressions
(Message 100211)
Posted 4 Aug 2020 by Richard Haselgrove Post: Just tried again to see how they were getting on, and got a big 504 Gateway Time-outfrom their url. Anyone know what they've morphed into this time? |
|
998)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 100205)
Posted 3 Aug 2020 by Richard Haselgrove Post: It's very very strange. It worked again for me this morning on my daily driver (Windows 7 x64, Chrome), but is failing now on both Windows 7 and Windows 10 - different test machines. But then again, the whole of SETI (and BOINC, for that matter) is very strange at the moment. |
|
999)
Message boards :
Questions and problems :
Not able to join - E-mail error???
(Message 100204)
Posted 3 Aug 2020 by Richard Haselgrove Post: It's very messy - I'm not surprised the OP was confused. What he saw was  which is a pretty firm turn-off. If you go to the CPDN link from the Projects Page of the BOINC website, you end up at https://www.cpdn.org/. If you're not signed-in already, that says: Join climateprediction.netOther links on the main CPDN site, like https://www.climateprediction.net/getting-started/, similarly guide you down the 'Attach to project' route - having told you to launch the Client, it shows you a screenshot of the Manager wizard. Tch, tch - sloppy. Einstein does it better:  Even if CPDN hasn't got much work at the moment (why not?) they really ought to pay more attention to the details and their website. I'll summarise the above to their email list. |
|
1000)
Message boards :
GPUs :
second gpu not in uses
(Message 100198)
Posted 3 Aug 2020 by Richard Haselgrove Post: I'm more worried by 8/2/2020 12:02:30 PM | | [coproc] No NVIDIA library foundThis could be our old friend 'Microsoft supplies incomplete drivers with Windows 10'. Yes, do check the placement and name of your cc_config.xml file - it should be listed as an 'XML file' in File Explorer, whatever name you see. 'Text Document' would indicate a problem. But some GPU projects require CUDA drivers, not just OpenCL. If you have a problem with getting work from a project you know supports NVidia cards under Windows, you may need to download a new driver direct from nvidia.com |
|
1001)
Message boards :
Questions and problems :
Not able to join - E-mail error???
(Message 100196)
Posted 3 Aug 2020 by Richard Haselgrove Post: MESSAGE FROM THE SERVER:Surely we should have better wording than that to cover the 'Must join via website' situation? I thought I'd seen better - GDPR has been around for a couple of years now - so I'm not sure if it's CPDN or BOINC that's responsible for the choice of words. I'll check the BOINC sources this afternoon. |
|
1002)
Message boards :
The Lounge :
These are no longer with us, may they have peace.
(Message 100193)
Posted 3 Aug 2020 by Richard Haselgrove Post: John Hume: Nobel Peace Prize winner and SDLP leader in Northern Ireland. We could use his skills today. |
|
1003)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 100178)
Posted 2 Aug 2020 by Richard Haselgrove Post: However, the BETA cert is still screwed upThe Beta cert worked for me this morning... Now to see if my clients can see the scheduler. Edit - Nope, still no scheduler. And no reply to my email, either. |
|
1004)
Message boards :
Questions and problems :
[OSX 10.14.6] Advanced View->Projects & Tasks UI Color Issues
(Message 100168)
Posted 31 Jul 2020 by Richard Haselgrove Post: Which version of BOINC are you using? I think that you may find this has been fixed in the recommended v7.16.6 |
|
1005)
Message boards :
Questions and problems :
Can't login
(Message 100101)
Posted 23 Jul 2020 by Richard Haselgrove Post: Each project is separate and distinct. Any account you had at SETI is unique: it can't be used at any other project. Simply sign up to any project that interests you as a 'new user'. It does help if you use a consistent email address across all projects, but it isn't essential. Just use the one that you are most comfortable with at the moment. |
|
1006)
Message boards :
Questions and problems :
Missing start tag in cc_config.xml
(Message 100079)
Posted 22 Jul 2020 by Richard Haselgrove Post: Under Windows 7, the UTF-8 file seems to be three bytes longer than the ANSI file. The extra three bytes come at the beginning, and are EF BB BF. Now to try the same thing with Windows 10 2004... And under Windows 10 2004 (the same build 19041.388 I was using yesterday), the UTF-8 file is the same length as the ANSI file, and there are no visible differences. Adding the three extra bytes, Notepad now recognises it as 'UTF-8 with BOM', and BOINC can't find <app_config> in app_config.xml. I think that concludes the examination of Windows file types, but still doesn't explain the OP's error message for the wrong file. |
|
1007)
Message boards :
Questions and problems :
Missing start tag in cc_config.xml
(Message 100076)
Posted 22 Jul 2020 by Richard Haselgrove Post: My test was using the Notepad from Windows 7. I'll try again since I got an error and you didn't, and poke at the resulting file with a hex editor to see exactly what damage UTF-8 does. |
|
1008)
Message boards :
Questions and problems :
Missing start tag in cc_config.xml
(Message 100071)
Posted 22 Jul 2020 by Richard Haselgrove Post: Re-saving the SETI app_config.xml file in ANSI format, using Notepad, resolved the error. |
|
1009)
Message boards :
Questions and problems :
Missing start tag in cc_config.xml
(Message 100070)
Posted 22 Jul 2020 by Richard Haselgrove Post: Renamed cc_config.xml so it wouldn't get in the way. 22/07/2020 14:40:10 | | Re-reading cc_config.xml 22/07/2020 14:40:10 | | cc_config.xml not found - using defaults 22/07/2020 14:40:10 | | log flags: file_xfer, sched_ops, task 22/07/2020 14:40:10 | Einstein@Home | Found app_config.xml 22/07/2020 14:40:10 | GPUGRID | Found app_config.xml 22/07/2020 14:40:10 | SETI@home | Found app_config.xml 22/07/2020 14:40:10 | SETI@home | Missing <app_config> in app_config.xml |
|
1010)
Message boards :
Questions and problems :
Missing start tag in cc_config.xml
(Message 100069)
Posted 22 Jul 2020 by Richard Haselgrove Post: Interesting. Fortunately, I have plenty of machines with working app_config.xml files for SETI@Home, currently unused. So I saved one in UTF format, and re-read all config files. Result? 22/07/2020 14:32:52 | | Re-reading cc_config.xml 22/07/2020 14:32:52 | | Config: allow multiple clients 22/07/2020 14:32:52 | | Config: GUI RPCs allowed from: 22/07/2020 14:32:52 | Einstein@Home | Config: excluded GPU. Type: NVIDIA. App: einstein_O2MDF. Device: 1 22/07/2020 14:32:52 | | Config: use all coprocessors 22/07/2020 14:32:52 | | log flags: file_xfer, sched_ops, task, cpu_sched, sched_op_debug 22/07/2020 14:32:52 | Einstein@Home | Found app_config.xml 22/07/2020 14:32:52 | GPUGRID | Found app_config.xml 22/07/2020 14:32:52 | SETI@home | Found app_config.xml 22/07/2020 14:32:52 | SETI@home | Missing <app_config> in app_config.xmlSo, what am I doing differently? |
|
1011)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 100062)
Posted 22 Jul 2020 by Richard Haselgrove Post: They worked years with a self signed certificate.I think that was before the browsers beefed up their security warnings. |
|
1012)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 100061)
Posted 22 Jul 2020 by Richard Haselgrove Post: But he does! Sunday night is his prime |
|
1013)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 100059)
Posted 22 Jul 2020 by Richard Haselgrove Post: I'm keeping my powder dry with David, waiting for him to finish the BOINC v7.16 release process. 27 days and counting since he was recorded saying "within the next month". |
|
1014)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 100056)
Posted 22 Jul 2020 by Richard Haselgrove Post: The invalid private certificate means that it takes three clicks to get anywhere on Main. And I'm getting a 404 Not Found for the Beta front page. I just wish they'd given us some inkling of what they're trying to achieve. |
|
1015)
Message boards :
Projects :
ASTEROIDSATHOME.NET distributing trojans?
(Message 100040)
Posted 21 Jul 2020 by Richard Haselgrove Post: Important update I have received a direct reply from a (named) 'Senior Malware Removal Specialist' at Malwarebytes. He reports: I have asked the researchers to check this site for you and I have been told that this block will be lifted.His advice was to perform a manual update to your copy of Malwarebytes any time after reading this message, and it should then be possible to visit the site normally. It worked for me: I'm not having any problems accessing Asteroids@Home any longer. I've given you the exact wording of the message from Malwarebytes: my interpretation is that this was a false alarm all along and there was no actual malware to be removed. If it rears its ugly head again, please notify Malwarebytes directly and mention support ticket 3108345 for the previous report. |
|
1016)
Message boards :
Questions and problems :
resource sharing
(Message 100033)
Posted 21 Jul 2020 by Richard Haselgrove Post: Unsure how to interpret this.The length of time it takes to shift half-way to the new value. https://en.wikipedia.org/wiki/Half-life I'm unfamiliar with this flag and with the process for using it.Or you could let Ctrl+Shift+F do it for you. |
|
1017)
Message boards :
Questions and problems :
Laptop stops crunching WUs when lid closes
(Message 100024)
Posted 20 Jul 2020 by Richard Haselgrove Post: I have a Windows 10 laptop running 2004 pro, build 19041.388 - it claims to be fully up-to-date. I've just tried closing the lid, while monitoring on a remote Manager. It kept running - as it does when I close the lid to keep the light down overnight. |
|
1018)
Message boards :
Questions and problems :
resource sharing
(Message 100017)
Posted 20 Jul 2020 by Richard Haselgrove Post: REC is tracked in the client_state.xml file, i.e.locally. For example, <project>
<master_url>http://einstein.phys.uwm.edu/</master_url>
<project_name>Einstein@Home</project_name>
...
[snip ~20 lines]
...
<rec>7688.584041</rec>
<rec_time>1595244898.740450</rec_time>
<resource_share>100.000000</resource_share>
...
|
|
1019)
Message boards :
Questions and problems :
resource sharing
(Message 100015)
Posted 20 Jul 2020 by Richard Haselgrove Post: Background is at https://boinc.berkeley.edu/trac/wiki/ClientSchedOctTen#Proposal:credit-drivenscheduling. Note that we are using the second part of that proposal (estimated credit - REC and its half-life), having rejected the use of actual (granted) credit as unreliable. |
|
1020)
Message boards :
Questions and problems :
resource sharing
(Message 100011)
Posted 20 Jul 2020 by Richard Haselgrove Post: <rec_half_life_days> adjusts the rate of change of the relative scheduling priorities of different projects. Explore the results with the <priority_debug> event log flag. |
|
1021)
Message boards :
Questions and problems :
How to run multiple GPUs from BOINC
(Message 99975)
Posted 15 Jul 2020 by Richard Haselgrove Post: This is what I'm talking about... So BOINC does creat cc_config.xml in a matter of speaking. But is the new version of BOINC v7.16.7 deleting this file??? If any one out there has the cc_config file and find out it's missing then you need to report as a bug on this.That's a fair question, but in my experience, the answer is 'NO'. I have two laptops on the desk in front of me - one Windows 7, the other Windows 10. They are both running v7.16.7. They both have a cc_config.xml file, and they both have a copy of the v7.16.7 downloadable installer file. That file was first made available on 31 May 2020. The Windows 7 machine downloaded the installer at 11:24 om 01 June 2020. It's cc_config.xml file is datestamped at 16:24 on 9 March 2019. So, no delete/recreate there. The Windows 10 machine downloaded the installer at 09:49 on 12 June 2020. It's cc_config.xml file is datestamped at 10:23 on 12 June 2020. I was closely involved in the events of 30 May - 01 June that led to the creation and release of v7.16.7. The only significant change (and the reason for the release) was https://github.com/BOINC/boinc/commit/ee17a911a43d54613363a776e1c65a59e8b8fd2b. That change did not affect cc_config.xml in any way. |
|
1022)
Message boards :
Questions and problems :
How to run multiple GPUs from BOINC
(Message 99970)
Posted 15 Jul 2020 by Richard Haselgrove Post: No, it is not. It's been there from the start. I remember the cc_config.xml codes in there that took at least 3 pages of coding. And no cc_config.xml never ones showed up in "Event log" that the cc_config.xml was missing.Can you give us an indication of when that anecdote relates to? Originally, (2005 or thereabouts) BOINC had no way of auto-creating cc_config.xml Round about 2015 (ten years later - subject to checking) and version 7.4, a dialog was added allowing users to change their Event Log settings visually, rather than creating/editing the file manually. Usage of the 'Event log options' dialog re-writes the entire cc_config.xml file, and thereby creates a new default file if none existed before. This is the first and only situation where BOINC creates the file - and it happens during usage of BOINC Manager, not at installation. |
|
1023)
Message boards :
Projects :
nanoHUB_at_home has failed every single task -- work time exceeded
(Message 99956)
Posted 14 Jul 2020 by Richard Haselgrove Post: The "Output file absent" error simply means that the science application finished - either by failure, or by forced closure - before it had time to write out the scientific result. I've never seen the minuscule difference in timing between a classic hard disk and an SSD cause this error - either the scientific data is present, or it isn't. More likely, the nanohub application isn't really making any real progress. BOINC invents a 'pseudoprogress' to reassure the user that something is happening, when in reality the science has got stuck. That possibility should be checked, but (for me) not at this time of night. I'll take a look in the morning. |
|
1024)
Message boards :
Questions and problems :
How to run multiple GPUs from BOINC
(Message 99939)
Posted 14 Jul 2020 by Richard Haselgrove Post: Our knowledge of cc_config.xml comes from three places: The User Manual Detailled knowledge of the source code Experience (15 years, in my case) of running BOINC. cc_config.xml is an optional file. From the user manual, "If the file is absent, the default configuration is used." The BOINC installers for Windows and Linux do not create it. I cannot speak from experience of Mac OS X. 'Skinned' installers for other versions of BOINC (WCG, CharityEngine) may include it, but that's beyond our scope here. As Jord says, uninstalling BOINC (whether manually, or as an automated part of the upgrade procedure) does not delete any user data. cc_config.xml is user data. It is possible that a new installation of BOINC may create different file locations for programs and data than the ones you previously used. This would have been displayed during installation, but not everybody checks. Normally, the previous locations are remembered and reused. |
|
1025)
Message boards :
Projects :
News on Project Outages
(Message 99930)
Posted 13 Jul 2020 by Richard Haselgrove Post: Asteroids@home is down.They must have been rebooting it. It's just accepted five uploads which have been waiting since 9/10 July. Not validated yet, though. The website feels very sluggish, but that's as might be expected after several days of constant errors. |
|
1026)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 99923)
Posted 13 Jul 2020 by Richard Haselgrove Post: Fifth - remove from public circulation (kidnap) any individuals who might have had early access to any data which might have given early warning of what was about to happen. |
|
1027)
Message boards :
Android :
What does "waiting for download" mean?
(Message 99893)
Posted 12 Jul 2020 by Richard Haselgrove Post: The WiFi RF stages on phones are often not as well designed as other devices when it comes to very high signal strengths - one of my old phones used to drop it's entire RX capability in very high RF fields - annoying when working near airport radar systems....Not just phones. That probably explains why my DAB car radio always cuts out just as I pass the highest point on my route out to the south-east - a hill crowned by several transmitter antenna towers. |
|
1028)
Message boards :
Projects :
Folding@Home and BOINC
(Message 99880)
Posted 11 Jul 2020 by Richard Haselgrove Post: Inter-University rivalry? |
|
1029)
Message boards :
Android :
What does "waiting for download" mean?
(Message 99878)
Posted 11 Jul 2020 by Richard Haselgrove Post: Asteroids is another story, their server is sending but not receiving.... (has been like that for a week) 11/07/2020 20:01:59 | Asteroids@home | [error] Error reported by file upload server: can't open log file '../log_project1/file_upload_handler.log' (errno: 9)Absentee Administrator syndrome. Until a tecchie sorts out that log file, nothing will get through - no point in retrying. |
|
1030)
Message boards :
BOINC Manager :
Some (Potential) New/Sharpened Tab Icons [Diff Included]
(Message 99862)
Posted 11 Jul 2020 by Richard Haselgrove Post: Image display is available if you follow the instructions. 
|
|
1031)
Message boards :
News :
SUNET client released
(Message 99833)
Posted 9 Jul 2020 by Richard Haselgrove Post: Well, I always get sent six numeric digits as the verify code. If I make a mistake when entering it, the site tells me that I haven't entered six digits as required. I've tried doing it with a copy'n'paste, and I've tried doing it by keying the digits manually. All fail. |
|
1032)
Message boards :
News :
SUNET client released
(Message 99820)
Posted 9 Jul 2020 by Richard Haselgrove Post: Well, I must have sent them my gmail address at least a dozen times over the last three weeks, and they've sent me the same number of verify codes in return. Every single one of them has been 'incorrect or expired' before it even reached me. |
|
1033)
Message boards :
GPUs :
Problem after NVIDIA driver update
(Message 99817)
Posted 9 Jul 2020 by Richard Haselgrove Post: I'm wondering what format he'll use to maintain compatibility, and still reconcile https://boinc.berkeley.edu/trac/wiki/AppPlanSpec#GPUapps https://github.com/BOINC/boinc/blob/master/sched/sched_send.cpp#L1103 <min_driver_version>x</min_driver_version> if (version) {
if (version < req.min_driver_version) {
sprintf(buf,
"%s: %s",
rsc_name,
_("Upgrade to the latest driver to process tasks using your computer's GPU")Is version 44101 going to be greater or smaller against 440100?
|
|
1034)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 99813)
Posted 9 Jul 2020 by Richard Haselgrove Post: David's internet is then really good to be able to post about Rosetta from there, wherever there is.You'd expect their internet to be better than ours! It still took him 12 days to post a limited version of what I'd already posted. |
|
1035)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 99810)
Posted 9 Jul 2020 by Richard Haselgrove Post: My guess is that the devs have found E.T., and they've all hitched a ride back to her home planet to find out what it's like. |
|
1036)
Message boards :
Questions and problems :
BOINC no longer requesting work.
(Message 99800)
Posted 9 Jul 2020 by Richard Haselgrove Post: Sorry, I'll have to pass on that one - I'm still a novice when it comes to the difference between various flavours of Linux. |
|
1037)
Message boards :
GPUs :
Problem after NVIDIA driver update
(Message 99799)
Posted 9 Jul 2020 by Richard Haselgrove Post: Confirmed. I now have Thu 09 Jul 2020 14:17:47 BST | | Starting BOINC client version 7.17.0 for x86_64-pc-linux-gnu Thu 09 Jul 2020 14:17:47 BST | | CUDA: NVIDIA GPU 0: GeForce GTX 1660 SUPER (driver version 440.10, CUDA version 10.2, compute capability 7.5, 4096MB, 3974MB available, 5153 GFLOPS peak) Thu 09 Jul 2020 14:17:47 BST | | OpenCL: NVIDIA GPU 0: GeForce GTX 1660 SUPER (driver version 440.100, device version OpenCL 1.2 CUDA, 5943MB, 3974MB available, 5153 GFLOPS peak) Thu 09 Jul 2020 14:40:12 BST | Asteroids@home | Sending scheduler request: To fetch work. Thu 09 Jul 2020 14:40:12 BST | Asteroids@home | Requesting new tasks for NVIDIA GPU Thu 09 Jul 2020 14:40:12 BST | Asteroids@home | [sched_op] NVIDIA GPU work request: 881.05 seconds; 1.00 devices Thu 09 Jul 2020 14:40:13 BST | Asteroids@home | Scheduler request completed: got 0 new tasks Thu 09 Jul 2020 14:40:13 BST | Asteroids@home | [sched_op] Server version 707 Thu 09 Jul 2020 14:40:13 BST | Asteroids@home | Message from server: NVIDIA GPU: Upgrade to the latest driver to process tasks using your computer's GPUThe same truncation of the CUDA version number, and the same refusal to issue new Asteroids tasks. However, a pre-downloaded task ran just fine with the new driver. And, Thu 09 Jul 2020 14:41:16 BST | GPUGRID | [sched_op] NVIDIA GPU work request: 37886.32 seconds; 0.00 devices Thu 09 Jul 2020 14:41:17 BST | GPUGRID | Scheduler request completed: got 2 new tasksSo the 'problem' appears to originate on the project server, when - presumably - the truncated version number is compared with the minimum version number specified. But we can't see that directly, at any project. Should we be able to? Meanwhile, the project has yet more problems: Thu 09 Jul 2020 14:35:32 BST | Asteroids@home | Started upload of ps_200624_input_98144_18_2_0 Thu 09 Jul 2020 14:35:34 BST | Asteroids@home | [error] Error reported by file upload server: can't open log file '../log_project1/file_upload_handler.log' (errno: 9) Thu 09 Jul 2020 14:35:34 BST | Asteroids@home | Temporarily failed upload of ps_200624_input_98144_18_2_0: transient upload error |
|
1038)
Message boards :
GPUs :
Problem after NVIDIA driver update
(Message 99797)
Posted 9 Jul 2020 by Richard Haselgrove Post: I have a Linux install with - currently - GeForce GTX 1660 SUPER (driver version 440.82, CUDA version 10.2, compute capability 7.5). That accepted Asteroids work without any complaint about not being a current driver. I see I have driver 440.100 queued up ready for installation. At a convenient point later this afternoon, I'll update and see what happens. |
|
1039)
Message boards :
Questions and problems :
BOINC no longer requesting work.
(Message 99796)
Posted 9 Jul 2020 by Richard Haselgrove Post: Crossed in the post. The difference between the repo versions and the BOINC version is that the repo ones are designed to run as a service, and BOINC's sh script is designed to run in user mode. The permissions are different, and - more importantly - the file locations are different. Since you have a repo distribution already, I wouldn't mix'n'match - use Gianfranco's PPA for consistency. |
|
1040)
Message boards :
Questions and problems :
BOINC no longer requesting work.
(Message 99795)
Posted 9 Jul 2020 by Richard Haselgrove Post: Can't help with that particular variant of Linux, but I have Linux Mint (derivative of Ubuntu) running v7.17.0 - is that close enough? I get it from https://launchpad.net/~costamagnagianfranco/+archive/ubuntu/locutusofborg-ppa: it's designed to 'upgrade in place' without losing work. |
|
1041)
Message boards :
Questions and problems :
BOINC no longer requesting work.
(Message 99788)
Posted 9 Jul 2020 by Richard Haselgrove Post: Before going much further, can I ask you to do one more thing? Yes, there are bugs in BOINC. We do our best to resolve them, when we are presented with sufficient and relevant diagnostic data. Keith Myers and I successfully argued for two successive improvements to the scheduling logic last time round. So, please update your copy of BOINC from the rather old v7.9.3, to one of the development versions v7.16.6 or v7.17.0 - who knows - we may even have solved your problem even before you posted it ;-) And with the new version, use the 'priority_debug' event log setting to show us a little more of what's going on with the work fetch logic. As a developer yourself, you will be aware that legacy versions of software eventually drop out of long term support. |
|
1042)
Message boards :
Projects :
ASTEROIDSATHOME.NET distributing trojans?
(Message 99786)
Posted 9 Jul 2020 by Richard Haselgrove Post: I'm not trying to access downloaded data - I'm trying to access a website. And now I've followed the rather convoluted procedure for raising a support ticket with Malwarebytes. I'll let you know what they say. |
|
1043)
Message boards :
Projects :
ASTEROIDSATHOME.NET distributing trojans?
(Message 99784)
Posted 9 Jul 2020 by Richard Haselgrove Post: I've also got a machine with Malwarebytes Premium, which I use to test potentially dodgy sites. That blocks Asteroids@Home, too, but unhelpfully simply reports "a trojan", with no diagnostic information. The link on the BOINC projects page takes you to http://asteroidsathome.net/boinc/, which seems clean. but there is also a home page http://asteroidsathome.net/ That home page contains source links <a href="http://astro.troja.mff.cuni.cz/projects/asteroids3D">DAMIT database</a> <a href="http://astro.troja.mff.cuni.cz/index_en.html">Astronomical Institute</a>Could they be the source of the problem? I'm not going to visit those pages myself, but I may be able to refer them to Malwarebytes for further inspection. |
|
1044)
Message boards :
News :
Rosetta@home expands COVID-19 research
(Message 99777)
Posted 9 Jul 2020 by Richard Haselgrove Post: The talk that these slides were prepared to illustrate can be heard by following the instructions in the Coronavirus Projects thread. |
|
1045)
Message boards :
Questions and problems :
BOINC no longer requesting work.
(Message 99774)
Posted 9 Jul 2020 by Richard Haselgrove Post: I agree with BobCat: these two lines together explain the 'problem'. Wed 08 Jul 2020 06:29:11 PM CDT | | [work_fetch] target work buffer: 8640.00 + 604800.00 secAny time the first entry for 'target work buffer' is less than 'saturated', BOINC will say that no work is needed, and won't fetch any by itself. Set the first figure for 'store at least' to a large enough figure to tide you over the gaps in your internet connection: add a small 'additional' figure if you wish, to tide you over any unexpected problems. But keep the two figures together well below the minimum deadline for any of your projects. BOINC will run much more smoothly then. |
|
1046)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 99749)
Posted 8 Jul 2020 by Richard Haselgrove Post: Well, they certainly have communication difficulties. I haven't heard a peep out of them for months! |
|
1047)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99745)
Posted 8 Jul 2020 by Richard Haselgrove Post: I got it running under Windows, something over 10 years ago - so my memory is rusty. I'd been doing some complex office automation work using Access and SQL Server, so my mind was full of VBA syntax - YMMV. Under those circumstances, I used the Visual Basic component of Visual Studio 6, running under Windows XP (I think - probably). That led me to a Winsock control, which exposed the required methods - probably similar to https://docs.microsoft.com/en-us/windows/win32/winsock/windows-sockets-start-page-2. My code has elements like: Winsock_Connect [HostIP, HostPort] Winsock.state Winsock_DataArrival Winsock.GetData Winsock.SendData Winsock.CloseI haven't used for, well, 10 years, but it ran for many months collecting status logs not otherwise available. I see a line of code: nonce_hash = LCase$(MD5_string(nonce + HostPwd))which matches the documentation I directed you to. Good luck. |
|
1048)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 99741)
Posted 8 Jul 2020 by Richard Haselgrove Post: It could also be that the University went on summer vacation and they just pulled the plug on all non-essential hardware.Then they shopuld have put a message to that effect on the front page, to replace the "The SETI@home message boards will continue to operate" statement that's currently visible. (as it wouldn't be if the server had been turned off) |
|
1049)
Message boards :
Questions and problems :
BOINC no longer requesting work.
(Message 99740)
Posted 8 Jul 2020 by Richard Haselgrove Post: All those "can't resolve hostname" entries point to something being seriously mis-configured in your computer's internet connection - specifically, your (or your internet provider's) Domain Name Service resolver. Take a good, hard, look at your TCP/IP stack and sort that out, before dealing with the secondary issue. P.S. 'www.cosmologyathome.org' resolves to 194.57.221.148 without problems, from here. |
|
1050)
Message boards :
GPUs :
Client not using GPU
(Message 99734)
Posted 8 Jul 2020 by Richard Haselgrove Post: BOINC itself doesn't use your GPU. It provides the facility for science projects - if they so wish, and if their branch of science can be expressed in a form suitable for GPU processing - to use the facilities you are offering. The two projects you have attached to so far - Rosetta and World Community Grid - unfortunately don't fall into that category. It's possible that they will in the future, but programming a GPU for scientific use is slow and difficult: don't hold your breath. In the meantime, look through the list of projects which use BOINC, paying specific attention to those which have a green NVidia icon in the right-hand column. See if you can find one you could assist, while you wait for your current projects to develop their GPU programs. |
|
1051)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99723)
Posted 8 Jul 2020 by Richard Haselgrove Post: How does the RPC work, is it just HTTP requests? Making my own handler to directly talk to that should be pretty easy. Any documentation for how to access that?I gave you that link in my first answer, but here it is again: https://boinc.berkeley.edu/trac/wiki/GuiRpcProtocol |
|
1052)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 99704)
Posted 7 Jul 2020 by Richard Haselgrove Post: Well, the place of sacrifice has re-opened in the last week, and I went down and had them sacrifice a chicken on my behalf - and a couple of libations on the side. Seems not to have worked, but at least I had fun trying. |
|
1053)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99691)
Posted 7 Jul 2020 by Richard Haselgrove Post: That's up to you. either you ask boinccmd to get information, which will provide human-readable answers in a fairly sparse, predictable format - some people are able to parse that output with scripting tools. or you write your own RPC handler, and work at the XML level. boinccmd may use XML RPCs internally, but it is not designed to output the raw XML to the user. |
|
1054)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99684)
Posted 7 Jul 2020 by Richard Haselgrove Post: The boinccmd utility contains the library routines for converting between XML and human formats as required. |
|
1055)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99655)
Posted 6 Jul 2020 by Richard Haselgrove Post: If you can't find it any other way, it's certainly in the "(Unix command-line version)" on the download_all page. |
|
1056)
Message boards :
BOINC Manager :
BOINC Indicator for (Ubuntu) Linux?
(Message 99648)
Posted 6 Jul 2020 by Richard Haselgrove Post: The system tray icon for BOINC relates specifically to the Manager component, not to BOINC itself. The scientific calculations can continue whether or not the Manager is active. The Linux platform developers have decided that for their purposes, the tray icon introduces complications that make it too hard to maintain and run BOINC on their platform. See Disabled Manager system tray icon for Linux builds for further information. |
|
1057)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99628)
Posted 5 Jul 2020 by Richard Haselgrove Post: It should be in the same folder/directory as the other main BOINC components - bonc(.exe) and boincmgr(.exe) On my Linux machines, that's /usr/bin On my Windows machines, that's D:\BOINC (personal choice in the installer) - the default is C:\Program Files\BOINC I don't have a Mac to check. |
|
1058)
Message boards :
Questions and problems :
Max amount of host limit?
(Message 99616)
Posted 3 Jul 2020 by Richard Haselgrove Post: Every project that I know of allows you (and provides tools for you to use) to inspect the outcome and final status of each task completed by each computer. If the task is 'successful' (completes normally) and 'valid' (sufficiently accurate, when compared to other work), it will be awarded credit. If not, not. |
|
1059)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99589)
Posted 2 Jul 2020 by Richard Haselgrove Post: boinccmd still exists, and is documented here. The full RPC specification is documented at https://boinc.berkeley.edu/trac/wiki/GuiRpcProtocol |
|
1060)
Message boards :
Questions and problems :
why does boinc use all my bandwidth (100% upload and download) while the manager is running even if everything's suspended and there's nothing in transfers?
(Message 99579)
Posted 1 Jul 2020 by Richard Haselgrove Post: There's only one other possibility I can think of: is either of our complainants running tasks which require a Virtual Machine? If the VM is communicating directly with its project base (as I know LHC does), this would show up in a network traffic scan but not in BOINC. Could either of you supply any precise detail about the projects, tasks and applications involved, and the measuring tools used to observe the bandwidth usage? |
|
1061)
Message boards :
Questions and problems :
old hosts don't show up as active?
(Message 99561)
Posted 30 Jun 2020 by Richard Haselgrove Post: We don't know, because we're not BOINCstats. Willy runs his own, third-party, independent, site. Ask on his own message board - he does monitor that (he replied and fixed a problem which I reported to him at the beginning of this month). |
|
1062)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 99510)
Posted 27 Jun 2020 by Richard Haselgrove Post: But I'd have had to book it weeks in advance! |
|
1063)
Message boards :
Questions and problems :
why does boinc use all my bandwidth (100% upload and download) while the manager is running even if everything's suspended and there's nothing in transfers?
(Message 99508)
Posted 27 Jun 2020 by Richard Haselgrove Post: No, BOINC doesn't use public, external, bandwidth - except when you see transfers in process. How are you measuring the traffic? If you're measuring it at the fone connection, then you have some other problem - I'd check for malware. But if you're using internal monitoring on the host computer, be aware that the Client and the Manager are in constant internal communication over TCP/IP 'loopback' address 127.0.0.1, port 31416. That traffic is innocuous - it doesn't use any external (paid-for) bandwidth. |
|
1064)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 99507)
Posted 27 Jun 2020 by Richard Haselgrove Post: It's not me, it's the accommodation. I don't think you can just walk up to a barn in the corner of a field and try the door, any more. They've all been converted into bijou bunkhouses... |
|
1065)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 99504)
Posted 27 Jun 2020 by Richard Haselgrove Post: Yes, 1976. I had a holiday booked (overland to Morocco), but the tour company rang up at the last minute and bumped me back a week. Didn't feel like working the extra days, so - literally - caught the service bus at the end of the road. Took me 65 miles from Bradford, via Hawes, to Keld at the head of Wensleydale. Then I spent the spare week walking back - eating in pubs, sleeping in unlocked barns on piles of hay (winter fodder in storage). It worked then, in that summer, but I don't think I'd try it again today. |
|
1066)
Message boards :
Projects :
Coronavirus projects?
(Message 99468)
Posted 26 Jun 2020 by Richard Haselgrove Post: BOINC held one of its conference calls yesterday, primarily as an information exchange between different projects. We heard an extended description of the work that Is being done by Rosetta@Home, and how the volunteers fit into the picture. All the resources are available in a GoogleDrive folder: BOINC Projects Call 2020_06_25 The audio recording is in 'Recording_BOINC_Projects_Call_2020_06_25.mp3' The Rosetta segment starts with an introduction at 35:15 - refer to slides in 'robetta_boinc.pptx' The description of the actual work you're participating in starts at 37:40 (combatting the viral infection by preventing the viral spike protein binding to cell receptors), and a second part devoted to combatting the body's cytokine storm response (the problem which causes the severe late-stage illness in some patients) starts at 45:35. Refer to 'Baker lab Boinc call.pptx' for slides on both these sections. The Rosetta segment ends at 57:10 - well worth 22 minutes of your time. |
|
1067)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 99467)
Posted 26 Jun 2020 by Richard Haselgrove Post: But the SSP is still stuck on groundhog day. |
|
1068)
Message boards :
Questions and problems :
New tasks not loading automatically
(Message 99374)
Posted 21 Jun 2020 by Richard Haselgrove Post: 6/21/2020 8:45:18 AM | | Config: fetch minimal workwould have the effect you describe in the thread title. That could have been set either in the optional cc_config.xml file, <fetch_minimal_work>0|1</fetch_minimal_work>(from Client Configuration - Options) or on the command line when the BOINC client first starts up: --fetch_minimal_work fetch only 1 job per device |
|
1069)
Message boards :
Questions and problems :
SUNET - first impressions
(Message 99341)
Posted 17 Jun 2020 by Richard Haselgrove Post: No, I couldn't get that far. The first thing you're presented with is their own login screen (which is more like signing in to a phone/tablet app), and if you haven't got an account already, you click the 'signup' link - that takes you out to a browser to validate your email. With the results I described. I've set the test machine back to normal BOINC crunching for now, but I'll keep trying to setup an account, and if I get one, I'll try their app again. They have some advice for what to do if the email doesn't arrive: I get the emails, copy and paste the code, and all I get is this: 
|
|
1070)
Message boards :
Questions and problems :
SUNET - first impressions
(Message 99333)
Posted 17 Jun 2020 by Richard Haselgrove Post: Well David claims everything is fixed now the servers have their certs updated, so that old client shouldn't have a problem with an outdated ca-bundle.crt - but for that all my Android devices still refuse to contact any of the projects with the updated certs.The SUNET install tree contains a ca-bundle.crt dated 10 February 2020, and it contains the AddTrust External Root certificate (though I haven't been able to check the expiry date yet *). The whole bundle is dated: ## Certificate data from Mozilla as of: Fri Jan 26 21:30:21 2018 GMT * Yup - it's the 30 May 2020 expired one. And it also has v1.0.2a of the OpenSSL libraries. We have v1.0.2s. |
|
1071)
Message boards :
News :
SUNET client released
(Message 99331)
Posted 17 Jun 2020 by Richard Haselgrove Post: Discussions and observations in SUNET - first impressions, please. |
|
1072)
Message boards :
Questions and problems :
SUNET - first impressions
(Message 99330)
Posted 17 Jun 2020 by Richard Haselgrove Post: Well, I thought I'd give it a whirl. First, the installer is not digitally signed, and comes from an 'unknown publisher' - install at your own risk. I'm using a sacrificial copy of Windows on a spare, test, partition. Second, this is not an upgrade from BOINC - it's a whole new application. If you have BOINC installed already, both programs could continue to run. Unless someone can decode this error message:  I'll keep prodding, and keep you posted. Edit - not going to get very far with this. You need to fill in a (sacrifical!) email address, ask for a verification code, collect it from email, paste it in the signup form, and create the account. The timeout counter runs down from 30 seconds when you request the verification code: I got it pasted within, I think, 22 seconds at the fourth attempt, but it still rejected it as 'expired or timed out'. Verification emails come from 'noreply@boinc.io', which, with the error message above, confirm that BOINC Planet, boinc.io, and sunet.cc are all the same organisation. Edit 2: It won't go away! Every time I tell it to shut down, it immediately restarts! Look what I found, buried a long way down the installation folder (C:\SUNET\resources\app.asar.unpacked\build\boinc_core\)  Note the date and version number. Last one before I run out of edit time:  It installs this to autorun (without asking permission) |
|
1073)
Message boards :
Questions and problems :
How to detach a project ?
(Message 99329)
Posted 17 Jun 2020 by Richard Haselgrove Post: The button is greyed out if you attached to the project via an Account Manager (BAM!, Science United, or similar). Go back to the Account Manager website, and use the tools there to remove it. |
|
1074)
Message boards :
BOINC client :
Amd OCL2 and OCL1.2 drivers won't co-exist on Boinc Manager but seems to work elsewhere
(Message 99288)
Posted 14 Jun 2020 by Richard Haselgrove Post: Since FAH and Boinc are very similar, both based on Gromacs (I believe), Boinc may have a similar limitation."GROMACS is a versatile package to perform molecular dynamics, i.e. simulate the Newtonian equations of motion for systems with hundreds to millions of particles." So no, BOINC isn't based on Gromacs. One or more of the projects that use BOINC may do so, but we don't. |
|
1075)
Message boards :
Questions and problems :
Second manager window.
(Message 99277)
Posted 13 Jun 2020 by Richard Haselgrove Post: Working now. Cheers.Glad to hear it. For the record, and in case anyone else needs help from this thread in the future, 'Controlling BOINC remotely' requires a hostname/password, not user name. |
|
1076)
Message boards :
Questions and problems :
Upgrade wiped all projects
(Message 99233)
Posted 10 Jun 2020 by Richard Haselgrove Post: Tell us your operating system, and we might be able to answer. |
|
1077)
Message boards :
BOINC client :
BOINC 7.16 Change Log
(Message 99179)
Posted 8 Jun 2020 by Richard Haselgrove Post: Because the old version put them there - which was a bug. The new version has the bug removed, so it's surprised by something that it knows isn't supposed to be there. Make a trivial change to the Event Log flags (Ctrl+Shift+F) and save it. Revert it again, if you wish. That writes a whole, new, clean copy of the file - and the problem's gone. |
|
1078)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99128)
Posted 5 Jun 2020 by Richard Haselgrove Post: The single explanation I have - 2-years-old-files missed some junk that was added later...Yes, you were probably using a 6-years-old file... ;-) |
|
1079)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99127)
Posted 5 Jun 2020 by Richard Haselgrove Post: Sorry, if that one is a problem, can it be removed as a download ? Or replaced with the latest correct version ?To explain how GitHub works: Normally, 'master' has the latest and greatest version. Access to master is controlled, peer-reviewed, and automatically tested - so usually, it's good. But in this case, because of the urgency of the situation, the priority was to get out a quick'n'dirty fix. So, the best copy (currently) is actually in https://github.com/BOINC/boinc/tree/client_release/7/7.16/curl (note the date and comment) But it's slightly difficult to download a single file, so it's easier to make your own file from https://raw.githubusercontent.com/BOINC/boinc/client_release/7/7.16/curl/ca-bundle.crt |
|
1080)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99124)
Posted 5 Jun 2020 by Richard Haselgrove Post: Hm... at least Rosetta uploaded, downloaded and running again on this host.Rosetta and LHC have done server-side updates. NumberFields, I suspect, is still broken and has lost a lot of computing power because of this foul-up. I seem to be doing mostly other people's cast-offs at the moment. |
|
1081)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99122)
Posted 5 Jun 2020 by Richard Haselgrove Post: Richard, maybe it's updated already? Check the dates, and read the comments I've put on #3802 this morning. |
|
1082)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99120)
Posted 5 Jun 2020 by Richard Haselgrove Post: Well, I't try to extract new file from latest package then report.I'll save you the trouble: it'll fail. Master has not yet been updated. There are two contenders for the honour: https://github.com/BOINC/boinc/pull/3791 (good) https://github.com/BOINC/boinc/pull/3802 (bad) |
|
1083)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99116)
Posted 5 Jun 2020 by Richard Haselgrove Post: NO That's the two-year-old one that caused all the trouble in the first place. |
|
1084)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99114)
Posted 5 Jun 2020 by Richard Haselgrove Post: Or maybe properly updated ca-bundle.crt is available somewhere?The file already released in the hotfix v7.16.7 of BOINC is at https://raw.githubusercontent.com/computezrmle/boinc/patch-1/curl/ca-bundle.crt It's plaintext, so the easiest way of making it is to 'select all' on that RAW page, and paste it into an empty ca_bundle.crt (text) file. But watch out for outdated OpenSSL library files, too. Ask separately if you fall foul of those. |
|
1085)
Message boards :
Questions and problems :
Incompetent BOINC Programmers
(Message 99084)
Posted 4 Jun 2020 by Richard Haselgrove Post: Well, I took my courage in both hands, donned PPE, and installed it.Nightly tarball?No, a fully fledged system update. 2020-06-04 09:33:07 upgrade boinc-client-nvidia-cuda:amd64 7.16.6+dfsg+202005240954~ubuntu18.04.1 7.16.15+dfsg+202006021756~ubuntu18.04.1 2020-06-04 09:33:32 status installed boinc-client-nvidia-cuda:amd64 7.16.15+dfsg+202006021756~ubuntu18.04.1 -- Logs begin at Mon 2020-02-03 12:48:12 GMT, end at Thu 2020-06-04 09:54:09 BST. -- Jun 04 09:34:29 Rosa systemd[1]: Started Berkeley Open Infrastructure Network Computing Client. Jun 04 09:34:29 Rosa boinc[628]: 04-Jun-2020 09:34:29 [---] Starting BOINC client version 7.17.0 for x86_64-pc-linux-gnu Jun 04 09:34:29 Rosa boinc[628]: 04-Jun-2020 09:34:29 [---] log flags: file_xfer, sched_ops, task, cpu_sched, sched_op_debug, unparsed_xml Jun 04 09:34:29 Rosa boinc[628]: 04-Jun-2020 09:34:29 [---] Libraries: libcurl/7.58.0 OpenSSL/1.1.1 zlib/1.2.11 libidn2/2.0.4 libpsl/0.19.1 (+libidn2/2.0.4) nghttp2/1.30.0 librtmp/2.3 Jun 04 09:34:29 Rosa boinc[628]: 04-Jun-2020 09:34:29 [---] Data directory: /var/lib/boinc-client Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [---] OpenCL: NVIDIA GPU 0: GeForce GTX 1660 SUPER (driver version 440.82, device version OpenCL 1.2 CUDA, 5943MB, 5943MB available, 5153 GFLOPS peak) Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [---] OpenCL: NVIDIA GPU 1: GeForce GTX 1660 SUPER (driver version 440.82, device version OpenCL 1.2 CUDA, 5945MB, 5945MB available, 5153 GFLOPS peak) Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [SETI@home] Found app_info.xml; using anonymous platform Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [---] App version needs CUDA but GPU doesn't support it Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [---] App version needs CUDA but GPU doesn't support it Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [---] App version needs CUDA but GPU doesn't support it Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [GPUGRID] Application uses missing NVIDIA GPU Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [GPUGRID] Missing coprocessor for task 1neeA01_379_4-TONI_MDADex2sn-8-50-RND2183_0 Jun 04 09:34:31 Rosa boinc[628]: 04-Jun-2020 09:34:31 [GPUGRID] Missing coprocessor for task 1svuA02_413_4-TONI_MDADex2ss-8-50-RND6622_0So it's v7.17.0 (master), but without the pre-release tag. When I install a package called "boinc-client-nvidia-cuda", I expect it to wait for the CUDA driver to load before launching... Off to find a bug reporting site. |
|
1086)
Message boards :
Questions and problems :
Incompetent BOINC Programmers
(Message 99073)
Posted 3 Jun 2020 by Richard Haselgrove Post: I could, I suppose, but I think I'd just be getting even further out of my depth. I'll sleep on it. |
|
1087)
Message boards :
Questions and problems :
Incompetent BOINC Programmers
(Message 99071)
Posted 3 Jun 2020 by Richard Haselgrove Post: Nightly tarball?No, a fully fledged system update. 
|
|
1088)
Message boards :
Questions and problems :
Incompetent BOINC Programmers
(Message 99069)
Posted 3 Jun 2020 by Richard Haselgrove Post: The latest BOINC version, as known to BOINC at the time of writing, is v7.16.7 So why is my Linux machine being offered v7.16.15 by Gianfranco's PPA? |
|
1089)
Message boards :
Questions and problems :
Incompetent BOINC Programmers
(Message 99067)
Posted 3 Jun 2020 by Richard Haselgrove Post: OK, we should have said "it will be fixed in the next planned update". That's still the plan: https://github.com/BOINC/boinc/pull/3749 But in the meantime, the incompetent BOINC administrators got caught out by an expired security certificate, and made an unplanned update which didn't include the planned updates. So the 'next' update became the 'next but [one?] [two?] [who knows?] update'. |
|
1090)
Message boards :
Questions and problems :
boincmgr won't start with Windows 10
(Message 99045)
Posted 3 Jun 2020 by Richard Haselgrove Post: There is an option for "Run Manager at login?" in the Options menu, 'Other options...' That is preferred over the startup folder technique, these days. |
|
1091)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99017)
Posted 1 Jun 2020 by Richard Haselgrove Post: 01/06/2020 21:25:15 | | BOINC can't access Internet - check network connection or proxy configuration.That's not a debug message - we all see that from time to time. Enable DEBUG logging to see WHY it failed. |
|
1092)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99010)
Posted 1 Jun 2020 by Richard Haselgrove Post: The error log says "blah... transient http error".That is exactly when you should activate http_debug logging - that's what it's there for. It will give you details, including what sort of transient error it was. |
|
1093)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99008)
Posted 1 Jun 2020 by Richard Haselgrove Post: I noticed that credits from one of my affected projects hadn't been added to my account at BOINCstats. Willy has confirmed that the same certificate expiry had prevented him collecting the exported credit stats from the affected projects. Willy's site is working again now (https://www.boincstats.com/forum/5/12516,1), and I've written to the stats mailing list, but please check your own stats totals on your own preferred stats aggregation site, and tell the operator if you find a problem. |
|
1094)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99007)
Posted 1 Jun 2020 by Richard Haselgrove Post: Windows 8.1, BOINC 7.16.5x64. Einstein has perhaps a dozen "Ready to report" on this machine, I just tried again, same. there are others, WCG and Rosetta cannot upload.That's obviously a problem, but it's a different problem from the one which sparked this thread. My advice would be to enable the <http_debug> Event Log flag (you can do that from BOINC Manager with a mouse click these days), and retry just one report or upload to each of the affected projects. We should be able to help you diagnose the problem from the resulting mess in the log. |
|
1095)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 99004)
Posted 1 Jun 2020 by Richard Haselgrove Post: I can't "Report" Einstein and can't upload Rosetta and World Community Grid. Are these symptoms of this problem? I run Windows 8.1.What version of BOINC are you using? What OS? Einstein have their own (different) reason for not being able to communicate with very old versions. WCG aren't affected by the new problem. |
|
1096)
Message boards :
Android :
2 bugs in Android beta
(Message 98986)
Posted 1 Jun 2020 by Richard Haselgrove Post: I reported that bug on 8 December 2019 - https://github.com/BOINC/boinc/issues/3397 They were aware of it even before I posted, but there's a huge re-write going on - expect it to be a while yet. |
|
1097)
Message boards :
Questions and problems :
BOINC projects have been stuck on Uploading for over a day now
(Message 98961)
Posted 31 May 2020 by Richard Haselgrove Post: See the adjacent thread 'Peer certificate cannot be authenticated with given CA certificates'. Rosetta@home is one of the projects where all communications are currently being blocked by an expired security certificate. There are workrounds in the other thread which will enable you to communicate with Rosetta by editing or replacing a file on your computer. There's also an active discussion on the Rosetta message boards. BTW, the terminology we use here is: you are attached to one project, Rosetta@Home. The jobs you see downloading and uploading are called tasks, not projects. |
|
1098)
Message boards :
Questions and problems :
Can't add a Rosetta@home project. Is it available?
(Message 98953)
Posted 31 May 2020 by Richard Haselgrove Post: See the adjacent thread 'Peer certificate cannot be authenticated with given CA certificates'. Rosetta@home is one of the projects where all communications are currently being blocked by an expired security certificate. There are workrounds in the other thread which will enable you to communicate with Rosetta by editing or replacing a file on your computer. There's also an active discussion on the Rosetta message boards. |
|
1099)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98950)
Posted 31 May 2020 by Richard Haselgrove Post: No problems on my two Linux machines on CPDN which uses https. No reports of this problem on the CPDN boards from users though uploading zips to Oz is proving a problem for Windows users at the moment but that is a server in Tasmania that needs size 10 therapy.I've just asked CPDN users to use extra logging detail on that Oz server to decode the "transient HTTP error" - we need to be certain exactly what's going on. The new report yesterday (that you asked for a log for) is within the time bracket for the certificate expiry. |
|
1100)
Message boards :
GPUs :
GPU Computation Suspending
(Message 98939)
Posted 30 May 2020 by Richard Haselgrove Post: Because it's always the same configuration file, whether you use GUI or terminal to edit it. |
|
1101)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98927)
Posted 30 May 2020 by Richard Haselgrove Post: Einstein is OK on my Linux and Windows machines. Edit - my Android tablet @ Einstein has been OK while all this has been going on. Phew. |
|
1102)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98924)
Posted 30 May 2020 by Richard Haselgrove Post: The second Linux machine is always easier than the first - whichever one you start with. All done, all running. Now for those emails. And is it beer o'clock yet? |
|
1103)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98923)
Posted 30 May 2020 by Richard Haselgrove Post: The ca-bundle.crt file is, as the name suggests, a bundle of many different certificates. Microsoft only shows you the details for the one on the top of the pile. Some projects rely on certificates buried further down in the pile. One of those other certificates expired today, and the projects that rely on it got cut off. We know of LHC, NumberFields, and Rosetta so far - there may be others. If you are running one of those projects you're in trouble - completed work can't be uploaded, new work can't be issued. But if you're not running one of the affected projects, nothing has changed - their certificates are still in the pile, and still valid. |
|
1104)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98918)
Posted 30 May 2020 by Richard Haselgrove Post: Uh oh. This does not bode well for when I eventually get round to testing Linux (which I've never used before) to see if it really is 3 times (!) faster at processing than Windows.Not in general - just the late-lamented special sauce at SETI. |
|
1105)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98916)
Posted 30 May 2020 by Richard Haselgrove Post: I did the full update/reboot, and the files uploaded. But this time I lost both CUDA and OpenCL. Faffed around for a while trying to reinstall them - no go. Eventually, simply stopped the client and started it again - GPU was back (I think there's an issue somewhere about needing a delay for the drivers to load, before starting the service). Then had to find a full-feature Manager, because the update had sent me a Debian one with half the menus missing. I think all that's left is to re-enable my backup OpenCL project. And do the other machine... |
|
1106)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98913)
Posted 30 May 2020 by Richard Haselgrove Post: Not immediately working, here. I've tried editing the file, restarting BOINC, rebooting machine. Nada. Running update-ca-certificates by itself did nothing ('no files changed'), but with -f things did happen. No change in BOINC, though. Copied the newly-generated ca-certificates.crt, renamed the copy as ca-bundle.crt, and put it where the BOINC file was (in data directory). Nada again, though without restart so far this time. Restarted BOINC, no upload - but I did get my CUDA driver back this time (it went awol, and downloaded backup OpenCL tasks instead). So now I'm doing a full update/reboot. |
|
1107)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98908)
Posted 30 May 2020 by Richard Haselgrove Post: Linux isn't or shouldn't be affected by this issue in most cases because the client looks for system-wide shared certificates (e.g. "/etc/ssl/certs/ca-certificates.crt" on Debian based distributions).My Linux Mint (based on Ubuntu 18.04) got caught, but should be fixable the same way. I'll report back. Maybe I haven't updated recently enough. |
|
1108)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98907)
Posted 30 May 2020 by Richard Haselgrove Post: Thanks all. GitHub issue at https://github.com/BOINC/boinc/issues/3789. NumberFields admin had started checking just before I reached him, but understandably didn't find any problem. I'll start firing out some emails to key people. |
|
1109)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98902)
Posted 30 May 2020 by Richard Haselgrove Post: BINGO! I hacked that new certificate into the bundle, and did a complete work cycle at NumberFields - Upload, report, download. All successful. It's actually easy to do - the certificate is a cryptographic string, but just stored in text format. The COMODO download comes with Linux format line-endings - use Notepad++ to flip those to Windows. Then use the text editor of your choice to splice it into the bundle file. De-activate the old bundle file in your BOINC program directory, replace it with the new one, and simply retry the failed uploads - no need for a restart. I'll do the GitHub issue next - feel free to pass the news round the boards while I'm busy. |
|
1110)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98901)
Posted 30 May 2020 by Richard Haselgrove Post: Many thanks - I see it (#6 of 133). I'll do a bit of playing around, then certainly pass it on. |
|
1111)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98899)
Posted 30 May 2020 by Richard Haselgrove Post: I've looked through the Linux certificate directory. All sorts of things in there, including what looks like a duplicate of our ca-bundle.crt under another name. It didn't work either. The NFaH log I posted was from Linux, so it isn't as simple as 'Linux & Mac good, Windows bad'. I've posted some extras in your LHC thread. Since moderator Toby Broom is paying attention (I think he's pretty clued up), that might be the best place. But I'll certainly report upstream once we have a clear idea where the problem lies. |
|
1112)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98894)
Posted 30 May 2020 by Richard Haselgrove Post: I got a bit of an impromptu introduction to the subject a couple of years ago. I've checked through all those "133 certificates in ca-bundle.crt", and - although a few of them have expired - none of them died anywhere near 11 am today. I'll look at those '402 certificates in /etc/ssl/certs' on my Linux machine, and see if they help - otherwise I''ll refer this one upstairs. I've left a message for the NFaH admin (Arizona - probably not minding the shop at this time on a Saturday morning): he was "thinking of switching from http to https when this Pentathlon is over" earlier this month, so may have something to add. |
|
1113)
Message boards :
Questions and problems :
Peer certificate cannot be authenticated with given CA certificates
(Message 98887)
Posted 30 May 2020 by Richard Haselgrove Post: Add NumberFields@Home to that list. I'm getting 30/05/2020 14:12:06 | NumberFields@home | [http] [ID#3878] Info: Connected to numberfields.asu.edu (129.219.51.76) port 80 (#1791)I don't normally check their logs as closely as that - it's one of the projects which runs very smoothly - but somebody else reported a problem around 11:30 UTC today. With three projects affected, I wonder if something in BOINC's ca-bundle.crt has expired? |
|
1114)
Message boards :
Questions and problems :
Virtualbox has active connections when shutting down Windows 10
(Message 98872)
Posted 28 May 2020 by Richard Haselgrove Post: Do you know a way to tell Boinc on a remote computer to quit? I use Boinctasks on this one to control the others, but I don't think I can get remote Boincs to close completely, just pause.If BoincTasks can see the remote machine, so can BOINC Manager - same hostname and password for all variants of Controlling BOINC remotely. File menu, "Shut down connected client", certainly works when sent to a remote computer. |
|
1115)
Message boards :
Questions and problems :
Running the client on a different machine than the computation
(Message 98870)
Posted 28 May 2020 by Richard Haselgrove Post: The idea is called a 'SuperHost', and as this document says, has been proposed many times before. https://boinc.berkeley.edu/trac/wiki/SuperHost "not yet been implemented" it says, and boy, is that still true - even though that wording was in the very first version of the document, 13 years ago. Best of luck - we'd all like to see this put into effect. |
|
1116)
Message boards :
Questions and problems :
Virtualbox has active connections when shutting down Windows 10
(Message 98856)
Posted 27 May 2020 by Richard Haselgrove Post: Virtualbox wasn't designed to have BOINC control it in such a fashion. You will also find if you pause BOINC that virtualbox does not pause. Also you will find vitrualbox uses all your cores. Remember it is an O/S.Both BOINC and CERN have put in a lot of effort over 10 years to ensure that information can flow out of VB to BOINC, and control signals can flow inwards. |
|
1117)
Message boards :
Questions and problems :
Virtualbox has active connections when shutting down Windows 10
(Message 98852)
Posted 27 May 2020 by Richard Haselgrove Post: Windows won't know exactly what VirtualBox is doing, but it's the LHC virtual machine which will have the connections open to CERN. Ask about it at LHC - BOINC advice can't go that deep into an individual project. |
|
1118)
Message boards :
Android :
Android BOINC 7.4.53 too old for Einstein@Home
(Message 98832)
Posted 26 May 2020 by Richard Haselgrove Post: If I can miss it, so can anyone else :P |
|
1119)
Message boards :
Android :
Android BOINC 7.4.53 too old for Einstein@Home
(Message 98829)
Posted 26 May 2020 by Richard Haselgrove Post: For information: Oliver's project is Einstein@Home. |
|
1120)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98820)
Posted 25 May 2020 by Richard Haselgrove Post: David made a PR about 3 hours ago: Web: if define('DARK_MODE', true) in project.inc, support dark mode #3762 There's development work going on to support 'dark mode' on Mac and Android, both for the Manager and the website. David may have been testing that. If so, I missed it - it's displaying as normal here now. |
|
1121)
Message boards :
Questions and problems :
How to use 100% from CPU
(Message 98811)
Posted 24 May 2020 by Richard Haselgrove Post: %age of the full CPU, so on a quad-core you're lucky to see 25%. Unfortunately, my dual CPU workstation died, so I can't check that one. |
|
1122)
Message boards :
Questions and problems :
Current Hardware Usage Column Removed?
(Message 98805)
Posted 24 May 2020 by Richard Haselgrove Post: Well, LHC sees your GPU: https://lhcathome.cern.ch/lhcathome/show_host_detail.php?hostid=10468215 But they don't have any GPU applications: LHC applications (in fact, I don't think they've ever had any GPU apps) They do use VBox virtual machines, and you have one of those. Might that be what you're referring to? Edit (again!): LHC says you have Application: All (42) · ATLAS Simulation (5) · CMS Simulation (0) · SixTrack (31) · sixtracktest (0) · Theory Simulation (6) The ATLAS and Theory simulations both run on VMs and should show that in the status column: SixTrack is for CPU only and won't bother. I'm better at GPUs than VMs, so I'll sign off here and hope a VM specialist can pick up. |
|
1123)
Message boards :
Questions and problems :
Current Hardware Usage Column Removed?
(Message 98803)
Posted 24 May 2020 by Richard Haselgrove Post: When BOINC starts up, does the Event Log show that you have a GPU/driver combination which is currently usable? Edit - and while we're on that question, are you attached to any project which is currently making work available for the type of GPU you have? |
|
1124)
Message boards :
Questions and problems :
Current Hardware Usage Column Removed?
(Message 98800)
Posted 24 May 2020 by Richard Haselgrove Post: Neither - it shows in full on all my machines, both Windows and Linux. You may need to widen the column by clicking on the column divider in the header row, and sliding it to the right. (Be careful to click on the divider, and not the column header itself - that applies a sort, and it's damned difficult to get rid of). |
|
1125)
Message boards :
Questions and problems :
How to use 100% from CPU
(Message 98782)
Posted 24 May 2020 by Richard Haselgrove Post: Quite possibly your CPU has entered a self-protective thermal state, and throttled down to prevent damage or loss of accuracy. In short, it's too hot. Check temps, check cooling system, clean out dust bunnies. |
|
1126)
Message boards :
Questions and problems :
account, userids, projects.
(Message 98754)
Posted 23 May 2020 by Richard Haselgrove Post: When you made the new account here, you increased you total account number to 3. First, you had a SETI@Home account. Then, you made a Rosetta@Home account Finally you made a BOINC account here. You have two project accounts (SETI and Rosetta), and one BOINC account. Each can only be used on the single project or website it was designed for. |
|
1127)
Message boards :
Projects :
GPUGRID
(Message 98753)
Posted 23 May 2020 by Richard Haselgrove Post: No, it doesn't. Reattaching via BOINC Manager is different. 1) The url data box is editable - you can add the extra 's' if it makes you feel better, but if it can find any part of the project, it makes no difference. 2) All the supplied url does is invoke http://www.gpugrid.net/get_project_config.php (note no 's' in this test) which will attach you to <master_url>https://www.gpugrid.net/</master_url> (with an 's') And once you've done that, the warning messages go away. At least, they've gone away on all my 7 GPU machines. |
|
1128)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98747)
Posted 23 May 2020 by Richard Haselgrove Post: You may not have read my 'recurring nightmare' on the SETI@Home Politics board. I have a recurring nightmare - and this is purely fictional, from my brain alone - that while Boris Johnson was safely tucked up in isolation/hospital, Dominic Cummings secretly crept out at night to seed the care homes. Reason? - everybody who has looked at the much-needed future reform of the British care system has shuddered at the cost, and put the file back on the shelf for the next government to deal with. So, let coronavirus cull the current occupants: no one will dare to move into the vacant rooms: the fees will dry up: the management will go bankrupt: Dom and his friends will be able to buy them up cheap in the resulting fire sale: and make a killing (sorry!) from the increased fees they'll be able to charge for the 'enhanced biosecurity' they'll be able to boast about when we've all forgotten about this and the next age-group need to be housed.That was written four days ago, in response to the Guardian's 'Agency staff were spreading Covid-19 between care homes, PHE found in April' story. I've asked the Guardian, not to publish my nightmare, but to investigate it as a theory. |
|
1129)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98745)
Posted 23 May 2020 by Richard Haselgrove Post: I've already written to the Guardian to make exactly that point. "Was Dominic Cummings the 'Typhoid Mary' of the north-east?" |
|
1130)
Message boards :
Questions and problems :
Boinc automatically assigning CPU projects when CPU WUs are disabled! BUG!
(Message 98742)
Posted 23 May 2020 by Richard Haselgrove Post: The old advice (10 years or more ago) was something like: Look at your active projects, and find the one that sets the shortest deadlines - mine is GPUGrid, at 5 days. Divide that number by the number of active projects you have sharing the same resource - CPUs only, in those far-off days. So, if you had five projects sharing, and the shortest was 5 days, your absolute maximum cache size should be 1 day, and work down from there. The reasoning behind this advice was to prevent resource share becoming unbalanced by episodes of EDF ('earliest deadline first') - the technique by which BOINC attempts to reconcile the unreconcilable demands placed on it by users ("cache 20 days of work and return it all within 5 days, when I only turn the computer on for a couple of hours a day"). EDF is how BOINC prioritises work, and runs tasks out of sequence, to get the science done when it's wanted. But people took no notice then, and I don't expect them to take any notice now. |
|
1131)
Message boards :
Projects :
GPUGRID
(Message 98722)
Posted 21 May 2020 by Richard Haselgrove Post: Servers are back up, and new work is flowing. One or two glitches along the way - we may yet find more - but I think they're ready for you to do what they suggest: when convenient (when you have no tasks at all from the project, using 'no new tasks' if needed), just detach and re-attach. |
|
1132)
Message boards :
GPUs :
Microsoft WSL 2 will support gpu acceleration
(Message 98711)
Posted 21 May 2020 by Richard Haselgrove Post: CERN, in particular, have invested a lot of effort over more than 10 years in developing a VM image which can communicate direct with the CERN servers in one direction, and can also communicate with the BOINC client on your machine, and hence with BOINC Manager, to keep you up-to-date with what's going on. VMs don't usually communicate with software running on the host machine, so that was non-trivial. You can suggest a change of platform to CERN, but I think you might find that they need some heavyweight persuading to change from a known open-source solution to a very recent arrival on the commercial scene. Edit - much of my knowledge on this subject comes from hearing Ben Segal's talk at the 2010 BOINC Workshop in London, England and talking with him afterwards. See also BOF on VM applications |
|
1133)
Message boards :
The Lounge :
I found the oldest message on boinc
(Message 98704)
Posted 20 May 2020 by Richard Haselgrove Post: Rom Walton (what happened to him)?He moved on to a new job when the money ran out. |
|
1134)
Message boards :
Questions and problems :
XFS File system.
(Message 98679)
Posted 20 May 2020 by Richard Haselgrove Post: I agree that Could not read directory attributes: Value too large for defined data typeindicates that the volunteer operating environment has outgrown the assumptions made when compiling the CPDN applications for BOINC compatibility. That happens: BOINC was caught out when SETI task numbers and later workunit numbers outgrew the 32-bit integer storage space previously allocated. At this stage, that's a problem for CPDN to sort out. But it reminds me of an older CPDN problem. CPDN downloads a *lot* of data for each task, and to save time downloads it in .zip (compressed) format. When starting, the first thing it does is to decompress the data into a task-specific sub-directory in the project folder. As I remember it in the past, the CPDN app deleted the sub-directory on successful completion, but sometimes failed to clean it up when the app crashed - leaving a lot of unnecessary data lying around, wasting space. Looking at my new-ish Linux machines, that doesn't seem to be a problem now - I know I crashed at least one task while testing whether I had the 32-bit libs (I didn't). There no trace of that crashed task now - or maybe the 32-bit lib crash happens so early that the files never get decompressed. Whatever. Maybe checkdir: cannot create extraction directory: hadam4h_a21t_209911_4_867_012014556
File existsis something similar. The app gets far enough to decompress the input data, then crashes, and exits leaving the data behind. If it tries to re-start, the data is still there, and the directory can't be created for the reason stated. That would be another problem for CPDN to solve: BOINC looks after the slot directories (cleaning them up after use), but the project has to manage its own private project directory.
|
|
1135)
Message boards :
Questions and problems :
Unable to get tasks / projects running on WIndows 10 PC
(Message 98676)
Posted 19 May 2020 by Richard Haselgrove Post: "Windows is suspending operations" sounds like something in the Windows Power Management area - sleep through to hibernation. Look in Settings --> System --> Power & sleep (--> Additional power settings if needed). That sequence goes on down into really fine detail if you want to. |
|
1136)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 98648)
Posted 19 May 2020 by Richard Haselgrove Post: Well, NOT having an outage has disappeared from, or not appeared on, the front page, so they've changed something... |
|
1137)
Message boards :
Questions and problems :
Install error 1706
(Message 98639)
Posted 19 May 2020 by Richard Haselgrove Post: I think your problem may be switching between the WCG-branded and the plain vanilla BOINC-branded versions of BOINC (or vice versa). Either: try to stick with the same installer branding - WCG or BOINC. Or: try to uninstall the old version you're using, and install the new one 'clean'. It won't matter - either version can be used to run all of your current and future projects. If your current version has been damaged by the failed upgrade, and either won't run or won't uninstall, try re-installing the version you were running before, ideally from your previous download file (it should offer to 'repair' the current installation - allow it to do that). Then, you should be able to make progress again. |
|
1138)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98603)
Posted 18 May 2020 by Richard Haselgrove Post: I've just checked the spec pages on a trade supplier I used to use when I was working. They're offering Twisted Pair Patch Cable with RJ45 connectorswith the explanation The Cat.7 cable material for operating frequencies up to 600MHz with total braided shielding and foil single-wire shielding (S/FTP) was developed for use in data centres, server rooms, rack mounting of network and control cabinets etc. for networks up to 10GBaseT. These cables are manufactured and tested according to the ISO/IEC 11801 specification. |
|
1139)
Message boards :
Questions and problems :
I need to log in every time I start up my Mac
(Message 98578)
Posted 17 May 2020 by Richard Haselgrove Post: Which version of BOINC, and which version of the Mac operating system? There have been changes in both recently, and this may be a side-effect of one or the other. We're trying to get everything re-aligned, but we need as much detail as possible. |
|
1140)
Message boards :
Questions and problems :
Please, help - too many downloaded jobs/too many jobs in progress
(Message 98573)
Posted 17 May 2020 by Richard Haselgrove Post: Even better to store 0.1 plus 0.1, or even less. Normally I would suggest looking in the Event Log, to work out exactly what work was requested, and how much was received in response. That could be done for a sample machine, using boinccmd - though refer also to the command line help, which is slightly different: --get_message_count show largest message seqno --get_messages [ seqno ] show messages > seqnoYou can control the level and nature of the detail shown with client configuration. |
|
1141)
Message boards :
Questions and problems :
RPi 1 B+ Not Getting Work (RE: Science United Message 488)
(Message 98559)
Posted 16 May 2020 by Richard Haselgrove Post: Side question,With BAM!, you choose which projects you run. With Science United, David Anderson chooses which projects you run. Edit - forgot to add With BAM!, you can participate in the BOINC community at each project - write on the message boards, provide or ask for help, send and receive private messages, make friends with other users. Fine tune your preferences at each project (which applications to run, which devices to use, whether to participate in Beta testing). With Science United, you have no access to any of these things. |
|
1142)
Message boards :
BOINC Manager :
Manager not starting since upgrade to ubuntu 20.04
(Message 98550)
Posted 16 May 2020 by Richard Haselgrove Post: Gianfranco's PPA Manager has the same menu redactions as the other distros, which makes it less useful for non-corporate users. I'm using the basic PPA framework, which seems fine, but I've replaced the Manager component with one from the SETI GPUUG builders - Ian&Steve, IIRC. But I'm still back at Mint, based on Ubuntu 18.04. I suppose I really ought to test this upgrade lark sometime... |
|
1143)
Message boards :
BOINC Manager :
Manager not starting since upgrade to ubuntu 20.04
(Message 98545)
Posted 16 May 2020 by Richard Haselgrove Post: Which version of wxWidgets are you using with Ubuntu 20.04? The recommendation seems to be wx-common_3.0.4+dfsg-15build1_amd64.deb https://pkgs.org/download/wx-common |
|
1144)
Message boards :
Questions and problems :
possible malware flaged in wcgrid files
(Message 98539)
Posted 16 May 2020 by Richard Haselgrove Post: I also believe that AV scanners have moved on a long way beyond simple pattern-matching of bit strings in files stored on your hard disk. Two other things they do (there may be more): 1) Monitor internet traffic, especially file downloads. If they spot a problem there, the file will never reach your hard disk. No amount of folder exclusion will make any difference, but your BOINC tasks will fail. 2) Test-run executable files in a sandbox environment. Some of the things that the BOINC client does (communicates over the internet; downloads and runs executable files; doesn't have a visible user interface - that's in the separate Manager) look to AV very suspicious and very much like a virus. Things you can do: a) If AV objects to a file, submit it for further analysis. b) Whitelist BOINC and project domains as trusted download sources. c) Wait a few days before you download any new version of anything. Particularly the 'heuristic' (test-run) warnings are often triggered when the AV company has not seen this version of a program before. It goes into lockdown until further information is available. Well-known and popular applications are allowed to pass unhindered - it takes a while for BOINC and project applications to reach this status. |
|
1145)
Message boards :
Questions and problems :
BOINC always "Suspended - computer is in use", when not in use
(Message 98501)
Posted 14 May 2020 by Richard Haselgrove Post: Reported upstream as #3715 - I'll keep an eye on it, and pass back any comments, suggestions - or solutions! |
|
1146)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98500)
Posted 14 May 2020 by Richard Haselgrove Post: I read that as if your neighbor had drunk all while you were out. 🤣Well, she didn't drink it all - it still feels reassuringly heavy and gluggy. I wouldn't begrudge it if she (or her father) took a wee taste as rent - I might even get a commission discount on my next order if they like it and decide to support local enterprise similarly. |
|
1147)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98497)
Posted 14 May 2020 by Richard Haselgrove Post: And glory here too - my first casklet of local ale has arrived, and was taken care of by my neighbour while I was out for a walk. A little time in the fridge, I think, and then I'll raise a housewarming toast to Jord's new gaff. ('best before 02 July", it says on the box. Fat chance of it lasting that long!) |
|
1148)
Message boards :
Questions and problems :
You have received a private message *then* No account with email address...
(Message 98485)
Posted 14 May 2020 by Richard Haselgrove Post: BOINC is designed to help scientific researchers around the world. These researchers set up 'projects' which supply work which needs to be done, and BOINC manages the flow of work between volunteers and projects. In order for you to do work for a project, somebody or something has to set up an account for you to use. Normally, you set up your own account, using your own email address and password, and that gives you access to things like private messages and friendship requests through the project website. From your description, you haven't done that. So I surmise that you joined BOINC through an alternative route called 'Science United'. Science United makes these project accounts for you, but does so anonymously. Although the account has an email address and a password, they are random and secret - nobody tells you what they are, and there is no way of finding out. If you did join BOINC through Science United, I'm afraid you'll never know what was in that private message, or be able to accept the offer of friendship. It's a strange world. |
|
1149)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98460)
Posted 12 May 2020 by Richard Haselgrove Post: Free local delivery is "within 10 mile Radius" - not far on the Australian scale, but close enough for me. Longer distances, they use a commercial carrier and charge you £7. Apparently it comes in a bib - any translators? I thought a bib was something you might need towards the end of a long session... http://www.bridgehousebrewery.co.uk/product-page/aired-ale-5l-bib OK, I've found my BiB. |
|
1150)
Message boards :
GPUs :
CUDA: NVIDIA GPU 1 (not used): Quadro 2000
(Message 98459)
Posted 12 May 2020 by Richard Haselgrove Post: Ma 12 mai 2020 19:39:04 +0300 | | CUDA: NVIDIA GPU 0: Quadro K4000 (driver version 390.13, CUDA version 9.1, compute capability 3.0, 3011MB, 2904MB available, 1245 GFLOPS peak)And it's worked. You do have GPU 0 and GPU 1 - device 0 and device 1. All you have to do is to persuade whichever NVIDIA-enabled project on your list, that has a higher priority than GPUGRID, to send you some work. |
|
1151)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98457)
Posted 12 May 2020 by Richard Haselgrove Post: Glory is scheduled to arrive here on Thursday or Friday, too. Not only did I receive a fresh consignment of café cooking today, I also discovered that the Barn is now delivering 5L mini-casks of my preferred beverage locally. The weekend, as they say, is sorted. |
|
1152)
Message boards :
Questions and problems :
Remove and Then Add
(Message 98456)
Posted 12 May 2020 by Richard Haselgrove Post: That's not the point. If you use an Account Manager or boinccmd, what you see or what you type is what you get. But if you use BOINC Manager, there's an extra step. BOINC Manager (only) goes to look up what the right answer should be. It contacts the project, with a http://boinc.bakerlab.org/rosetta/get_project_config.phpNote that's a simple (insecure) http: request, but it returned (a couple of minutes ago) <project_config> <name>Rosetta@home</name> <master_url>https://boinc.bakerlab.org/rosetta/</master_url> <web_rpc_url_base>https://boinc.bakerlab.org/rosetta/</web_rpc_url_base> ...You will be attached to the "master url", as shown in that reply - the secure one they're asking you to use from now on. |
|
1153)
Message boards :
GPUs :
CUDA: NVIDIA GPU 1 (not used): Quadro 2000
(Message 98443)
Posted 12 May 2020 by Richard Haselgrove Post: Two things. (1) before we start: Ma 12 mai 2020 14:11:43 +0300 | | Creating new client state fileYou are missing hugely important - critical - working files and directories. I'm assuming that this is because of your efforts to solve the perceived problem - but I'd urge you to verify that you can save and preserve these files while closing down and restarting BOINC. Without the filing system being stable, the rest won't work. (2) Only using one GPU, when two dissimilar GPUs are present, is a design decision which can be over-ridden by configuration. Read the description of the optional cc_config.xml file in the User Manual, paying particular attention to this line in the <options> section: <use_all_gpus>0|1</use_all_gpus>That should sort out your K4000. |
|
1154)
Message boards :
Questions and problems :
Run BOINC using Windows' Task Scheduler
(Message 98421)
Posted 10 May 2020 by Richard Haselgrove Post: I think somebody has to be logged in to gain access to the GPU driver. |
|
1155)
Message boards :
Questions and problems :
Run BOINC using Windows' Task Scheduler
(Message 98419)
Posted 10 May 2020 by Richard Haselgrove Post: You're over-thinking the problem. This is what 'Run as a service" is designed for - it's an installation option. |
|
1156)
Message boards :
Questions and problems :
Certain Projects Hijack Boinc
(Message 98415)
Posted 10 May 2020 by Richard Haselgrove Post: Understood and accepted. But we are stuck under that title until either the OP or a moderator changes it. We - collectively - need to debunk it.I would never, ever, suggest that it was malicious - the hijack title was not mine, I was trying to advise how the op could mitigate his perceived problem.As I personally dislike this behaviour I have set this value down from 10 days to 1 day.I agree - I set mine down to one day many years ago. |
|
1157)
Message boards :
Questions and problems :
Certain Projects Hijack Boinc
(Message 98413)
Posted 10 May 2020 by Richard Haselgrove Post: As I personally dislike this behaviour I have set this value down from 10 days to 1 day.I agree - I set mine down to one day many years ago. But this is a - configurable - default value in BOINC. It is wrong to give the impression that it is a malicious action by any project or projects. |
|
1158)
Message boards :
Questions and problems :
multiple cpu core issue
(Message 98400)
Posted 9 May 2020 by Richard Haselgrove Post: Which version of BOINC are you using? There was a change between v7.14 and v7.16 which may address this. |
|
1159)
Message boards :
Questions and problems :
kepp getting ths 'error'? message
(Message 98393)
Posted 8 May 2020 by Richard Haselgrove Post: If you connect directly using BOINC Manager, the correct new master url will be detected automatically, whatever you type (within reason!) in the 'attach' url box. If you connect using boinccmd, you have to be careful to supply the new url recommended by the project. If you connect via an Account Manager, you're quite possibly s*****d until the administrator of that AM wakes up and smells the coffee. |
|
1160)
Message boards :
Questions and problems :
Event Log
(Message 98370)
Posted 8 May 2020 by Richard Haselgrove Post: There was a change to the default flags for the event log in the latest incarnation of BOINC. Not sure if these flags will be enabled by default again in the next round of releases.It rather depends what you mean by 'next' release. We're working on David, trying to plant the idea of a hotfix bug release in the short term: there were signs of possible acceptance a week ago, but no material action since then. But for the record: the removal of the default flags was a mistake, not a deliberate action. |
|
1161)
Message boards :
Questions and problems :
Question: move project to another BOINC -instance-
(Message 98342)
Posted 5 May 2020 by Richard Haselgrove Post: I've not heard of anybody doing exactly that - they usually have too much of a struggle simply to set up two independent BOINC instances - but in principle it should work. I wouldn't attempt to migrate any actual 'work in progress' (running tasks) - that would involve moving a slot directory as well, and you'd probably have to renumber it. Don't even go there. So, suspend any tasks you have cached for the project you're moving. Let any running tasks complete, upload, and then report the results. Stop both BOINC instances, move the project folder, move all the named files. AND THERE'S MORE. Carefully open the file 'client_state.xml' in a text editor - in the root of the BOINC instance you are removing the project from - and find this entire section: <project>
<master_url>http://asteroidsathome.net/boinc/</master_url>
<project_name>Asteroids@home</project_name>
...
</project>
<app>
...
</app>
<file>
...
</file>
<app_version>
...
</app_version>
<workunit>
...
</workunit>
<result>
...
</result>
<project_files>
...
</project_files>Not every section need be there, but keep reading until you reach the next<project>Leave that one alone. Now, CUT the whole section out of this version, and PASTE the whole thing into the other client_state.xml file - conventionally, they go in alphabetical order of project name. It would be easier to describe if it had been written in conventional XML, with everything related to a project within the <project>...</project> tags, but it's too late for that. But I hope you can see the general 'shape' of the contiguous block you need to move. Ask again if not. |
|
1162)
Message boards :
Questions and problems :
Limit CPU Speed
(Message 98340)
Posted 5 May 2020 by Richard Haselgrove Post: Some projects need high CPU activity to feed GPUs, especially RTX GPUs.I was afraid you thought that. No. It's the other way round. Say, for example, you have a quad-core CPU plus one or more GPUs. The BOINC client (not the manager) allows over-commitment up to 4.99(999?) CPU cores - I'm not sure how many decimal places are significant. Say it's 4.99, as in your advice. That will allow all four cores of the CPU to run a full CPU app, plus the asserted 99% support required by the GPU app. If the GPU app really is that greedy (not all are). you will see the 'pure CPU' applications drop to around 80% throughput. If, on the other hand, you set a single GPU app to use 100% of a CPU, or several GPU apps to use %ages adding up to >= 100%, then BOINC would be entering forbidden territory with an over-commitment of 5 or more CPUs. At this point, BOINC will stop one of the CPU apps from running. In short: Setting GPU to need 1% to 99% --> 4 CPU apps running, plus 1 GPU app - over committed, liable to thermal throttling. Setting GPU to need 100% -->3 CPU apps running, plus 1 GPU app - fewer tasks, but cool, quiet, and efficient. It's that final tick from 99% to 100% that makes the difference. Edit - we've been using the phrase 'reserve a core' for GPU support to describe this process at SETI, for at least 8 years. |
|
1163)
Message boards :
Questions and problems :
Limit CPU Speed
(Message 98309)
Posted 3 May 2020 by Richard Haselgrove Post: Depending on what project, you may need to run your CPU this high just to feed the GPUs.How? Pray explain? |
|
1164)
Message boards :
Questions and problems :
Limit CPU Speed
(Message 98282)
Posted 2 May 2020 by Richard Haselgrove Post: No. That would be a function of the operating system, the BIOS, or a motherboard utility. |
|
1165)
Message boards :
Projects :
Incorrect CPID on one of my attached projects
(Message 98273)
Posted 2 May 2020 by Richard Haselgrove Post: I think because they're mainly a German outfit, that they think it doesn't apply to them.It was the German wing of Einstein that was most insistent on adding GDPR compliance - California was much more laid back (until the lawyers got in on the act). I don't think it's primarily a nationalistic thing. |
|
1166)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98272)
Posted 2 May 2020 by Richard Haselgrove Post: God just loves standards - obviously, since She created so many of them. |
|
1167)
Message boards :
Questions and problems :
BOINC always "Suspended - computer is in use", when not in use
(Message 98257)
Posted 1 May 2020 by Richard Haselgrove Post: BOINC's definition of 'computer in use' is that BOINC has detected input from a HID - a 'Human Interface Device'. The traditional HIDs are keyboard and mouse, and they usually don't cause problems. But we have had people report problems when they introduce a super-sensitive joystick or game controller - some of them have picked up slight traces of environmental vibration and treated it as input. So, do you have any unusual hardware like that? If so, could you try disconnecting it for a test, to see if the problem goes away? |
|
1168)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98236)
Posted 1 May 2020 by Richard Haselgrove Post: It's fascinating how the past and the present are constantly intertwining. My interest in the Luddite Trail was sparked by the murder of Jo Cox MP, who was shot by a constituent during the week before the Brexit referendum in 2016. The local Civic Society opened the Jo Cox Community Wood in her memory just at the start of the CV lockdown, and I went to pay my respects. The second part of the Luddite Trail passes right down the side of the new wood, linking the major political events, two centuries apart. The local Luddite story is also told here. |
|
1169)
Message boards :
The Lounge :
Grumbles, Glory and All Your Off Topic Discussions
(Message 98233)
Posted 1 May 2020 by Richard Haselgrove Post: The Luddites proved you can't stop technology and some people want to live free and die.Funny you should mention them. I went out for a walk yesterday, and followed the first half of the Luddite Trail - from their assembly point at 'dumb steeple' up on to Hartshead Common. It's all about 20 miles from my house. They were met by armed soldiers at Rawfolds, and two of them were shot and critically injured. They were taken to the Star Inn at Roberttown (a surprisingly long way away - maybe close to the soldiers' encampment on the common), and interrogated. A plaque on the pub wall says that a local priest acted as observer: after having nitric acid poured into his wounds, the protester beckoned the priest closer and whispered, "can you keep a secret"? The priest said yes. The protester said "so can I", and died: the soldiers never got the names of the other plotters, which was what they were after. |
|
1170)
Message boards :
Questions and problems :
Data directory
(Message 98222)
Posted 30 Apr 2020 by Richard Haselgrove Post: The default location for Windows 7 and Windows 10 is C:\ProgramData\BOINC Unfortunately, Microsoft has made C:\ProgramData a hidden folder :-( You can either search for 'File Explorer Options' in settings, and make all hidden folders visible (on View tab in File Explorer Options dialog) or paste C:\ProgramData into the address bar in File Explorer. |
|
1171)
Message boards :
Questions and problems :
remaining time to next checkpoint
(Message 98182)
Posted 29 Apr 2020 by Richard Haselgrove Post: I don't think that will work. Some project applications may checkpoint at specific progress %age points, but you still have the time estimation problem - different devices run at different speeds. Other project applications are free to checkpoint at almost any time, and the actual interval is determined by the preferred interval set in BOINC. I think your best bet is to set <checkpoint_debug> event logging, and monitor the actual behaviour of your preferred project applications on your particular system for a day or two. That should give you an idea of what to expect in the future. |
|
1172)
Message boards :
Questions and problems :
Data directory
(Message 98168)
Posted 29 Apr 2020 by Richard Haselgrove Post: Yes, you can, but it's a little involved. I'm assuming you know exactly where the data folder is on drive C: (usually C:\ProgramData\BOINC). 1) Do an orderly shutdown of BOINC, so no science apps are running. 2) Uninstall the BOINC program, using Settings --> Apps. This removes the program, but leaves your data unchanged. 3) Move the whole data folder, including all subfolders, to your new location. It doesn't have to have the same name - mine is D:\BOINCdata 4) Re-run the BOINC installer. Pause on the page where the folder locations are displayed. 5) Click the 'Advanced' button on the bottom-right corner of that page. 6) Choose the new location for the data folder (second item). If you want, you can choose a new location for the programs too. 7) Continue through the normal installation process. Next time you start BOINC, it should pick up exactly where it left off, including tasks in progress. The only thing likely to go wrong is if you don't tell BOINC exactly the right place, to match where you put the data folder. They must be the same! |
|
1173)
Message boards :
BOINC Manager :
What controls the page tab label fonts?
(Message 98164)
Posted 29 Apr 2020 by Richard Haselgrove Post: I'll have a look. But I doubt there's anything specifically setting the font property, especially if the only thing which has changed is the OS version. It's more likely that BOINC is simply using the default font, and the default setting has changed in wxWidgets or one of the libraries. Which doesn't narrow it down much. |
|
1174)
Message boards :
Questions and problems :
setup virtualbox
(Message 98162)
Posted 29 Apr 2020 by Richard Haselgrove Post: I think it's perfectly possible to install VirtualBox for the projects that issue work for VBs, and GPUs for projects that issue work for GPUs. But it isn't possible to run GPU work inside a VB, even if the host has a GPU outside the VB. Not until VirtualBox successfully implements 'PCI passthrough', and the last time I looked they had abandoned the effort. Until then, the VirtualBox container acts as an event horizon for GPUs. |
|
1175)
Message boards :
BOINC client :
BOINC will not uninstall after hard drive removed.
(Message 98151)
Posted 29 Apr 2020 by Richard Haselgrove Post: Yes, that's a known problem with the Installshield/msi installer. Did you install the programs (INSTALLDIR) to the secondary drive as well? The uninstaller should be leaving the user information (DATADIR) alone. |
|
1176)
Message boards :
GPUs :
I can't get Intel HD Graphics 4000 to work on Ubuntu 18.04
(Message 98149)
Posted 29 Apr 2020 by Richard Haselgrove Post: But if boinc sees it, you'll have to edit your config.xml file and enable all GPUsEr, no. Admittedly, that's a badly worded bit of documentation: <use_all_gpus>0|1</use_all_gpus>In fact, that rule only applies to dissimilar GPUs within each manufacter's range. The best available GPU from each manufacturer is automatically activated. |
|
1177)
Message boards :
BOINC Manager :
Uploads not being reported in the log
(Message 98146)
Posted 29 Apr 2020 by Richard Haselgrove Post: Yes, that's a known problem, affecting all platforms. The default log flag options (file_xfer, sched_ops, task) were inadvertently broken by an unrelated update. You can re-enable them yourself through the Event Log options dialog in BOINC Manager (Ctrl+Shift+F). The problem has been corrected at source, but we are awaiting a decision as to whether the faulty applications will be re-issued. |
|
1178)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 98128)
Posted 28 Apr 2020 by Richard Haselgrove Post: Whereas, the auto-message first thing this morning still said they were foregoing maintenance this week. They've forgotten to change that back again (as, I suspect, an awful lot of things have been forgotten or neglected). |
|
1179)
Message boards :
Questions and problems :
Project requested delay... after update
(Message 98112)
Posted 28 Apr 2020 by Richard Haselgrove Post: Since this seems to be becoming a thing, let's be clear about it. The 'delay' in the message is: Please don't ask this project for new work for xxx seconds. It's absolutely routine, and is sent every time you contact any project, for any reason. Each project sets their own delay value, and they vary from 7 seconds to (at least) 3636 seconds - 1 hour plus 1%. It says nothing at all about the processing of tasks on your computer. It also says nothing at all about whether the project has sent you any new tasks or not. For whether or not tasks have been received and are running, look at the tasks page in BOINC Manager. To enable normal Event Log display, open the Event Log options page (Ctrl+Shift+F), and click 'Save'. |
|
1180)
Message boards :
BOINC client :
Project requested delay of 121 seconds
(Message 98078)
Posted 26 Apr 2020 by Richard Haselgrove Post: Now all of the sudden all the projects are stating:"Project requested delay of 121 seconds"The message has always been sent by the projects, but has only been visible on your computer if you choose the <sched_op_debug> logging flag. I asked for that to be changed, because people (not least on this message board) kept getting confused because they weren't issued new work if they clicked 'update' before the delay expired - and they didn't know how long the delay was until after they'd clicked. The message should be displayed with the other scheduler contact messages with <sched_op> logging, but I made a mistake in preparing the patch and it currently appears constantly - sorry about that. I've asked for that to be corrected, but it's still in limbo - #3607. These are all part of the changes in the recent v7.16.5/6 release of BOINC, which has a much bigger bug - the normal default flog flags aren't operational. We got that fixed yesterday, but as yet there's no decision to release a hotfix. If a hotfix is approved, I'll try to get my error corrected at the same time. |
|
1181)
Message boards :
BOINC Manager :
Manager not starting since upgrade to ubuntu 20.04
(Message 98048)
Posted 24 Apr 2020 by Richard Haselgrove Post: Well done. We probably need to report this upstairs to the repo packagers. Do you know exactly which package you got this version from, and can you report any other error messages apart from the one you've told us about? |
|
1182)
Message boards :
Questions and problems :
Remove and Then Add
(Message 98047)
Posted 24 Apr 2020 by Richard Haselgrove Post: How did you add (attach to, start computing for) the Einstein project in the first place? I assume you were shown a list of projects to choose from when you first ran BOINC? On that assumption, 0) Finish any work for Einstein that may be running, and report it. The following steps are destructive. 1a) In BOINC Manager, switch to 'Advanced' view (Ctrl+Shift+A) 1b) On the 'Projects' tab, click on the Einstein project. 1c) Click on 'Remove' in the list of command buttons to the left of the pane. 2a) From the 'Tools' menu, select 'Add project...'. 2b) Select the Einstein project in the list that appears 2c) Read the Project URL in the text box below the list of projects.Confirm it is exactly the same as the url you were given in the message (it is on mine) 2d) follow the chain of 'next' buttons, filling in the boxes as a 'returning user'. And that's it. If you attached a different way, please tell us about it. |
|
1183)
Message boards :
BOINC Manager :
Manager not starting since upgrade to ubuntu 20.04
(Message 98041)
Posted 24 Apr 2020 by Richard Haselgrove Post: Some clues in https://boinc.berkeley.edu/trac/wiki/SoftwarePrereqsUnix. That looks like some part of wxWidgets 3.0 |
|
1184)
Message boards :
BOINC Manager :
Manager not starting since upgrade to ubuntu 20.04
(Message 98039)
Posted 24 Apr 2020 by Richard Haselgrove Post: no stderrdae.txt file in /var/lib/boinc-client.The repo versions treat BOINC as part of the operating system, and put BOINC's event log into the main system journal. Ugh. I'm not sure about error messages, but the equivalent of stdoutdae can be extracted and inspected - I'll look up the command I used if you like. But neither of those files will explain why the Manager isn't working. Try ldd? |
|
1185)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 98027)
Posted 23 Apr 2020 by Richard Haselgrove Post: Compatibility is always slow?But I can't see why anyone would think proprietary things are good. Compatibility is always good.Peter when used properly Cuda is more efficient than open CL. |
|
1186)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 98023)
Posted 23 Apr 2020 by Richard Haselgrove Post: Distributed Computing. What you do on your GPU. |
|
1187)
Message boards :
Questions and problems :
Problems limiting CPU usage on headless linux box
(Message 98017)
Posted 23 Apr 2020 by Richard Haselgrove Post: That's why I added boinccmd :-) But yes, if you've restarted the machine, that should have done it. You obviously have a second machine, with GUI, to post here. If they're on the same LAN, it's quick and easy to set up Controlling BOINC remotely - that would be worth doing, because then you could see the Event Log messages using the normal Manager across the LAN. That would make it much easier to see what was happening. |
|
1188)
Message boards :
Questions and problems :
Problems limiting CPU usage on headless linux box
(Message 98014)
Posted 23 Apr 2020 by Richard Haselgrove Post: Do you have GUI control (BOINC Manager) available on that machine? If not, you may need to have access to the Boinccmd tool manual. Two possible problems: You need to read in the override file you've created - options --> Read local prefs file from the Manager, --read_global_prefs_override from boinccmd You (may) have to set Activity to 'run based on preferences' - 'Activity' menu in Manager, --set_run_mode auto from boinccmd. [That may not be needed for this change, but it's worth remembering it for some of the other options] |
|
1189)
Message boards :
Questions and problems :
BoincTasks cannot connect to BOINC 7.16.5
(Message 97981)
Posted 22 Apr 2020 by Richard Haselgrove Post: Updated versions of libeay32.dll and ssleay32.dll (version 1.0.2s).Have you tested that with an actual Atom? My testbed Celeron J3455 couldn't run 64-bit v7.14.2, but is running v7.16.5 64-bit version just fine. A fix was applied, and worked in that case.What fix was applied? |
|
1190)
Message boards :
Questions and problems :
Why does BOINC take tasks in advance?
(Message 97979)
Posted 22 Apr 2020 by Richard Haselgrove Post: If you set your cache levels to the absolute minimum (0+0 days), BOINC will start requesting a new task three minutes before the previous task is estimated to finish. That's to allow time for connection, the scheduler request to complete, and any files needed for the next task to download. (BOINC still thinks at dialup speeds) |
|
1191)
Message boards :
Questions and problems :
Where can I find a version history documented for users ?
(Message 97969)
Posted 22 Apr 2020 by Richard Haselgrove Post: That page also had a link to Version History ten days ago, but rather than update the Wiki, somebody has removed the link from the web page. |
|
1192)
Message boards :
BOINC Manager :
The Boinc manager fails to connect if it was placed under the "startup applications"
(Message 97965)
Posted 22 Apr 2020 by Richard Haselgrove Post: And it looks I could not attach the image here in this message...You need to upload the image to an internet-accessible website, and reference it as [img] https://your.image.site [/img] |
|
1193)
Message boards :
GPUs :
How to limit GPU Usage?
(Message 97955)
Posted 22 Apr 2020 by Richard Haselgrove Post: Sadly, none that I know of - and nothing has been suggested by any of the GPU programmers that I've spoken to. The design of the programming environment for GPUs seems to allow any GPU to grab, and use, any GPU resources it can: there doesn't seem to be any control interface exposed to the programmer to limit that. Mind you, the programmers I was speaking with were intent on getting as much performance as possible out of the devices: they may not even have looked for a way of slowing them back down again. |
|
1194)
Message boards :
Questions and problems :
BoincTasks cannot connect to BOINC 7.16.5
(Message 97954)
Posted 22 Apr 2020 by Richard Haselgrove Post: Have you tested that with an actual Atom? My testbed Celeron J3455 couldn't run 64-bit v7.14.2, but is running v7.16.5 64-bit version just fine. A fix was applied, and worked in that case. |
|
1195)
Message boards :
Questions and problems :
Lost Username and Password after install
(Message 97931)
Posted 21 Apr 2020 by Richard Haselgrove Post: I do not believe the password is wrong. I think this is a false error message. I've seen that flashing of the screen and (repeated) partial loading of the display in BOINC Manager just recently. It was caused by setting one of the very aggressive Event Log options - the Manager simply couldn't cope with the flood of information being sent by the client. Shut everything down, and find your BOINC data directory - by default, at C:\ProgramData\BOINC. Look for a file called cc_config.xml - it's an optional file, so it may not be there. If you haven't got one, we'll have to go back and think again. If it's there, you could simply rename it so it can't be found. If BOINC then starts normally, we've found the problem. Or you could look inside it with a plain text editor. The Event Log settings are at the top. The one which killed it for me was <gui_rpc_debug>. If that line is present and set to 1, change it to 0 - worked for me. For further information, see #3592 |
|
1196)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97890)
Posted 20 Apr 2020 by Richard Haselgrove Post: Well, Charlie has said some nice things about my proposed solution - but David is more interested in adding yet more new features than fixing the ones he broke. I despair sometimes. |
|
1197)
Message boards :
Questions and problems :
Smallest CPU WUs?
(Message 97864)
Posted 19 Apr 2020 by Richard Haselgrove Post: Well, I had a bit of a poke around after posting that, and if you go to http://wuprop.boinc-af.org/results/delai.py and hover over 'afficher le menu', you can drill down and find things like runtimes for different projects running on the same hardware configuration as you have. You can look at other people's experiences without having to contribute yourself - although I think the single thread used is non-compute-intensive, so can monitor in the background without impacting on the science.People suggest looking at https://wuprop.boinc-af.org/, but I've not gone into details myself.Looking at the site you’ve linked it appears to take at least 1 thread continuously to record what the other threads are doing. A bit wasteful? |
|
1198)
Message boards :
Questions and problems :
Issue in work fetch schedule?
(Message 97844)
Posted 19 Apr 2020 by Richard Haselgrove Post: No problem. Glad to be of help. That 'no applications' really leaps out at you, doesn't it? Actually, no, it doesn't. It takes the eye of a programmer to find it amidst all the noise. Although I routinely set <sched_op_debug> on all installations, and given the choice I'd have that active by default, I would adamantly argue against defaulting <work_fetch_debug>. It's huge, it's frequent, it's complex, and it overwhelms everything else. To the average user, it's a waste of space (literally - it bloats their disk storage). It is, however, very valuable in cases like this. I guided you to it, and went to bed. I saw that 'no applications' first thing this morning, but by then you'd sorted it out by yourself. Job done. But while you're here - you said you're running BOINC v7.4.42. That's - erm - ancient. We're in the middle of trying to shake the bugs out of v7.16.5. One new thing in that version - arguably, one of the bugs - is extra massages at startup which might have flagged up this type of problem. People hate them. We still haven't worked out the answer to that one. |
|
1199)
Message boards :
Questions and problems :
Computer dead
(Message 97841)
Posted 19 Apr 2020 by Richard Haselgrove Post: Do you still have, not just the projects folder, but the whole upper level BOINC Data folder? If you still have that, with all the sub-folders, you have a chance. Set 'No new tasks' on your temporary re-installation, for all projects. Wait until all the replacement tasks finish, and allow them time to report (or you can hurry that up by manually updating a project, even if it's in the middle of a backoff). Move your whole new data folder structure somewhere safe - it may be easier to stick with that, if the subsequent steps go pear-shaped. Bring back your original data folder, but don't start BOINC just yet. You mention that you had restored you computer IDs. You also need to make one more change. In client_state.xml, in each <project> section, you will find the <hostid> - the original one. A couple of lines above that, you'll find <rpc_seqno>. Go to that project's website, look at your computer's detail page for that project, and see "Number of times client has contacted server". Set the <rpc_seqno> to one more than the website number. For example, this machine has contacted SETI 187547 times, and client_state has <rpc_seqno>187548</rpc_seqno>. Be careful when editing client state.xml - don't change anything else. It's an important and fragile file. Best of luck. |
|
1200)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 97839)
Posted 19 Apr 2020 by Richard Haselgrove Post: one of the first things that need to be done is to restrict all air travel, rich and famous or someone like me.The Sunday Times (UK newspaper) has a story today that 545 private jets have landed at UK airports since the lockdown commenced, with some hirers having lied about the purpose of their travel. The UK is one of the few countries not to have restricted entry into the country that way. "The UK operates an open borders policy", the paper says, which might come as a surprise to people outside the rich and famous. (or so the BBC image in their newspaper review shows. The paper itself is behind a paywall, so I can't confirm it.) |
|
1201)
Message boards :
Questions and problems :
Issue in work fetch schedule?
(Message 97829)
Posted 18 Apr 2020 by Richard Haselgrove Post: That hasn't really helped, has it? We'll probably have to move on to the whole <work_fetch_debug>, but that's horribly verbose. Set it, let it run once, then unset it again. Then, snip that single iteration from start to finish, and post it here. Edit - what version of BOINC are you running, anyway? we put some fairly serious work fetch fixes into v7.16.5 |
|
1202)
Message boards :
Questions and problems :
Lost Username and Password after install
(Message 97821)
Posted 18 Apr 2020 by Richard Haselgrove Post: What is trying to communicate with what? Is the the Manager - the control interface - trying to communicate with the client - the workers behind the scenes on your computer. Or is the client trying to communicate with a project over the internet? |
|
1203)
Message boards :
Questions and problems :
Issue in work fetch schedule?
(Message 97815)
Posted 18 Apr 2020 by Richard Haselgrove Post: Come on, you've been running BOINC as long as I have! Set <sched_op_debug> and let's see what's going on in a bit more detail. You can even do that from the GUI these days - Ctrl+Shift+F |
|
1204)
Message boards :
News :
New BOINC server
(Message 97812)
Posted 18 Apr 2020 by Richard Haselgrove Post: downloaded the new boinc and now the client is asking for my password? Can't connect to the host.That normally means that the client isn't running, rather than any real problem with the connection or the password. Try switching to 'Advanced View' (if you aren't there already) - Ctrl+Shift+A From the file menu, choose 'Select computer...' Leave both boxes blank, and click 'OK' If that doesn't work, come to the 'Questions and problems' section of this message board, and ask again, with a little information about your computer, operating system, and what you've upgraded from. |
|
1205)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 97806)
Posted 18 Apr 2020 by Richard Haselgrove Post: Looking at the way the virus was originally transmitted it would appear unrestricted air travel was the main cause.Some of that was pure, hedonistic, unproductive, travel. But a lot of it wasn't - especially, the first 'out of China' viral exodus. I don't see the 'elite' - business people, politicians, sports stars, movie stars - accepting a total shutdown of their air travel in that way. Our package holidays may have gone for good, but that won't stop the initial spread of the next pandemic round the world, triggering little local clusters in every corner of the globe. |
|
1206)
Message boards :
News :
New BOINC server
(Message 97804)
Posted 18 Apr 2020 by Richard Haselgrove Post: Why does the whole BOINC development wiki appear on this url? https://scienceunited.org/trac/wiki/SoftwareDevelopment |
|
1207)
Message boards :
Questions and problems :
Smallest CPU WUs?
(Message 97803)
Posted 18 Apr 2020 by Richard Haselgrove Post: People suggest looking at https://wuprop.boinc-af.org/, but I've not gone into details myself. |
|
1208)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97797)
Posted 17 Apr 2020 by Richard Haselgrove Post: And it seems to work in Linux, too: Fri 17 Apr 2020 16:57:41 BST | | cc_config.xml not found - using defaults Fri 17 Apr 2020 16:57:41 BST | | Starting BOINC client version 7.16.6 for x86_64-pc-linux-gnu Fri 17 Apr 2020 16:57:41 BST | | log flags: file_xfer, sched_ops, task ... Fri 17 Apr 2020 16:58:57 BST | | Re-reading cc_config.xml Fri 17 Apr 2020 16:58:57 BST | | Config: GUI RPC allowed from any host Fri 17 Apr 2020 16:58:57 BST | | Config: GUI RPCs allowed from: Fri 17 Apr 2020 16:58:57 BST | | Config: don't check file sizes Fri 17 Apr 2020 16:58:57 BST | | Config: don't use VirtualBox Fri 17 Apr 2020 16:58:57 BST | | Config: event log limit 3000 lines Fri 17 Apr 2020 16:58:57 BST | | log flags: file_xfer, sched_ops, task, sched_op_debug, unparsed_xmlI'll break for coffee, but it's looking good. |
|
1209)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97796)
Posted 17 Apr 2020 by Richard Haselgrove Post: Jason:Well, I've made that change, and set the three standard values to true as in post 97743 It seems to start OK with or without cc_config.xml, showing appropriate inital config report and Event Log messages; I can change settings through the GUI, and they work properly, too. I'll try Linux next: what else do I need to test before I make a PR? |
|
1210)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97793)
Posted 17 Apr 2020 by Richard Haselgrove Post: But where would the actual defaults be set? In the STRUCT definition in the header, as in my second attempt yesterday? I think that might be in the narrow-column-formatted one, but it's hard to read. I'll switch to a development machine, and try it out. |
|
1211)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97789)
Posted 17 Apr 2020 by Richard Haselgrove Post: Yes, I'm getting the copies now. Probably best not to interrupt the thought processes - I'll nip out for my allowed daily exercise (and pick up the newspaper), catch up when I get back. |
|
1212)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97782)
Posted 17 Apr 2020 by Richard Haselgrove Post: I have some (very slight) sympathy with David over this one. v7.16.5 was actually made available for download (for Windows only) towards the end of February. I noticed it, and downloaded it at the beginning of March. No announcement was made, but David later said that he was waiting for someone (himself?) to make a Mac build. I tested it, but I was checking for fixes to the problems we already knew about. My machines already have cc_config.xml files, so this problem didn't show up - even when I deliberately went looking for it (I deleted the log flag entries in the file, not the file itself). I'm afraid I don't have the patience and mindset to go back to ground zero and test new releases on clean machines, as the very bureaucratic 'box ticking' alpha test regime requires. So I missed it. David's test strategy almost relies on a continual turnover of testers - old ones retire, new one arrive with new machines. I do agree with you about missing Rom, though. |
|
1213)
Message boards :
BOINC Manager :
Waiting To Run
(Message 97779)
Posted 17 Apr 2020 by Richard Haselgrove Post: If it happens again, could you please have a look in the Event Log (Ctrl+Shift+E) and see if there's any explanation? (although there are problems with the Event Log in v7.16.5, too, just at the moment, so it's a long shot) |
|
1214)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97773)
Posted 17 Apr 2020 by Richard Haselgrove Post: Thanks for getting us back on topic - we could use some lateral thinking, or in-depth knowledge of C++, right now. Juha used to pop in at times like this, but I haven't seen him around for a while. I think the proximate cause - removing memset - is probably an unnecessary red-herring. The original compiler warning - "clearing an object of non-trivial type ‘struct PCI_INFO’; use assignment or value-initialization instead" probably didn't need such drastic action (LOG_FLAGS is a simpler structure: we are initialising, not clearing). "But the moving finger writes; and, having writ, moves on: nor all thy piety nor wit shall lure it back to cancel half a line, nor all thy tears wash out a word of it." (Omar Khayyam). So, we need a replacement that works, and that will pass the scrutiny of the gatekeeper of the code himself. Back of a postcard, anyone? |
|
1215)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97754)
Posted 16 Apr 2020 by Richard Haselgrove Post: Well, the trail is #3245, #3363, #3364, with the final one being responsible for this problem. Enter the jungle at your peril (and remember your PPE). |
|
1216)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97746)
Posted 16 Apr 2020 by Richard Haselgrove Post: Some people don't like 'memset' - eradicating that was the stated reason for the change. I tried setting x.task = true - the syntax passed, but x.task was read-only. |
|
1217)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97743)
Posted 16 Apr 2020 by Richard Haselgrove Post: But this, on the other hand, struct LOG_FLAGS {
// If you add anything, you must add it to parse() and write()
// on by default; intended for all users
//
bool file_xfer = true;
// file transfer start and finish
bool sched_ops = true;
// interactions with schedulers
bool task = true;
// task start and finish, and suspend/resumedoes at least generate16/04/2020 11:13:02 | | log flags: file_xfer, sched_ops, taskDare I suggest it? |
|
1218)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97742)
Posted 16 Apr 2020 by Richard Haselgrove Post: OK, I've been staring at this - what's wrong with it (except that it doesn't work)? The problem is to create and initialise a structure of variables (not constants), and to set the initial values of three of them. What could be simpler? 'static' I understand - it's universal, can be used anywhere in the program, and the values are the same everywhere you look at them. 'this' seems fine: we could be explicit and say 'this->task' - it compiles, but doesn't change the behaviour. So we're stuck with 'const'. Why is that there? Again, taking it out compiles, but doesn't change the behaviour. *const is badly described in cppreference, but the only other reference is to const as a cv type qualifier. |
|
1219)
Message boards :
Questions and problems :
Bug - bad at meeting deadline - leaves until the very last minute
(Message 97710)
Posted 15 Apr 2020 by Richard Haselgrove Post: I would think all of us on here monitor what it's doing. But what proportion of Boinc users are not computer-savvy at all and just install it and leave it forever to run by itself. They probably have the (in my opinion wrong) defaults of stopping work whenever the computer is in use. The user probably turns it off at night or whenever they're not using it, they probably have it go to sleep after half an hour, very unpredictable. Boinc needs to handle this. And it should default to doing way more work, asking the user on install if they want to stop the computer from sleeping, run it during use, etc.BOINC does actually track its own availability, quite closely. This is my daily driver: <time_stats>
<on_frac>0.999871</on_frac>
<connected_frac>0.999992</connected_frac>
<cpu_and_network_available_frac>0.999987</cpu_and_network_available_frac>
<active_frac>0.999994</active_frac>
<gpu_active_frac>0.999994</gpu_active_frac>
<client_start_time>1586108916.836021</client_start_time>
<total_start_time>1452717932.503635</total_start_time>
<total_duration>134204793.748473</total_duration>
<total_active_duration>134203482.360486</total_active_duration>
<total_gpu_active_duration>134203482.360486</total_gpu_active_duration>
<previous_uptime>864154.903479</previous_uptime>
<last_update>1586973069.731384</last_update>
</time_stats>total_active_duration is about four and a quarter years - sounds about right, since I last replaced it. Many of the work fetch requests are scaled according to the availability fraction. The one thing BOINC can't measure is your future holiday plans (and at the moment, I can't predict my own, either).
|
|
1220)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97705)
Posted 15 Apr 2020 by Richard Haselgrove Post: Well, I've decided it's close enough to a smoking gun (code change in November 2019) to open #3606. We'll see what happens. And I need to change my own #3402 to put the scheduler delay under sched_op control (but not sched_op_debug, as it was). |
|
1221)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97702)
Posted 15 Apr 2020 by Richard Haselgrove Post: OK, I think this might be the problem. https://github.com/BOINC/boinc/blob/master/lib/cc_config.cpp#L42 void LOG_FLAGS::init() {
static const LOG_FLAGS x;
*this = x;
// on by default (others are off by default)
//
task = true;
file_xfer = true;
sched_ops = true;
}In v7.14.2, it was https://github.com/BOINC/boinc/blob/client_release/7/7.14/lib/cc_config.cpp#L41LOG_FLAGS::LOG_FLAGS() {
init();
}
void LOG_FLAGS::init() {
memset(this, 0, sizeof(LOG_FLAGS));
// on by default (others are off by default)
//
task = true;
file_xfer = true;
sched_ops = true;
}but new lines like "*this = x;" are meaningless to me - horrible language.
|
|
1222)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97700)
Posted 15 Apr 2020 by Richard Haselgrove Post: Missing the log flags line.This is the line which should print the missing flags: https://github.com/BOINC/boinc/blob/master/client/client_state.cpp#L477 Now, where's that log_flags.show() function? |
|
1223)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97699)
Posted 15 Apr 2020 by Richard Haselgrove Post: Tried doing the same thing, on my new testbed: 15/04/2020 16:00:26 | | Processor: 4 GenuineIntel Intel(R) Celeron(R) CPU J3455 @ 1.50GHz [Family 6 Model 92 Stepping 9]- it need rebooting to apply patches, so I renamed cc_config.xml Same missing log flags line, and also missing 'Initialization completed', which should come after 15/04/2020 16:00:26 | | Setting up GUI RPC socket 15/04/2020 16:00:26 | | Checking presence of 76 project filesAfter doing benchmarks, there was only one line from the first scheduler request 15/04/2020 16:00:58 | | 4985 integer MIPS (Dhrystone) per CPU 15/04/2020 16:13:57 | NumberFields@home | Project requested delay of 61 secondsThat's 'new this version', and I had a hand in requesting it - we can check the differences. Then I opened and saved the Event Log flag chooser, without making any changes. The next scheduler request was complete. 15/04/2020 16:19:53 | | Re-reading cc_config.xml 15/04/2020 16:19:53 | | log flags: file_xfer, sched_ops, task 15/04/2020 16:30:04 | NumberFields@home | Sending scheduler request: To fetch work. 15/04/2020 16:30:04 | NumberFields@home | Reporting 1 completed tasks 15/04/2020 16:30:04 | NumberFields@home | Requesting new tasks for CPU 15/04/2020 16:30:06 | NumberFields@home | Scheduler request completed: got 1 new tasks 15/04/2020 16:30:06 | NumberFields@home | Project requested delay of 61 secondsSo, what went wrong? I don't think it the double reference to cc_config.h header - both references have been there for 8 years. More likely client: get rid of the use of memset() to initialize structs to zero. (5 months ago) client: changes to GUI RPC file fetch mechanism (8 months ago) client: add <ignore_tty> config file option (Unix) (14 months ago) All touched this file, but I think that's a decreasing order of likelihood, as well as age. I'll start looking from the top. |
|
1224)
Message boards :
Projects :
ScienceUnited Not Working On One Device
(Message 97672)
Posted 15 Apr 2020 by Richard Haselgrove Post: That's your problem. Science United isn't a project, it's an Account Manager. Science United doesn't (itself) provide any work: it attaches you to other projects with work that needs doing.master URL: https://scienceunited.org/ attached via Account Manager: noAny help getting this device running with Science United would be much appreciated; it is listed in the GUI as an attached project but does not successfully communicate and grab work. Remove both SETI (which has gone into hibernation) and Science United from your list of projects, and re-attach to Science United as an Account Manager. |
|
1225)
Message boards :
Documentation :
Installing boinc client on Amazon Linux2 (EC2 spot instance)
(Message 97638)
Posted 14 Apr 2020 by Richard Haselgrove Post: I did not imagine boinc could be ran from any cloud services; hence the question.Then you're lacking in imagination. It's even documented: https://boinc.berkeley.edu/trac/wiki/CloudServer |
|
1226)
Message boards :
Questions and problems :
Bug - bad at meeting deadline - leaves until the very last minute
(Message 97635)
Posted 14 Apr 2020 by Richard Haselgrove Post: Because the first figure I copied out of the Windows calculator was 30,000 - I've no idea how I got that. 60*24*30 hasn't got much to go wrong.[Edit - rechecked minutes/month. I thought that felt odd.]Why did it feel odd? |
|
1227)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97630)
Posted 14 Apr 2020 by Richard Haselgrove Post: Einstein has an interesting technology structure. Either the web server is at the University of Wisconsin, Milwaukee, and the database is in Hanover, Germany, or the other way round. Everything has to go one way or the other over the private line linking the two sites. Sometimes it glitches. |
|
1228)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97627)
Posted 14 Apr 2020 by Richard Haselgrove Post: Congrats Richard Way to go!Thanks all. I've celebrated with a local cafe meal, and a glass (probably more to come) of wine. In these days of lockdown, the cafe has closed for the duration. But the owner is still cooking, and delivering door-to-door in the local area. Makes a pleasant change from supermarket ready meals. |
|
1229)
Message boards :
Questions and problems :
Bug - bad at meeting deadline - leaves until the very last minute
(Message 97626)
Posted 14 Apr 2020 by Richard Haselgrove Post: I've had an LHC Theory task (1 CPU core) for about 2 weeks. Since I've set Rosetta at a much higher priority, it hardly gets runI think this might be the key to the problem. What are your actual Resource Shares for each of the two projects? Have you done the maths for, say, how many minutes per month each project would be allowed to run at that ratio? Starting point - there are 43,200 minutes in this month. Divvy them up (but remember to allow time for gaming). I suspect the answer is that BOINC has to break one of your rules. Either it breaks Resource Share, or it breaks deadline. As far as BOINC is concerned, RS is desirable, deadline is required. So it will keep to resource share as long as possible, but eventually sacrifice it. [Edit - rechecked minutes/month. I thought that felt odd.] |
|
1230)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97619)
Posted 14 Apr 2020 by Richard Haselgrove Post: Well, it loaded all my tasks at the first ask - but that's under 4,000, less than a tenth of what it was a few days ago. And I got my 200 mill - 200,000,853, at the last look. |
|
1231)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97618)
Posted 14 Apr 2020 by Richard Haselgrove Post: That was fast, i was only able to participate in the sacrifices with a pair of Irish coffees.That's what a lean, mean, database does for you - and it should be even leaner and meaner now. The big test is how long it takes to fully warm up everything and start sending out tasks again. Er, maybe not. |
|
1232)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97614)
Posted 14 Apr 2020 by Richard Haselgrove Post: I think the 'frozen SSP' started around the time of one of the botched upgrades, in the run up to Christmas. We've seen it, but not for very long. Timing depends on the person throwing the switch. Jeff (Cobb) is an early bird, up with the larks. Eric and David - aren't. |
|
1233)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97613)
Posted 14 Apr 2020 by Richard Haselgrove Post: I don't think we've tracked down the cause of the problem enough, yet, so I'm just free-associating for places to look. New install, or upgrade? If upgrade, with or without cc_config.xml? If cc_config.xml, no flags, standard flags, or custom flags? BOINC started by installer, at machine start, or by user? And so on. |
|
1234)
Message boards :
BOINC client :
BOINC 7.16.5 for Win, 7.16.6 for Mac released to the public
(Message 97611)
Posted 14 Apr 2020 by Richard Haselgrove Post: Replied by PM. |
|
1235)
Message boards :
Questions and problems :
New Boinc version - very sparse event log
(Message 97607)
Posted 14 Apr 2020 by Richard Haselgrove Post: By going to the Event Log Options menu you only made sure that cc_config.xml was written ...... and is readable by the new client? |
|
1236)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97594)
Posted 14 Apr 2020 by Richard Haselgrove Post: Well done Richard, I'm sure your pendings will push that over the line.Thanks - I've been living off pendings for the last fortnight, and I've still got plenty left - only the timing is in question. While SETI is down, please could you take a look at my recent report in https://boinc.berkeley.edu/forum_thread.php?id=13490 and point me in the right direction for Android feedback on a BOINC Beta problem ?Replied in thread. I've described my own experiences, but they go back many years - not related to this Beta. |
|
1237)
Message boards :
Android :
New Version 7.16.5
(Message 97593)
Posted 14 Apr 2020 by Richard Haselgrove Post: Unfortunately I received over 100 Einstein tasks before I managed to hit NNT.Now that it's happened, probably all you can do is wait it out. I also have an Android tablet (Samsung A) which I use for similar testing: my biggest problem is that it seems sometimes to lose internet connection, and I wake it up to see all tasks uploading, no movement - and I can't find 'retry transfers' anywhere. The system awakening seems to sort out wifi, and they upload/report/replace soon enough, but it's a bit of a waste. To your problem. Look at your machine in this view: https://einsteinathome.org/host/12824479/tasks/0/0?sort=asc&order=Sent (sorted by date, oldest first). I see 1 @ 12 Apr 2020 10:30:58 5 @ 12 Apr 2020 10:32:02 8 @ 12 Apr 2020 10:34:14 7 @ 12 Apr 2020 10:35:21 and later some @ 14 Apr 2020 13:12:53 some @ 14 Apr 2020 13:13:56 These batches are separated by a bit over a minute - which means it is your client which is making repeated requests, not Einstein making unsolicited deliveries, as some imply. I've seen (and reported) that behaviour with desktop Windows machines attached to Einstein (only): the client recognises the first arrival of new work, but doesn't adjust its assessment of the current cache level, and asks again. And again. And again... It's a very difficult bug to track down, and I seem to be the only person worried about it. When it starts, that problem seems to continue until 'something' happens to change the state. Even setting a log flag for diagnosis seems to be enough, or the project deciding I've had enough. My Android device is set on a venue which I've specified to cache 0.01 days only, and that seems to work. My desktop machines are controlled by local preferences, so there's no conflict. |
|
1238)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97583)
Posted 14 Apr 2020 by Richard Haselgrove Post: Project down for maintenance! Is it Tuesday already? And I'm getting twitchy on 199,999,798 credits - don't leave us now! |
|
1239)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97582)
Posted 14 Apr 2020 by Richard Haselgrove Post: To repeating myself: It is a new (for me) BM behaviour after upgrading from 7.14.2 to 7.16.5 !That's OK - and I did read right down to the bottom. This may be a small point, but the messages we're talking about are displayed by the Manager, but generated by the client. I assume you did update, and subsequently stop/start, both components? (obviously, you would with a machine restart) If it's any consolation, at least one of the core developers has written to say that he shares you irritation. I think some middle way will be found, but it may take a while. |
|
1240)
Message boards :
Questions and problems :
Questions about running BOINC
(Message 97580)
Posted 14 Apr 2020 by Richard Haselgrove Post: You would need to use, at least, the <app_version> template in app_config.xml - see Project-level configuration Set both the <avg_ncpus> (to schedule the right number of cores) and the --nthreads cmdline (to limit what the app grabs for itself) - using the same value for each. Even so, I'm not certain that Virtual machines (which LHC uses) follow the same rules as other MT applications. |
|
1241)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97572)
Posted 14 Apr 2020 by Richard Haselgrove Post: Thanks for coming back and trying to explain. Just a suggestion: you say you have "changed all profiles/preferences on all relevant projects". That's the first step. What you have to do next is to ensure that your own copy of BOINC contacts all of those projects, so that the new preferences are transferred from the project servers to your machine. You can use the 'Update' button on the Projects tab in BOINC Manager to do that. If you could complete all those updates, and then restart BOINC, I think you should find that all the messages and notices go away - it would be helpful if you could test that out. In the meantime, have a look at Virtualbox 6.1.4 Not Utilized by BOINC 7.14.2x64 on Windows 10 Pro. That guy (running the older v7.14.2 without the annoying notices) had precisely the opposite problem - he wanted to use a resource - his new VM - and couldn't work out why it wasn't happening. I don't think the new messages would have helped in that case - VMs are not treated in the same way as GPUs - but it's an indication that there's a balance to be struck somewhere. |
|
1242)
Message boards :
BOINC client :
BOINC 7.16.5 for Win, 7.16.6 for Mac released to the public
(Message 97548)
Posted 13 Apr 2020 by Richard Haselgrove Post: It's definitely David's sea Manager which is pre-release:And having done some more tinkering, I've now got BOINC Manager to open, and display my running service client. It says it's a pre-release copy, but it appears to work - and it's got the standard File menu and the exit options dialog straight from the BOINC code, whereas the repo Manager has had useful bits stripped out. That's enough for today.Is the manager extracted from the shell archive describing itself as pre-release or it it telling you it is connected to a pre-release client? Or perhaps even both situations are happening? The file menu of the manager will show you what the manager is. The "connected to ..." message in the bottom RH corner should tell you about the client.  That's actually correct, because it's also labelled as pre-release on https://boinc.berkeley.edu/download_all.php - he's consistent. The pre-release flag is controlled by https://github.com/BOINC/boinc/blob/master/version.h#L24 - that, plus the 7.17.0 version number, means he must have built this package from master, rather than the release branch tag. I have actually built a client from sources, but I failed with the Manager - I don't think I've got all the dependencies properly installed yet. |
|
1243)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97536)
Posted 12 Apr 2020 by Richard Haselgrove Post: I did say that in the line above the one you quoted! |
|
1244)
Message boards :
BOINC client :
BOINC 7.16.5 for Win, 7.16.6 for Mac released to the public
(Message 97530)
Posted 12 Apr 2020 by Richard Haselgrove Post: And having done some more tinkering, I've now got BOINC Manager to open, and display my running service client. It says it's a pre-release copy, but it appears to work - and it's got the standard File menu and the exit options dialog straight from the BOINC code, whereas the repo Manager has had useful bits stripped out. That's enough for today. |
|
1245)
Message boards :
BOINC client :
BOINC 7.16.5 for Win, 7.16.6 for Mac released to the public
(Message 97527)
Posted 12 Apr 2020 by Richard Haselgrove Post: How does one install the Linux (Ubuntu) version? Clicking the download link results in garble.Being very much a newbie in Linux myself, I had a bit of a play. The file downloaded is boinc_ubuntu_7.16.6_x86_64-pc-linux-gnu.sh, which seems to be a 27.6 MB shell script. Us Windows types are used to simple scripts which operate on data stored in a separate file: this one appears to have a very minimal script at the top #!/bin/sh ( read l; read l; read l; exec cat ) < "$0" | gunzip | tar xf - && /bin/sh BOINC/binstall.sh sea.tar boinc_7.16.6_x86_64-pc-linux-gnu.sh BOINC/binstall.sh exitand then contains its own data. By analogy with the NVidia driver installer, I typed sh boinc_ubuntu_7.16.6_x86_64-pc-linux-gnu.sh --infoin terminal, having first done a CD into the download directory. That failed - folder names are case-sensitive in Linux. And I don't think the --info switch is implemented. I ended up with a BOINC folder (looks complete) one level lower down, and the instruction in terminal "use /home/richard/Downloads/BOINC/run_manager to start BOINC". Which I'm not going to do, because I've already got a fully configured and operational copy of BOINC running as a service from the PPA. But there may be something useful in there. |
|
1246)
Message boards :
Questions and problems :
Ubuntu & Android BOINC Version Questions
(Message 97521)
Posted 12 Apr 2020 by Richard Haselgrove Post: Excellent. Be aware that v7.17.0 (and any other version you may be offered in updates) is a development version, and half-baked changes in future may be problematic - I hit a very nasty bug in v7.16.1 (now fixed). At the moment, the development line is very close to the newly-released v7.16.5, and all should be good - and it's got you started, which is the main thing. But it might be wise to avoid further updates unless you see a good reason for them, and try eventually to get on to an even-numbered release version, if it ever appears in your distro's repository. |
|
1247)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97506)
Posted 11 Apr 2020 by Richard Haselgrove Post: The notices themselves, I copied from the BOINC Manager display on the Notices tab. (These internally generated notices don't seem to be stored in the data directory, only the news items sent by projects) I've also linked the guilty commit on your #3590. |
|
1248)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97504)
Posted 11 Apr 2020 by Richard Haselgrove Post: Another datum: 25-Oct-2019 19:11:00 [---] Version change (7.13.0 -> 7.16.3)Prior to that change, there were no warnings about non-fetchable device/project combinations in the event log. |
|
1249)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97503)
Posted 11 Apr 2020 by Richard Haselgrove Post: It may be significant, or maybe not. I got just one notice for each device type (CPU, NVidia, Intel_GPU) in that order, although multiple projects were 'prefer not' for each device. |
|
1250)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97502)
Posted 11 Apr 2020 by Richard Haselgrove Post: For the record, and since you've escalated it as a bug: The particular machine that exhibited three notices is 05-Apr-2020 18:48:36 [---] Starting BOINC client version 7.16.3 for windows_x86_64It has 12 projects attached, though few of them are currently active. I'll check for other warnings in the startup log, to see if any others were active but failed to generate a notice. The full set is: 05-Apr-2020 18:48:38 [Albert@Home] Your settings do not allow fetching tasks for CPU. To fix this, you can change Project Preferences on the project's web site.So that's seven projects, ten warnings, but only three notices. |
|
1251)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97498)
Posted 11 Apr 2020 by Richard Haselgrove Post: Since the question has come up, I've looked at the notices on this machine (I rarely do - I look at the Event Log much more often). I find three at the top: SETI@home: Notice from BOINCSo, it's not just one project - it's every project where I've set an active "Don't fetch for ..." preference. Note that the timestamps are identical: they correspond (to within 2 seconds) with the last time I started BOINC on this computer. |
|
1252)
Message boards :
Questions and problems :
Ubuntu & Android BOINC Version Questions
(Message 97485)
Posted 11 Apr 2020 by Richard Haselgrove Post: Upon review of the available BOINC versions located on the download page at the link below:Version 7.16.5 is listed on the download_all page as a beta test for 64-bit Linux. This is the self-extracting archive version mentioned in the post that Rob links. David hasn't made an sea version for a while, so that's why it's marked as a beta. The Linux builds are distributed through repositories nowadays: your repo should get it soon, or you could attach https://launchpad.net/~costamagnagianfranco/+archive/ubuntu/costamagnagianfranco-ppa and download/install via your package manager from there - Gianfranco's PPA is automatically updated as soon as new versions are seen (and reverted if they turn out to be buggy). |
|
1253)
Message boards :
Questions and problems :
Virtualbox 6.1.4 Not Utilized by BOINC 7.14.2x64 on Windows 10 Pro
(Message 97483)
Posted 11 Apr 2020 by Richard Haselgrove Post: 4/11/2020 3:04:49 AM | LHC@home | Scheduler request completed: got 0 new tasksLooks like you may have the VM tasks switched off through project preferences at LHC. 'Sixtrack' is their normal CPU application - you would normally only see these message if you were requesting CPU/Sixtrack only. |
|
1254)
Message boards :
Projects :
Asteroids@home multiple compute errors
(Message 97480)
Posted 11 Apr 2020 by Richard Haselgrove Post: Just found a Notice from Asteroids@Home in my Manager (I have pop-ups turned off), announcing the release of a CUDA 10.2 application. News release dated 23 March 2020. The advice and specifications in the news release look sound. I'm giving it a try now. |
|
1255)
Message boards :
Questions and problems :
BOINC Manager V7.16.5 is issuing "annoying" messages which I would like to suppress
(Message 97479)
Posted 11 Apr 2020 by Richard Haselgrove Post: What did you upgrade from? I think this behaviour has existed for several versions now. You can't turn off the Event Log lines, or prevent the Notices being generated. But if you find them annoying, you can turn off the reminders: In BOINC Manager, either Simple or Advanced view, go to the 'Options' menu and choose 'other options'. In the first tab, there's a drop-down list for 'Notice reminder interval'. Set that to 'never', and it'll shut up. |
|
1256)
Message boards :
Questions and problems :
BOINC has stopped using my second GPU
(Message 97466)
Posted 10 Apr 2020 by Richard Haselgrove Post: It's possibly due to the total core count calculation. You have a N core processor (is N 4 or 8? "8 AuthenticAMD AMD Ryzen 5 1400 Quad-Core" is a bit ambiguous). Anyway, GPU tasks need a bit of CPU support, and the rule is "tasks can be run until N-and-a-bit cores are in use". 'A bit' is deliberately left vague: it can be made up of multiple fractional parts for different GPUs, but the fractions have to add up to strictly less than 1. For the new GPUGrid application, 'a bit' is unfortunately very large. I'm seeing 0.973 and 0.975 here - run two of those, and you're definitely going past one. Next question: what's happening on the CPU: Is the task that's running using a single CPU core, or is it marked 'MT' (multi-threaded) and using all N of them? If there were N single-threaded CPU tasks running, the 'and a bit' rule would allow one GPU task to run. The solution would be to pause one of the CPU tasks (run N-1), and let the GPU task take its place. We can show you how to do that. If, on the other hand, it's running MT, 'and a bit' is only going to allow one GPU to run at a time: you can't simply stop one of the threads. Well, you can tell it to run on fewer threads, but it's a bit fiddly. Again, we can show you how to do it. |
|
1257)
Message boards :
Android :
BOINC projects running under Android 10
(Message 97461)
Posted 10 Apr 2020 by Richard Haselgrove Post: If you look at the Einstein application list, you'll see that there are four different application versions available for Android on ARM - (NEON, VFP, ASIMDPIE, NEONPIE). That's not as bad as SETI@Home, where there are 12 versions, but still... It may be as simple as incrementing the version check on one of the last two, but I rather doubt it. |
|
1258)
Message boards :
Questions and problems :
BOINC has stopped using my second GPU
(Message 97453)
Posted 10 Apr 2020 by Richard Haselgrove Post: Event Log is accessed from BOINC Manager. Tools menu (in either Simple or Advanced view), bottom item. Keyboard shortcut Ctrl+Shift+E It's best to look at it immediately after (re-)starting BOINC: for situations like this, the initial lines after startup are most useful. |
|
1259)
Message boards :
Questions and problems :
BOINC client not detecting VirtualBox installation
(Message 97439)
Posted 9 Apr 2020 by Richard Haselgrove Post: When the client starts, depending on project work, the client process (BOINC) runs at least 90% busy for minutes.I've recently had a private communication suggesting that, in the author's opinion, the process which reads, parses and sorts into order the tasks in client_state.xml is coded inefficiently, and in particular takes O(n²) to perform the sort. They have coded an alternative procedure, but I am not sufficiently skilled to verify it is fit for purpose in this context. Anybody available with coding skills to committer level like to take this on? |
|
1260)
Message boards :
Questions and problems :
BOINC client not detecting VirtualBox installation
(Message 97426)
Posted 9 Apr 2020 by Richard Haselgrove Post: Check to see if a line <dont_use_vbox>1</dont_use_vbox> or <dont_use_vbox/> has been created in configuration file cc_config.xml It's a long shot, but at one time the BOINC client created that value to save repeated error messages when things went wrong. |
|
1261)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 97424)
Posted 9 Apr 2020 by Richard Haselgrove Post: Can the government charge the battery remotely, if you don't?You mean the battery can be removed?The app switches the phone on, at least the GPS and CPU.Not if I remove the battery which is what I would do if forced to install such an app. |
|
1262)
Message boards :
Questions and problems :
Detection of stalled projects?
(Message 97416)
Posted 9 Apr 2020 by Richard Haselgrove Post: I ran a Dell Precision workstation, circa 2007 - the first that could be fitted with dual Xeon quad-core CPUs. CPDN tasks took about 4 months on that, even after I'd upgraded the RAM to quad-channel. There's an awful lot of ocean and atmosphere to be modelled. |
|
1263)
Message boards :
The Lounge :
Wishing important projects would start supporting GPU crunching!
(Message 97412)
Posted 9 Apr 2020 by Richard Haselgrove Post: I'll see your COBOL and raise you LISP ;-) My dissertation project was mainly AlgolW, but I deliberately wrote a couple of subroutines in FORTRAN to prove that I knew about cross-language linkers. After that, I took a number of years away from computers, and my next machine came with an 8K BASIC interpreter mounted in a re-purposed 8-track cartridge case:   Trouble was, you couldn't read the manual when it was plugged into the computer!
|
|
1264)
Message boards :
The Lounge :
Wishing important projects would start supporting GPU crunching!
(Message 97409)
Posted 9 Apr 2020 by Richard Haselgrove Post: Snap, snap, and BCPL! |
|
1265)
Message boards :
The Lounge :
Wishing important projects would start supporting GPU crunching!
(Message 97403)
Posted 9 Apr 2020 by Richard Haselgrove Post: Have you ever tried to program a GPU to do anything - even a simple mathematical task like finding primes? |
|
1266)
Message boards :
Questions and problems :
resources allocation - cpu
(Message 97402)
Posted 9 Apr 2020 by Richard Haselgrove Post: That can be done for projects which send their executables direct to your computer for native execution. Remember to set the --nthreads command line as well. But I hesitated from offering that advice in this case because the project is LHC. They have the added abstraction layer of sending a virtual machine image, and then running their multi-threaded science application inside that. It's not as simple as you make it sound. |
|
1267)
Message boards :
The Lounge :
Don't know where it should go? Stick it here!
(Message 97386)
Posted 8 Apr 2020 by Richard Haselgrove Post: I've just been watching a new BBC documentary, and I can't make up my mind what to think about the events depicted. Life with a Russian billionaire: money and death threats iPlayer: The Countess and the Russian Billionaire There are multiple strands to consider here, and all are at least touched on in the film. The evolution of the USSR into current Russia (with the characters including Boris Yeltsin, Vladimir Putin, sundry oligarchs, and a cast of thousands). A simple human love story. The courts and legal systems across Europe. Sundry enforcers, both judicial and extra-judicial. All of those are still very much a 'work in progress' - a first draft of history. It'll take a century or more to write the whole story, and conclude who were the goodies and who were the baddies (I suspect the baddies will outnumber the goodies). But in this time of pandemic coronavirus, my mind was taken up by the lifestyles of this group of people in the good times. "We have a PA, two drivers, two housekeepers, an English nanny, and a Russian nanny as well as a French teacher for homework", counts Alexandra as she gives a tour of her home. "He'd give me his credit card and I'd go shopping, I could do what I liked," she says. "I had a private jet. I just had to pack my suitcase and go." The couple split their time between an array of properties; including a £12m family home in Battersea, a 200-acre estate in Hertfordshire, and a beach-front villa in the Caribbean, worth $40m. [not to mention the Chateau in the south of France]If they'd kept that up into the first three months of this year, they'd have made the lead in a different sort of documentary. |
|
1268)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97362)
Posted 8 Apr 2020 by Richard Haselgrove Post: When I first logged on this morning, it was still showing the latest automatic Tuesday message: "No outage this week, to let the database catch up". Now, all pages are showing a much sparser display than during a planned (or manually triggered) maintenance. I agree with Andy - the Master decided to have a short break. AKA crashed. |
|
1269)
Message boards :
Questions and problems :
All my points were deleted on one of my servers?
(Message 97343)
Posted 7 Apr 2020 by Richard Haselgrove Post: Rather than 'discounted', it could be that they were just 'transferred' to another host.Have you checked the project yet? It's not impossible that your hardware change (different motherboard, you said) has triggered a new host ID number in the project database. That is a known operation in BOINC server circles - as is the way to recover the old ID number. I know how to do that, but I won't bother to explain until we know whether it's needed or not. |
|
1270)
Message boards :
Questions and problems :
"Notices" window says I have Private Message, but there is none.
(Message 97330)
Posted 7 Apr 2020 by Richard Haselgrove Post: In general, the principle behind BOINC is: 1) Download and install the BOINC software. No account or signup needed. 2) From within BOINC, choose and attach to individual projects. The attachment process creates the project account for you. And that, more or less, is that. In recent years, people (and lawmakers) have become more concerned about rights to privacy. Some projects now prefer you to create your account via their webpage, so they can obtain and record your informed consent to their terms of use. But I think that's still the exception, rather than the rule. |
|
1271)
Message boards :
Questions and problems :
All my points were deleted on one of my servers?
(Message 97327)
Posted 7 Apr 2020 by Richard Haselgrove Post: The only authoritative credit stat is the one on the website of the project awarding the credit. Everything else is secondary and derivative. Check it out at the source first. |
|
1272)
Message boards :
Questions and problems :
"Notices" window says I have Private Message, but there is none.
(Message 97315)
Posted 6 Apr 2020 by Richard Haselgrove Post: The big problem here is that the private message will have been sent to the account which Science United created for you. The account you created for yourself is a different account, and there's no communication between them. So far as I know, there's no way of retrieving a PM sent to an SU account. It will have to remain a mystery, but it probably didn't contain anything important. |
|
1273)
Message boards :
Projects :
Coronavirus projects?
(Message 97307)
Posted 6 Apr 2020 by Richard Haselgrove Post: Hopefully GPU app comes soon after launch.I'm sure apps will be developed to meet the needs of the science. Any other consideration is irrelevant. |
|
1274)
Message boards :
Projects :
Coronavirus projects?
(Message 97300)
Posted 5 Apr 2020 by Richard Haselgrove Post: WCG also only runs CPU-only jobs.WCG runs the computing platform, but they don't provide the science. They host other people's science. If the Scripps Institute - who are the science partner for this IBM/WCG effort - code a GPU application, WCG will be ready and willing to host it. https://www.scripps.edu/news-and-events/press-room/2020/20200401-OpenPandemics-covid19.html |
|
1275)
Message boards :
Questions and problems :
Seeking Driver for GeForce GTX 1050 on Windows 7 64bit
(Message 97246)
Posted 3 Apr 2020 by Richard Haselgrove Post: I'm always nervous when I hear vague terms like 'they' have advised 'the latest' driver. Without knowing the level of technical knowledge 'they' have, and the reason they are suggesting 'the latest' driver, I usually prefer to suggest a driver by version number, and to pick one which is known to be good. Jord has linked Version: 445.75, Release Date: 2020.3.23 - this driver is known to have caused problems at SETI, and NVidia have admitted that they 'forgot' to include a previous bug fix in this release. For the moment, I'm preferring Version: 442.74, Release Date: 2020.3.19 unless the adviser specifically mentions the 445 driver range. https://www.nvidia.com/download/driverResults.aspx/158753/en-us (or your local German NVidia site) |
|
1276)
Message boards :
News :
New BOINC server
(Message 97235)
Posted 2 Apr 2020 by Richard Haselgrove Post: After a period of invisibility, everything seems to be back in place. One small problem: the directory listing for https://boinc.berkeley.edu/dl/ - the download archive for tools and earlier versions - can't be displayed (although downloads of individual files still work). It would be most helpful to be able to see what's available. |
|
1277)
Message boards :
BOINC client :
Docker Boinc Client v7.17.0 ?
(Message 97140)
Posted 27 Mar 2020 by Richard Haselgrove Post: I'm a member (volunteer, small scale) of the boinc dev team. The source code at https://github.com/BOINC/boinc-client-docker is certainly part of the collective BOINC project - different volunteers work on different part of the project, according to their interests and skill levels. You can see that my contribution was accepted by the team at places like #3183, if you want confirmation. The discussion here about current version numbers is accurate, as far as it goes. |
|
1278)
Message boards :
Server programs :
Checkpointing when using wrapper
(Message 97134)
Posted 27 Mar 2020 by Richard Haselgrove Post: There is also a description of checkpointing (which involves configuring it in the Job Description file) at https://boinc.berkeley.edu/trac/wiki/WrapperApp |
|
1279)
Message boards :
BOINC client :
Docker Boinc Client v7.17.0 ?
(Message 97133)
Posted 27 Mar 2020 by Richard Haselgrove Post: We are in an unfortunate interim phase where release version 7.16 is slowly, ever so slowly, crawling towards the official release line. The Docker team probably had their own imperative for making an interim release without the wait. |
|
1280)
Message boards :
Questions and problems :
Boinc.exe Doesn't Start Automatically 7.16.5
(Message 97132)
Posted 27 Mar 2020 by Richard Haselgrove Post: It's always worth checking. Stranger things have happened at sea. |
|
1281)
Message boards :
Questions and problems :
BOINC does not respect suspend settings in computing preferences with low priority processes
(Message 97131)
Posted 27 Mar 2020 by Richard Haselgrove Post: Regular users should not need to know about priorities at all ...And they don't. But I don't think that examples like 'using 7-zip to compress large amount of data' or 'backround rendering apps' count as regular uses - I'd call them specialist or advanced uses. And, like you, specialist or advanced users can work these things out for themselves, without adding a tl;dr text string that will have to be translated into 50-odd languages. And will confuse the regular users who, as you say, won't understand the terms referenced. |
|
1282)
Message boards :
Questions and problems :
Boinc.exe Doesn't Start Automatically 7.16.5
(Message 97120)
Posted 27 Mar 2020 by Richard Haselgrove Post: There is a checkbox in recent versions of BOINC Manager - menu Options --> other options, either Manager view - which asks Run the client?Please verify that this is checked, and test again. |
|
1283)
Message boards :
Questions and problems :
BOINC does not respect suspend settings in computing preferences with low priority processes
(Message 97117)
Posted 27 Mar 2020 by Richard Haselgrove Post: This is probably a deliberate choice. BOINC sets its own processes, like the Rosetta application, to run at low priority:it should not take any cpu time away from other processes on the computer at any other priority like 'below normal'. The oddity would seem to be that you have significant processes active on the machine, which you would want to continue running, but which are also set to low priority. Modern operating systems are used to running 100 processes or more: processes are swapped into and out of the available cores many hundreds of times a second, so that each of them can make progress. If you have significant processes which you want to proceed at 'faster than shared' speeds, but which do not run at an appropriate priority, then perhaps BOINC is not for you. But for most people, the various processes, at their various and appropriate priorities, coexist quite happily. |
|
1284)
Message boards :
Questions and problems :
BOINC 7.14.2 x64 - can't install on new host.
(Message 97096)
Posted 26 Mar 2020 by Richard Haselgrove Post: What CPU? Late generation Intel core range, low power version? We know about that one - SSL (secure sockets layer - https) crashes. Use https://boinc.berkeley.edu/dl/boinc_7.16.5_windows_x86_64.exe or https://boinc.berkeley.edu/dl/boinc_7.16.5_windows_x86_64_vbox.exe instead. |
|
1285)
Message boards :
Questions and problems :
GPU tasks reset to 0% when they restart from idle time
(Message 97085)
Posted 26 Mar 2020 by Richard Haselgrove Post: There haven't been any short tasks for years, and the recent long tasks were a mistake and issued for a licence-expired app - they all failed. There is only one - new - app at the moment. No choice. |
|
1286)
Message boards :
Questions and problems :
GPU tasks reset to 0% when they restart from idle time
(Message 97072)
Posted 26 Mar 2020 by Richard Haselgrove Post: GPUGrid tasks (specifically) do checkpoint, but not very often. The ones I've seen recently only checkpoint when they reach 10%, 20%, 30% and so on - irrespective of how long it's taken them to reach those points. And the tasks have varying run times, so it's hard to predict when the checkpoint is going to happen. Before it reaches the first 10% progress (but after the first 60 seconds), BOINC invents a 'pseudo progress' indication just to reassure you that the task is still alive. The pseudo progress is a guesstimate, and can't be relied on to predict when the real 10% point will be reached. If I were you, I'd try to let a GPU task run for as long as possible, and see how far it gets if you leave the machine switched on but idle. Then, decide if GPUGrid is right for your computer and your way of working. Their tasks tend to require a lot of computation, and are best advised for a modern, powerful, GPU. If you decide that you can't run them, you can always set 'no new tasks' locally via BOINC Manager - use 'Project commands' in simple view, or the projects tab in advanced view. I don't think Science United over-rides that, but I could be wrong :-( |
|
1287)
Message boards :
Projects :
Rainfall Rescue Project
(Message 97068)
Posted 26 Mar 2020 by Richard Haselgrove Post: It's not BOINC, but it might catch the imagination of some people here. Help needed to rescue UK's old rainfall records At a loss to know what to do with your self-isolation time? |
|
1288)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 97057)
Posted 25 Mar 2020 by Richard Haselgrove Post: Yes, they should have minted those when the put the prices up, in a single jump, from 50p per hour last year to 70p per hour. You can only buy whole hours... I don't really mind paying £1.50 for my two-hour Saturday lunch, but none of the other permutations really work. Best is 50p + 20p, but it requires careful management and regular small cash purchases. |
|
1289)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 97056)
Posted 25 Mar 2020 by Richard Haselgrove Post: Our country entering on a total lockdown quarantine and the sell of any liquors is forbidden...We feel your pain - and so, surprisingly, does our government. This just in: Off-licences added to list of 'essential' retailers 'Off-licence' in British-speak is an ordinary shop, allowed to sell alcohol "for consumption off the premises". And my local barn, previously discussed, has swiotched all its production to the bottling line. |
|
1290)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 97048)
Posted 25 Mar 2020 by Richard Haselgrove Post: Just completed my weekly 'dash and grab' food run to the CoOp. Glory - no traffic (to speak of). Empty car park. No queues to get in or out, just wait in line (I was #3, behind 'scanning and packing' and 'unpacking basket to conveyor belt'). Everything I normally buy was in stock (including milk) - just had to change to an alternate breakfast cereal. Back at car in well under 20 minutes. Grumble - no small change! Town centre store, shares car park with the rest of the town. Pay at meter, coins only, no change. Get voucher, redeem in store - and pay the rest by debit card. I'm running out of 70p pieces - had to raid the piggy bank. I'm actually quite enjoying this way of life. Back out now for the permitted 'exercise walk' to the paper shop. |
|
1291)
Message boards :
Questions and problems :
Rosetta@home computation error
(Message 97032)
Posted 24 Mar 2020 by Richard Haselgrove Post: The first line of the event log that you posted contains a task name, Rosetta has the 'task name search' feature enabled, which led me to Rosetta task 1131821072. That has the finish reason Exit status -1073741819 (0xC0000005) STATUS_ACCESS_VIOLATION- which is actually a Windows code, not Rosetta or BOINC. There is a brief report from the 'Windows Runtime Debugger', but it's much shorter than most I've seen - the fuller ones have a 'stack trace' towards the end, which is often the most useful part. Yours is missing. I think you may have to write this one off to experience. You can look around the Rosetta forums to see if anyone else has reported this kind of problem recently, You can also ask there for advice - quote that task ID number above. Finally, keep an eye on Rosetta workunit 1019438559 - a duplicate copy of your task has been sent out to another volunteer. It sometimes happens that a data error causes a task to crash, and if that happens, all the other copies fail as well. At least that would give you reassurance that it wasn't your - or your machine's - fault. |
|
1292)
Message boards :
Questions and problems :
Rosetta@home computation error
(Message 97029)
Posted 24 Mar 2020 by Richard Haselgrove Post: Usually, it's the other way round. If an application crashes, or otherwise ends abnormally, then it doesn't reach the code that writes the output file. 'Output file absent' is a sign that the app finished, but didn't finish normally - most often, a crash. |
|
1293)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 97006)
Posted 23 Mar 2020 by Richard Haselgrove Post: No, I don't have any knowledge about that, either way. But I thought it best to be prepared - and I had three suitable containers! |
|
1294)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 97004)
Posted 23 Mar 2020 by Richard Haselgrove Post: A small glory, in these hard times. The pub across the road - now closed for the duration - is selling off its surplus draught ale as a carry-out, for a pound a pint. Bring your own container. So I've got a couple of pints of gnat's piss for tonight, and a pint of the decent stuff for tomorrow's outrage party. |
|
1295)
Message boards :
Projects :
Folding@Home on BOINC
(Message 96965)
Posted 21 Mar 2020 by Richard Haselgrove Post: And with a lot more tech-savvy people working from home on (probably) modern employer-supplied kit (and when not working, bored out of their skulls), the next three months are the time to do it. Jord's post to the mailing list has got one positive endorsement, so far: Steffen Möller wrote: How about their lead programmer attending the upcoming workshop so we |
|
1296)
Message boards :
Projects :
Folding@Home on BOINC
(Message 96963)
Posted 21 Mar 2020 by Richard Haselgrove Post: You would have to overcome generations of rivalry between the Cardinals and the Bears to solve that one. See Big Game |
|
1297)
Message boards :
Projects :
Folding@Home on BOINC
(Message 96961)
Posted 21 Mar 2020 by Richard Haselgrove Post: 'zero resource share' is probably the wrong tool for this job. I have that set for Einstein GPU jobs, for when a SETI hiccup last longer than my cache. It fetches when dry (good for latency), but the last time it kicked into action I was left with unstarted work when the primary project came back to life. And with very low resource share ==> low priority, the surplus tasks hung around until within 24 hours of deadline. 13 day latency. Not good (but I think Einstein will survive). If that's the problem, there are probably other options we could pick to solve it - some of these may not have been available the last time they evaluated BOINC to see if it was fit for their purpose. Try: <fetch_minimal_work>0|1</fetch_minimal_work> <report_results_immediately>0|1</report_results_immediately>The problem is that these are set by the user, and - outside a corporate environment - they won't be. GPUGrid has the same working principle, and gets round it by setting - in the sched_reply from the server, so under their control - * Maximum two tasks per GPU - one to run, one spare to start after * Short deadlines - 5 days maximum, 50% bonus gollum points for finishing within 24 hours * Return results immediately (server version of the above) Some combination of those might be enough, or might be tweakable to be enough. If they want our user-base, they might be willing to lend a programmer or two to do the tweaking. But - big question - is their current server setup compatible with BOINC clients? Are the sched_request and sched_reply formats compatible? If not, do they want to throw away their current servers, or do they want to run two different server farms to support the two different platforms? That would be a nightmare to administer, and for researchers who want to submit work and retrieve the results. |
|
1298)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96957)
Posted 21 Mar 2020 by Richard Haselgrove Post: It strikes me that most of the measures announced so far have been designed with the needs of urban areas in mind. My village - population somewhat over 1,000, I think - has nothing like that. For 10 years, we didn't even have a shop (closed when the Post Office counter was pulled), but fortunately a small volunteer-run community shop opened last year. Selling mostly local produce, but essential basic foods as well, they will be able to stay open - and they already do simple takeaway snacks like soup and sandwiches as well. I think they're going to be very busy! |
|
1299)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96954)
Posted 20 Mar 2020 by Richard Haselgrove Post: Not quite. I've just come back from having my Last Supper in the barn. But I think they've closed now, and I left to a chorus of "See you on the other side" from the staff. |
|
1300)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96938)
Posted 19 Mar 2020 by Richard Haselgrove Post: Three. But you can't do it retrospectively if you didn't need to keep a copy in the first place. Good. Posted immediately, and index updated within 7 seconds. Going to watch some telly. |
|
1301)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96936)
Posted 19 Mar 2020 by Richard Haselgrove Post: Forum index here is taking 8 minutes to update after a new post. I think a Berkeley fiddler has been busy on both this and SETI's database configuration/settings. Email notification of PM took 42 minutes to arrive (although that could be Yahoo's fault - I've not got much email from anywhere today). |
|
1302)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96935)
Posted 19 Mar 2020 by Richard Haselgrove Post: Yup, SETI is broken too. Just replied to a PM that still hasn't been notified to me by email yet. I sent the reply PM, didn't get the completion screen, and of course with no 'sent' folder (despite repeated requests), you can't be sure it's actually been processed. So - browser back button from spinning wheel screen (to composition screen): change addressee to self: post again: open inbox in new tab: verify PM has arrived. So I'm assuming Rob's got his copy too. |
|
1303)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96933)
Posted 19 Mar 2020 by Richard Haselgrove Post: OK, I didn't spend the whole 40 minutes talking to the call handler - she was much too efficient for that - but the call lasted 40 minutes, and I typed the first message while I was waiting in the queue. But then I posted the second one, and waited, and waited, over 7 minutes for the confirmation to appear. Eventually, opened a new tab and it was there already, while the timer was still spinning. Set out to do my shopping, by then into the middle of the school run. Yay! That's something I will happily do without next week. But why is there no milk in Yorkshire? Edit - again. Posting this edit in a second tab, while the first tab hasn't completed the original post. |
|
1304)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96930)
Posted 19 Mar 2020 by Richard Haselgrove Post: wait while I hold for BT ...Well, that was worth doing. Just spent 40 minutes talking to a very friendly and helpful - and knowledgeable, and efficient - lady call handler, who has booked an 'up to five hours' home visit from a specialist engineer, on Monday morning. See if we can get to the bottom of this once and for all. |
|
1305)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96928)
Posted 19 Mar 2020 by Richard Haselgrove Post: You're lucky you got an order. I'm self-isolating in the box with Schrödinger's cat. I'm (nearly) 67¾, so on one list I'm not vulnerable because I'm under 70. But on another list, I'm offered a free flu jab every year because I'm over 65 (and for no other reason). And anyone who qualifies for a flu jab is automatically vulnerable, and has to stay in. I'll have to ask that darn cat - when I can find her. And when I get back from shopping. And the shopping will have to wait while I hold for BT, to find out why they cancelled the callback that they booked for this morning. |
|
1306)
Message boards :
Questions and problems :
Rosetta@home Can't create login "Captcha Failed"
(Message 96909)
Posted 18 Mar 2020 by Richard Haselgrove Post: I'm not a regular user of the Rosetta forum - I looked at a couple of possible sub-fora before posting, but decided to post in an active forum where there might be a better chance of it being read. And sure enough, a user has posted a reply there which might be or use to the originator of this thread: I also had this issue. Signing up directly through the BOINC Manager worked fine though since it doesn't require a captcha. |
|
1307)
Message boards :
Questions and problems :
Rosetta@home Can't create login "Captcha Failed"
(Message 96904)
Posted 18 Mar 2020 by Richard Haselgrove Post: There was a recent code change: "fix ReCaptcha problem" #3461. One of you could report that to the project - uh, if you could log in. I'll do it. Done. http://boinc.bakerlab.org/rosetta/forum_thread.php?id=13533&postid=92057#92057 |
|
1308)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96885)
Posted 17 Mar 2020 by Richard Haselgrove Post: Plenty of space for selfPlenty of space for self-abuse of any kind. It's got a gin garden, as well as the beer barn. |
|
1309)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96883)
Posted 17 Mar 2020 by Richard Haselgrove Post: Chance would be a fine thing! Dave would probably be able to say how long this lasted (it's his home town), but I doubt it was more than an hour or two:There are those in England who would accuse you of heresy.At >36C most of the day under high energy tropical sun (UV factor >10) you need a very cold beverage, why that is heresy? I use beer to cooling myself! The 38.7°C recorded in Cambridge Botanic Garden on 25 July 2019 during the recent European heatwave has now been confirmed to be the highest temperature ever recorded in the UK.By contrast, I live in the town which had the highest ever rainfall (by comparison with the local average - > +400%) last month. Most Brits are grateful to extract warmth from any beverage going. Speaking of which, my local barn is still providing beer, food, and plenty of space for self-isolation. They say they'll keep going until the Government actively orders them to shut down, but tonight it looked like it was going to be a struggle. |
|
1310)
Message boards :
Questions and problems :
Increase task transfer?
(Message 96874)
Posted 17 Mar 2020 by Richard Haselgrove Post: I inserted the line in cc_config.xml.For future reference, it's better to make a modest/minor change to the logging options via the GUI - Ctrl+Shift+F. This will create a complete cc_config.xml template, with tags in alphabetical order, and pre-filled with current or default values. Then if you want something new, find the right line and edit the value. Saves mistakes typing tag names, and avoids unexpected results from duplicated lines. We've had people complain that something doesn't work, only to find they've added a duplicated line and had the effect negated by the original. Last value to be read in from the file wins. |
|
1311)
Message boards :
Questions and problems :
Bad overclock, GPU no longer used?
(Message 96858)
Posted 17 Mar 2020 by Richard Haselgrove Post: "Doesn't get..." begs a question. Is your client asking for work? |
|
1312)
Message boards :
Questions and problems :
New way to login
(Message 96854)
Posted 17 Mar 2020 by Richard Haselgrove Post: If the mgr can't connect to the client, is sometimes because the client isn't running.Or sudo systemctl start boinc-clientMind you, if the client isn't running because it crashed last time, it'll probably crash again. You'll have to work out whether, and if so why, it isn't running before you try either of these. |
|
1313)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96837)
Posted 16 Mar 2020 by Richard Haselgrove Post: But what about the pub quizzzz?That's on Wednesdays. You bumped into a private quiz run by a dining party. |
|
1314)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96834)
Posted 16 Mar 2020 by Richard Haselgrove Post: I've just been to - and emerged safely from - my local branch. I don't foresee any problems with tomorrow's observance: the current venue is a big old barn of a place, set up for dining with the tables a respectful distance apart. I can self-isolate behind my newspaper at the top end, no bother at all. |
|
1315)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 96824)
Posted 16 Mar 2020 by Richard Haselgrove Post: Grumble. Telling people to isolate themselves at home will be an absolute goldmine for the phone scammers. Had to put the phone down on two already today. |
|
1316)
Message boards :
The Lounge :
That other thread
(Message 96823)
Posted 16 Mar 2020 by Richard Haselgrove Post: The good news is that my router is due to fail again around 11 am Wednesday, while I'am awake - it normally does it at 3 am, when I'm home but not interested. And BT are due to ring on Thursday morning to see how it went. It's incredibly easy to work round, just irritating. |
|
1317)
Message boards :
The Lounge :
That other thread
(Message 96819)
Posted 16 Mar 2020 by Richard Haselgrove Post: I have a British Telecom Home Hub 6 router which does that by itself every 2 weeks, whether I'm at home or not. :-( |
|
1318)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96745)
Posted 13 Mar 2020 by Richard Haselgrove Post: As a post on Faceache I just saw said in a cartoon of someone watching the news, "God I miss Brexit!"Default (English) government position: look at what European governments are doing, and do the exact opposite. |
|
1319)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96736)
Posted 13 Mar 2020 by Richard Haselgrove Post: England (as distinct from Britain) is very strange. There has been no central decision to cancel anything, yet sporting events are dropping like flies - football went earlier today, and the Wales-Scotland rugby match has just gone. The only advice we're being given is to take a full duvet week if we get a sniffle, but don't tell anyone. Eh? They made coronavirus a notifiable disease earlier in the outbreak, but now we can stop notifying NHS 111? How are they going to know how bad the epidemic is, where the hotspots are, and when it's all over, if people can keep their infection secret? |
|
1320)
Message boards :
Questions and problems :
Increase task transfer?
(Message 96716)
Posted 12 Mar 2020 by Richard Haselgrove Post: Client configuration. Bookmark it. |
|
1321)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96702)
Posted 12 Mar 2020 by Richard Haselgrove Post: Has the Formula One been canceled yet now McLaren decided not to drive?The BBC thinks it has: Coronavirus: Australian Grand Prix postponed |
|
1322)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96670)
Posted 11 Mar 2020 by Richard Haselgrove Post: He wasn't my friend, but someone in a video in the news. And no, he did not talk about soap or where to stick it. Why?British TV has been going on and on (and on) about washing our hands. The virus may not be alive, but it is infectious - and the lipid membrane is a critical part of the infective pathway. Disrupt the lipids, and you don't get infected. Or so our TV pundits say. |
|
1323)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96664)
Posted 11 Mar 2020 by Richard Haselgrove Post: Did your virologist friend talk about the specific role of soap and ethanol in disrupting the lipid membrane surrounding the coronavirus? |
|
1324)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 96646)
Posted 10 Mar 2020 by Richard Haselgrove Post: Thank goodness we've got an all-purpose thread again - I can't resist this one. Health minister Nadine Dorries tests positive for coronavirus Couldn't have happened to a nicer person - not. |
|
1325)
Message boards :
Documentation :
Boinc project list needs update:
(Message 96644)
Posted 10 Mar 2020 by Richard Haselgrove Post: Bam manager does use BM (Boinc Manager). Boinc Manager is the one that interfaces with BAM.Really? Methinks you need to read Account Management. And remember the role of the client. |
|
1326)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 96643)
Posted 10 Mar 2020 by Richard Haselgrove Post: Water off a duck's back. I've been spanked before, and I like to think that I'm as close to shitproof as the capitalist world permits. |
|
1327)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 96639)
Posted 10 Mar 2020 by Richard Haselgrove Post: Personally, I think that the coronavirus is the Earth's (or Gaia's, if you prefer) way of re-balancing the earth's ecology, by removing the super-spreaders (or Frequent Fliers). Nice to see a disease which affects the rich more than the poor. |
|
1328)
Message boards :
Documentation :
Boinc project list needs update:
(Message 96622)
Posted 10 Mar 2020 by Richard Haselgrove Post: The 'Science projects' list on this website and the list in BOINC Manager are both generated from the same source. Named projects are added to the list (on request) by David Anderson. The devices supported by each project are generated automatically from the plan_class names exported by each project. Having updated the code which generates the 'supported device' icons in BOINC Manager (#2643), I'm pretty sure it's robust. I'm only aware of one false positive - LHC's 'native_theory' and 'native_mt' plan classes falsely trigger the 'ATI supported' flag for the project. BAM! is an independent third party initiative, not under the control of BOINC. You'll have to take up any issues with them directly. |
|
1329)
Message boards :
Projects :
Low on Nvidia tasks
(Message 96611)
Posted 10 Mar 2020 by Richard Haselgrove Post: I believe a lot of people (especially Americans) pause high-powered computers when the hot summer weather arrives - they don't see the point of heating their houses up, and then paying for air conditioning to cool them down again. |
|
1330)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96610)
Posted 10 Mar 2020 by Richard Haselgrove Post: Spoilsports. Not only do they take away our project: now they've taken away our excuse for a party, too. |
|
1331)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96562)
Posted 9 Mar 2020 by Richard Haselgrove Post: We saw this list last night:  'GetDecics' are a BOINC CPU app ''MB8_win_x86_SSE3_' are a BOINC NVidia GPU app 'hsgamma_' is a BOINC intel GPU app from Einstein. Here are two more versions. This is a different machine, with only one NVidia GPU - but all four cores are still committed.And I'm showing you the einsteinbinary_BRP4 app, again for intel_gpu, because these tasks take less time, so it's easier to show you changes. The machine is Einstein host 8864187 - Intel i5 with HD 4600 GPU. This one is running under BOINC, but with the Process Lasso rule removed:  And this one is with the Einstein app running from the command line, outside BOINC:  I think we can confirm that BOINC is capable of setting process priorities to 'below normal' (in this GPU case), and that this particular Einstein developer chose to let BOINC do its thing. That's just one example: it is not possible to generalise that every developer at every project will make the same choice. |
|
1332)
Message boards :
Questions and problems :
List of projects not found in BoincManager?
(Message 96561)
Posted 9 Mar 2020 by Richard Haselgrove Post: The list of projects shown in BOINC Manager is (in principle) the same as the list of projects shown on Choosing BOINC projects - which is probably easier to work with. The list does change periodically, as new projects put themselves forward for inclusion, or old projects run out of data or energy. So at any given moment, a project might appear on one version of the list but not on the other - BOINC Manager only fetches a new list once every 14 days. |
|
1333)
Message boards :
BOINC Manager :
BOINC 7.16.5 not auto starting!
(Message 96556)
Posted 9 Mar 2020 by Richard Haselgrove Post: Both Cliff and I have now rebooted and got a normal restart - probable false alarm. |
|
1334)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96536)
Posted 8 Mar 2020 by Richard Haselgrove Post: Did it decide to set realtime by itself? I don't use Intel GPU apps, as I get the same power from a CPU app, and the Intel GPU takes a core away from the CPU.No, I deliberately choose to run Process Lasso (from which the screengrab is taken) to lock that particular application to RealTime - note the 'R' in the rules column. It's a particular special case which needs very little CPU support - but by god, does it need it quickly. This is a quad core - I'm running 2x integer CPU, 2x OpenCL needing 100% CPU, and the Intel_gpu using next to nothing. Keeps it all within the TDP power limit, so no throttling. Edit - notice that there are no rules controling the CPU and GPU apps. If you want, I can run those two apps (tomorrow!) in a command window or bench test, outside the BOINC app_start environment. That should settle it. |
|
1335)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96532)
Posted 8 Mar 2020 by Richard Haselgrove Post: Not the first time the code has diverged from the documentation, over time.Gotta wonder though which one it is now. I'm not going to ask. 'GetDecics' are a BOINC CPU app ''MB8_win_x86_SSE3_' are a BOINC NVidia GPU app 'hsgamma_' is a BOINC intel GPU app from Einstein, which is an interesting divergence from normal - Peter might be interested. |
|
1336)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96524)
Posted 8 Mar 2020 by Richard Haselgrove Post: But what I was observing was if the CPU is too busy with unrelated tasks (eg CPU WUs), then the GPU slowed down. This suggests to me that the CPU part of that GPU WU was not being given enough CPU time. But it should have been, as its priority in the OS was higher than the CPU WUs.Not necessarily. It might be that the CPU had to refetch a lot of data that wasn't in local cache memory, at every 'context switch' (look up that one up too). |
|
1337)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96523)
Posted 8 Mar 2020 by Richard Haselgrove Post: I prefer to look at the actual code:It's done when the BOINC client is starting the science app, as part of its management role.It's done when you compile the program using the BOINC API: https://github.com/BOINC/boinc/blob/master/client/app_start.cpp#L570 // Start a task in a slot directory.Not the first time the code has diverged from the documentation, over time. |
|
1338)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96516)
Posted 8 Mar 2020 by Richard Haselgrove Post: 'Utilisation' is not the same thing as 'priority'. A GPU, in particular, will only be 100% utilised if the problem is amenable to parallelisation, and the programmer is clever enough to program it that way. |
|
1339)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96511)
Posted 8 Mar 2020 by Richard Haselgrove Post: But BOINC doesn't set run priority of applications, as it doesn't run any work. BOINC is just the managing software. It's the project's science applications that do the work and it's the project that sets the run priority of the applications when they compile said applications. BOINC documentation does advise on what to set it to but it's still up to the project to follow that advice.'Priority' is a difficult word, because it's used in so many settings to mean so many things. But if you mean the thread priority of the CPU (part of a) project science app - yes, it does. It's always set full-CPU apps to run at idle priority, so as not to get in the way of the foreground apps the user is using. And when GPUs came along, their apps were set to 'below normal'. It's done when the BOINC client is starting the science app, as part of its management role. |
|
1340)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96493)
Posted 7 Mar 2020 by Richard Haselgrove Post: Perhaps we can persuade another worthwhile project to have a weekly outage?I think they'll need it if we arrive in force... |
|
1341)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 96470)
Posted 6 Mar 2020 by Richard Haselgrove Post: And solar doesn't even feature according to that site unless I am going blind?It's there, in red in the third column. Daytime only, for perhaps obvious reasons. |
|
1342)
Message boards :
BOINC Manager :
Two small translation errors for German
(Message 96469)
Posted 6 Mar 2020 by Richard Haselgrove Post: That's a sad page. "There is no activity the last weeks". The top displayed languages are complete, but they tail down to "Norwegian 0.4% 3,786 strings to translate". Which reminds me - I haven't seen any translation activity in preparation for v7.16.x |
|
1343)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 96455)
Posted 5 Mar 2020 by Richard Haselgrove Post: And I can't read that article, they want paid.Sorry, they must allow Goggle searches one free hit - I can't see it now, either. This one should work better: onshore wind |
|
1344)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 96452)
Posted 5 Mar 2020 by Richard Haselgrove Post: If wind is so cheap, why is my electricity bill going up and up?Largely because David Cameron blocked onshore wind to appease the nimbys in his own party. |
|
1345)
Message boards :
BOINC Manager :
maxos BOINC Manager: register or login not possible as new user
(Message 96443)
Posted 5 Mar 2020 by Richard Haselgrove Post: Boinc accounts are separate from project accounts (and each project is separate from each other). At Rosetta, you were an existing user. Here, you were a new user. But you are an existing member now - welcome. |
|
1346)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96433)
Posted 5 Mar 2020 by Richard Haselgrove Post: But how would we reach 10K posts if nobody is posting? ;-) |
|
1347)
Message boards :
Questions and problems :
how to get closest deadlines computed first?
(Message 96404)
Posted 4 Mar 2020 by Richard Haselgrove Post: Since my ears are burning (though I can't be sure you were referring to me) - why would you want to do this, anyway? |
|
1348)
Message boards :
News :
Top 100 user list
(Message 96400)
Posted 4 Mar 2020 by Richard Haselgrove Post: Again, my question was: floating point maths/operations, or integer ditto? How do you know that "it rates the gflops PPD correctly"? What measure of floating point speed are you using? Might you be using the 'GFLOPS Peak' reported by BOINC, which is a theoretical maximum known not to be achievable in real-world scientific maths: its only value is for advertising. |
|
1349)
Message boards :
Questions and problems :
Error connecting to localhost
(Message 96332)
Posted 3 Mar 2020 by Richard Haselgrove Post: Good to know. Please keep us posted with any oddities you encounter, or any positive feedback you have. It would be good to get this one out of the door, it it's good enough. |
|
1350)
Message boards :
Questions and problems :
Error connecting to localhost
(Message 96323)
Posted 3 Mar 2020 by Richard Haselgrove Post: Thanks. That is a Sunny Cove range CPU - specifically "Ice Lake-U" and - significantly - ultra-low power. I think it's highly likely to be the same bug. You have two choices, both of which are known to work. 1) Safety first. the 32-bit version of BOINC runs on those machines. Download from https://boinc.berkeley.edu/dl/boinc_7.14.2_windows_intelx86.exe 2) Be adventurous. We have been waiting for the next version of BOINC for a long time, and it might finally be here. https://boinc.berkeley.edu/dl/boinc_7.16.5_windows_x86_64.exe I've been running it for a couple of days, and so far I haven't seen any problems - including on one of the affected CPUs. But it is always possible (all-too-possible) that bugs still exist and are only revealed when we properly test the more obscure features - which I haven't done yet. It's up to you. |
|
1351)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 96319)
Posted 3 Mar 2020 by Richard Haselgrove Post: Another whole bottle of single malt? My my.I'm saving mine for the end of the month. |
|
1352)
Message boards :
Questions and problems :
Error connecting to localhost
(Message 96317)
Posted 3 Mar 2020 by Richard Haselgrove Post: Look in the Windows task list to see if the client - boinc.exe - is actually running. There have been many reports that BOINC client v7.14.2 (which is actually quite old) crashes at startup on some recent Intel CPUs - and if it's crashed, you can't connect to it. You mention that it's a laptop - could you give us make, model, and - specifically - CPU model number, please? |
|
1353)
Message boards :
BOINC Manager :
Manual suspend keeps getting reset
(Message 96264)
Posted 2 Mar 2020 by Richard Haselgrove Post: If there's one paused task in queue, the client still redownloads more, until it has enough tasks to satisfy it's daily task buffer (eg preload tasks for 1 day, and there'll be a whole list of tasks it will download).I've seen people report that, but the key is that 'RE-downloads' - tasks which you have lost through some mishap may be re-sent while some other task is suspended. It should never happen in normal usage. |
|
1354)
Message boards :
Questions and problems :
When do the boinc stats update?
(Message 96263)
Posted 2 Mar 2020 by Richard Haselgrove Post: Each project publishes its own stats, usually once or twice per day but sometimes more, sometimes less. It's configurable. The timing usually starts on the 24 hour or 12 hour anniversaries of the last time the project servers were restarted - so it's effectively random. Each of the many statistical aggregation sites will have their own policies and timetables. Probably the best known, https://www.boincstats.com/, polls the projects throughout the day, and collects any new data available. Incremental updates are posted whenever available in the 'today' column, but the definitive 'Credit/Day' total is usually published in the afternoon, UTC. |
|
1355)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96249)
Posted 1 Mar 2020 by Richard Haselgrove Post: Yes I'm already doing that. But the problem is different tasks require different amounts of CPU. If you're running several projects it can become quite complicated. And other things use the CPU, like the user using the computer. It's very variable and would be better monitored 24/7 by the computer itself. It must be fairly easy for Boinc to monitor the CPU usage, and if it's hitting 100%, lower the number of CPU tasks running. Then no GPU tasks will ever get throttled, and the user will never experience a slowdown of the interface.I think you're asking for something like #2949 |
|
1356)
Message boards :
GPUs :
Big ask - auto throttling of CPU tasks when using GPUs?
(Message 96235)
Posted 1 Mar 2020 by Richard Haselgrove Post: Yes. Use an app_config.xml file and 'declare' that the GPU tasks require 100% of a CPU (core) in support. Doesn't change the actual GPU application behaviour, but does change BOINC's scheduling. |
|
1357)
Message boards :
Questions and problems :
False positive network errors detection due to goggle CAPTCHA on reference site
(Message 96232)
Posted 1 Mar 2020 by Richard Haselgrove Post: Tested with a later test version - Windows v7.16.5. Internet availablity test worked properly from here in the UK, but I don't think that's due to any change in BOINC - more likely a difference in our patterns of use. I'm a regular user of Google Chrome, so Google sees my machines quite a lot and knows I'm not a robot. 01/03/2020 08:21:59 | | Project communication failed: attempting access to reference site |
|
1358)
Message boards :
Questions and problems :
Stalled downloads
(Message 96217)
Posted 29 Feb 2020 by Richard Haselgrove Post: I tried pasting that download url into a browser, and got the same thing:  It's still stalled in the browser, after all the time it took me to capture and upload that screenshot, and even after a pause/resume. So it seems to be a real problem of the Rosetta server - feel free to copy and show them that image. Next question - how to get BOINC to handle it with a simple 'Abort transfer'? |
|
1359)
Message boards :
BOINC Manager :
Manual suspend keeps getting reset
(Message 96202)
Posted 29 Feb 2020 by Richard Haselgrove Post: Or suspend a task, after which Boinc decides to load more tasks from that same project.No. Work fetch from a given project is inhibited if a task from that project is suspended. |
|
1360)
Message boards :
Questions and problems :
Boinc unable to connect to the core client
(Message 96201)
Posted 29 Feb 2020 by Richard Haselgrove Post: Does that mean adding that PPA boinc will successfully connect to the core client?The Manager should connect to the core client if the core client is capable of running on your CPU, and is actually running. |
|
1361)
Message boards :
BOINC Manager :
File-Select Computer
(Message 96199)
Posted 29 Feb 2020 by Richard Haselgrove Post: If necessary with a VGA HDMI adapter:VGA image quality also depends critically on the quality of the VGA cable. Use shielded cable for high resolutions where possible, or at the very least a cable with ferrite cores at each end - before those, I used to get visible ghosting. I've now switched to digital signalling throughout. |
|
1362)
Message boards :
Questions and problems :
A easy question???
(Message 96167)
Posted 28 Feb 2020 by Richard Haselgrove Post: Does the word 'Breakthrough' also appear on screen, alongside the word 'Listen'? If so, it'll be a reference to the Breakthrough Listen initiative. Breakthrough Listen supplies source data (recordings of radio signals picked up by radio telescopes) to the SETI@Home project. In this case, it's a figurative use of the word, as in "the human race is listening to the stars, trying to find out if we are alone". Sorry, no audio component. |
|
1363)
Message boards :
Projects :
News on Project Outages
(Message 96161)
Posted 28 Feb 2020 by Richard Haselgrove Post: I haven't been able to contact GPUGrid the entire day today. All uploads are not going through and backing off.No. My last successful contact was just before 21:00 UTC last night. They're in Barcelona, so make that 22:00 local. My guess is that the server crashed when there was no-one in the lab, and they're just now coming in to a cold office and beginning to wonder what happened. Edit - bits of it are working again, but they still have more to fix. |
|
1364)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 96147)
Posted 27 Feb 2020 by Richard Haselgrove Post: How do you exclude certain GPUs on the same system from running a certain project?See the Client configuration - Options page in the User Manual. |
|
1365)
Message boards :
News :
Top 100 user list
(Message 96132)
Posted 26 Feb 2020 by Richard Haselgrove Post: And the majority of the featured users have a large pie-slice of Collatz Conjecture processing. How accurate is it to describe that as floating point operations, when the conjecture itself concerns integer mathematics? |
|
1366)
Message boards :
GPUs :
"Application has been blocked from accessing graphics hardware" in Windows 10 notifications.
(Message 96064)
Posted 25 Feb 2020 by Richard Haselgrove Post: How can it not be fresh when I only installed it a week ago? I've run DISM and SFC anyway, and they both found no problems.Depends how old your installation medium is. Even if you downloaded it the same day, it might have been sitting on the server for weeks, months, years. Did you keep forcing Windows updates until it said there was nothing left to do? |
|
1367)
Message boards :
Questions and problems :
200 tasks downloaded for a project with a resource share of 0?
(Message 96038)
Posted 24 Feb 2020 by Richard Haselgrove Post: So if you change for instance that from now on only GPU work can be had, and your client requests work for CPU and GPU, the server will a) send the new GPU only preferences, and b) only give work for the GPU (if available).I'd like to believe that, and it's what the documentation says should happen. But my nasty cynical mind is saying that it hasn't always happened, at all projects, in the past. |
|
1368)
Message boards :
Questions and problems :
Account Management (BAM, Science United)
(Message 96033)
Posted 24 Feb 2020 by Richard Haselgrove Post: The argument for BAM! (from the users) is that a single administration point makes it easier to make significant changes to a large number of computers on an account. I wonder if that correlates to participation in challenges and events? The argument for SU (from the administrator) is that it attracts a different kind of volunteer, who "wants to help" but doesn't want to get involved in the detail. |
|
1369)
Message boards :
Questions and problems :
Account Management (BAM, Science United)
(Message 96030)
Posted 24 Feb 2020 by Richard Haselgrove Post: Which begs the question - why did you feel the urge to use an account manager in the first place? You seem to have a clear idea of what projects you would like to help - just attach to those projects directly, and cut out the middleware. You are recommended to use the same email address at each project (in case you ever wish to view consolidated statistics in the future), but you can use a different password at each one. The only limit is your ability to remember them all. |
|
1370)
Message boards :
Questions and problems :
200 tasks downloaded for a project with a resource share of 0?
(Message 96027)
Posted 24 Feb 2020 by Richard Haselgrove Post: Doesn't have to involve a venue change - that applies to any change to settings made on the project website. Remember that no project server can unilaterally send you work. It wouldn't get through your firewall. Instead, project servers respond to your computer's request for work. So for every website change, you have to make TWO project updates. The first transfers the change in settings from the website to your computer: the second acts on the new settings. If you want to stop something happening, * Set 'no new tasks' * Make the change on the website * Update the project * Allow new tasks |
|
1371)
Message boards :
Questions and problems :
Stalled downloads
(Message 96017)
Posted 23 Feb 2020 by Richard Haselgrove Post: Another way of cutting down the log and making it more readable is to set <dont_contact_ref_site>1</dont_contact_ref_site> in cc_config.xml - that removes the debug clutter of the check-call to google.com after a failed BOINC connection. |
|
1372)
Message boards :
Questions and problems :
BOINC in Windows 10
(Message 95997)
Posted 21 Feb 2020 by Richard Haselgrove Post: Dave is probably assuming that you are using 'Advanced View'. If you have a multicoloured display with images representing the project you're working on -Could you go to options, event log options and enable http debug then post what appears there when you try and connect to a project. That may give a few more clues.I have little clue on how to do what you asked... No idea. Look at the top of the display, find the word 'View', click it with your mouse or trackpad, and click again on the line 'Advanced View' which appears. After that, the sequence of words to click on will make more sense. |
|
1373)
Message boards :
Questions and problems :
What's the deal with VirtualBox?
(Message 95991)
Posted 21 Feb 2020 by Richard Haselgrove Post: Did you find that the 64-bit versions crash in their (somewhat elderly) libeay32.dll file, on those new lower-power processors? This file is a component of the OpenSSL network security framework. The next version of BOINC will be shipped with a newer version of the library which has been tested (by me, among others) to solve the problem. |
|
1374)
Message boards :
Questions and problems :
What's the deal with VirtualBox?
(Message 95988)
Posted 21 Feb 2020 by Richard Haselgrove Post: No, you don't need to install VirtualBox. The only side-effect is that you will not be able to run work from projects that have configured their work to run inside a VB container. |
|
1375)
Message boards :
BOINC Manager :
File-Select Computer
(Message 95963)
Posted 19 Feb 2020 by Richard Haselgrove Post: It's not running a whole new copy of the Manager - it's just a new window spawned by the already-running app. File menu, top item: "New BOINC Manager window..." |
|
1376)
Message boards :
BOINC Manager :
File-Select Computer
(Message 95959)
Posted 19 Feb 2020 by Richard Haselgrove Post: I currently have five different BOINC Manager windows available on this machine - one for the local host, and four for different remote machines on my network (two Windows, two Linux). When I reboot locally (perhaps once a month), I re-setup those connections by computer name - BOINC Manager (I presume) gets the IP addresses from the DHCP or DNS servers in my router. Once set up, I can see the computer name for each instance in three places - the title bar of the main window on screen, the miniature image popped up from the Windows task bar, and the tray icon balloon in the notification area. It's one of the subtle little touches that the early developers were quite good at embedding in BOINC - and it makes life a lot easier. |
|
1377)
Message boards :
Questions and problems :
Stalled downloads
(Message 95954)
Posted 19 Feb 2020 by Richard Haselgrove Post: We won't be able to solve this unless somebody catches one in the act and can display some diagnostic test results. Then we'll at least know which bit of the system is broken - though even that doesn't guarantee we can fix it. |
|
1378)
Message boards :
BOINC Manager :
File-Select Computer
(Message 95919)
Posted 18 Feb 2020 by Richard Haselgrove Post: If you saw how much my 8-port DVI model cost (GBP 'ouch'), you'd know why I keep using it, with HDMI <--> DVI adapters as needed (free with some card models, a few years ago - never throw anything away!) |
|
1379)
Message boards :
Questions and problems :
Selecting consecutive tasks
(Message 95918)
Posted 18 Feb 2020 by Richard Haselgrove Post: You could always switch to Linux - then the settings are stored in hidden files. |
|
1380)
Message boards :
Questions and problems :
Selecting consecutive tasks
(Message 95914)
Posted 18 Feb 2020 by Richard Haselgrove Post: I said it was a wild thought in my first reply - I was clutching at straws, because I've never, ever, seen this. I've now set an experiment running, in a 'New BOINC Manager window...' - because experience tells me that changes made in a secondary window aren't permanent. Which brings me on to one of my pet peeves. The only way I've found to remove a sort, once applied, is to close BOINC Manager entirely (you can leave the client running), and edit the registry. In regedit, navigate to [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager\Tasks], and set "SortColumn"=dword:ffffffffThat's equivalent to column -1 - used as a null column value. |
|
1381)
Message boards :
Questions and problems :
Selecting consecutive tasks
(Message 95910)
Posted 18 Feb 2020 by Richard Haselgrove Post: Your screenshot shows that you still have tasks sorted by (ascending) deadline - so it may still be implicated, even if not directly by 'new task insertion'. |
|
1382)
Message boards :
BOINC Manager :
File-Select Computer
(Message 95908)
Posted 18 Feb 2020 by Richard Haselgrove Post: I see the cc_config option is missing from the 'remote control' page - see Client configuration - Options instead. I suppose we should update that doc - the option may be more recent.It's there now. |
|
1383)
Message boards :
BOINC Manager :
File-Select Computer
(Message 95906)
Posted 18 Feb 2020 by Richard Haselgrove Post: Are the contents of gui_rpc_auth.cfg just the characters of the password or do you need, for example, password=xyz.Just the password. Best not even to have an end-of-line character. After changing the contents of the file, you should restart the BOINC client so it's expecting the new password in the future. That explains adrianxw's point as well (no waiting time, just the restart). Is the system name literally the pc’s hardware name or it’s IP address or ...Can be either - I prefer to use the hardware name. And coming in from a different location do you have to set up a hosts file or is it better to leave that blank (for dynamic ip addressing).Whatever suits your networking setup. I prefer to use DHCP. Because IP addresses can change on a dynamic network, it's easier to allow remote control via cc_config.xml, rather than listing them all, or a wildcard, in remote_hosts.cfg I see the cc_config option is missing from the 'remote control' page - see Client configuration - Options instead. I suppose we should update that doc - the option may be more recent. |
|
1384)
Message boards :
BOINC Manager :
File-Select Computer
(Message 95885)
Posted 17 Feb 2020 by Richard Haselgrove Post: 1) The box under your desk is known as a KVM, or 'keyboard-video-mouse' switch. They are available for VGA, DVI and (probably) HDMI - I've only used the first two. And the KM part might be PS/2 or USB. 2) For the "Select Computer" part, read Controlling BOINC remotely. The 'password' in question is the contents of gui_rpc_auth.cfg on the remote computer. The default values are empty (Linux) or a random 32-character string (Windows). Neither is very helpful. This value is only used for controlling BOINC, and for any domestic home network is not a high-security item. My own is an eight character string, starting with 'p' and ending with 'd'. To get control of your primary machine back, use the menu item for 'select computer', and leave both boxes blank. That should give you control of the local machine. |
|
1385)
Message boards :
Questions and problems :
Stalled downloads
(Message 95880)
Posted 17 Feb 2020 by Richard Haselgrove Post: Sorry - no, I didn't. (The morning was getting a bit busy with other tests around that time). I'll try and remember next time. It did remind me of SETI message 1343415 from 2013: SETI was having major problems with stalled downloads at the time, and somebody found RFC 1323 - an official Internet standard for dealing with congested internet links. Linux and OS X have it enabled by default: Windows supports it, but it has to be enabled manually. It worked a treat for the SETI problems at that time. I haven't looked at it again for Windows 10. It works in Windows 7, and is enabled on this machine - so it didn't prevent this morning's glitch. But it's worth knowing about. |
|
1386)
Message boards :
Questions and problems :
Stalled downloads
(Message 95874)
Posted 17 Feb 2020 by Richard Haselgrove Post: Just had another 'rb_' download. This time it allocated me the 64-bit version of the app, so two more big downloads. Both of them stopped at 99.65%. That sounds to me like a dropped packet somewhere in the middle, and a wait for the resend. I tried the usual trick for that - suspend networking for a few seconds via the 'Activity' menu in BOINC Manager, and then revert to networking 'always'. Worked a treat - the task is now ready to run. |
|
1387)
Message boards :
Questions and problems :
Selecting consecutive tasks
(Message 95873)
Posted 17 Feb 2020 by Richard Haselgrove Post: I use multi-select quite a lot (usually mouse-click, shift-click at end of a range, ctrl-click to add or remove single tasks). I've never seen this problem. Wild thought - do you have the tasks displayed in any special sort order? I'm careful to keep all my displays in 'natural' (FIFO) order. |
|
1388)
Message boards :
Questions and problems :
Stalled downloads
(Message 95865)
Posted 16 Feb 2020 by Richard Haselgrove Post: My line is running at 71.17 Mbps, but it still took 20 minutes. The files I've seen stick always seem to be associated with WUs with a name beginning multistate, like http://boinc.bakerlab.org/rosetta/workunit.php?wuid=1010934932Thanks - it's helpful to know that. Mine all start 'rb_02' |
|
1389)
Message boards :
Questions and problems :
Stalled downloads
(Message 95863)
Posted 16 Feb 2020 by Richard Haselgrove Post: Well, I thought I'd try and see what happens for myself. Allowed new work for an old test attachment - got four new Rosetta v4.07 tasks. All downloaded cleanly, although there wasn't a 3KB file amongst them (nothing between 1KB FLAGS and 7KB robetta.zip files). Download server was slow nearly 20 minutes for the database file - my internet link is a lot faster than theirs). Nothing obviously wrong in client_state.xml They'll run in due course - I'll watch what happens. |
|
1390)
Message boards :
Questions and problems :
Stalled downloads
(Message 95861)
Posted 16 Feb 2020 by Richard Haselgrove Post: Your log said 16-Feb-2020 10:30:15 [Rosetta@home] Not requesting tasks: some download is stalledThat's entirely a client decision - the server isn't involved at all (and can't over-rule it). The problem seems to be that the download isn't completely forgotten about when it's cancelled. We need to find out why not. |
|
1391)
Message boards :
Questions and problems :
Stalled downloads
(Message 95858)
Posted 16 Feb 2020 by Richard Haselgrove Post: Cancel and request new work shouldn't be prevented - that sounds more like a client issue to me.Happens on all 4 of my computers. Only with Rosetta@home. Others experiencing the same thing judging by their forums.If already raised on the Rosetta forums, my guess is you will just have to wait till someone kicks the appropriate server. |
|
1392)
Message boards :
Questions and problems :
Stalled downloads
(Message 95857)
Posted 16 Feb 2020 by Richard Haselgrove Post: Unfortunately, the log doesn't show you cancelling the download - which is probably the log's fault, not yours. Two suggestions: 1) if you see it happening, set <http_debug> in Event Log options, and retry the transfer - find out what's happening behind that 'transient HTTP error'. 2) make a careful and exact note of the file name in question. Cancel the download, and make sure it disappears from the transfers tab. Restart the client, and if the 'stalled download' message reappears, have a very careful 'read only' (no edits) peek inside client_state.xml - same folder. Find the reference (if any) to the file you cancelled, and post the whole of the <file> ... </file>section it's enclosed in. |
|
1393)
Message boards :
Questions and problems :
Stalled downloads
(Message 95854)
Posted 16 Feb 2020 by Richard Haselgrove Post: Which operating system? And if it's Linux, how was it installed? If Linux, try variations on journalctl --boot=-1 --unit=boinc-clientIf Windows, file 'stdoutdae.txt' in data directory. |
|
1394)
Message boards :
Questions and problems :
Stalled downloads
(Message 95852)
Posted 16 Feb 2020 by Richard Haselgrove Post: Which version of BOINC? |
|
1395)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95837)
Posted 14 Feb 2020 by Richard Haselgrove Post: Nvidia drivers can decide to run OpenCL tasks in CUDA if they want...Yes. OpenCL is an intermediate level, cross-platform, programming language. Every manufacturer's driver - not just Nvidia's - compiles the OpenCL source code into machine code primitives to match the hardware in use. And since Nvidia's hardware runs CUDA primitives, that's what the Nvidia compiler implementation will be designed to output. It's not a 'decision' by the driver: it's a deterministic pathway defined by the programmer. |
|
1396)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95784)
Posted 9 Feb 2020 by Richard Haselgrove Post: Do grow up.It's trolling if you post assertions on a technical board like this without being willing to back up your assertions with answers to factual questions when asked. What was the year of manufacture of the NVidia card "which broke in under a week"? |
|
1397)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95772)
Posted 9 Feb 2020 by Richard Haselgrove Post: And as Ian&Steve C. said, it would be helpful to know when this anecdote dates from. I think the mainstream manufacturers have put a lot more effort into designing their cooling systems, once the problems with overheating became known.Nvidia selling me a card which broke in under a week is not acceptable. They can't make up for thatWas it a GPU that was manufactured by nVidia, or was a GPU manufactured by a third party using an nVidia chipset. Since the vast majority of the GPUs we use are actually manufactured by a third party with the chipset supplied by either nVidia or AMD I would hazard a guess at your "nVidia" GPU was actually packaged and manufactured by a third party (Asus, MSI etc...). Thus you should be blaming that third party manufacturer for your woes - they got the package wrong. 200th. post in this thread. That's a lot said about PCIe risers! |
|
1398)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95761)
Posted 8 Feb 2020 by Richard Haselgrove Post: I thought Nvidia actively helped making Cuda apps for the likes of Seti?Yes, they did. They wrote the application announced and released on 18 Dec 2008. Which was poorly tested and buggy, causing BSODs on VLAR tasks. After testing and revision, I ran the first successful VLAR task on 15 Jan 2009. After that, NVidia supplied a slightly revised version compatible with Fermi GPUs in early 2010, but after that - nothing. Note that ATI/AMD/Intel never supplied SETI applications - all that porting was contributed by SETI volunteers. Anyway, if the OpenCL version works better, that probably means it was easier to program in.I wouldn't say it works 'better', but it is easier for the programmers because the same program code (with minor changes) can be used on all three platforms. OpenCL is pretty much obligatory for Intel GPUs, and has been for ATI/AMD since they dropped support for their proprietary CAL platform. OpenCL is supposed to be a common, unified language available for all platforms, but that doesn't mean that it currently has a competitive, qualitative, advantage over the closed, expensive, platforms. |
|
1399)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95758)
Posted 8 Feb 2020 by Richard Haselgrove Post: Cuda is designed for crunching.It still needs to be programmed - by a programmer. Not all programmers are created equal. If the project's application is open source, you are invited to improve it (as the SETI apps mentioned by Ian&Steve C. have been improved by crunchers, with more or less success). |
|
1400)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95754)
Posted 8 Feb 2020 by Richard Haselgrove Post: Factor in the efficiency / proficiency of the programmer, too. |
|
1401)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95751)
Posted 8 Feb 2020 by Richard Haselgrove Post: So the Nvidia uses a lot more CPU? Kinda points to the Nvidia GPU being rubbish.It's a reflection on the software (programming) language chosen, not on the hardware. |
|
1402)
Message boards :
Questions and problems :
BOINC client crashes on Windows 10
(Message 95731)
Posted 6 Feb 2020 by Richard Haselgrove Post: Actually, you should be able to run - either the 32-bit version of v7.14.2 or the 64-bit version of v7.16.4 (or later) |
|
1403)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95651)
Posted 4 Feb 2020 by Richard Haselgrove Post: Any idea of opening time this week?About ... now. We're down. |
|
1404)
Message boards :
BOINC Manager :
Uninstalled but its still there
(Message 95632)
Posted 3 Feb 2020 by Richard Haselgrove Post: At least thanks to your help I hopefully have stopped it from starting automaticly from the options --> other options.Unfortunately, yes. For that trick to work, each user would have to uncheck 'Run at startup' in their own account's running environment. |
|
1405)
Message boards :
BOINC Manager :
Uninstalled but its still there
(Message 95631)
Posted 3 Feb 2020 by Richard Haselgrove Post: Can you work out whether it was installed 'as a service' or in user mode? You might be able to find an old stdoutdae.txt in the data directory (although you'd have to find that first). If that still has a BOINC startup section (might not if the server is rarely rebooted, but probably has because of the BSODs), it will be followed line a line saying 'Running as a Daemon' (=service mode) 'Running under account Username' (=user mode) On a server, I'd expect to see Service mode. Go to the service control panel, see if BOINC is listed. If it is stop it, and then set startup mode to disabled. If it's User mode, Jord's technique will work. |
|
1406)
Message boards :
BOINC Manager :
After Update to BOINC Manager 7.16.4, System Tray Icon Missing (Linux)
(Message 95624)
Posted 3 Feb 2020 by Richard Haselgrove Post: It's a conscious decision made by the Linux distribution managers / repo maintainers. They appear to have a world view which says that Linux is used by workers inside large, centrally directed, institutions (business, academia, science). Their model assumes: 1) BOINC runs as a service, requiring administrator access to stop and restart. 2) End users are not administrators - they have a technical support team for that. The home / soho sector doesn't fit that model, but they have no knowledge of that sector using Linux. |
|
1407)
Message boards :
Questions and problems :
How to change project priority without account manager?
(Message 95617)
Posted 2 Feb 2020 by Richard Haselgrove Post: On each separate project website, in your account. [project url]/prefs.php?subset=project You can set 'Resource share' (as it is properly known) for each of the four venue presets independently. |
|
1408)
Message boards :
BOINC client :
7.16.4 release info?
(Message 95588)
Posted 29 Jan 2020 by Richard Haselgrove Post: Still crashes under WINE on Linux as soon as you go to advanced view.It's unlikely to be the libeay32.dll issue at that point. Do you get a crash dump? |
|
1409)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95562)
Posted 28 Jan 2020 by Richard Haselgrove Post: Big earthquake near Jamaica. 7.7 Not sure if they felt it in Puerto Rico. First I hope all are safe and well, but I also hope all is good for our Aricebo .Caribbean earthquake of 7.7 triggers tsunami warnings suggests it's far enough away (other side of the Dominican Republic) not to be too damaging and the Tsunami shouldn't reach that high. Not good for the people and infrastructure of Puerto Rico, though - bang goes the power supply again, I expect. |
|
1410)
Message boards :
Questions and problems :
Computer doesn't like BOINC
(Message 95516)
Posted 24 Jan 2020 by Richard Haselgrove Post: Have tested and it does work. |
|
1411)
Message boards :
Projects :
Astro Pulse
(Message 95510)
Posted 24 Jan 2020 by Richard Haselgrove Post: About Astropulse |
|
1412)
Message boards :
Questions and problems :
Not downloading tasks
(Message 95507)
Posted 24 Jan 2020 by Richard Haselgrove Post: But it has become very common for BOINC to simply say there is no work to be had.BOINC itself is just a management tool - it doesn't provide any work itself. Any message about no work being available will have come from a project. Projects vary greatly in their consistency and reliability in providing work. If you want continuous work, attach to multiple projects. Alternatively, tell us which projects are failing to supply work to you, and somebody here will have some idea of their relisability and any problems they may have been suffering. |
|
1413)
Message boards :
BOINC Manager :
Boinc manager does not connect in openSUSE 15.2
(Message 95504)
Posted 24 Jan 2020 by Richard Haselgrove Post: cc_config.xml is a useful, but optional, file. If it exists on your machine, it's probably there because some user created it manually - and if it's missing the start tag, then they possibly created it badly. I'd suggest you look in the BOINC data directory, and see if it's there: if it is, delete it. BOINC should run without it, and you can re-create it later if you wish. That should eliminate one variable. It's hard to imagine what "No protocol specified" is telling us without some additional context. The only thing I know of that happens once per second is that the Manager attempts to contact the client (over TCP/IP) to obtain status updates. |
|
1414)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95495)
Posted 23 Jan 2020 by Richard Haselgrove Post: And Annie is back. Welcome to the outrage party. |
|
1415)
Message boards :
Questions and problems :
Computers signed up to project via Science United
(Message 95483)
Posted 23 Jan 2020 by Richard Haselgrove Post: The SU issue was raised on the Projects call yesterday by CPDN, by David Wallom (Andy's boss). David Anderson was one of several people not attending or calling in sick form home - there must be a lot of flu around. But the central BOINC team are going to discuss it at their next meeting. Bernd Machenschalk of Einstein is trying to contact Andy, because he thinks there might be (or ought to be) a way of preventing clients without the 32-bit libs being sent work. But Bernd's email has bounced - does anyone know if Andy has a particularly strong spam filter, or an alternate email address? |
|
1416)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95426)
Posted 22 Jan 2020 by Richard Haselgrove Post: That might not be the right solution for all. Resource share 1 will allow BOINC to request a full multi-day cache, and when it arrives, BOINC will panic that it won't be able to finish it in time.So adjust the share on your preferred projects so that the emergency project is 1 in a thousand or more.IIRC there are some projects which only allow a a resource of 1 or greater, I don't know what they are.Didn't know that. Interesting ... If resource share 0 is available, it should do a more manageable fetch. Zero is a relatively recent addition, so it'll be the projects which are on the ball and update their servers regularly which have it: projects which leave their servers untouched for years - won't. |
|
1417)
Message boards :
Questions and problems :
716.4 plans.
(Message 95400)
Posted 21 Jan 2020 by Richard Haselgrove Post: Bernd's first analysis did include both Windows 8 and Windows 8.1: 16 Microsoft Windows Server 2008 18 Microsoft Windows Server 2003 "R2" 21 Microsoft Windows Server 2012 23 Microsoft Windows Server 2003 65 Microsoft Windows 8 93 Microsoft Windows Vista 118 Microsoft Windows Server 2008 "R2" 162 Microsoft Windows Server 2016 167 Microsoft Windows Server 2012 R2 369 Microsoft Windows XP 751 Microsoft Windows 8.1 4406 Microsoft Windows 7 20274 Microsoft Windows 10(connections to Einstein@Home in the last 60 days) |
|
1418)
Message boards :
BOINC Manager :
Boinc manager does not connect in openSUSE 15.2
(Message 95370)
Posted 21 Jan 2020 by Richard Haselgrove Post: There's a reference to 'boinc-client.service' in an older post. I have no experience of openSUSE, but in Ubuntu service installs the BOINC message archive is blended in with the system log (and rather overwhelms the actual system messages). If you could open your local system log viewing tool, and filter on 'boinc', you might be able to extract something useful. |
|
1419)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95358)
Posted 20 Jan 2020 by Richard Haselgrove Post: Eric says (new news item): "We'll be doing a full duration outage ... tomorrow." Unfortunately, I won't be able to join you in the ceremonies - my branch of the outrage pub chain is ... having an outage! Three days off to recover from Christmas, I reckon. I've laid in a small case of emergency supplies, but I'd better remember the neighbours - no howling at the moon, I'm afraid. |
|
1420)
Message boards :
BOINC Manager :
Boinc manager does not connect in openSUSE 15.2
(Message 95324)
Posted 19 Jan 2020 by Richard Haselgrove Post: Implying that the client has not started or has crashed. |
|
1421)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95310)
Posted 18 Jan 2020 by Richard Haselgrove Post: But the rig in question here has no monitor. I can plug one in while setting things up for troubleshooting, but usually it's easier to access it remotely. Boinctasks to control Boinc and Remote Utilities for accessing Windows (the built in Windows remote program shuts off the GPU!) So would the limit still apply?The general principle of BOINC is that all science apps are generally written to be downloaded and run on general-purpose, consumer grade electronics. Some projects have a particular 'need for speed' and may recommend minimum limits, but SETI - since we've mentioned it already - is designed to be accessible to all-comers and their kit (hence the extremely long deadlines). If, against the whole spirit of BOINC, you choose to build your own supercomputer for bragging rights - you have to develop your own super-software to run on it. As the GPU User Group have done at SETI. Both BOINC and SETI - and some other projects as well - have published their canonical source code under the GPL, and (provided you comply with the licence terms under GPL) you are free to grab it and use it. Remembering that your scientific results will only be accepted and recognised if they pass the quality-control verification provided by validation. |
|
1422)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95301)
Posted 18 Jan 2020 by Richard Haselgrove Post: Why do they have the 5 second limit? Is it a real limitation of the GPU, or a silly decision made by the CUDA inventors? and why can't everything be stored in GPU memory, with just calls by the CPU for the GPU to access it? I thought that was the whole point of a GPU having memory on board, so it didn't have to access main RAM and a) get in the way of the CPU trying to use it for something else, and b) wait on bus transfers.Googling around (an exercise I suggest you try sometime), it seems that the execution limit is applied by the operating system, and only applies to GPUs with video output ports. Perhaps it's the working limit that the operating system 'look and feel' developers think that a human being is prepared to watch the eggtimer (or modern equivalent) while her search results load. I don't know. The 5-second limit doesn't appear to apply to Tesla-class professional GPUs without video outputs, or presumably to coin mining frivolities - though I think you have to use the professional drivers to get that benefit. But since BOINC is designed to be used on volunteers' home computers with - presumably - screens attached, it's a moot point. |
|
1423)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95290)
Posted 18 Jan 2020 by Richard Haselgrove Post: Just how big are those SETI tasks? Surely you can't need half a GB to be transferred every second, that would severely max out your internet connection giving the result back. Can't the memory on the GPU be used during the task? Or is the CPU doing a lot of assistance and needs to communicate with the GPU regularly?GPUs - taking NVidia CUDA as an example - are limited to 5 seconds runtime per program launch, with a tighter TDR limit of 2 seconds under Windows. You will have noticed that SETI tasks take longer than this... The whole art (and it is an art) of GPU programming is to break the original task into a myriad of tasklets, or kernels - as many as possible of which are running in parallel. The GPU itself doesn't have the decision-making hardware to manage this: the CPU is responsible for ensuring that each new kernel - and its associated data - is present and correct at the precise millisecond (or is that microsecond?) that the shader becomes available to process it. You can't just measure the data flow as a one-time load: you have to know the average memory used by each kernel, and the number of kernel launches needed to process the whole meta-task. I can't begin to estimate those numbers. |
|
1424)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95282)
Posted 18 Jan 2020 by Richard Haselgrove Post: additionally since no one here seems to know how to check PCIe bus utilization with an in-progress WU. I checked on my system so I can answer for you. I run Linux Ubuntu 18.04.3 and I used a program called 'gmonitor' which is similar to htop and gives graphical and numerical info in a terminal window. Under Windows, I would assume that a program like GPUz or MSI afterburner might show you a bus utilization. but my crunchers do not run windows so you'll have to find a program yourself.Windows 10 (later releases only) has an enhanced Task Manager, with GPU monitoring on the Performance tab. The GPU monitor can visualise many different performance metrics (11 on the Intel iGPU I've just checked), but only has 4 mini-graphs to display them. The components used for distributed computing aren't among the default four shown at startup - you have to search through the drop-down lists to find the applicable one(s). |
|
1425)
Message boards :
Questions and problems :
7.16.3 has idle GPU: which parameter is causing the delay?
(Message 95270)
Posted 18 Jan 2020 by Richard Haselgrove Post: ... not a lot of help from the "project properties" toward diagnosing the problem.No, it's not a helpful diagnostic tool. Better to see whether you have, or have not, requested work - and if you have, what for and how much. |
|
1426)
Message boards :
Questions and problems :
7.16.3 has idle GPU: which parameter is causing the delay?
(Message 95269)
Posted 18 Jan 2020 by Richard Haselgrove Post: I was hoping to see some actual numbers: 18/01/2020 16:44:37 | SETI@home | Sending scheduler request: To fetch work. 18/01/2020 16:44:37 | SETI@home | Reporting 17 completed tasks 18/01/2020 16:44:37 | SETI@home | Requesting new tasks for NVIDIA GPU 18/01/2020 16:44:37 | SETI@home | [sched_op] CPU work request: 0.00 seconds; 0.00 devices 18/01/2020 16:44:37 | SETI@home | [sched_op] NVIDIA GPU work request: 41239.70 seconds; 0.00 devices 18/01/2020 16:46:49 | SETI@home | Scheduler request completed: got 0 new tasks 18/01/2020 16:46:49 | SETI@home | Project has no tasks availableThe last line is getting increasingly common. |
|
1427)
Message boards :
Questions and problems :
7.16.3 has idle GPU: which parameter is causing the delay?
(Message 95267)
Posted 18 Jan 2020 by Richard Haselgrove Post: Set <sched_op_debug> and see what you're actually asking for. You need to distinguish between "SETI doesn't have any work available" and "I didn't even ask for any work, available or not". |
|
1428)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95260)
Posted 18 Jan 2020 by Richard Haselgrove Post: I do have about 1 year of experience observing the phenomenon in Folding, Boinc, Bitcoin mining, and Games.BOINC doesn't use the GPUs: each separate (GPU using) project uses the GPU in its own way. You can't generalise from one project to another. GPUs have been used by BOINC projects for over 10 years now, and the way they are used has changed over time. |
|
1429)
Message boards :
Questions and problems :
Boinc start on boot using Linux Mint.
(Message 95258)
Posted 18 Jan 2020 by Richard Haselgrove Post: As far as the PPA sources, what's wrong with them?The newest versions (still in the Beta test stage - more changes are expected) have been modified by the Linux community in a way that makes them more suitable to institutional security environments. The changes haven't been universally welcomed by private individuals more used to having total control over their own computers. Boinc is a more advanced version, and shows more projects listed.The list of projects available is dynamic data, not tied to any particular version of BOINC. 'all_projects_list.xml' is derived from the same sources as https://boinc.berkeley.edu/projects.php. |
|
1430)
Message boards :
Questions and problems :
Computer doesn't like BOINC
(Message 95257)
Posted 18 Jan 2020 by Richard Haselgrove Post: I had the same. Some windows 10 versions (64 bit) for an unexplained reason, don't connect.I think it has been explained. The Manager can't connect to the client, because the client has crashed. The client crashes because the communications security library (libeay32.dll) crashes on some low-power CPUs in 64-bit applications. BOINC will be deploying a newer version of libeay32.dll in the next release, in an attempt to cure this problem. |
|
1431)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95195)
Posted 16 Jan 2020 by Richard Haselgrove Post: That sounds like the sort of device I helped Eric install in Muarae2 in the summer - 4 M.2 SSDs on a single PCIe card, lying flat in a 1U server case. Very fiddly. I doubt that would work on the end of a riser cable. Mind you, I think Eric is still having difficulty getting it to work in the datacenter.Since there aren't too many PCIE 4.0 devices out...But for all AMD GPUs available now, RX 5500, 5600 and 5700 series. Intel added support to its Optane SSDs. Asrock and Asus have launched own versions of M.2 cards in which you can slot up to 4 M.2 SSDs. There's enough PCIe 4.0 around, if only you care to look. |
|
1432)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95192)
Posted 16 Jan 2020 by Richard Haselgrove Post: Just checked in my parts bin. An IDE cable is 2 inches across (like all electronics specified in the USA, I'm sure it's non-metric). So the wire pitch in a 40 conductor cable is 0.05 inches, for 80 conductors it's 0.025 inches. The difference is clear to the naked eye. Seems like the guard wire technology came in with the ATA-66 specification around 2000. |
|
1433)
Message boards :
Questions and problems :
Not downloading tasks
(Message 95189)
Posted 16 Jan 2020 by Richard Haselgrove Post: I think it does, yes.The pressure is coming from the big Linux distribution managers, which seem to be academic and research based. They took the menus out of the BOINC code, and after some pressure I got them put back into BOINC. So the Linux people took them out at the distribution packaging stage...Presumably that means rolling our own still keeps them in? |
|
1434)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95188)
Posted 16 Jan 2020 by Richard Haselgrove Post: I'm of an age to remember when we went through a similar transition for IDE hard drives. The actual pinout required a 40 conductor cable, but the fastest motherboards and drives switched to using 80 conductors in the same form factor - thinner wires, so the ribbon cable felt much smoother to the touch. I don't know how the 80 conductor cables were wired - they used the same 40-pin connectors. They might have been two wires per signal, or they might have been grounded guard wires between each signal wire. Rob, any idea? |
|
1435)
Message boards :
Questions and problems :
Not downloading tasks
(Message 95184)
Posted 16 Jan 2020 by Richard Haselgrove Post: I've also noticed that a lot of the options have disappeared from the menu under File, including Exit BOINC Manager.Not 'commercialised' as such, but 'institutionalised'. The pressure is coming from the big Linux distribution managers, which seem to be academic and research based. They took the menus out of the BOINC code, and after some pressure I got them put back into BOINC. So the Linux people took them out at the distribution packaging stage... The argument is that Linux people work in large organisations like CERN, and that the decision to install BOINC is taken by the organisation, not by the individual volunteer. [Not a description I recognise, but it's how they think]. The effect is that 'their' BOINC is running on computers which users can't control - no sudo, no root. If they 'Shut down connected client...', they have to call technical support to start it again (or just pull the power cord, and restart everything that way). When David pulls his finger out, and restarts testing on the new release which was urgent six months ago, I'll restart the battle. |
|
1436)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95165)
Posted 15 Jan 2020 by Richard Haselgrove Post: Another bullet point: the science. Look at Einstein's Gravity Wave application (now available for GPU). I downloaded 130 MB of data this morning - all of that has to get to the GPU somehow. Compare with a maths app: no data, just a parameter set. Is number X prime, or does number Y disprove the Collatz Conjecture? They place different demands on the GPU. |
|
1437)
Message boards :
GPUs :
BOINC on new PC says "GPU Missing waiting to run"
(Message 95150)
Posted 15 Jan 2020 by Richard Haselgrove Post: Well, that's exactly what it should look like. That would normally be ready to compute under BOINC, but for the pesky problem with the OpenCL part of driver version 441.66, as previously discussed. |
|
1438)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95141)
Posted 15 Jan 2020 by Richard Haselgrove Post: That may require some nifty work with - probably not - a hacksaw, or a fine cutting tool like a Dremel. By opening the end of the PCIe x1 slot, the x16 riser could be physically inserted into the x1 slot. He'd still have to watch out for power consumption: the sense pin should tell the motherboard that a card is present. But he would still have to investigate and manage the card's actual power draw from each input. It should be possible: his cards have nominal power inputs for 375W (75W from the PCIe slot, plus 2x 150W 8-pin supplementary inputs). The cards he has are rated at 250W average total board power. So there's headroom - it's just a question which input has the spare capacity, and that depends on the manufacturer. To my mind, fitting the dual 8-pin suggests that the bulk of the power will be taken from them: if the full 75W was taken from the motherboard, they could have got away with 1 8-pin and 1 6-pin. But I am not a circuitry designer: it's all supposition, and it might still fry the motherboard. Proceed with extreme caution, and keep a fire extinguisher close at hand. |
|
1439)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95127)
Posted 15 Jan 2020 by Richard Haselgrove Post: I am now officially surprised. Having said that, it's very close to 24 hours since it started, so whoever got up early yesterday to stop them (Jeff?), perhaps got up early to start them? Edit - scheduler is now accepting reports, very slowly. I'll be doing 'report only' (with NNT set) until it's woken up a lot more. |
|
1440)
Message boards :
Questions and problems :
Not downloading tasks
(Message 95123)
Posted 15 Jan 2020 by Richard Haselgrove Post: Several clicks later (and a few security updates while I'm down here), both 'number of times contacted' and rpc_seqno seems to be updating normally. So I'm even more stumped, unless it's our old friend from SETI - replica database not keeping up with master. |
|
1441)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95122)
Posted 15 Jan 2020 by Richard Haselgrove Post: I have a suspicion they are having another try at the upgraded database software that cause so many problems with the anonymous platform just before Christmas. In any case, with all the large crunchers out of work, the recovery is going to be one for the ages.Although we have one report of a fix (working at LHC-dev), I'd be surprised if they tried it with the database in this state. But I've been surprised before. |
|
1442)
Message boards :
Questions and problems :
Not downloading tasks
(Message 95115)
Posted 15 Jan 2020 by Richard Haselgrove Post: Well, you have contacted the server (Les has contacted CPDN), because you've had a reply with version numbers and everything. I attached Linux Mint CPDN host 1497236 yesterday after the conversation with Dave, and connected / got a task straightaway. So I have to admit I'm stumped on this one. Edit - my machine has contacted the server 0 times as well, even though I've just clicked update and the 'last contact' time has updated. I think we have to conclude that the 'number of contacts' stat is broken. I'll go downstairs and check on client_state.xml and see if the local anti-fraun bumber is accurate. |
|
1443)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95109)
Posted 15 Jan 2020 by Richard Haselgrove Post: Gone 1am in Berkeley - I don't think it's coming back until this evening. Time to fire up the backup projects. |
|
1444)
Message boards :
Questions and problems :
Why doesn't Boinc schedule earlier deadlines first?
(Message 95107)
Posted 15 Jan 2020 by Richard Haselgrove Post: Multiple issues, pointing in multiple different directions, here. Seems I over-simplified them.That's one specific failure by a project developer that should have known better, not a general rule. In fact, I think the project in question may have lost the ability to write their own applications (staff turnover), and bought in a replacement app not originally designed to run under BOINC (they're now using the wrapper). 1) The licence. It needed renewing periodically, and the project failed to do that in a timely fashion. Buy a diary! But more seriously, it shows that they were using somebody else's software, and - presumably - paying them money for the privilege. Their software isn't home-written by the project. There are different ways of licensing external software - source code, precompiled library - and they may have switched models. I don't know. 2) Can't switch devices. We looked into that. The new Science app compiles CUDA code at startup, for the specific device it's running on. We could work round that, by forcing re-compile at restart. But the new app also bakes the device name into the checkpoint (restart data) file: that's the blocker. 3) Using the wrapper. Nothing to do with science and devices - the wrapper handles the communications between the science app and the BOINC client. Previously, this was done by linking the communications tool - the BOINC API library - into the science app at compile time, and calling the API functions directly when needed. That requires in-house code-level knowledge of BOINC (previously held by MJH, but I think he's left the project). The wrapper is a kludge for people who don't have access to the application source code, or who can't/won't learn how to program the API. Those are the separate straws in the wind. I hope that clarifies the separate indicators that led me to my conclusion. |
|
1445)
Message boards :
Questions and problems :
Not downloading tasks
(Message 95104)
Posted 15 Jan 2020 by Richard Haselgrove Post: Who's asking about which project here? |
|
1446)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 95060)
Posted 14 Jan 2020 by Richard Haselgrove Post: It is 4:10 pm here in the Southern Midwest ... it's time to set some hair on fire while outside ...Make sure you don't set fire to the whole of the South-West while you're out there... |
|
1447)
Message boards :
Questions and problems :
Computers signed up to project via Science United
(Message 95045)
Posted 14 Jan 2020 by Richard Haselgrove Post: Well, the CPDN task has been running for an hour and a half so far (out of an estimated 12 days). You'd have thought it would have noticed by now if the libs were missing. |
|
1448)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95034)
Posted 14 Jan 2020 by Richard Haselgrove Post: Why do I have to plug them into 16x slots if they are only 1x plugs?Although all the power pins are concentrated in the short section before the key, and are common to all sizes, the motherboard manufacturer is at liberty only to provide enough power tracks between PSU and slot to provide the PCI spec power for that size of slot. And if they are at liberty, they will have saved money by exercising that liberty. Copper is expensive. |
|
1449)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95027)
Posted 14 Jan 2020 by Richard Haselgrove Post: Try https://en.wikipedia.org/wiki/PCI_Express#PowerExactly which AMD GPU cards are you trying to use? Do they have sockets for supplementary power input? |
|
1450)
Message boards :
Questions and problems :
Why doesn't Boinc schedule earlier deadlines first?
(Message 95021)
Posted 14 Jan 2020 by Richard Haselgrove Post: This interests me. Surely there is only one correct answer to a mathematical calculation!! How on earth could different results be produced?Talk to Eric Mcintosh at CERN: IEEE 754 as intended, especially 'ETH Paper on Bit-Reproducible Portable HPC Applications'. |
|
1451)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 95019)
Posted 14 Jan 2020 by Richard Haselgrove Post: Exactly which AMD GPU cards are you trying to use? Do they have sockets for supplementary power input? If they don't have sockets, they'll take all the power they need - up to 75 watts - from the motherboard via the 16-bit slots. A 1-bit slot won't supply enough power, and you'll have to check that your riser can carry enough power. |
|
1452)
Message boards :
Questions and problems :
Computers signed up to project via Science United
(Message 95014)
Posted 14 Jan 2020 by Richard Haselgrove Post: Having started this hare running (Andy says he'll raise it on the Projects call), I thought I'd better run some CPDN - now I have a Linux box to run it on. First question, what version of Ubuntu do I have? No matter how nicely I ask it, all it will say is "Linux Mint 19.1 Tessa". Didn't like the curses6 incantation for Ubuntu 19.1, but seems to have accepted curses5 for Ubuntu 18.04 / 19.04 |
|
1453)
Message boards :
GPUs :
BOINC on new PC says "GPU Missing waiting to run"
(Message 95003)
Posted 14 Jan 2020 by Richard Haselgrove Post: In the Event Log, immediately below your login, but above the processor line you've shown us already, there should be one or more lines about GPUs. It might be 'No usable GPUs found', or it might show the details of the GPUs it has found. Which have you got? |
|
1454)
Message boards :
Questions and problems :
Computers signed up to project via Science United
(Message 94998)
Posted 14 Jan 2020 by Richard Haselgrove Post: Andy has created #3420, saying "In climateprediction.net we now have account creation disabled through the BOINC Manager". Dies anyone know - either way - whether this has fixed the SU problem, or can SU still create accounts? [Edit - SU did attach me to CPDN when I ran a test on 25 Nov 2018, so that would come into the problem category. I've got it deselected in preferences.] |
|
1455)
Message boards :
Questions and problems :
Computers signed up to project via Science United
(Message 94997)
Posted 14 Jan 2020 by Richard Haselgrove Post: Or some lightweight probe app that tests for 32-bit libs, and sends back a 'blacklist' message if none found - all before committing a substantial model download. |
|
1456)
Message boards :
GPUs :
PCI express risers to use multiple GPUs on one motherboard - not detecting card?
(Message 94995)
Posted 14 Jan 2020 by Richard Haselgrove Post: I did a quick search on PCIe electrical compliance testing. The PCI Special Interest Group seems to be the place to go. I'm not getting anything about the sort of cables which might be used to make a riser. And if there's no standard, then there's no chance any manufacturer will be building to standard. Timing, crosstalk, waveform all come to mind as imponderables at the sort of speeds we're talking. |
|
1457)
Message boards :
Questions and problems :
Computers signed up to project via Science United
(Message 94985)
Posted 14 Jan 2020 by Richard Haselgrove Post: What I didn't see on their fora was much in the way of answers.I think it's about time that we declared that the Emperor of SU has no clothes. |
|
1458)
Message boards :
Questions and problems :
Computers signed up to project via Science United
(Message 94979)
Posted 14 Jan 2020 by Richard Haselgrove Post: I emailed him about this before posting here. :)And I saw that before replying here :) The point is that I think we ought to think about this as a broader, BOINC-wide, problem with Science United. Did you see the user at SU who asked "How do I accept a Friend request at a project?" [that I can't log in to]. No effective reply. David is very proud that SU processes a million tasks a month. SETI does that three times a day. But people like Andy (nice chap, met him several times at workshops) tend only to think INside the box, in Andy's case CPDN. I'm busy trying to sort out a cross-project problem with Anonymous Platform, which has existed and been talked about in public and private for 9 months. But everybody regarded it as a private, project, problem, not important for my particular project - until it blew up spectacularly at SETI, just before Christmas. We need to join the dots, better. I can ride shotgun if he'll only broach the subject, but I think he should be the one to introduce it. |
|
1459)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94976)
Posted 14 Jan 2020 by Richard Haselgrove Post: Depressing ...I think the most depressing thing is probably that Eric gave in to pester-power and increased the per-device limits. |
|
1460)
Message boards :
Questions and problems :
Computers signed up to project via Science United
(Message 94968)
Posted 14 Jan 2020 by Richard Haselgrove Post: Andy Bowery is due to speak at the BOINC PROJECTS CALL #6, a week tomorrow. I'd like to hear him raise problems like this (including the inability of Science United users to access project preferences, or to contribute to project message boards) in front of the wider BOINC project community. |
|
1461)
Message boards :
Questions and problems :
Why doesn't Boinc schedule earlier deadlines first?
(Message 94963)
Posted 14 Jan 2020 by Richard Haselgrove Post: Keith's comment probably related to one specific application on one specific project that he and I both run. That's one specific failure by a project developer that should have known better, not a general rule. In fact, I think the project in question may have lost the ability to write their own applications (staff turnover), and bought in a replacement app not originally designed to run under BOINC (they're now using the wrapper).Keith's comment about applications trashing tasks if they start on the "wrong" processor ...How common are computers with differing GPUs? If there's quite a few, perhaps Boinc should try to make them run on the same one again? Using multiple GPUs is quite common amongst enthusiasts, and they can make it work well. This project's problem is unusual - it's the first time I can remember it happening in 10 years, and the only part which is BOINC's fault is the failure to predict bad programming. |
|
1462)
Message boards :
Questions and problems :
Why doesn't Boinc schedule earlier deadlines first?
(Message 94947)
Posted 13 Jan 2020 by Richard Haselgrove Post: That algorithm won't help for computer shut down, phone unplugged from charger, exclusive application running.No, the human factor can never be predicted, but it's pretty good for the events under BOINC's control. There's some good design in there, if you can be bothered to look. |
|
1463)
Message boards :
Questions and problems :
Why doesn't Boinc schedule earlier deadlines first?
(Message 94945)
Posted 13 Jan 2020 by Richard Haselgrove Post: That's a terrible system which means we're all wasting huge amounts of processing power if we run more than one project, as every time a processor is swapped from project A to project B, project A is likely to lose calculations. Data should ALWAYS be written to disk when an application is paused for whatever reason (computer shut down, phone unplugged from charger, exclusive application running, another project taking the processor).I think the algorithm is 1) Wait until Task Switch Interval has expired. 2) Then start looking for a good time to switch. 3) Wait until task has just checkpointed. 4) SWITCH You can set debug message log flags that will show you all that happening. |
|
1464)
Message boards :
Questions and problems :
Not downloading tasks
(Message 94930)
Posted 13 Jan 2020 by Richard Haselgrove Post: I think that's covered by the sched_op in the same log: 13-Jan-2020 19:34:40 [LHC@home] [sched_op] Starting scheduler request 13-Jan-2020 19:34:40 [LHC@home] Sending scheduler request: To fetch work. 13-Jan-2020 19:34:40 [LHC@home] Requesting new tasks for CPU 13-Jan-2020 19:34:40 [LHC@home] [sched_op] CPU work request: 51840.00 seconds; 1.00 devices 13-Jan-2020 19:34:43 [LHC@home] Scheduler request completed: got 0 new tasks 13-Jan-2020 19:34:43 [LHC@home] [sched_op] Server version 715 13-Jan-2020 19:34:43 [LHC@home] No tasks sent 13-Jan-2020 19:34:43 [LHC@home] Project requested delay of 6 seconds 13-Jan-2020 19:34:43 [LHC@home] [sched_op] Deferring communication for 00:00:06 13-Jan-2020 19:34:43 [LHC@home] [sched_op] Reason: requested by project all of which is normal. The only thing still missing is which sub-applications are allowed in preferences, and whether those sub-applications have work available at the moment - although the sched_reply doesn't say anything about no tasks being avaible [for xxx]. Edit - I get 13/01/2020 18:52:05 | LHC@home | Requesting new tasks for CPU 13/01/2020 18:52:05 | LHC@home | [sched_op] CPU work request: 1161.08 seconds; 2.00 devices 13/01/2020 18:52:05 | LHC@home | [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 13/01/2020 18:52:05 | LHC@home | [sched_op] Intel GPU work request: 0.00 seconds; 0.00 devices 13/01/2020 18:52:06 | LHC@home | Scheduler request completed: got 0 new tasks 13/01/2020 18:52:06 | LHC@home | [sched_op] Server version 715 13/01/2020 18:52:06 | LHC@home | No tasks sent 13/01/2020 18:52:06 | LHC@home | No tasks are available for SixTrack 13/01/2020 18:52:06 | LHC@home | No tasks are available for sixtracktest LHC server status confirms. The other three apps (at bottom) all require some sort of VM. |
|
1465)
Message boards :
Questions and problems :
Not downloading tasks
(Message 94923)
Posted 13 Jan 2020 by Richard Haselgrove Post: Both ALPN failures are followed by a successful connection 13-Jan-2020 19:34:42 [---] [http] [ID#0] Info: SSL certificate verify ok. 13-Jan-2020 19:34:42 [LHC@home] [http] [ID#1] Info: SSL certificate verify ok. using the Manchester certificate instead of the CERN certificate. I think we can check that off as 'working OK' (although it was worth checking). |
|
1466)
Message boards :
Questions and problems :
Why doesn't Boinc schedule earlier deadlines first?
(Message 94916)
Posted 13 Jan 2020 by Richard Haselgrove Post: You can also use the 'Switch between tasks every ...' control in the global computing preferences section of any project. Notes:
|
|
1467)
Message boards :
Questions and problems :
What does "run only the selected applications" mean?
(Message 94902)
Posted 12 Jan 2020 by Richard Haselgrove Post: I'm glad the UK is leaving.Ah, well - each to their own. Enjoy the ride. We'd better stop here, or Jord will get cross. |
|
1468)
Message boards :
Questions and problems :
What does "run only the selected applications" mean?
(Message 94900)
Posted 12 Jan 2020 by Richard Haselgrove Post: Oh my god, seriously?! The EU has forbidden SETI from asking us for donations? The EU needs to be shut down, now.SETI - unlike some Americans - is operating with an excess of politeness. It's been on their front page since the end of November: Because of our continued effort to be GDPR compliant, we're not sending emails to anyone with an EU/EAA email addresses, anyone who listed an EU/EAA country, or anyone who listed "International" as their country. And, of course, anyone who has opted out of our emails won't get them.What has the EU ever done for us? Provided a spam filter. I am Spartacus. |
|
1469)
Message boards :
Questions and problems :
What does "run only the selected applications" mean?
(Message 94898)
Posted 12 Jan 2020 by Richard Haselgrove Post: Checked a couple of other projects, and it seems as if the Is it OK for [project] and your team (if any) to email you?formulation is part of the standard BOINC website template. |
|
1470)
Message boards :
Questions and problems :
What does "run only the selected applications" mean?
(Message 94896)
Posted 12 Jan 2020 by Richard Haselgrove Post: At SETI, I see: Is it OK for SETI@home and your team (if any) to email you? It's in https://setiathome.berkeley.edu/prefs.php?subset=project. There isn't an actual 'newsletter', as such, but they do email many users once a year with a fundraising request (*), usually with a progress report. It looks as if Einstein were preparing to do the same in https://einsteinathome.org/account/prefs/subscriptions, but never got round to it. * not to users currently in the EU and subject to the GDPR. |
|
1471)
Message boards :
Questions and problems :
Memory limit ignored?
(Message 94868)
Posted 12 Jan 2020 by Richard Haselgrove Post: The project "declares" how much CPU resource they think (often wild guess) a GPU application requires.It's even worse than that. The relevant configuration value is <cpu_frac>x</cpu_frac> - default 0.1, and it's optional. Remember that BOINC projects are operated by scientists, and they have three priorities - science, science, and science. Most of them have little interest in how BOINC works for us volunteers, and even less knowledge. (I once made a statement like that during a conference call, and received a hearty endorsement from the other end of the line. It came from one of the most experienced and involved project administrators of all - Eric Korpela of SETI) Consider that '<cpu_frac>' again - "the fraction of total FLOPs that are done by the CPU". What's the speed difference between those devices? Often more that 1000 to 1, these days - aggravated by GPU speed being recorded at Peak, but the CPU speed measured by benchmark (unoptimised). In general, the <cpu_frac> you see in the BOINC client is arbitrary, and has zero connection with the real world - it has no relationship with the programming choices (OpenCL or CUDA) made in developing the app. I have an open request to review the situation: Scheduler: estimate GPU app needs better. In just over a year, that request has attracted precisely one comment, and that came from a volunteer familiar on these boards. The BOINC programming team is rather bogged down in internal restructuring at the moment: only small changes are being attempted, and this would be a big job. Only when the message changes from "they should fix this" to "WE are going to damn well fix this" will anything change. |
|
1472)
Message boards :
Questions and problems :
Not downloading tasks
(Message 94862)
Posted 11 Jan 2020 by Richard Haselgrove Post: I think something is wrong with this new version of BOINC, so I'm going to remove it and try to find something older.BOINC client version 7.9.3 (from your log) is quite old - around March 2018. It might be worth looking to see what other people at CPDN are using. |
|
1473)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94860)
Posted 11 Jan 2020 by Richard Haselgrove Post: The client detects each and every GPU in the system, but it only notifies the server about the 'best' one (plus a count of all the others) - look inside any sched_request file at each separate <coproc_xxx> section.Perhaps it could be set to only send tasks out if the client had > x GB of RAM? Or could identify what GPU(s) are on the client? I think both these pieces of data are available to the project server.If that special app is so much faster, ought Seti not incorporate that programming into the mainstream ones? They'd get huge amounts more work done.Partly because the extra speed comes at the expense of much higher memory usage: the special app can't be used on every GPU. Managing the distribution to compatible cards only is not handled easily or well under the BOINC framework, especially when two (or more) dissimilar cards are installed in the same computer. That decision was made in 2008, and by 2014 it was acknowledged as a mistake - but it would be a hugely complex bit of programming to report and store details of each GPU separately, whilst maintaining compatibility. It hasn't been attempted, and it is unlikely to be in the near future. |
|
1474)
Message boards :
Promotion :
Dreamlab works on IOS, BOINC could take the same route to work on Apple mobile devices - BOINC devs/management please read!!
(Message 94854)
Posted 11 Jan 2020 by Richard Haselgrove Post: There's a possible glimmer of hope in this area. A new volunteer developer, Angelo Koutsogiannis, participated in the contributor call this week, and from what he said has significant practical experience of negotiating with Apple to overcome problems in gaining access to the IOS app store, on behalf of an employer with significant activity in this field. Angelo spoke at the July 2019 BOINC workshop: there's a link to a video recording of his talk under 'Volunteers' on that page. It'll take time for him to get up to speed with the specific BOINC requirements, but he should be able to piggyback on the renewed Android development activity recently. |
|
1475)
Message boards :
BOINC Manager :
Manual suspend keeps getting reset
(Message 94851)
Posted 11 Jan 2020 by Richard Haselgrove Post: Ah, but under the Activity tab on the BOINC GUI there is a Suspend option, which does exactly what the OP says (and has caught me out a couple of times - behaves as one would expect Snooze to)That's a bug, then. See the semantics under the Boinccmd tool: --set_run_mode {always | auto | never} [ duration ]and the same for GPUs and networking. We use that in, for example, the SETI bench test suite - stop normal production while a test is being run, and restart it at the end. BOINC Manager uses the same control tools, although the nuts and bolts are hidden under the hood. When I get some spare time, I'll look in the code to see how those menu entries are hooked into the GUI RPC procedure calls. |
|
1476)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94849)
Posted 11 Jan 2020 by Richard Haselgrove Post: If that special app is so much faster, ought Seti not incorporate that programming into the mainstream ones? They'd get huge amounts more work done.Partly because the extra speed comes at the expense of much higher memory usage: the special app can't be used on every GPU. Managing the distribution to compatible cards only is not handled easily or well under the BOINC framework, especially when two (or more) dissimilar cards are installed in the same computer. |
|
1477)
Message boards :
BOINC Manager :
Manual suspend keeps getting reset
(Message 94835)
Posted 10 Jan 2020 by Richard Haselgrove Post: Suspend is different from snooze. Suspend - which you do from the main BOINC Manager screen, for either projects or single tasks - is supposed to be permanent. Snooze - which you do from the notification area icon in Windows - is supposed to be for one hour only. |
|
1478)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94812)
Posted 10 Jan 2020 by Richard Haselgrove Post: Tonight's the night for a breakdown and mass howling at the moon... January Full Moon 2020: The 'Wolf Moon' Arrives with a Lunar Eclipse https://www.timeanddate.com/astronomy/moon/wolf.html |
|
1479)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 94805)
Posted 9 Jan 2020 by Richard Haselgrove Post: Shit, what a day. I think I'm coming down off the wall now. |
|
1480)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94787)
Posted 8 Jan 2020 by Richard Haselgrove Post: TDP = Total Dispersed Power - electrical input, in watts. The Intel GPUs run at full speed if there is a free core, same as other GPUs. They slow down 7-fold if all cores are busy: but you get that back with real-time priority (I use Process Lasso). That's enough for tonight, UK time :-) |
|
1481)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94785)
Posted 8 Jan 2020 by Richard Haselgrove Post: How come OpenCL was never written as well for NV?Marketing (monetization). They want to promote their proprietary technology - CUDA. I see no point in using Intel GPUs. Might aswell use the CPU as a CPU, it can't do both at once, one will slow down.Intel GPUs work fine if the CPU cores are doing low-power work (integer maths). You need to keep the entire die within TDP to avoid throttling. SEVENfold? How is that possible? Do you mean if your CPU is also doing other projects so the GPU tasks never get a chance?Probably voodoo, but it works. Partly because the app is kept in on-die cache memory, and the time overhead of a full context switch is enough to kill the performance of the GPU app. |
|
1482)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94782)
Posted 8 Jan 2020 by Richard Haselgrove Post: I'm told that the OpenCL runtime libraries on NVidia only support the primitive spin-wait synchronisation for all the parallel kernels on the GPU. That means that that the CPU is constantly 'active' (clocking up time), but it's doing nothing useful - just twiddling its thumbs, waiting for something to happen. The CUDA environment is far more sophisticated, with callback synchronisation available, but it's different. Projects prefer to code their science apps only once, so they prefer to program in OpenCL, since it supports all three GPUs - AMD (efficient runtime), NV (lousy runtime), and Intel_gpu (special runtime - needs little but fast). I run my Einstein Intel_GPU apps at real-time priority - sounds dangerous, but speeds them up seven-fold, with no downside even on 100% utilised CPUs, except briefly during task end swapover. |
|
1483)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94780)
Posted 8 Jan 2020 by Richard Haselgrove Post: I notice your CPU time is about the same as the task time. Mine is a lot lower. Have you checked your CPU isn't throttling the GPU?Peter has mentioned a RX560 GPU (AMD?): Keith a GTX 1070 Ti (NV). The two manufacturers have supplied very different programming and runtime support environments. CPU usage is one of the big differences. |
|
1484)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94763)
Posted 7 Jan 2020 by Richard Haselgrove Post: Unless they're all concentrating so hard that they haven't noticed how late it's getting?No, they're not. David and Eric are both working on Nebula. And I'm sacrificing some Glenlivet. |
|
1485)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94762)
Posted 7 Jan 2020 by Richard Haselgrove Post: Eric confirmed this morning (in a limited-distribution email) that Beta was down because of "muarae2 SSD problems" - muarae2 being the new upload server. They want to have Beta available to diagnose the Anonymous Platform problem - David can't be bothered to work with LHC. So that all makes sense, but it doesn't explain why Main is still down. Unless they're all concentrating so hard that they haven't noticed how late it's getting? |
|
1486)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94752)
Posted 7 Jan 2020 by Richard Haselgrove Post: The moon is shining brightly enough to light my way to the place of sacrifice - and will shortly do so. |
|
1487)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94750)
Posted 7 Jan 2020 by Richard Haselgrove Post: Very good Richard. I hope you get paid one way or another for all the debugging you do for SETI (and probably other projects too).Thanks for the kind words. Actually, I'm hopeless at the actual code (I never learned C++, and I can't disentangle it for toffee - nor do I want to). But I did undergo a proper "computer science" training - I understand the way computers think (follow logic) in general, without concentrating on any single programming language. That's what I bring to the party. |
|
1488)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94749)
Posted 7 Jan 2020 by Richard Haselgrove Post: Before I pester the obviously busy Einstein folk, can I get this straight? There is only one DCF (for Einstein as they use the older coding) which my client adjusts itself every time it runs a task. But it's based on an estimate of runtime given by the server. Therefore they could just change that estimate for one of the applications to make it closer? Or does it depend a lot on the GPU in use - somebody else might have completely different timings for the two applications?The project server doesn't actually calculate the runtime estimate you see in the BOINC Manager. The project admins say how much work needs to be done to complete the task: <rsc_fpops_est>17500000000000.000000</rsc_fpops_est> for BRP4 tasks <rsc_fpops_est>525000000000000.000000</rsc_fpops_est> for FGRPB1G tasks Those figures are intrinsic to the job itself, and will be the same for every computer. The server also keeps track of your computer's speed for each type of application: <flops>16474211923.891272</flops> for BRP4 <flops>134145015228.871060</flops> for FGRPB1G Dividing one by the other - in your client - gives (Floating Point Operations) divided by (Floating Point Operations per second), or seconds: about 1,062 seconds per task for BRP4, and 3,913 seconds per task for FGRPB1G on this machine. To which, a DCF of 4.5 is applied. All of which is completely fubar'd. That DCF is derived from FGRPB1G tasks, and the estimate is about right (just under 5 hours). But BRP4 is estimating 80 minutes, when they run in nearer 10 minutes. I'm not sure exactly where Einstein gets the different speed estimates for the two applications from: both are running on the same hardware. You may be able to see some of the workings next time I request new work, at https://einsteinathome.org/host/1001562/log. |
|
1489)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94747)
Posted 7 Jan 2020 by Richard Haselgrove Post: Well, they found out about that problem in May already but they obviously didn't fix it back then. Instead they seems to have been tangles up in some "vboxwrapper" stuff that they seem to think is the problem. At least according to these threads on their board:Thanks, that's very interesting - and useful. Note in particular message 38913 in the first thread, written by Nils Høimyr. Seconds after I wrote the post you quoted, I had an email from Nils, asking me to try reproducing the problem on their dev site, from cold (he sent me the invitation code for the 'no new members' server). After some struggle, I did exactly that, from first principles. I'll probably be speaking to Laurence (posted in the dev thread you linked) in a teleconference on Thursday: if they still need educating, they'll be educated. |
|
1490)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94740)
Posted 7 Jan 2020 by Richard Haselgrove Post: Another early outage, it's not too soon to speculate on what this bodes.My betting (spoiler alert - I'll probably jinx it) is on 'quick and easy' - no fancy extra work. The server guys from CERN are back from their holidays, and beginning to think about the Anonymous Platform problem: that one isn't going to be either quick or easy. |
|
1491)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94737)
Posted 7 Jan 2020 by Richard Haselgrove Post: DCF adjusted downwards by 10% of the error, per task, and upwards by the full amount immediately. The modern server-based runtime estimation requires of the order of 100 completed tasks to normalise, in either direction. |
|
1492)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94735)
Posted 7 Jan 2020 by Richard Haselgrove Post: And one extra snafu is that under WINE BOINC somehow gets a much lower number for the crunching capacity of your cpu so estimates on all projects will be a lot higher than reality.Under DCF, that wouldn't be a problem. DCF adjusts, relatively quickly, to the actual measured runtime of your real tasks - since CPDN isn't planning to use GPUs any time soon, you might ask them to turn off the '<dont_use_dcf/>' flag in their scheduler replies. (*) That cannot work on GPU projects, because there is no fixed ratio of the speeds of the CPU and the GPU in any given host. Mass-market manufacturers use relatively decent CPUs, but pair them with cheap and slow bottom-end GPUs. Enthusiast home builders use a basic CPU, but throw all their cash into the best GPU money can buy. * They'll have to hack the server code. It should be possible to set that via a Project configuration option, but it appears to be missing. |
|
1493)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94714)
Posted 5 Jan 2020 by Richard Haselgrove Post: And we're back up and running. My pending sacrifice will be replaced by a toast to Eric. |
|
1494)
Message boards :
Questions and problems :
Inaccurate "time left"
(Message 94706)
Posted 5 Jan 2020 by Richard Haselgrove Post: This is a well-known consequence of a decision taken by the Einstein project in 2010 (yes, ten years ago). Before 2010, the central BOINC code (and hence every project) kept track of task speed estimated locally, through a single variable called the Duration Correction Factor. But after GPU computing was introduced in 2008, it quickly became clear that the single-value approach couldn't cope with multiple applications and multiple device speeds - just as you are now reporting. Code for handling multiple DCF values was developed by a BOINC volunteer, but it was rejected centrally and replaced instead by server-based code incorporating a new system for calculating credit and for runtime estimation. Einstein decided not to adopt the new credit system, and in the process rejected the runtime estimation system too. They never bothered to develop an alternative. |
|
1495)
Message boards :
Questions and problems :
Why doesn't Boinc schedule earlier deadlines first?
(Message 94701)
Posted 5 Jan 2020 by Richard Haselgrove Post: BOINC is not designed for single project users. BOINC is designed to handle, as safely as possible, the complexities introduced by running multiple projects with deadlines ranging from hours to years, and with runtimes taking anything from seconds to months. If 'Earliest Deadline First' running became the standard, some forms of scientific research (in particular, climate research) would become impossible under BOINC - their work would keep getting deferred until it became too stale to be useful. That's why BOINC preferentially runs work in FIFO order - in the order in which it's allocated. There are safeguards in place which endeavour to prioritise tasks which are in danger of missing their individual deadline. |
|
1496)
Message boards :
BOINC client :
Support for Visual Studio versions newer than 2013?
(Message 94641)
Posted 1 Jan 2020 by Richard Haselgrove Post: The section of code you need for this purpose isn't the client code as a whole. It's one specific library, known as the 'API' (Application Programming Interface), which handles the communications between the application and the client (stop, start, how far, how long, etc.) Search the BOINC docs for API - I think it's designed to be built separately in the build tool of your choice, and linked to the science code in the final stage. It should be a relatively simple build with no user interface components, so you should probably expect to use the scientific compilers, rather than the heavyweight Visual Studio. |
|
1497)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94619)
Posted 31 Dec 2019 by Richard Haselgrove Post: I suppose someone ought to go and fetch Juan back from the outrage pub, so he can go and get a proper hangover at midnight? |
|
1498)
Message boards :
Questions and problems :
problem setting up anonymous platform - need help
(Message 94585)
Posted 28 Dec 2019 by Richard Haselgrove Post: If you have existing stock work for the application you are trying to mimic, I find the easiest way is to look inside client_state.xml for the matching <app_version> segment, and copy the string values from that as you make up your new app_info.xml file. The syntax is not exactly the same, but once you get the general shape right, it falls into place. |
|
1499)
Message boards :
Questions and problems :
problem setting up anonymous platform - need help
(Message 94581)
Posted 28 Dec 2019 by Richard Haselgrove Post: The project name doesn't appear in app_info - it's inferred from which project folder you put the file in. There are application names and file names. They all appear at least twice, because you define them first, and then specify haw to use them. You really have to get a feel for the rhythm of how those components fit together. In general, every project will accept anonymous platform once you've pieced the jigsaw together. I've found two exceptions recently: 1) One of my projects has actively disabled anonymous platform: it posts an explicit message in the Event Log to say that it's done so. 2) Shortly before Christmas, it appeared that the most recent version of the server code has a bug which causes an internal server error when you try to fetch work with an otherwise working app_info.xml file. I doubt this affects MilkyWay, and it should be fixed soon after everyone gets back from their hols. If you're still having problems getting to grips with it, I suggest you do as Ageless suggested - post your attempt so far, and we can proofread it. |
|
1500)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94532)
Posted 24 Dec 2019 by Richard Haselgrove Post: Well, my trip to the pub doesn't seem to have worked. I'll have to see what I've got in the cellar. Let's hope they're actually testing the reinstatement this time before they switch it on... |
|
1501)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94514)
Posted 24 Dec 2019 by Richard Haselgrove Post: Well, let's hope that's the backup done. Now to put it all back together again... Fingers crossed. |
|
1502)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94490)
Posted 24 Dec 2019 by Richard Haselgrove Post: Well, they've started early again. I wonder how long it'll take them to revert the database? |
|
1503)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94468)
Posted 22 Dec 2019 by Richard Haselgrove Post: Well, going to the pub has made me feel better in the interval between work and bed :-) |
|
1504)
Message boards :
Questions and problems :
Unable to revert to "legacy" driver
(Message 94467)
Posted 22 Dec 2019 by Richard Haselgrove Post: At the "normal" download site, https://www.nvidia.com/Download/index.aspx the only types listed are GRD and Studio.On that page, scroll down to "Beta and Older Drivers". That's a link, although it doesn't look like it. Click through, and you have a better search tool. |
|
1505)
Message boards :
Questions and problems :
Unable to revert to "legacy" driver
(Message 94461)
Posted 22 Dec 2019 by Richard Haselgrove Post: If you do a full search for Beta / older drivers at NVidia.com, and specify Windows 10, you get a new search term box for "Windows Driver Type:", with a choice of Standard or DCH. Driver 431.60 is available for download in DCH format. |
|
1506)
Message boards :
Questions and problems :
New and n00bish
(Message 94458)
Posted 22 Dec 2019 by Richard Haselgrove Post: Argh - CORRECTION. My mind was elsewhere. Don't choose any driver later than 431.60 - it's the 436 series that introduced the problem. |
|
1507)
Message boards :
Questions and problems :
New and n00bish
(Message 94457)
Posted 22 Dec 2019 by Richard Haselgrove Post: 2019-12-22 4:13:03 AM | | CUDA: NVIDIA GPU 0: GeForce GTX 960 (driver version 361.43, CUDA version 8.0, compute capability 5.2, 2048MB, 1636MB available, 2412 GFLOPS peak)One specific point from that: you are using a relatively old driver for your graphics card. That, in itself, is GOOD: scientific computing doesn't, in general, need the constant updates that gaming does. But you are missing a component often used by BOINC projects: there should be a second, similar, line giving details for an 'OpenCL' driver for your GTX 960. Go to nvidia.com (or your local country equivalent), and download one of NVidia's own drivers: that will have all components included. The very latest drivers have a technical bug which causes problems at some projects. Don't select anything newer than 436.48 - and when you install it, choose the installation which excludes 'GeForce Experience', and choose a 'custom install' and exclude other optional components. All unless you know you need a specific tool for your other, non-BOINC, uses of the machine. |
|
1508)
Message boards :
Questions and problems :
Error 1327
(Message 94455)
Posted 22 Dec 2019 by Richard Haselgrove Post: This is likely to be an error resulting from the Windows installer's memory of where the the last version of BOINC was installed to. (You did say you were running Windows, didn't you?) Use the Windows Registry Editor 'regedit' to navigate to [HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup] and remove any references to drive D: There'll probably be two: "INSTALLDIR"="D:\\BOINC\\" "DATADIR"="D:\\BOINCdata\\" |
|
1509)
Message boards :
GPUs :
no usable GPUs found fedora 31 workstation
(Message 94397)
Posted 19 Dec 2019 by Richard Haselgrove Post: Do you have either of stderrgpudetect.txt stdoutgpudetect.txt in your boinc-client directory? Do they help? |
|
1510)
Message boards :
Questions and problems :
RIP Android users?
(Message 94373)
Posted 18 Dec 2019 by Richard Haselgrove Post: Thanks. Confirms 32-bit (sideways - won't switch from portrait to landscape). So armv7 is right - any thoughts on pie? |
|
1511)
Message boards :
Questions and problems :
RIP Android users?
(Message 94371)
Posted 18 Dec 2019 by Richard Haselgrove Post: Well, I've searched the tablet itself, developer mode tools, the shop blurb, the Samsung UK blurb, ... none of it says 32-bit. But none of it says 64-bit, either. Bother. OK, what do we know about pie / nopie, then? That's what the error message said - nothing about library bitness. Edit - the Einstein server reports 2019-12-18 17:46:40.8721 [PID=23677] [version] Best version of app einsteinbinary_BRP4 is 1.46 ID 643 NEONPIE (3.05 GFLOPS), and the successful (valid) tasks report Application:Binary Radio Pulsar Search (Arecibo) v1.46 (NEONPIE)SETI has a v8.00 (armv7-neon-nopie) and a v8.00 (armv7-neon). Why did it send me v8.00 (armv7-vfpv3d16-nopie)? |
|
1512)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94367)
Posted 18 Dec 2019 by Richard Haselgrove Post: I think all that happens locally, not on the server. As I see it, the BOINC client (which does know about the different GPUs installed) just picks 'first free card', and tells the project's app to get on with its task on that card. For GPUGrid, the first process is to compile the application source code into a format suitable for that card - that's the point at which the s_52, s_60 etc come into the picture. [I haven't looked at all the files to find the raw source code yet, but I might. The compiler output is an intermediate p-code similar to assembly language, which is why we can read it: there'll be a later interpreter stage where the p-code is rendered into binary op-codes] |
|
1513)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94365)
Posted 18 Dec 2019 by Richard Haselgrove Post: Sorry, I was unclear. I was numbering the tasks as they appear in an unsorted FIFO list. We work on tasks 1, then 2, then 3. When task 1 is reported as complete, it is removed from the list, and everything below moves up one line. Task 2 becomes the new task 1, task 3 becomes the new task 2, and so on. Those new list positions are the only ones known to BOINC after the restart.I think it's more likeI do not follow this. I would think that if task 1 completed then it is gone and done with and d1 would get task 3,4,5 etc Question: How does the project know there is more than one type of GPU? If the scheduler request identifies what is available then that accounts for different classes being sent. In that case the GPU identification could be "faked" to indicate that all the GPU were lower class and all would get s_52 instead of a mix. That could easily be done as the SETI people already fake the number of GPUs and all that is necessary is force all the identities to be the weaker GPU. just a guess and it would only work if the checkpoint file contains science data only and not unique gpu parameters.No: the server does not know that you have GPUs of different specifications. Look in the file 'sched_request_www.gpugrid.net.xml' - that's the only way that GPUGrid (or any project) gets information about our machines. The file contains all the data you see on the website (and then some), but it goes on to say <coproc_cuda> <count>2</count> <name>GeForce GTX 970</name>- and that's from one of my GTX 970 + GTX 750 Ti combos. There's no reference to the lesser card at all. |
|
1514)
Message boards :
Questions and problems :
RIP Android users?
(Message 94364)
Posted 18 Dec 2019 by Richard Haselgrove Post: You probably have ...Don't think so - I think I'm savvy enough to watch out for things like that. The error messages I got at SETI were all of the form "../../projects/setiathome.berkeley.edu/setiathome_8.00_arm-android-linux-gnu__armv7-vfpv3d16-nopie": error: Android 5.0 and later only support position-independent executables (-fPIE).I didn't hang around long enough to see if the server would eventually send me any of the pie apps they have installed. Ideally, the server plan class spec should know already that Android 9 excludes non-pie apps. |
|
1515)
Message boards :
Questions and problems :
RIP Android users?
(Message 94354)
Posted 18 Dec 2019 by Richard Haselgrove Post: Because v7.16.3 is still in testing, still has bugs, and will be replaced soon by at least v7.16.4 But it works better than yours appears to work currently, so it might be worth a try. I'm told the play store has a beta access route, but a quick Google / APK search worked better for my needs. |
|
1516)
Message boards :
Questions and problems :
RIP Android users?
(Message 94352)
Posted 18 Dec 2019 by Richard Haselgrove Post: I got my v7.16.3 from APKMirror - I'm sure other sources are available. Are you sure your phone's Android version is fully up-to-date? There have been developments since the days of v7.4.53, such as more projects requiring support for SSL (https). |
|
1517)
Message boards :
Questions and problems :
RIP Android users?
(Message 94350)
Posted 18 Dec 2019 by Richard Haselgrove Post: I bought an Android 9 tablet earlier this month, so I could join in the testing. I logged in to both SETI and Einstein with no problem at all (didn't try WCG). SETI's applications haven't kept pace with new developments in the Android landscape, and all the tasks I've been sent have failed. But Einstein has been sending tasks, and the tablet has been returning them, consistently. Specs are at Einstein host 12798421 Edit - when you say 'latest google play boinc version', can you put a number on that? You'll see from the link above that I'm running the test version 7.16.3 |
|
1518)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94347)
Posted 18 Dec 2019 by Richard Haselgrove Post: boinc looks at priorities to decide which to run first: the short tasks have short deadlinesI'm not sure that is the cause "d0" starts working on data-2 in slot-1 whichbut that's certainly the mechanism that causes the error. I think it's more like d0 starts working on task 1 d1 starts working on task 2 d0 finishes task, starts working on task 3 task 1 reports, and everything shifts up one: d1 is working on task 1, d0 is working on task 2 restart d0 is allocated to the new task 1 d1 is allocated to task 2 - which is a swap from before the restart. I think it's a simple 'first come, first served' for each device, each task. |
|
1519)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94332)
Posted 17 Dec 2019 by Richard Haselgrove Post: Why wait till then? I reckon we've reached beer o'clock, local time, so I'm off to find a temperate beer and a real pub fire. I may be some time... |
|
1520)
Message boards :
Questions and problems :
Not sending work: Last request too recent - what does this mean?
(Message 94324)
Posted 17 Dec 2019 by Richard Haselgrove Post: According to the repository, those words were added to the server code 11 years ago. The message comes from the server, so downloading a new client won't change it: and knowing how often some projects update their servers, we'll probably go on seeing it for another 11 years. Oooh, we could have a birthday party: "... committed on Dec 22, 2008" Edit - it's even older than that. It first appeared as "last RPC too recent" - committed on Dec 23, 2003. Put back the birthday party until Monday! |
|
1521)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94314)
Posted 17 Dec 2019 by Richard Haselgrove Post: Sadly, no go. Found an error message: ERROR: src\mdsim\context.cpp line 322: Cannot use a restart file on a different device!so I'll have to put all those backup files back in place. Ugh. Well, it's running. File restart.chk is generated at the first checkpoint, and the only readable bits are: 25000 GeForce GTX 970 OpenMM Binary Checkpoint- the rest is binary. (And there's a lot of it - 24,186,281 bytes. I don't think we'll be able to get much further.) |
|
1522)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94313)
Posted 17 Dec 2019 by Richard Haselgrove Post: Which I'll try to do next time I get allocated a task during working hours.Well, I got one - just in time for SETI maintenance, which is convenient. Having backed up all the files, taken a copy of client_state.xml, and disabled networking - let's see what we can do. First, I can confirm that running on a GTX 750 Ti, the compiler output files start // // Generated by NVIDIA NVVM Compiler // // Compiler Build ID: CL-26218862 // Cuda compilation tools, release 10.1, V10.1.168 // Based on LLVM 3.4svn // .version 6.4 .target sm_50 .address_size 64- so they are specific to the card in use. I'll run this to the first checkpoint (~15 minutes), and then see if it'll pick up on the GTX 970 if I delete those files (testing to see if they get re-generated automatically). |
|
1523)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94311)
Posted 17 Dec 2019 by Richard Haselgrove Post: Early start to maintenance - not surprising after the server problems this morning. I predict a long one. |
|
1524)
Message boards :
Questions and problems :
How much does RAM speed play a role?
(Message 94305)
Posted 17 Dec 2019 by Richard Haselgrove Post: As with so many of your questions, 'it depends on the project'. Does their science - and their application - need large amounts of memory, and is it constantly swapping data in and out of memory? |
|
1525)
Message boards :
Questions and problems :
Not sending work: Last request too recent - what does this mean?
(Message 94296)
Posted 16 Dec 2019 by Richard Haselgrove Post: If using manager under status it says communication deferred and counts down to when you can request work again. under the projects tab.I get caught out by that sometimes when using a local Manager to view a remote client. If the clocks on the two machines are out of synchronisation (and most drift slightly over the course of a week), the clock difference is applied to the countdown timer. |
|
1526)
Message boards :
Questions and problems :
Not sending work: Last request too recent - what does this mean?
(Message 94290)
Posted 16 Dec 2019 by Richard Haselgrove Post: As Les says. The calculation of 'too recent' is done on project servers, so will only change as projects update their server code. Every project also sends a message saying how long you should wait between contacts, but at the moment you can only see it if you enable <sched_op_debug> Event Log messaging. I've got a request in at the moment to change both of those: so far, two separate programmers have misunderstood 'asking will reset the timer', so maybe third time lucky. |
|
1527)
Message boards :
GPUs :
BOINC stopped recognizing GPU
(Message 94274)
Posted 15 Dec 2019 by Richard Haselgrove Post: There's this post about it.I see I wrote 'testing of the new client should start either over this weekend, or during the week commencing 17 June 2019.' Well, testing did start (rather later than that), and there is a version 7.16.3 available for the strong of heart. But it isn't finished yet, and the latest information I heard this past week is that another round of testing may start around this next Tuesday, or 17 December 2019. |
|
1528)
Message boards :
BOINC Manager :
Where does boincmgr store the list of remote hosts on Linux?
(Message 94261)
Posted 14 Dec 2019 by Richard Haselgrove Post: They are stored in a hidden file in your personal home folder, called .BOINC Manager(note the initial '.') The section you are looking for is called '[ComputerMRU]'. |
|
1529)
Message boards :
Questions and problems :
Are Risc CPUs supported?
(Message 94254)
Posted 14 Dec 2019 by Richard Haselgrove Post: Apart from Android/ARM, you will probably have to compile (and possibly tweak) BOINC for your specific CPU / OS combination. A few of the projects provide the source code for their science applications, so you would be able to do the same there: but for most projects, you would need to persuade the developers to compile a special app for you. That's unlikely to be welcomed unless you're bringing a huge farm of identical CPUs online for them. If your RISC machine can run a VM, some of those problems go away, but a simple CPU emulating a complex one doesn't sound like a recipe for efficiency. |
|
1530)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94246)
Posted 13 Dec 2019 by Richard Haselgrove Post: There is just one app - acemd3.exe, which is invoked by wrapper_6.1_windows_x86_64.exe The wrapper defines how acemd3 is deployed, using job-win64.xml: <job_desc>
<task>
<application>acemd3.exe</application>
<command_line>--boinc input --device $GPU_DEVICE_NUM</command_line>
<setenv>TMPDIR=$PWD</setenv>
<setenv>TEMP=$PWD</setenv>
<setenv>TMP=$PWD</setenv>
<setenv>HOME=$PWD</setenv>
<setenv>ACEMD_PLUGIN_DIR=$PWD</setenv>
<setenv>SystemRoot=C:\Windows</setenv>
<setenv>ComSpec=C:\Windows\system32\cmd.exe</setenv>
<stdout_filename>progress.log</stdout_filename>
<checkpoint_filename>restart.chk</checkpoint_filename>
<fraction_done_filename>progress</fraction_done_filename>
</task>
</job_desc>(that's the one which is missing the 'priority' line)That also shows how acemd3 gets the device number from init_data.xml (via the wrapper setting $GPU_DEVICE_NUM) That much is certain. Now for the speculation. There are also a lot of DLLs ('Dynamic Link Libraries', under Windows). An interesting one is 'OpenMMCudaCompiler.dll': just at the moment, I don't have a GPUGrid task running, so I can't identify the source code that the compiler is going to work on. It may be downloaded with the task. My guess is that when the task starts, OpenMMCudaCompiler.dll compiles the source code to suit the GPU initially specified by BOINC. On restart, the compiled code is found, or assumed, to be already present - but if the device has changed, the old code is no longer appropriate for the new device. We need to find a way of forcing re-compilation to suit the new device. Which I'll try to do next time I get allocated a task during working hours. |
|
1531)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94243)
Posted 13 Dec 2019 by Richard Haselgrove Post: No slot contents got moved during the course of this conversation. It's the device that matters, not the storage location. |
|
1532)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94240)
Posted 13 Dec 2019 by Richard Haselgrove Post: The problem is that ACEMD spits out so many blasted files - I was trying to find the right ones. To me, a "checkpoint" file is written out (or added to) as the science progresses. I'd expect that device hardware enumeration would take place only once, at the start of the run - and the most likely candidate is the compilation stage. If we can prove that, we have something to offer the admins. |
|
1533)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94237)
Posted 13 Dec 2019 by Richard Haselgrove Post: Now, here's a thought. All those hexadecimal hash files in the slot directory are actually plain text content, and they start // // Generated by NVIDIA NVVM Compiler // // Compiler Build ID: CL-26218862 // Cuda compilation tools, release 10.1, V10.1.168 // Based on LLVM 3.4svn // .version 6.4 .target sm_52 .address_size 64Just as a test, we could try deleting those for a paused task. My guess is that the app will re-compile them if it finds they're missing. And if the .target sm_52 is different on a different device, the binary compiler output might be different, and might run on the new hardware. Worth a punt? (edit - that target value is on my GTX 970. Is yours different, for a different card?) |
|
1534)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94235)
Posted 13 Dec 2019 by Richard Haselgrove Post: OK, so the real problem seems to be that GPUGrid's 'New version of ACEMD' app evaluates the hardware it's running on when it first starts, and then remembers it. If restart hardware doesn't match the original evaluation, it crashes. That means their new application is "not fit for BOINC". We need to convince them that the hardware evaluation has to be re-done from scratch when resuming from a pause, so that computation can continue. Is that a fair form of words? If so, we have to work out whether they "don't care", or "don't understand". I suspect it's the latter, for which the appropriate penalty is re-education. |
|
1535)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94233)
Posted 13 Dec 2019 by Richard Haselgrove Post: Going back to the very beginning of this conversation, you said I am guessing that when the system reboots boinc has lost track of which app was running in which slot and assign the first device it finds to the first suspended app.If you'd written "... boinc has lost track of which task was running on which device ..." I'd have agreed with you. BOINC doesn't lose track of which task's files are in which slot. |
|
1536)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94231)
Posted 13 Dec 2019 by Richard Haselgrove Post: Again, no. It's BOINC which manages the slot directories, not the project.Richard: I don't have a problem with anything you have written here. It is the project that has to manage the resume and it is not working correctly due to some designed fault on their part. I assume it is finding the wrong checkpoint files and that is causing the failure. BOINC tells the app which GPU to use. BOINC may say 'use device 0' when it first starts, and it may say 'use device 1' after a restart. But the files will be is slot 17 or whatever, both times. Both the initial data files, and the checkpoint files, will be in that slot the whole time - neither BOINC nor I moved them. The problem with GPUGrid is that their new app wants to run on the same model of card after a restart, and BOINC doesn't guarantee that: all it guarantees is that CUDA tasks will run on 'a' NVidia GPU - any NVidia GPU. The machine I'm pulling examples from has a GTX 970 and a GTX 750 Ti. The second card is too slow to use for GPUGrid, so I have <exclude_gpu>
<url>http://www.gpugrid.net/</url>
<device_num>1</device_num>
<type>NVIDIA</type>
</exclude_gpu>in cc_config.xml - that avoids the problem you're describing. GPUGrid runs on device zero, period.
|
|
1537)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94229)
Posted 13 Dec 2019 by Richard Haselgrove Post: Again, no. It's BOINC which manages the slot directories, not the project. When a task is first started, all the files are put the the first directory which BOINC determines is currently empty. In this case, it was slot 1: 13-Dec-2019 08:58:53 [GPUGRID] [cpu_sched] Starting task initial_1706-ELISA_GSN4V1-35-100-RND9825_0 using acemd3 version 210 (cuda101) in slot 1And everything stays there until the task finally fishes, at which point BOINC deletes everything so it's clean for the next task (which might be a completely different project). If there's no empty directory, BOINC creates a new one: 13-Dec-2019 14:25:18 [GPUGRID] task initial_1706-ELISA_GSN4V1-35-100-RND9825_0 suspended by user 13-Dec-2019 14:25:24 [SETI@home] [cpu_sched] Starting task 11dc19aa.13973.1294.14.41.146.vlar_2 using setiathome_v8 version 800 (opencl_nvidia_SoG) in slot 5BOINC couldn't re-use slot 1, because it was still full of the files copied at 08:58. After SETI had finished with slot 5, BOINC cleaned it up: Directory of D:\BOINCdata\slots\5
13/12/2019 14:39 <DIR> .
13/12/2019 14:39 <DIR> ..
0 File(s) 0 bytes
2 Dir(s) 940,707,196,928 bytes freeI should point you to the init_data.xml file in the GPUGrid slot (slot 1): <gpu_type>NVIDIA</gpu_type> <gpu_device_num>0</gpu_device_num> <gpu_opencl_dev_index>0</gpu_opencl_dev_index> <gpu_usage>1.000000</gpu_usage>and in my normal SETI working slot (slot 2): <gpu_type>NVIDIA</gpu_type> <gpu_device_num>1</gpu_device_num> <gpu_opencl_dev_index>1</gpu_opencl_dev_index> <gpu_usage>1.000000</gpu_usage> |
|
1538)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94227)
Posted 13 Dec 2019 by Richard Haselgrove Post: No, if it did that it would probably try to start a task from a different project.And the slot number is of no effect. It's just the first scratch folder which happened to be available when the task was first launched - which may have been hours earlier.Actually, I think that is the problem. on restart the app is assigned the first slot and gets the wrong checkpoint file. As it happens, I've got a GPUGrid task on this machine. The xml files as I start this reply are: 13/12/2019 14:19 511 boinc_task_state.xml 13/12/2019 08:58 14,807 init_data.xmlAfter suspending it and allowing SETI to take a turn on the GPU, they are 13/12/2019 14:25 511 boinc_task_state.xml 13/12/2019 14:39 14,815 init_data.xml'Task_state' has been updated, because I waited until the next checkpoint before pausing it: init_data was re-written on restart to reflect the new device number - although on this machine GPUGrid always runs on device 0, and this time ran in slot 1. 13/12/2019 14:39:08 | GPUGRID | [cpu_sched] Restarting task initial_1706-ELISA_GSN4V1-35-100-RND9825_0 using acemd3 version 210 (cuda101) in slot 1 |
|
1539)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94223)
Posted 13 Dec 2019 by Richard Haselgrove Post: Check the timestamps and other clues. --device n is an old way of doing things. With GPUGrid, you might be looking at the start instruction given by the wrapper to the science app. In turn, BOINC will have passed the device instruction to the wrapper in init_data.xml - which will probably have been re-written at the start of the 'resume' session. And the slot number is of no effect. It's just the first scratch folder which happened to be available when the task was first launched - which may have been hours earlier. |
|
1540)
Message boards :
The Lounge :
VirtualBox 6.1.0 released
(Message 94207)
Posted 11 Dec 2019 by Richard Haselgrove Post: Linux host: Drop PCI passthrough, the current code is too incomplete (cannot handle PCIe devices at all), i.e. not useful enoughThis is a significant (non-) feature for BOINC. No GPUs. |
|
1541)
Message boards :
Projects :
GPUgrid not always resuming tasks correctly
(Message 94206)
Posted 11 Dec 2019 by Richard Haselgrove Post: It will have written the normal checkpoint file, if the project has that capability. That enables calculations to resume from the point reached prior to the checkpoint. If the task is to be resumed, BOINC will resume it when 'a' GPU becomes available. BOINC will direct that the task runs on the free GPU - by a command line, or by an init_data.xml file, depending on the API version in use. BOINC does not wait until the previously used, or even an identical, GPU becomes available. |
|
1542)
Message boards :
BOINC client :
gpu ONLY tasks not executing on Boinc
(Message 94200)
Posted 11 Dec 2019 by Richard Haselgrove Post: The Manager might be the control interface, but it's the client which actually decides what to run. Similarly, for a GPU task, the control interface runs on the CPU, but despatches the heavy work (in tiny fragments) to the GPU. The developers decided on a purist approach: even GPU tasks do actually require the CPU to run, so 'GPU only' is an invalid specification. Out in the real world, we know what you mean, and many of us would like to be able to do what you want. In the meantime: don't accept new tasks from CPU-only projects, and switch off CPU jobs on the project website, where projects have both CPU and GPU applications. |
|
1543)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 94193)
Posted 11 Dec 2019 by Richard Haselgrove Post: Well, maybe I could get a nano-computer next Black Friday. But at the moment I'm a bit busy with the Android tablet I bought this time. |
|
1544)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 94191)
Posted 11 Dec 2019 by Richard Haselgrove Post: Your report was on Apr 16, 2018, nearly 20 months ago - I think more projects might have switched to SSL since then. Else, what is strange about The BOINC server? The BOINC SSL certificate? The 'get info' process/code? that makes that fail, but allows everything else? |
|
1545)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 94188)
Posted 11 Dec 2019 by Richard Haselgrove Post: What happens when a 64-bit client with <no_info_fetch> tries to contact a project over SSL? That fix might just push the problem a bit further down the road. |
|
1546)
Message boards :
GPUs :
Mali G31 support?
(Message 94184)
Posted 11 Dec 2019 by Richard Haselgrove Post: I'm teaching myself Android on a tablet with a Mali-G71 GPU. BOINC detects it just fine, and requests work for it, but no project I know of has written any science applications for it yet. |
|
1547)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94175)
Posted 10 Dec 2019 by Richard Haselgrove Post: Is this why we needed the extended caches? Edit - that did the trick. We're back. |
|
1548)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94161)
Posted 10 Dec 2019 by Richard Haselgrove Post: Down the hatch! |
|
1549)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 94158)
Posted 10 Dec 2019 by Richard Haselgrove Post: Pass it widdershins, of course. |
|
1550)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 94152)
Posted 10 Dec 2019 by Richard Haselgrove Post: NO NO NO - https://boinc.berkeley.edu/dl/boinc_7.14.2_windows_intelx86.exe |
|
1551)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 94148)
Posted 10 Dec 2019 by Richard Haselgrove Post: Thanks goodness the Alzheimer's hasn't kicked in yet. Have a look at message 85887. |
|
1552)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 94146)
Posted 10 Dec 2019 by Richard Haselgrove Post: Might be worth a try with a 32-bit BOINC while we look up 'Celeron(R) N4000 CPU'. |
|
1553)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 94144)
Posted 10 Dec 2019 by Richard Haselgrove Post: No local host connection = no boinc.exe running (at least as far as I can tell).Yup, that's a good takeaway. Conversely, boinc.exe has to be running in order for connection to be made. 10-Dec-2019 04:47:11 [---] Starting BOINC client version 7.14.2 for windows_x86_64So boinc has run, for just over two minutes. That behaviour can happen if the cooling system is completely broken (enough time to reach shutdown temperature), but it's unlikely in this case - (a) overheating usually crashes the whole computer, not just one app, and (b) with no science apps, there's not really much heat generated - not even benchmarks. @other helpers - didn't we have a problem with one of those low-power Intel chips failing to run 64-bit SSL routines? Might be worth a try with a 32-bit BOINC while we look up 'Celeron(R) N4000 CPU'. |
|
1554)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 94139)
Posted 10 Dec 2019 by Richard Haselgrove Post: If boinc.exe is now running, you will be able to access the Event Log. Copy the first 30 lines or so, including all the startup and identification lines, and paste them here - we can take a look. |
|
1555)
Message boards :
BOINC Manager :
Einstein@home results not showing in BAM
(Message 94103)
Posted 8 Dec 2019 by Richard Haselgrove Post: There is an explicit Do you consent to exporting your data to BOINC statistics aggregation Web sites?on https://einsteinathome.org/account/prefs/privacy when logged in. |
|
1556)
Message boards :
BOINC Manager :
10x average points per day??
(Message 94102)
Posted 8 Dec 2019 by Richard Haselgrove Post: Yes. there are some projects (which shall remain nameless) which give away points like candy - completely disregarding the formal definition of a Cobblestone (see section labelled Claimed and granted credit). That horse bolted a long time ago. More significantly, points are not awarded immediately upon the return of completed work. Sometimes they are, but at most projects you have to wait for 'validation' - where your work is compared for accuracy against another user. If you're unlucky, this can take weeks, or even months. |
|
1557)
Message boards :
BOINC Manager :
Einstein@home results not showing in BAM
(Message 94097)
Posted 8 Dec 2019 by Richard Haselgrove Post: It is normal, if you haven't given explicit GDPR permission for the project to export your stats. Visit Einstein's preferences page. |
|
1558)
Message boards :
GPUs :
Cuda or OpenCL?
(Message 94096)
Posted 8 Dec 2019 by Richard Haselgrove Post: It depends on the project. Some write their applications in CUDA, some write them in Open CL, some write them in both (different applications, obviously). BOINC handles both, so you'd be safest installing both. But I believe there's some doubt about OpenCL compatibility for some drivers at some projects... |
|
1559)
Message boards :
Questions and problems :
BOINC not seeing 3rd GPU
(Message 94092)
Posted 8 Dec 2019 by Richard Haselgrove Post: Make sure that your cc_config.xml file is called exactly that. Windows has a habit of calling it a text file, named cc_config.xml.txt, but with the final three letters hidden. Make sure that the file is described as an XML Document in File Explorer. Edit - or put it in the data directory :-) |
|
1560)
Message boards :
GPUs :
GPU 0 suspended and GPU 1 always computing
(Message 94085)
Posted 8 Dec 2019 by Richard Haselgrove Post: The 4 files in /etc/boinc-client/ are sym linked to the ones in /var/lib/boinc-client/ so while it appears you have one in each folder they are the same files. You can edit it in either folder.Bear with me - I'm still a newbie in Linux. Since I wrote that, I found that I can go through the symlink from /var/lib/boinc-client/ and edit the file as myself. If I edit the 'real' file in etc/boinc-client/, I have to use sudo. Given that cc_config.xml is an optional file, does anyone happen to know whether the file and the symlink are created 'empty' by the installer, or whether they appear after first use? |
|
1561)
Message boards :
GPUs :
GPU 0 suspended and GPU 1 always computing
(Message 94068)
Posted 7 Dec 2019 by Richard Haselgrove Post: I prefer the old method, as it's the same as Folding@Home.You are free to use whatever method you want, provided you check that it actually works the way you want it to. |
|
1562)
Message boards :
GPUs :
Can't start GPU 0 to compute after cc_config.xml modification
(Message 94056)
Posted 6 Dec 2019 by Richard Haselgrove Post: I've written it up as "Linux client: the OOBE for volunteers" - #3396 |
|
1563)
Message boards :
GPUs :
Can't start GPU 0 to compute after cc_config.xml modification
(Message 94038)
Posted 5 Dec 2019 by Richard Haselgrove Post: I originally got my v7.16.3 from LocutusOfBorg PPA, although I'm now experimenting with a bespoke version. The PPA installs the new version as a systemd service - hence my alternative stop|start commands - but otherwise you'll be familiar with it. Be aware that v7.16.3 is still officially an alpha test version. I think it works OK, but the debian package maintainers have decided to make their own changes from the central BOINC code. In particular, they are packaging for a specific target audience - workers in a secured, controlled, corporate environment, where the decision to deploy BOINC has been made at a corporate level. In particular, they assume that the user doesn't have either sudo or administrative rights to control the service - so if they stop it, they cannot restart it. Thus, they've removed the GUI control for shutting down the connected client, and a few others. That doesn't match my use case, and I suspect not yours either - I'm fighting behind the scenes towards a ensuring there will still be a distribution available with full controls for owner-users. It's a slow process. My primary knowledge is for Windows, and I've only started fiddling with Linux in the last 6 months - I haven't reached scripting yet. But I might give it a try. In return, if you don't mind, I'm thinking of writing up your case as an example of a perfectly reasonable set of user preferences which can't be accommodated in the current schema. |
|
1564)
Message boards :
News :
BOINC roadmap / wish list
(Message 94034)
Posted 5 Dec 2019 by Richard Haselgrove Post: It takes a certain amount of power to calculate a certain amount of work.No, it takes a certain amount of energy - which is the product of power and time. We can reduce the power by increasing the time - slowing the clock, using fewer components, pausing at intervals. If we want to accommodate users who want a quiet life (less fan noise), we have to tone down the 'need for speed'. |
|
1565)
Message boards :
GPUs :
Can't start GPU 0 to compute after cc_config.xml modification
(Message 94030)
Posted 5 Dec 2019 by Richard Haselgrove Post: OK, I've just tried it on my Ubuntu Mint / BOINC v7.16.3 box, which also has two NVidia GPUs. Using the cc_config.xml option we discussed this morning, adding the "ignore" line prevented one of the GPUs from running, and - more significantly - removing it allowed both GPUs to run again. I did it manually (stop client - edit and save file - start client), but it should be possible to script that. |
|
1566)
Message boards :
GPUs :
Can't start GPU 0 to compute after cc_config.xml modification
(Message 94029)
Posted 5 Dec 2019 by Richard Haselgrove Post: Sadly, what's wrong is the underlying design of this particular part of the BOINC configuration process. I'm not surprised that 'read config files' didn't clear the setting (that's part of the poor underlying design), but I am surprised that restarting the BOINC client didn't either. I need to have a think, and maybe do some reading and testing, about that one. Unless anyone else can chime in? Assuming we can clear this hurdle, I think the ultimate solution - at this stage of BOINC client configuration - will be a script (possibly under cron control, or launched manually - your choice), to stop the client swap between two different cc_config.xml files restart the client Again, unless somebody else knows better. |
|
1567)
Message boards :
News :
BOINC roadmap / wish list
(Message 94028)
Posted 5 Dec 2019 by Richard Haselgrove Post: Trixx, EVGA and MSI are all examples of what I was meaning by 'controlled externally'. I don't see how they could, even hypothetically, be integrated into a future BOINC client and controlled through a unified BOINC GUI. |
|
1568)
Message boards :
GPUs :
GPU 0 suspended and GPU 1 always computing
(Message 94024)
Posted 5 Dec 2019 by Richard Haselgrove Post: That's good - happy to help. Be aware that when BOINC v7.16.x is finally, fully, released, the restart command will change to sudo systemctl start boinc-client |
|
1569)
Message boards :
GPUs :
GPU 0 suspended and GPU 1 always computing
(Message 94021)
Posted 5 Dec 2019 by Richard Haselgrove Post: Refer to the User Manual. Create, or if you have one already, edit, a file called 'cc_config.xml'. The easiest way is probably to edit the logging options from BOINC Manager (Ctrl+Shift+F), which will create a complete template file for you to work from. You'll probable need an administrator/sudo editor to make changes. Then, add a block to the 'Options' section, as shown in the manual. Something like <ignore_nvidia_dev>N</ignore_nvidia_dev>I need to edit go downstairs to remind myself of the likely location of the new cc_config.xml file - I'll edit it in. The file on my Ubuntu system with BOINC v7.16.3 is in /etc/boinc-client |
|
1570)
Message boards :
News :
BOINC roadmap / wish list
(Message 94017)
Posted 5 Dec 2019 by Richard Haselgrove Post: Where are we going to actually "discuss" this? A lot is made of fan noise in the paper. I think we need to distinguish between CPU fans GPU fans Integrated laptop fans I think CPU fans are fairly well understood, and the paper talks about things like power state management. I think we're on track there. I suspect GPU fans are a big problem. So far as I know, the GPU manufacturers have only supplied software tools which generate apps which operate in 'fire and forget' mode: once a scientific application is launched, it grabs as many resources (shaders) as it can find, and runs them at a speed/power which can only be controlled externally, not by the program itself. In the spirit of the paper, BOINC needs to reach out to the GPU manufacturers and persuade them to implement internal power state management, similar to the CPU tools. And laptop fans are probably the biggest problem of all. They are physically closer to the user, and more intrusive if the user's foreground application has an audio component (movies or music). I think most user complaints are likely to come from this group, and I don't have any answers. |
|
1571)
Message boards :
Questions and problems :
optional arguments not being passed to boinc client in ubuntu
(Message 93989)
Posted 1 Dec 2019 by Richard Haselgrove Post: Fair enough. I've run several of Juan's mods - that's my main reason for stopping and starting BOINC. So systemctl works in that context, too. |
|
1572)
Message boards :
Questions and problems :
optional arguments not being passed to boinc client in ubuntu
(Message 93987)
Posted 1 Dec 2019 by Richard Haselgrove Post: A suggestion by Richard was to just use systemcrl to start boincAll I can say is that the systemctl call I posted 'works for me' on a recent Ubuntu-based installation: it starts and stops the client cleanly when I need to change things, and don't want to bother rebooting the whole machine. Why are you trying to pass arguments to the BOINC client, anyway? |
|
1573)
Message boards :
Questions and problems :
Boinc for Android computation error?
(Message 93973)
Posted 1 Dec 2019 by Richard Haselgrove Post: https://einsteinathome.org/it-it/host/12789879 shows a host at Einstein running Android 10, doing Arecibo tasks and finishing them without errors (https://einsteinathome.org/host/12789879/tasks/0/0).Unfortunately, the owner of that machine has posted I just upgraded on of my Android devices to Android 10 and it doesn't receive other WUs.I think the completed tasks must have been finished - or at least issued - before the Android version upgrade. The Einstein server is now rejecting requests for new work, with Couldn't match OS version '4.14.116 (Android 10)' with required regexp 'Android [987N65]|Android 4\.[1-9]|3\.4\.[0-9]' |
|
1574)
Message boards :
Questions and problems :
AVX 512 beneficial for Boinc projects?
(Message 93965)
Posted 30 Nov 2019 by Richard Haselgrove Post: I doubt any project scientist would do that, because absolute speed would be negated by all the caching, task swapping, and transmission overheads of a BOINC project. They'd just build a 'one size fits all' app and recruit more volunteers. But some volunteer optimiser might build one for an open-source project, and make it available under anonymous platform. |
|
1575)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 93949)
Posted 29 Nov 2019 by Richard Haselgrove Post: But is boinc.exe running? That has to happen first. By default, boinc.exe is started as a consequence of starting BOINC Manager. But mis-configuration can prevent that. You have to concentrate on getting it running first. |
|
1576)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 93946)
Posted 29 Nov 2019 by Richard Haselgrove Post: Concentrate on the first item first, then. If the client isn't running, nothing else can possibly work. Review where the program has been installed. In this day and age, you will probably be running the 64-bit version. You would expect to find that in C:\Program Files\BOINC, although you are at liberty to change that during installation. Go to that (or your chosen) directory, and double-click on the boinc application. That may reveal installation errors like missing DLLs. |
|
1577)
Message boards :
Questions and problems :
project list doesn't populate & URL entry does not work for addition of 1st project new machine
(Message 93943)
Posted 29 Nov 2019 by Richard Haselgrove Post: Do remember that BOINC consists of two components, the Client and the Manager. They communicate over TCP/IP, port 31416. First check - is the client running? Dies boinc.exe appear in Windows task manager? Second check - can that purely internal network traffic pass through your security device/software? Third check - can boinc.exe (only) communicate externally over both port 80 and port 443 (http and https respectively)? |
|
1578)
Message boards :
BOINC client :
Need help identifying bug(s): ubuntu or boinc or both?
(Message 93926)
Posted 27 Nov 2019 by Richard Haselgrove Post: Restarting the client caused a problemAre you sure that's the right command for that software combination? On my Mint box (derivative of Ubuntu), loaded from LoctusOfBorg PPA, the incantation is sudo systemctl stop boinc-clientwith 'start' and, I expect, 'restart' variants. |
|
1579)
Message boards :
GPUs :
Two projects on one GPU?
(Message 93834)
Posted 22 Nov 2019 by Richard Haselgrove Post: |
|
1580)
Message boards :
Questions and problems :
What do these 2 stats mean in BAM (for every project)?
(Message 93833)
Posted 22 Nov 2019 by Richard Haselgrove Post: RAC - 'Recent Average Credit' - is your current output. Incremental update is based on the total credit since the beginning of time, around 2005. |
|
1581)
Message boards :
Projects :
Einstein - PEER CERTIFICATE CAN'T BE AUTHENTICATED
(Message 93816)
Posted 20 Nov 2019 by Richard Haselgrove Post: I think they've fixed it. I've got that message in my overnight logs, but a manual retry 30 minutes ago went through properly. You may need to update the project manually to get things moving normally again. |
|
1582)
Message boards :
BOINC client :
found a "fix" for an issue in github but..
(Message 93778)
Posted 18 Nov 2019 by Richard Haselgrove Post: As far as I know, no documentation needs to updating. I am retired and have plenty of time available but am not wasting it. I am sorry if I am wasting your time by asking for advice here. I thought it was appropriate to ask here and not at GitHub.I'm in the same boat - retired, and with time on my hands. It's absolutely fine to ask questions here, or at any of the multiple other places where BOINCers hang out - but if you refer in one place to a problem described in another place, it's helpful to tie the two together with a link or a reference number. |
|
1583)
Message boards :
Questions and problems :
Restrict number of work units.
(Message 93774)
Posted 18 Nov 2019 by Richard Haselgrove Post: That won't reduce the number of tasks issued by the project, but will reduce the impact on your system. Does the project in question have a very heavy download burden, impacting on your internet access? If you need to avoid the downloads too, we'll have to think again. |
|
1584)
Message boards :
BOINC client :
found a "fix" for an issue in github but..
(Message 93773)
Posted 18 Nov 2019 by Richard Haselgrove Post: Without knowing for certain which issue you're talking about, it's hard to advise. Why not at least post the issue number here, in your question? From what you've actually said, I'd say: Leave the problem that you first encountered visible, for others to read. Remove all the guesses that turned out to be false trails. Write a clear statement about what the true solution was. Possibly make that the first thing that people will read, below the initial statement of the problem. Tell us - here, if you like - if any documentation needs updating, to avoid other people wasting their time as you appear to have done. |
|
1585)
Message boards :
BOINC client :
Suspend network activity does not work for "lost tasks"
(Message 93766)
Posted 17 Nov 2019 by Richard Haselgrove Post: What if you don't want the lost tasks to be downloaded as there was a legitimate reason in banishing them? I don't see a way to stop the download and even a detach had no effect.Just abort the tasks and report them. Then the tasks are no longer 'lost' but accounted for. |
|
1586)
Message boards :
BOINC client :
Suspend network activity does not work for "lost tasks"
(Message 93764)
Posted 17 Nov 2019 by Richard Haselgrove Post: FWIW: I would consider the project sending a "lost" file to the host to be a file transfer.What is 'resent' for a lost task after an update is the job specification - what application is to be used, what command line or processing parameters, etc. If a data file needs to be downloaded, that's a separate operation which follows later, and may on may not be delayed by network suspension. FWIW, I have a vague recollection from the early days of BOINC that manually updating a project during network suspension temporarily re-enabled networking - for five minutes, IIRC. But I haven't tested that or read about it happening for many years. The code may have been taken out. |
|
1587)
Message boards :
Questions and problems :
Requesting extended completion time
(Message 93706)
Posted 14 Nov 2019 by Richard Haselgrove Post: Task deadlines are set by each project individually. The only way to do this is to select a project with lengthy deadlines. The classic example is CPDN (Climate Prediction), with deadlines extending close to 1 year. SETI has some tasks with three week deadlines, but most tasks are allowed seven weeks or longer. Unfortunately, you can't select only to receive the longer tasks. RNA World has very long-running tasks, but I'm not familiar with the details of their deadlines. |
|
1588)
Message boards :
Questions and problems :
Bionic Manager Downloads
(Message 93704)
Posted 14 Nov 2019 by Richard Haselgrove Post: I think it would be best if you asked this on the SETI@Home message boards. They have a special Questions and answers section where anybody can post: you could ask in the specific operating system section that matches your particular computer, or in the general 'Getting started' area. But wherever you post, please be a little more explicit about your computer/OS and exactly what you're seeing: where exactly are these green check marks and red circles appearing? If you have a photo-sharing account anywhere, you could upload a screengrab and post a link to it - that would help. |
|
1589)
Message boards :
Questions and problems :
Change username?
(Message 93695)
Posted 13 Nov 2019 by Richard Haselgrove Post: The primary use of BOINC Manager is to display the status of the local client, and to enable certain control functions to be operated. But BOINC Manager is a fully network-aware tool, and can also be used to display/control remote clients instead. It's designed for low-latency networks, so the usefulness in reality is limited to clients on a Local Area Network - be that at home or an office/lab/school. There are other tools available to perform the same function for many local clients at once (BM can only handle one client at a time). The best known is BoincTasks. Again, there are independent statistics sites (mostly run by volunteers, who appreciate any contributions towards their costs) which can display all your collected statistics in one place. The best known of these is BOINCstats, but others are available. |
|
1590)
Message boards :
Questions and problems :
Change username?
(Message 93693)
Posted 13 Nov 2019 by Richard Haselgrove Post: I thought Boinc was the main program, not a project like eg: rosetta or Einstein?BOINC is a facilitator, not a controller. Volunteers interact with each project individually, and may choose whatever identity they wish (although for help-desk purposes, it's really helpful if they use the same name at each project). BOINC itself - this website - uses the same tools to identify posters, even though this website doesn't contain any of the other research/statistical tools. |
|
1591)
Message boards :
GPUs :
Intel IGP, how can I know when it's used?
(Message 93686)
Posted 13 Nov 2019 by Richard Haselgrove Post: Spend a little time learning your way around the BOINC interface. Most serious users switch to the 'Advanced' view. On the 'Tasks' tab you will find a complete list of all tasks issued to your machine: the 'Status' column is the key one, and you may need to expand the column width to see all the detail. Tasks are assigned to a specific hardware device when issued by the project server. They will only run on that device. A task issued for the Intel GPU will show that in the status, and if it's running, then your GPU is in use. Although Rob has correctly stated the general experience, I have a slightly different view. It's the combination of what your CPU is doing with both the traditional x64 cores and the Intel GPU that matters. If the x64 cores are all busy doing intensive work - floating point arithmetic using the SIMD operators - there won't be any power left over to run the GPU efficiently as well. Something will get throttled down. But if the x64 cores are on light duties - integer arithmetic, or ordinary light home/office use - then the GPU can be used at full speed. |
|
1592)
Message boards :
Questions and problems :
GPU dummy plug still needed?
(Message 93685)
Posted 13 Nov 2019 by Richard Haselgrove Post: I find a dummy plug is needed to activate the Intel GPU on machines where the display is connected to a discrete GPU - NVidia, in my case. |
|
1593)
Message boards :
Questions and problems :
Some projects get compensated by google?
(Message 93629)
Posted 11 Nov 2019 by Richard Haselgrove Post: My guess is they are looking for ET both at SETI and at Berkeley.Yes - they are searching for ET everywhere, and quite a lot of it happens at Berkeley. https://seti.berkeley.edu/ summarises how some of it hangs together: SETI@Home is just one part of the overall operation. Note that the SETI Institute, based in Mountain View, is not part of the Berkeley (University of California) operation. |
|
1594)
Message boards :
Questions and problems :
Seti@home not transfering w/VPN
(Message 93624)
Posted 11 Nov 2019 by Richard Haselgrove Post: Does your VPN handle secure (HTTPS) connections correctly? |
|
1595)
Message boards :
Questions and problems :
Beginner, with questions
(Message 93594)
Posted 9 Nov 2019 by Richard Haselgrove Post: And there is a huge range of speeds between different GPUs, too. The GT 1030 mentioned in the opening post is fairly new and efficient, but very much at the lower end of the performance scale. |
|
1596)
Message boards :
Questions and problems :
Reporting timer?
(Message 93566)
Posted 6 Nov 2019 by Richard Haselgrove Post: I'm not convinced by that.[EDIT]This is the setting that determines our normal 91 second rpc delay at MilkyWay. If you look at a sched_reply file from SETI, it contains <request_delay>303.000000</request_delay>That's recognisable as the standard 'shut up and wait' between contacts. The equivalent file from GPUGrid contains both <request_delay>31.000000</request_delay> <next_rpc_delay>3600.000000</next_rpc_delay>The second one triggers a 'phone home every hour', which is useful in their case to check if any new work has been created recently. I'll keep looking. If you do get a copy of the file, I'd be interested in taking a look at it - in private if necessary. Milkyway has <request_delay>91.000000</request_delay>I'm more interested in <min_sendwork_interval> N </min_sendwork_interval>I'm not yet certain that this is the one which emerges as <request_delay>, but I think it's a more plausible candidate. Edit - candidacy confirmed (I think) by https://github.com/BOINC/boinc/blob/master/sched/sched_types.cpp#L784 |
|
1597)
Message boards :
Questions and problems :
Reporting timer?
(Message 93565)
Posted 6 Nov 2019 by Richard Haselgrove Post: I guess you don't do MW or frequent their fora.I have run their tasks, and jousted with their 'admins', in the past. But I don't judge it as a project worthy of spending any more time or electricity on - unless anybody can supply evidence of sentient life at project management level. You might like to read messages 63074 - possibly relevant to the current problems - and 58550. |
|
1598)
Message boards :
Questions and problems :
Reporting timer?
(Message 93561)
Posted 6 Nov 2019 by Richard Haselgrove Post: Well, this would be the place to look: https://boinc.berkeley.edu/trac/wiki/ProjectOptions But I must say I've never heard anyone discussing the need for a project setting like that, nor remember seeing one when I've been looking through for something else. |
|
1599)
Message boards :
Questions and problems :
Reporting timer?
(Message 93558)
Posted 6 Nov 2019 by Richard Haselgrove Post: Except for the specific case that Keith mentions, BOINC doesn't 'report' tasks: it saves up everything it needs to do, and 'updates' the server all in one go. What you're seeing at Milkyway will be requests for new work (oh, and we'll report the old stuff while we're at it). John McLeod VII used to keep a boilerplate of six or seven reasons why an update might be triggered - I think some more have been added or changed since he stopped posting. |
|
1600)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93552)
Posted 5 Nov 2019 by Richard Haselgrove Post: ... who is the God of computing, anyway?Having consulted the oracles, I came up with Charles Babbage for hardware and Ada Lovelace for software. It's back. :-)And I'm glad my invocations worked :-) |
|
1601)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93548)
Posted 5 Nov 2019 by Richard Haselgrove Post: Well, the time is approaching for my time at the |
|
1602)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93540)
Posted 5 Nov 2019 by Richard Haselgrove Post: I would offer up my delivery driver as a human sacrifice, but he's given up for the night and scurried back to his depot. |
|
1603)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93534)
Posted 5 Nov 2019 by Richard Haselgrove Post: Sitting at home waiting for a delivery. If it had arrived during the original 1-hour estimate, I could have had it fitted and crunching by now. |
|
1604)
Message boards :
Questions and problems :
GPU idling just because I set from 10 days to 1 day of work storage
(Message 93518)
Posted 4 Nov 2019 by Richard Haselgrove Post: I think coproc_debug is mainly concerned with device detection. Despite the name, cpu_sched_debug gives more information about what tasks to run, or not run, on all devices. In particular, GPUs. |
|
1605)
Message boards :
Projects :
Some projects will not get tasks
(Message 93497)
Posted 3 Nov 2019 by Richard Haselgrove Post: I think you ought to look in the Event Log, preferably with the optional <sched_op_debug> enabled. What you've posted so far gives no idea of the individual transactions with each project server - what is requested, and what reply is received. |
|
1606)
Message boards :
BOINC client :
Multi-GPU multi-project scheduling oddity
(Message 93485)
Posted 2 Nov 2019 by Richard Haselgrove Post: Projects do not 'decide' to send you work: your BOINC client decides to 'Request' work. To get to the bottom of this, you're probably going to have to do some intense work with the Event Log. Start by setting <sched_op_debug> - that adds comparatively little to the total volume of messages, but will get you started with understanding how the process works. If you're running a recent version of BOINC (you mention neither that, not your operating system), you can set event log flags from Ctrl+Shift+F. Don't set too many options at once: too many massages can overwhelm the system. |
|
1607)
Message boards :
Questions and problems :
Detaching from projects?
(Message 93483)
Posted 2 Nov 2019 by Richard Haselgrove Post: I keep a few dead projects around so I can test questions like that ;-) 'Reset project' appeared in the Event Log before final detach, but there was no attempt to make any communication with the project. It's possible that there might have been an attempt to report any final tasks, but it wasn't needed in this case. Given that one of the main reasons for detaching a project is because it's stopped sending out work, and may have switched off its servers, or even had its domain name recycled to a new organisation, I don't think there's much to be gained by contacting the server - and it might even be dangerous. |
|
1608)
Message boards :
GPUs :
Two projects on one GPU?
(Message 93473)
Posted 31 Oct 2019 by Richard Haselgrove Post: That reminds me... BOINC v7.14.2 doesn't handle <max_concurrent> very well. You might find count (2) might work better under v7.16.3 For count (1) - I can't speak for ATI tasks, but some Einstein tasks are very much boosted by running at a greatly enhanced process priority. The machine I'm typing on has Einstein's intel_gpu app running at real time priority under Process Lasso. I notice a brief stutter each time a task finishes and another starts, but at once every five hours that's not a hardship. Use that factoid with care and at your own risk. |
|
1609)
Message boards :
GPUs :
Two projects on one GPU?
(Message 93469)
Posted 31 Oct 2019 by Richard Haselgrove Post: Dunno. You could be the first to try it! You'd need two app_config.xml files - one for each project - both with a <gpu_usage> value of 0.5 and a <max_concurrent> of 1. I'd suggest you avoid fetching CPU tasks from either project while you test. |
|
1610)
Message boards :
Questions and problems :
Sorting columns
(Message 93467)
Posted 31 Oct 2019 by Richard Haselgrove Post: I can't speak directly for Mac, but the Linux answer is that configuration settings are held in a hidden file called '.BOINC Manager'. In the Linux case, the file is in the user's home directory. When you find it, the file has sections for each tab - Projects, Tasks, Transfers etc. The first line in each section starts SortColumn=, and the value should be -1 for an unsorted display. I'd advise you to make sure that BOINC Manager is fully shut down before making manual edits. |
|
1611)
Message boards :
Questions and problems :
Why old virtual box comes with boinc download?
(Message 93466)
Posted 31 Oct 2019 by Richard Haselgrove Post: Don't worry - just touched a raw nerve. I've got an ongoing grumble with BOINC itself for sometimes emitting misleading error messages. |
|
1612)
Message boards :
Questions and problems :
Why old virtual box comes with boinc download?
(Message 93460)
Posted 31 Oct 2019 by Richard Haselgrove Post: From the opening post: I had to update it manually to run LHC properly.No hard feelings, but could all posters please be careful about the precise language they use? We do read these boards, and sometimes they're a very useful source of information about bugs or features that we need to pass on to the developers. In this case, it seems to have transpired that that phrase was, perhaps, a tad over the top? |
|
1613)
Message boards :
Questions and problems :
Why old virtual box comes with boinc download?
(Message 93459)
Posted 31 Oct 2019 by Richard Haselgrove Post: OK, I'll stand down my earlier warning. But suggest that it's something worth considering - Jacob Klein is another bleeding-edger, who might have a view. |
|
1614)
Message boards :
Questions and problems :
Why old virtual box comes with boinc download?
(Message 93457)
Posted 31 Oct 2019 by Richard Haselgrove Post: I'm in the process of sending an email to CERN, to ask that very question.I've had a reply. CERN wrote: We don't have a strict dependency on the VBox version. The minimum version we accept is v5. The volunteers seem to have a good sense for what a good version is.Which doesn't take us much further forward. Can anybody here point me to one of those volunteers with good sense? Or a message board thread? |
|
1615)
Message boards :
Questions and problems :
Why old virtual box comes with boinc download?
(Message 93455)
Posted 31 Oct 2019 by Richard Haselgrove Post: It happened with the v7.16.1 release of BOINC, too. But at least that was intended as a test release (although it rather leaked out). We expected it to have problems - though perhaps not quite as spectacular as aborting the entire cache of WUs, completed, running and waiting.Really?! I've never in my entire life had any new version of any software work worse than the older version. But I've lost count of how many times I've had unknown problems only to find there was a newer version to fix them that I didn't know about. That would be why almost all software now autoupdates.You have clearly not worked in the NHS. During my time there, more than once I found a new version of the software broke things. |
|
1616)
Message boards :
Questions and problems :
Why old virtual box comes with boinc download?
(Message 93451)
Posted 31 Oct 2019 by Richard Haselgrove Post: I'm in the process of sending an email to CERN, to ask that very question. |
|
1617)
Message boards :
Questions and problems :
Why old virtual box comes with boinc download?
(Message 93448)
Posted 31 Oct 2019 by Richard Haselgrove Post: I've posted a public note on the mailing lists about the need to choose a stable VBox version which meets the minimum needs of LHC for the next BOINC release, without necessarily being right on the bleeding edge. I've also suggested privately that BOINC and CERN (David and Laurence) should compare notes and choose a version. I'll keep an eye on how that goes. |
|
1618)
Message boards :
BOINC Manager :
CTRL-TAB doesn't work on the "Notices" tab.
(Message 93441)
Posted 31 Oct 2019 by Richard Haselgrove Post: I thought I'd post it to help the other readers who sometimes look over our shoulders while we sort things out here. And for readers who discover this thread in the future. |
|
1619)
Message boards :
Questions and problems :
Trying to maximize the performance of my Mac Pro
(Message 93440)
Posted 31 Oct 2019 by Richard Haselgrove Post: GPUGrid does checkpoint, but it also takes a long time to quit its science app when requested. When I've suspended it manually, BOINC Manager displays the status as 'suspended', but the counters continue to click up progress and time. And no new task starts. After what seems an age (probably about 10 seconds), the counters stop - and BOINC starts a new task. It's clunky, but I think it's true that BOINC monitors the 'suspended' task and doesn't switch until it really has left the room. |
|
1620)
Message boards :
Questions and problems :
Stuck in gear
(Message 93430)
Posted 30 Oct 2019 by Richard Haselgrove Post: It's bad programming to leave the end of a decision tree undefined.Hear, hear. I did my last formal computer training 45 years ago (and computers were different then). But it was computer science training, not computer programming: understanding the background to the way different programming languages were designed, and how and why they differed. But even I can work out the truth of that statement. |
|
1621)
Message boards :
Questions and problems :
Stuck in gear
(Message 93426)
Posted 30 Oct 2019 by Richard Haselgrove Post: Just to be on the safe side, I thought it might be wise to test that we'd diagnosed this one properly. So I set about half the debug log flags on one of my machines - all the ones in the right-hand column in v7.16.3, to be precise. That includes some biggies like <app_message_receive> and <rrsim_detail>. And clicked 'apply', with both Manager and Event Log windows visible. Sure enough, the first actual error message I got was "The password you have provided is incorrect, please try again.". Then, all hell broke out in both windows. Both views cleared, started to redraw, cleared again, flickered and flashed, showed partial information, etc. etc. Clearly couldn't keep up with the flow of GUI RPCs. I crashed everything down, killed the processes for the science apps which hadn't got the 'quit' messages, and rebooted the machine for luck. BOINC auto-restarts, of course (I hadn't stopped that), and the windows started flashing again. Another crash stop, another set of process kills, and this time I hand-edited the log flag section of cc_config.xml. This time, BOINC started cleanly. I've been baby-sitting it for about an hour, and no sign of trouble. The first two projects have reported a batch of tasks each (including some of the tasks which I'd hand-killed) and both sets of tasks validated - no errors. So, the core BOINC functionality is robust, and came up smelling of roses - full marks. But that "password incorrect" message is a liar: I do have a password set, but the contents of that file haven't changed since September 2013, when I set this machine up. |
|
1622)
Message boards :
Questions and problems :
What does (alt) mean
(Message 93424)
Posted 30 Oct 2019 by Richard Haselgrove Post: It's just a name given to distinguish between two alternative applications which perform the same work. Sometimes, a different approach can work faster on some machines but not on others - one might be faster on Intel CPUs, for example, and the other on AMD. Having the two alternatives to try enables the server (hopefully!) to pick the faster one next time. |
|
1623)
Message boards :
Questions and problems :
Local host connection, NOT
(Message 93422)
Posted 30 Oct 2019 by Richard Haselgrove Post: Are you using the same anti-virus solution on both laptops? |
|
1624)
Message boards :
BOINC Manager :
CTRL-TAB doesn't work on the "Notices" tab.
(Message 93414)
Posted 30 Oct 2019 by Richard Haselgrove Post: I don't have access to the nav bar either, that's a CSS file as far as I know.The source describes it as a "generated-sidebar portlet". <div class="generated-sidebar portlet" id="p-Resources" role="navigation"> <h3>Resources</h3> <div class='pBody'> <ul> <li id="n-BOINC-home"><a href="http://boinc.berkeley.edu" rel="nofollow">BOINC home</a></li> <li id="n-User-Manual"><a href="/wiki/User_manual">User Manual</a></li> <li id="n-Help"><a href="/wiki/BOINC_Help">Help</a></li> <li id="n-Download"><a href="http://boinc.berkeley.edu/download.php" rel="nofollow">download</a></li> <li id="n-Message-boards"><a href="http://boinc.berkeley.edu/dev/" rel="nofollow">Message boards</a></li> <li id="n-Web-sites"><a href="http://boinc.berkeley.edu/links.php" rel="nofollow">Web sites</a></li> <li id="n-Add-on-software"><a href="http://boinc.berkeley.edu/addons.php" rel="nofollow">Add-on software</a></li> <li id="n-Other-ways-to-help"><a href="/wiki/Other_ways_to_volunteer">Other ways to help</a></li> </ul> </div> </div>Presumably it's edited through id="n-Message-boards" - there must be a list somewhere. |
|
1625)
Message boards :
BOINC Manager :
CTRL-TAB doesn't work on the "Notices" tab.
(Message 93412)
Posted 30 Oct 2019 by Richard Haselgrove Post: Workround for the presenting problem: you can navigate via the 'View' menu with Alt-V. |
|
1626)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93410)
Posted 30 Oct 2019 by Richard Haselgrove Post: As did the 'going to bed and sleeping through it'. |
|
1627)
Message boards :
BOINC Manager :
CTRL-TAB doesn't work on the "Notices" tab.
(Message 93409)
Posted 30 Oct 2019 by Richard Haselgrove Post: The extra "dev" came from https://boinc.berkeley.edu/wiki/BOINC_Help where I followed the "Message Boards" link on the left.Thanks. I've edited the link at the foot of the help page, but I don't have access to the navigation bar - I'll have to pass that upstream. |
|
1628)
Message boards :
Questions and problems :
Adjust time between project contacts?
(Message 93396)
Posted 29 Oct 2019 by Richard Haselgrove Post: The other advantage of <sched_op_debug> is that you get to see exactly what's going on during the scheduler request: 29/10/2019 22:46:46 | NumberFields@home | Sending scheduler request: To report completed tasks. 29/10/2019 22:46:46 | NumberFields@home | Reporting 4 completed tasks 29/10/2019 22:46:46 | NumberFields@home | Requesting new tasks for CPU 29/10/2019 22:46:46 | NumberFields@home | [sched_op] CPU work request: 2856.25 seconds; 0.00 devices 29/10/2019 22:46:46 | NumberFields@home | [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 29/10/2019 22:46:46 | NumberFields@home | [sched_op] Intel GPU work request: 0.00 seconds; 0.00 devices 29/10/2019 22:46:49 | NumberFields@home | Scheduler request completed: got 1 new tasks 29/10/2019 22:46:49 | NumberFields@home | Project requested delay of 61 seconds 29/10/2019 22:46:49 | NumberFields@home | [sched_op] estimated total CPU task duration: 3230 secondsIf the 'estimated total task duration' is less than the work request (for whichever device), then you're asking for too much and hitting some other limit. |
|
1629)
Message boards :
Questions and problems :
Adjust time between project contacts?
(Message 93395)
Posted 29 Oct 2019 by Richard Haselgrove Post: I don't get that request for a delayTurns out you need <sched_op_debug> set in the logging options. That should show up as [sched_op], as in the other lines in the following example: 29/10/2019 23:10:26 | SETI@home | Scheduler request completed 29/10/2019 23:10:26 | SETI@home | [sched_op] Server version 709 29/10/2019 23:10:26 | SETI@home | Project requested delay of 303 seconds 29/10/2019 23:10:26 | SETI@home | [sched_op] handle_scheduler_reply(): got ack for task blc22_2bit_guppi_58692_03898_HIP79702_0124.28847.409.22.45.114.vlar_0 29/10/2019 23:10:26 | SETI@home | [sched_op] handle_scheduler_reply(): got ack for task blc22_2bit_guppi_58692_02278_HIP80797_0119.28898.0.22.45.227.vlar_1 29/10/2019 23:10:26 | SETI@home | [sched_op] handle_scheduler_reply(): got ack for task blc22_2bit_guppi_58692_04223_HIP79568_0125.28907.818.21.44.55.vlar_0 29/10/2019 23:10:26 | SETI@home | [sched_op] Deferring communication for 00:05:03 29/10/2019 23:10:26 | SETI@home | [sched_op] Reason: requested by projectThat's a bug. |
|
1630)
Message boards :
Questions and problems :
Adjust time between project contacts?
(Message 93391)
Posted 29 Oct 2019 by Richard Haselgrove Post: Can the server not set some kind of gap?Servers can and do set a minimum gap: you see it in the Event Log after every contact. 29/10/2019 22:32:22 | Milkyway@Home | Project requested delay of 91 secondsI find it hard to believe that Milkyway is clever enough to set a second, secret, delay before issuing work. |
|
1631)
Message boards :
Questions and problems :
Adjust time between project contacts?
(Message 93390)
Posted 29 Oct 2019 by Richard Haselgrove Post: In theory, you can do it by careful manipulation of cache settings. My slower machines have settings of 0.25 + 0.05 days: grab an extra hour or so when fetching, then mull things over for another hour or more before requesting a refill. My fastest has 0.5 + 0.01 days - keep well filled up, little and often. The only hard-wired value is 1 hour: if a completed task has been hanging around that long, report it and (probably) request new work. |
|
1632)
Message boards :
BOINC Manager :
CTRL-TAB doesn't work on the "Notices" tab.
(Message 93376)
Posted 29 Oct 2019 by Richard Haselgrove Post: Bollocks. Forget everything I've just said. It's a LOT more subtle than that. Starting on the Tasks tab, for example. Pressing the TAB key repeatedly steps the focus forward from 'Show active tabs', via all enabled command buttons, to the task list, and then back to 'Show active tabs'. As you might expect. Pressing Shift-TAB repeatedly steps the focus back round the same loop, until 'Show active tabs' - then it sets the focus on the tab label itself - and stops there. Pressing Ctrl-TAB, starting with the focus on the tab name, cycles through all the tabs including Notices. If you start with the focus on one of the command buttons or the list control, it cycles through except Notices. If you reach Notices, movement stops there. As the thread title says. Argh. Edit - all the above applies to Windows. Linux (Mint, based on Ubuntu) is different. TAB and Shift-TAB both cycle the focus through all active controls, in both directions - including the tab name label. But Ctrl-TAB doesn't shift between tabs. I don't have a Mac to test. |
|
1633)
Message boards :
BOINC Manager :
CTRL-TAB doesn't work on the "Notices" tab.
(Message 93375)
Posted 29 Oct 2019 by Richard Haselgrove Post: Problem confirmed on the v7.16.3 Windows Manager currently in testing (and probably on every version since the Notices tab was introduced!). I'll see if I can find a missing keyboard handler from one of the other tabs. Further detail: it's broken in v7.14.2 and v7.16.1 - but it seems to be fixed in a very recent (< 48 hours) test build. The cause of the change isn't obvious. |
|
1634)
Message boards :
BOINC Manager :
CTRL-TAB doesn't work on the "Notices" tab.
(Message 93374)
Posted 29 Oct 2019 by Richard Haselgrove Post: Oh yes, and the "Report Bugs" link on https://boinc.berkeley.edu/dev/ doesn't work, either. I tried that first."Report bugs" does work on https://boinc.berkeley.edu. Where did you get the extra /dev/ in your link form? I'll have a look into the keyboard shortcut issue. Have you considered a cheap USB mouse? |
|
1635)
Message boards :
Questions and problems :
Stuck in gear
(Message 93365)
Posted 28 Oct 2019 by Richard Haselgrove Post: I did as instructed. The computer is now set such that the stdoutdae.txt has been reset and the values in the registry are set to the default which is zero. I noticed that the gui_rpc_auth.cfg had been changed and the password did not match the authenticator line in account_einstein.phys.uwm.edu.xml. I don't know if that would have an effect or not, but I did change the gui_rpc_auth.cfg to the original configuration file when I set up the computer to run. I am assuming that this is the password and I have no idea how it got changed. I don't read hex so I'm not sure if this is my password or not. After all this I attempted to connect with BOINC manager again and got the same result, the password is not recognized. I am at a loss right now.The gui_rpc_auth.cfg password is local only - it's not meant to have any relationship with the authenticator. I think it's random. I think the client only reads the file at startup, and is then listening for that password for the rest of the session (next restart, days or weeks later). If you change it, the Manager will read the new one from the file, and offer that instead - and it won't match the old one. Moral: restart the client after changing the password. |
|
1636)
Message boards :
Questions and problems :
Stuck in gear
(Message 93358)
Posted 28 Oct 2019 by Richard Haselgrove Post: Bobby's machine contacted the Einstein server about 20 minutes ago - Tasks for computer 12791665. It reported a completed task (still pending validation), so it looks like the client is running properly. Completed tasks are showing a big discrepancy between elapsed time and CPU time, which correlates with a very busy CPU. Jord's probably on to something with the 'too many log flags' rendering the client incommunicado. |
|
1637)
Message boards :
Questions and problems :
Problems about temporary data and offline about BOINC
(Message 93357)
Posted 28 Oct 2019 by Richard Haselgrove Post: A few years ago now but I remember editing that file when restoring tasks to a point before a backup had been taken. It even worked occasionally. BOINC version at the time was about 4.5 and I installed it then in a folder in my home directory.It works the same now, but with GPUs it's got a little more complex and much, much longer. A recent Manager bug marked well over 1,000 tasks as 'aborted'. Fortunately, I was only reporting 128 tasks at a time: I managed to save about 800 of them at the third attempt. |
|
1638)
Message boards :
Questions and problems :
Problems about temporary data and offline about BOINC
(Message 93354)
Posted 28 Oct 2019 by Richard Haselgrove Post: BOINC also stores instructions about how each task is to be processed in a complex (and fragile) file called client_state.xml. Some projects only issue processing instructions (parameter strings) for that file, and don't download a separate data file. |
|
1639)
Message boards :
Questions and problems :
Stuck in gear
(Message 93353)
Posted 28 Oct 2019 by Richard Haselgrove Post: That's an alternate theory. Modern BOINCs only rotate the logs to enforce the 2 MB (or whatever) limit when the client restarts, so it may not be crashing. Maybe you have so many optional log flags that the client simply can't keep up, and doesn't have enough time to process comms with the manager at the same time. Ideally, in that case you would need to find a way of shutting the client down cleanly, so you can edit cc_config.xml and read the new version at startup. Best way is to navigate to BOINC's program directory (C:\Program Files\BOINC), open a command window, and issue boinccmd --quit |
|
1640)
Message boards :
Questions and problems :
Stuck in gear
(Message 93350)
Posted 28 Oct 2019 by Richard Haselgrove Post: It sounds like the client is crashing every 10 seconds, then restarting. You won't be able to post the whole 72 MB here (64 KB limit), and we wouldn't want to read it! Can you search through that file for the phrase "Starting BOINC client", and then copy out everything until the next appearance of the same phrase. Choose one of the most recent occurrences towards the end of the file. If the result is still over 64 KB, you'll have to upload it to a file sharing site and post us a link. It might be worth looking at the .txt files starting 'stderr...' in the same folder. Look at the datestamps first - not worth bothering with one that hasn't changed in the last couple of days. |
|
1641)
Message boards :
Questions and problems :
Stuck in gear
(Message 93339)
Posted 27 Oct 2019 by Richard Haselgrove Post: I think the problem starts at https://github.com/BOINC/boinc/blob/master/clientgui/BOINCBaseFrame.cpp#L479. There are a couple of genuine causes listed, then - at line 518 - there's an } else {If we ever reach this point, there's only one answer we can give - "The password you have provided is incorrect, please try again.": whatever the true failure.And that's as far as I'm going tonight - late evening, European time. I'll think again in the morning. |
|
1642)
Message boards :
Questions and problems :
Stuck in gear
(Message 93338)
Posted 27 Oct 2019 by Richard Haselgrove Post: Are you the same 'Bobby' as the 'BOBBY CONGER' that I replied to at Einstein, earlier this evening? If so, let's try and tie the two threads together. I see that the Einstein thread now has a brief snippet from an Event Log, with several debug flags set. It would probably help if you could post a longer snippet, from the top, showing the complete sequence of events from computer restart to the recurrence of the error messages. (Edit - since others are getting involved who don't usually read Einstein, the message in question is https://einsteinathome.org/content/stuck-last-gear-0#comment-174084) As I said at Einstein, I think the 'bad password' error message is bad programming: it's the fallthrough message when all known reasons for failure have been excluded. So we're dealing with something unknown to the programmers (although it's been around for years). The password in question is described in Controlling BOINC remotely. It's a string placed in the file gui_rpc_auth.cfg on the computer you're trying to control. Under Windows, it may be a random 32-character string created when BOINC is first installed, or you can place something simpler there yourself. If you are connecting to a BOINC client on the same machine as the Manager you're using, you should never need to supply a password: the Manager can look up the password from the same place as the client, and self-authenticate. The password field needs to be populated if you are genuinely needing to control a different, remote client on another machine. Then you would go to the 'Select computer...' menu entry on the File menu in BOINC Manager (Advanced View, if you're not using that already). You should fill in BOTH boxes (for a remote computer), or NEITHER box (for a local client). But as I've said, I think this whole message is a red-herring - we need to find a different cause. Hold on while I find the source code for the bad message, so I can explain why I think it's wrong. |
|
1643)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93333)
Posted 27 Oct 2019 by Richard Haselgrove Post: If someone gets real bored they can talk about the validate issues over at Einstein.Couldn't find that discussion, so I gave them a bellyache about passwords instead. |
|
1644)
Message boards :
Server programs :
bug in server: report to project or to boinc?
(Message 93318)
Posted 26 Oct 2019 by Richard Haselgrove Post: You'll probably find that the CPDN admins set a one hour delay exactly - and that the BOINC server code then added a 1% margin for error. Not every computer clock runs perfectly, and we wouldn't want a fast-running clock to send premature requests - and have them rejected. |
|
1645)
Message boards :
Server programs :
bug in server: report to project or to boinc?
(Message 93316)
Posted 26 Oct 2019 by Richard Haselgrove Post: And a note on timing. The three requests I just posted ended with 26/10/2019 04:32:30 | Einstein@Home | Project requested delay of 60 seconds 26/10/2019 07:10:32 | NumberFields@home | Project requested delay of 61 seconds 26/10/2019 08:12:28 | SETI@home | Project requested delay of 303 secondsDifferent projects, different delays. The client internally maintains a 60 seconds polling loop: unless something else happens, it may be up to 60 seconds after the end of the project requested delay before any work request is made. And the first request might be to a different project. There's usually a five second pause between requests to different projects. |
|
1646)
Message boards :
Server programs :
bug in server: report to project or to boinc?
(Message 93315)
Posted 26 Oct 2019 by Richard Haselgrove Post: Absolutely. Work is always allocated as the result of a REQUEST. Not by a whim of the server. The simplest way to monitor it is to set the <sched_op_debug> logging flag. 26/10/2019 07:10:29 | NumberFields@home | [sched_op] CPU work request: 1192.03 seconds; 0.00 devices 26/10/2019 07:10:29 | NumberFields@home | [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 26/10/2019 07:10:29 | NumberFields@home | [sched_op] Intel GPU work request: 0.00 seconds; 0.00 devices 26/10/2019 08:12:26 | SETI@home | [sched_op] CPU work request: 0.00 seconds; 0.00 devices 26/10/2019 08:12:26 | SETI@home | [sched_op] NVIDIA GPU work request: 8674.68 seconds; 0.00 devices 26/10/2019 08:12:26 | SETI@home | [sched_op] Intel GPU work request: 0.00 seconds; 0.00 devices 26/10/2019 04:32:28 | Einstein@Home | [sched_op] CPU work request: 0.00 seconds; 0.00 devices 26/10/2019 04:32:28 | Einstein@Home | [sched_op] NVIDIA GPU work request: 0.00 seconds; 0.00 devices 26/10/2019 04:32:28 | Einstein@Home | [sched_op] Intel GPU work request: 4325.95 seconds; 0.00 devicesYou only get work if you ask. And if you ask for work for a device that the project supports. And if the project has the right kind of work available at the instant you ask. Bit it starts with an ask. |
|
1647)
Message boards :
Server programs :
bug in server: report to project or to boinc?
(Message 93313)
Posted 25 Oct 2019 by Richard Haselgrove Post: An alternative way of handling that is to paste the search phrase "last request too recent" (with the quotation marks in this case) into the low-contrast search box at the top left of the GitHub screen (beside the 'black cat' logo). That gave me the two hits in sched/handle_request.cpp that you quoted. Most of the server code that BOINC users will be interested in is in that 'sched' folder. It also found that the phrase had appeared in three bug-report 'issues' - one recent, and the other two both over a decade old. There's no difference between Linux code and Windows code - it's all just ASCII text. One difference that trips Windows users up is in the line termination convention used in text files: Windows uses the character pair CRLF, whereas Linux uses a single CR character. Notepad++ is a useful Windows tool which can display either format without batting an eyelid - and convert between the storage formats on disk whenever needed. Going back to the query about "last request too recent": servers always issue a "Don't bother me for xxx seconds" instruction after every contact. The value of xxx is set by the project: it varies from 7 seconds at SETI Beta, to 303 seconds at SETI, to an hour at CPDN (last time I looked). It's there to protect against effective DDoS attacks by BOINC clients. If you do make a second request too soon, everything works as normal, except that you won't be allocated any new work. You have to wait for another timeout interval. The commonest time I see it is if I'm doing maintenance, and have to restart the BOINC client. The client doesn't remember where it is in the backoff cycle when restarted, and is capable of requesting new work before the server is willing to send it. |
|
1648)
Message boards :
BOINC Manager :
7.16.3 on Ermine
(Message 93289)
Posted 23 Oct 2019 by Richard Haselgrove Post: The reversion of the original BOINC code change can be found in https://github.com/BOINC/boinc/pull/3106 - there are links in that PR back to the original change and the motivation for it. There are links to the Debian and Fedora patch lists in https://github.com/BOINC/boinc/issues/3260.I got the initial BOINC source implementation reverted, but it was re-implemented as a downstream patch by the Linux deployment team.Can you point me to the right zip to download to get it back? I believe that the https://github.com/BOINC/boinc/tree/client_release/7/7.16 branch of BOINC's own code is clean in this respect, although it's still subject to testing and I don't think we've got all the bugs out of it yet. |
|
1649)
Message boards :
BOINC Manager :
7.16.3 on Ermine
(Message 93286)
Posted 23 Oct 2019 by Richard Haselgrove Post: I got the initial BOINC source implementation reverted, but it was re-implemented as a downstream patch by the Linux deployment team. |
|
1650)
Message boards :
Questions and problems :
Scheduler Concerns
(Message 93285)
Posted 23 Oct 2019 by Richard Haselgrove Post: It's true that there are scheduler changes in the new v7.16.3 client, but they are mainly to do with deciding which task(s) to run, not directly to do with work fetch - although they do have side effects on work fetch. I'd make two comments on the issues you describe. 1) Milkway - over-running initial runtime estimates, so there's no time left for other tasks to run. There are two causes for this. One is bad initial estimation pf the computational load of the task. That's down to the project, and I have to say that my impression is that Milkyway in particular (but other projects as well) pays too little attention to getting these details right. The second cause is that BOINC is particularly bad at estimation in the initial stages of running a new project and/or a new application version. The estimates will eventually become a lot more accurate, but it takes a long time - too long, in my view. Again the adjustments are made on the server, not under the control of the client. 2) Einstein - re-sending lost tasks, even though NNT is set. This is a known and deliberately intended policy, but it should only apply to genuinely "lost" tasks. These are tasks unknown in your current client cache, but still marked as 'in progress' on the project web site. If you had tasks cancelled because of over-running deadlines, they should have been marked as 'aborted' on the website and not been eligible for re-sending. If you can document any sample case of a task being cancelled and reported to the project, but still being resent as a 'lost' task, please report it here or directly to Einstein. Having said that, I concur with the advice you've received: when joining a new project, it's always wise to reduce your cache size initially until you get a feel for task performance and deadline. Also, be aware that initially resource share considerations will mean that your old projects have had more than their fair share of computer resources, and the new projects will appear to hog the machine until they've caught up. That's normal for BOINC. |
|
1651)
Message boards :
BOINC Manager :
7.16.3 on Ermine
(Message 93279)
Posted 23 Oct 2019 by Richard Haselgrove Post: There is a strong push coming from the Linux repository maintainers (not from BOINC) to remove a lot of the user controls from Linux versions of BOINC. Their view - I think coming from a 'large, secure, organisation' point of view - is that users don't and shouldn't have low-level administration rights over the computer on their desks: sudo is strongly frowned upon, and root access is a complete no-no. They want BOINC to run as a service, and for the minions in their sweatshops not to be able to stop BOINC from running under any circumstances. That's fine for their particular use-case, where running BOINC is a high-level corporate decision. But I am fighting a lone battle to persuade them that there are other Linux users out here - home users, who are the personal owners of the machine in question, with full authority, and perforce acting as their own sysadmin. There is some support for this view from the BOINC side of the table, but not enough yet. |
|
1652)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93264)
Posted 22 Oct 2019 by Richard Haselgrove Post: I'll cover the shift from 19:00 UTC onwards. |
|
1653)
Message boards :
Server programs :
bug in server: report to project or to boinc?
(Message 93256)
Posted 21 Oct 2019 by Richard Haselgrove Post: The server code is contained in the exact same BOINC/boinc repository at GitHub - you can also choose server_release/1/1.0 and server_release/1/1.2 branches if you want confirmed, stable, code. I'm not exactly sure what you're describing at a first, rushed reading. But I think it's a well-known and deeply embedded design decision, which it will be hard to change at this stage. The process goes: Make a change to project preferences, on the project website. Update the project in the client. It's wise to have 'No new tasks' set for this stage. This update ensures that the web preferences are transferred to the client. THEN allow new work, and update for a second time. The client will then request new work based on the new preferences. app_config.xml can't be used to control download choices. |
|
1654)
Message boards :
Questions and problems :
LHC@Home, GPU, and BOINC ProjectList
(Message 93251)
Posted 21 Oct 2019 by Richard Haselgrove Post: I think all the ATI detections are made automatically, including the project preferences. In which case, it's probably best that they're consistent, driven off the same data via the same rules. I did go into this in some detail when I was writing #2643, in an attempt to synchronise the web page here with the 'attach to project' wizard displayed in the client - they were badly out of sync. It's better now, but I've discovered that the automatic updating process has broken down: Universe@Home has added an Android application which doesn't show in BOINC Manager, for example. I agree the outcome has become bad programming, but in defence of the BOINC programmers, they're working at the cutting edge of computer science development. When BOINC coding started, we worked on CPUs only, on the main desktop operating systems and a few legacy OSs. Since then, we've added CUDA, ATI, OpenCL, intel_gpu, Android and probably more, and the original OSs have undergone huge changes. In general, BOINC works surprisingly well, under the circumstances. Going back to the specific ATI question: the decision on what devices to support is taken individually by each separate project. The rules for supported devices are specified in Plan class names: Plan class names encode information as follows:CERN's use of 'native' breaks the "and only if" component of those rules. I pointed this out to CERN while I was working on the wizard, but it's up to them to fix it. If you can think of a better rule which would correctly discriminate all of these, please let me know and I'll forward it. ati_opencl ati14 ati14_app13 BRP4G-Beta-opencl-ati BRP4G-Beta-opencl-ati-Lion BRP4G-opencl-ati BRP4G-opencl-ati-lion FGRPopencl1K-ati FGRPopencl-ati FGRPopencl-ati-mav GW-opencl-ati native_mt native_theory opencl_ati opencl_ati_100 opencl_ati_101 opencl_ati_AP27 opencl_ati_cat132 opencl_ati_gpu opencl_ati_mac opencl_ati_nocal opencl_ati_sah opencl_ati5_cat132 opencl_ati5_mac opencl_ati5_nocal opencl_ati5_sah opencl_ati5_SoG opencl_ati5_SoG_cat132 opencl_ati5_SoG_mac opencl_ati5_SoG_nocal opencl_atiapu_sah openclatiGFN15 openclatiGFN16 openclatiGFN17LOW openclatiGFN17MEGA openclatiGFN18 openclatiGFN19 openclatiGFN20 openclatiGFN21 openclatiGFN22 openclatiGFNEXTREME openclatiPPSsieve |
|
1655)
Message boards :
BOINC Manager :
VS2013 Boinc build: build tools missing
(Message 93250)
Posted 20 Oct 2019 by Richard Haselgrove Post: As I said in that thread, I did update Compile Client for Windows with a note on that limitation. I feel your pain in discovering it afresh - it took me some time to work out what was going wrong, too! |
|
1656)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93245)
Posted 20 Oct 2019 by Richard Haselgrove Post: I think it's safe for you to come out now. |
|
1657)
Message boards :
Questions and problems :
Screen "blanker" is being deactivated by BOINC activity
(Message 93229)
Posted 16 Oct 2019 by Richard Haselgrove Post: And also your hardware setup. I too have 'desktop' computers with LCD screens (seven computers, four screens, and a KVM). I've never seen an LCD turn itself back on because of software activity, after the power switch has been turned off. |
|
1658)
Message boards :
Questions and problems :
permission problem: if client cannot run the app (wrong owner) why does it delete the work unit?
(Message 93200)
Posted 14 Oct 2019 by Richard Haselgrove Post: I never understood why it is possible to spoof the number of cpu cores either. I think from memory when I tried it to see what happens it resulted in tasks crashing but that may just be my experience. I didn't actually try to download extra work with it though.It's always been described in the documentation as a programming and debugging aid for the project scientists that develop the apps to run under BOINC. Remember that BOINC was being developed in the earliest years of the new millenium: I guess that at that time, a research postgrad programmer would feel pretty blessed to be offered a Pentium D computer. But their institutions would have been looking forward to the coming developments in multi-core hardware and multi-threaded software. <ncpus> would have been provided to allow simulation of multi-core server or workstation hardware. It would have run unbearably slowly, and it would have thrashed memory as program contexts were swapped in and out - but it was better than nothing. We don't have any equivalent need for multi-GPU simulation. I've never heard of a scientific programming language (CUDA, OpenCL, or whatever) utilising SLI or CrossFire: BOINC (currently) doesn't have any support for any equivalent of multithreading in the GPU arena. Except, of course, for the zillions of kernels that are executing in parallel on the hardware within each single GPU. That's enough. The only pressure for spoofing GPUs has come, again, from the SETI@Home project - and the perceived need arises, again, from the self-defence mechanisms which that project has felt necessary to wrap around itself to keep the hordes of users at bay. SETI normally has a relatively steady eight million tasks stored in its database. They come and go at the rate of about 35 a second - day in, day out. Just roll those figures around on your tongue for a moment. Their servers can cope - just, most of the time. But they do try quite hard to stop things getting out of hand. There is a spoofed GPU client in existence, but its use is tightly restricted to a team which helps Eric Korpela (project scientist) with hardware and fundraising. I've been granted permission to test it out, and I can see the problem: my machine is a fairly standard commercial gaming machine, and it spits out a task every 50 seconds with no non-standard hardware at all. It's currently at #55 in the 'top hosts' list at SETI. Cost? under 1,000 GBP for the basic machine (assembled and 3-year warranty), plus another 270 GBP for a second GPU. Just the two in total. The only advantage of spoofing the client on a machine like that (it reports 16 GPUs) is to ride out the maintenance outages in a production environment. Unlike the ncpus spoof, it confers no programming assistance: I think GPU spoofing won't be accepted into production versions of BOINC any time soon. |
|
1659)
Message boards :
Questions and problems :
permission problem: if client cannot run the app (wrong owner) why does it delete the work unit?
(Message 93193)
Posted 14 Oct 2019 by Richard Haselgrove Post: I'd be a little bit careful about taking this discussion too far. BOINC is designed to work quietly and automatically in the background. If things go wrong, there are automatic recovery processes in place - like cleaning up lost tasks when no application can be found to run them. If you choose to drive it manually - you own your own mistakes. Hopefully, that teaches you not to make the same mistake next time. In this case, a mistake caused the loss of a number of tasks from a cache. Shit happens. On most projects, BOINC would automatically recover: Job retransmission is a normal and recommended project option: <resend_lost_results> 0|1 </resend_lost_results>In this case, the tasks were lost from the SETI@Home project. SETI happens to be a special case where "this will increase the load on your DB server" applies - it's been tried, and deliberately turned off because the load was unmanageable. As people have found over the years, there's a loophole - quite possibly a programming bug - which enables lost tasks to be resent by using the 'ghost task recovery protocol'. Keith's protocol is benign: it relies on precise timing, but doesn't involve editing any files. [editing files usually causes more problems than it solves] But as the discussion and revised protocol around 30 August makes clear, the actual trigger for the loophole is reporting the same task for a second time. Keith achieves this by deliberately not listening for the scheduler reply which contains the acknowledgement. But the converse is: not hearing a scheduler reply is exactly what causes a task to be 'ghosted' in the first place. If you miss a step in the protocol, you risk making things worse, rather than better. I'd be very careful about automating this, outside the control of a tight user group. If an automated tool became widely available and increases the server load as before, it could cause the devs to hunt down and fix the server programming bug which opened the loophole in the first place. But if you do want to automate it (which I did once consider doing myself), I think I'd handle it by discarding one of the 'acks' in the scheduler reply, so that the client retains one of the 'ready to report' tasks for re-reporting. You must NOT do this on every report: the DB load is caused by checking every entry in your (possibly long) list of other tasks in progress. Only do this when you have to - so it needs a manual trigger and automatic 'return to normal' afterwards. |
|
1660)
Message boards :
Questions and problems :
LHC@Home, GPU, and BOINC ProjectList
(Message 93170)
Posted 12 Oct 2019 by Richard Haselgrove Post: Unfortunately, the project page is generated automatically from the plan class names of the project's applications. LHC have a couple of applications they have named native_mt and native_theory. The word native contains the letters ATI, and that's what triggers the false positive GPU report. The same thing happens in the 'Attach to project' wizard in BOINC Manager, because that's driven from the same automatic list. |
|
1661)
Message boards :
BOINC client :
How to control number of CPUs assigned per task
(Message 93164)
Posted 11 Oct 2019 by Richard Haselgrove Post: Yes, you can use an app_config.xml file, documented in the User Manual The tasks that use all cores are called Multi Threaded tasks: the example given in the manual covers that case. You have to identify exactly which project and application version is associated with these tasks: it's possible that the project's own website has a discussion of the problem and example files. Then, fill in the lines for both <avg_ncpus> and <cmdline>--nthreads with the number of cores you wish to use. |
|
1662)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 93147)
Posted 9 Oct 2019 by Richard Haselgrove Post: British take on the power situation - reasons and outcomes. https://www.bbc.co.uk/news/world-us-canada-49982236 Having driven through the area affected by the Carr Fire on my American trip this summer, I have a slightly better idea of the scale and scope of these events - even though, a year after the burn (we drove through on the anniversary), much of the undergrowth was already re-greening. |
|
1663)
Message boards :
BOINC Manager :
What module contains the code to detect when the host computer has shifted to battery power?
(Message 93128)
Posted 8 Oct 2019 by Richard Haselgrove Post: I think it would be better to continue this discussion in the 'BOINC client' version of this thread. |
|
1664)
Message boards :
Questions and problems :
About the Manager 7.16.3
(Message 93104)
Posted 6 Oct 2019 by Richard Haselgrove Post: ... something was changed who make that stop working.Previously, clicking the X caused 'minimise to system tray'. Now that the system tray icon has been abolished on Linux builds, my guess is that the replacement pathway doesn't handle all scenarios cleanly. I'll try to take a look this afternoon. |
|
1665)
Message boards :
BOINC client :
What module contains the code to detect when the host computer has shifted to battery power?
(Message 93071)
Posted 5 Oct 2019 by Richard Haselgrove Post: Ideally, you would be querying their module via an API, rather than reading from a file (even though that's the antedeluvian method BOINC developers prefer). Please ensure that any code patches you supply are as generic as possible across the *nix domain, not just limited to the 'U' variant. Apart from the laptop case (has anyone tested that under Linux?), the devs will mainly think of UPSs in the context of servers powered by Scientific Linux (becoming obsolete - forget that one), CentOS, Redhat etc. |
|
1666)
Message boards :
BOINC client :
What module contains the code to detect when the host computer has shifted to battery power?
(Message 93067)
Posted 4 Oct 2019 by Richard Haselgrove Post: I think you need to start reading from https://github.com/BOINC/boinc/blob/master/client/hostinfo_unix.cpp#L195 There is a separate 'client/hostinfo_linux.cpp', but clearly 'on battery' isn't a "move common linux detection functions to separate file." |
|
1667)
Message boards :
BOINC Manager :
What module contains the code to detect when the host computer has shifted to battery power?
(Message 93062)
Posted 4 Oct 2019 by Richard Haselgrove Post: Sorry, I've had an unexpectedly busy day. I'll try and find time over the weekend to have a ferret around. |
|
1668)
Message boards :
BOINC Manager :
What module contains the code to detect when the host computer has shifted to battery power?
(Message 93059)
Posted 4 Oct 2019 by Richard Haselgrove Post: Detection will be done in the client. The Manager is simply a 'command and control' module, allowing the user to see data and configure settings (like whether battery status is allowed to change processing state). More seriously, I don't know enough about the internal workings of either Linux or Windows to verify whether the 'external UPS' case is handled through exactly the same pathway as the 'internal laptop primary power' case. That needs nailing down first. |
|
1669)
Message boards :
Questions and problems :
how to get closest deadlines computed first
(Message 93018)
Posted 1 Oct 2019 by Richard Haselgrove Post: It's not random. BOINC operates a strict 'First in, First out' policy - although that's hard to see once you've sorted the tasks into some different order. If any task is actually in danger of missing its deadline (and BOINC keeps a pretty strict watch on that), it will be promoted above its normal place in the queue. |
|
1670)
Message boards :
Questions and problems :
Manualy copy *.exe and *.dll to project folder?
(Message 93003)
Posted 30 Sep 2019 by Richard Haselgrove Post: I've noticed that the "client_state.xml" file has a record of exe & dll files downloaded successfully or not. Could that file be edited to tell Boinc that the files are downloaded? Does Boink parse that file on startup, or is it more a record the state, with no feedback to what might need to be done?I'd advise you not to try editing that file manually. It's the complete record of "Life, the Universe and Everything" as far as BOINC is concerned - a single slip and you're back to square one. In the SETI Q&A, you were advised by Keith Myers to consider downloading and running the Lunatics application installer. I wrote that, so I recuse myself from advising it - but it's the simplest way of achieving a comprehensive set of compatible applications. |
|
1671)
Message boards :
Questions and problems :
7.16.2 BOINC Installer.app on Mojave
(Message 92993)
Posted 29 Sep 2019 by Richard Haselgrove Post: BOINC v7.16.2 has been withdrawn. Please wait for v7.16.3, probably tomorrow. |
|
1672)
Message boards :
BOINC Manager :
boincmgr will not start on Ubuntu LINUX when left over file /home//BOINC Manager-
(Message 92986)
Posted 28 Sep 2019 by Richard Haselgrove Post: Having sorted out my networking, I can confirm that 'New BOINC Manager window...' is working for me, with both the full-feature v7.16 private build, and with the cut-down v7.16 PPA repo version. Attaching to Windows hosts, I seem to have to connect by IP address first, but after I've done that, attaching by name seems to work - for the current session, at least. |
|
1673)
Message boards :
BOINC Manager :
boincmgr will not start on Ubuntu LINUX when left over file /home//BOINC Manager-
(Message 92985)
Posted 28 Sep 2019 by Richard Haselgrove Post: I tried the "File -> New BOINC Manager window" option and not surprisingly it doesn't work.I just tried the same thing on Mint with a v7.16.1 manager, and it did work - sort of. A new blank window opened, but I couldn't get it to connect to any of my other (Windows) machines by name. I'll look up some IP addresses, and check my remote access settings, and let you know how I get on later. Edit - mind you, that was with a private build Manager with some of the missing features put back in. I'll test with the repo Manager, if I can find a copy. |
|
1674)
Message boards :
The Lounge :
That traveling Dutchman is coming back to Britain and is no longer going to Germany
(Message 92983)
Posted 28 Sep 2019 by Richard Haselgrove Post: Mine sure isn't "The South", either. But Jord went on up to Acomb (Northumberland), not the better-known Acomb (York). |
|
1675)
Message boards :
Questions and problems :
Manualy copy *.exe and *.dll to project folder?
(Message 92975)
Posted 27 Sep 2019 by Richard Haselgrove Post: I'm not quite sure why BOINC would delete those files - they should be persistent. It might happen if the versions you installed manually were slightly different (different signature, perhaps) from the ones supplied by the server. Does the Event log show any errors at startup? The best solution I can think of is to declare the files in an app_info.xml file, under the Anonymous Platform procedure. It's a bit messy to set up, although there are installation packages which would help - ask on the SETI message boards. You would have to do the same thing, probably without external help, for any other project you decide to join in the future. |
|
1676)
Message boards :
BOINC Manager :
boincmgr will not start on Ubuntu LINUX when left over file /home//BOINC Manager-
(Message 92973)
Posted 27 Sep 2019 by Richard Haselgrove Post: Excellent - you've just answered a problem I had about an hour ago, before I'd even asked! In my case, the cause was slightly different: I'd absent-mindedly closed the BOINC Manager with the green (for Mint - red for most other distros) cross in the top-right corner. With a v7.16.1 Manager, this is known to be problematic - it's better to use the file menu 'Close' or 'Exit' (if available). But the end result was the same - I couldn't reopen the Manager. Same result from strace (under my own account - running as admin had, er, 'interesting' results). rm'd the file you indicated, and it works fine again. That one's going in the memory banks. |
|
1677)
Message boards :
The Lounge :
That traveling Dutchman is coming back to Britain and is no longer going to Germany
(Message 92972)
Posted 27 Sep 2019 by Richard Haselgrove Post: Not to worry - our Yorkshire stone is hard enough to take those sort of knocks. I've done exactly the same (it's almost a ritual whenever I get a new car), and so has everyone else. My neighbour's house is still standing, 150 years on. I was sorry to hear about your head this morning. I hope rest and clean sea air will help to clear it before your birthday. |
|
1678)
Message boards :
Questions and problems :
BOINC preferences
(Message 92967)
Posted 26 Sep 2019 by Richard Haselgrove Post: But it does seem to be working now, although it would have been nice to know just what was going on before.To do that, you would have needed to enable the logging and capture the data before destroying the previous installation. |
|
1679)
Message boards :
Questions and problems :
BOINC preferences
(Message 92965)
Posted 26 Sep 2019 by Richard Haselgrove Post: It might be helpful to post the startup lines from the Event Log, where the CPU data would normally be registered following BOINC detection. It would also be helpful to enable the <sched_op_debug> Event Log option before your next work fetch attempt, and post the output from that: without being too verbose, that does provide more information about what the client is requesting, and the server response. |
|
1680)
Message boards :
The Lounge :
That traveling Dutchman is coming back to Britain and is no longer going to Germany
(Message 92901)
Posted 22 Sep 2019 by Richard Haselgrove Post: Actually, it's just gone very dark here. The equinoctial rainstorm has arrived, with a vengeance. |
|
1681)
Message boards :
Questions and problems :
Running Boinc GUI Manager from Headless Ubuntu 18.04
(Message 92893)
Posted 21 Sep 2019 by Richard Haselgrove Post: First check that the client is running on the headless machine (the client is 'boinc', rather than 'boincmgr'). Something like top should tell you. Once boinc is running - essential for boincmgr to connect to it - review Controlling BOINC remotely and check that you've gone through all the procedures necessary for a remote connection to work. |
|
1682)
Message boards :
The Lounge :
That traveling Dutchman is coming back to Britain and is no longer going to Germany
(Message 92882)
Posted 21 Sep 2019 by Richard Haselgrove Post: Just been outdoors for three hours plus: walking into town, eating lunch al fresco in the outdoor cafe courtyard, walking back. Bright sun, clear blue sky, literally not a cloud in the sky. Walked back past the local canal boat club, who had chosen today to celebrate their 60th. birthday. They obviously know a thing or two about local seasonal weather patterns. But Monday is the autumn equinox, and the usual disturbances are due to arrive - as they do every year, and have done since long before I had any role in influencing them. Another temptation for the weather gods is that the 2019 UCI Road World Cycling Championships start tomorrow in Harrogate, just up the road - though they have shown leniency for the para-cyclists today. There are many road closures, so be careful not to overshoot Wetherby: the route we worked out in the Spring should be clear, although it's likely to be busy with out-of-towners trying to find their accommodation and a good viewing spot. Whatever you do, don't go north of the River Wharfe: after what happened last time, you may not be able to get back for a year or more! |
|
1683)
Message boards :
Questions and problems :
boinc.berkeley.edu/dev [Bug Reports] takes you to https://boinc.berkeley.edu/dev/trac/wiki/ReportBugs which is not found
(Message 92858)
Posted 20 Sep 2019 by Richard Haselgrove Post: The pageHow did you find yourself on https://boinc.berkeley.edu/dev/? The normal landing page is https://boinc.berkeley.edu/, and if you follow the same link, you end up at https://boinc.berkeley.edu/trac/wiki/ReportBugs, which does work. Or you can post your bug here, and we can take it from there. Edit: there are several other links to /trac/wiki/ which inherit your rogue /dev/, and similarly can't be found. It would probably be best to remove (or redirect) the duplicate copy of the front page. |
|
1684)
Message boards :
Questions and problems :
About the Boinc 7.16.1
(Message 92855)
Posted 20 Sep 2019 by Richard Haselgrove Post: Github has been updated overnight to v7.16.2 That includes the fix we're talking about here, plus at least 60 others. There is currently a delay in updating the translations, so it isn't yet ready to begin full release testing - but English speakers with a sense of adventure are welcome to try rolling their own. Caveat emptor. |
|
1685)
Message boards :
Questions and problems :
Unexpected behaviour
(Message 92854)
Posted 20 Sep 2019 by Richard Haselgrove Post: You don't mention which version of BOINC you're using. This sounds like a problem I first reported with BOINC v7.6 three years ago (although it probably dates back to the first code allowing the use of max_concurrent in app_config.xml). Keith Myers hit the same problem last year, and with help from David Anderson, we think that the problem will be solved in v7.16 (several of us have been testing the code for months). After the conference call yesterday, I'm hopeful that the release of v7.16 will finally get underway in the next few days - although an email from David this morning, asking for help, suggests that he fell at the first hurdle. |
|
1686)
Message boards :
Questions and problems :
About the Boinc 7.16.1
(Message 92850)
Posted 19 Sep 2019 by Richard Haselgrove Post: Well, the right questions were asked and acknowledged, so I didn't have to go nuclear. We might see some progress towards v7.16.2 soon. |
|
1687)
Message boards :
Questions and problems :
About the Boinc 7.16.1
(Message 92848)
Posted 19 Sep 2019 by Richard Haselgrove Post: Edit: I inadvertently downloaded 7.15 so perhaps not.'master' is always indicated as a development version by the odd number in the middle position. It should have been revised to v7.17 by now, to indicate the new post-v7.16 cycle, but that hasn't been done either. |
|
1688)
Message boards :
Questions and problems :
About the Boinc 7.16.1
(Message 92847)
Posted 19 Sep 2019 by Richard Haselgrove Post: Yes.and if the offender doesn't pull his finger out and fix itBecause I wasn't sure what happens when you fiddle with things I downloaded a fresh copy of boinc-master.zip and the offending line seems to have been fixed. Master good, release branch bad. That's my main grouse: the bug was fixed on 13 August and merged into master the same day (#3249). It should have been copied into the release branch within hours of the patch being tested and approved - which you can see I did, again on the same day. You can also see that LoctutusOfBorg and Germano updated their releases within three days. Yet here we are, over a month later, and it still hasn't been done. Protocol says the Release Manager should do it: I did that role once, so I know the form. And I know it's been disregarded. See https://github.com/BOINC/boinc-policy/blob/master/Development_Documents/Client_Release_Process.md, especially the 'Process' section and the 'Prepare for a Release' and 'Release Branch' sub-sections. Back to your home build: if you simply took that one line, or even the whole file, from a 'master' download, you'll be fine. If you did the whole build from master, you run the risk of some other bug, as yet undiscovered, catching you and biting you on the ankles. See, for example, #3292: it took David almost a fortnight to notice and correct that one, but at least it was caught before it left master. |
|
1689)
Message boards :
Questions and problems :
About the Boinc 7.16.1
(Message 92845)
Posted 19 Sep 2019 by Richard Haselgrove Post: The fix is trivial, and you can do it on your own machine in seconds. Remove the comma from the end of https://github.com/BOINC/boinc/blob/client_release/7/7.16/lib/cc_config.cpp#L506 That's - Folder: lib File: cc_config.cpp Line: 506which currently reads " <allow_gui_rpc_get>%d</allow_gui_rpc_get>\n",I'm trying to play this by the book, but last fortnight I had my editor loaded with the offending line during the developer conference call. I'll do the same tonight, and if the offender doesn't pull his finger out and fix it (as he's been told to), I will break protocol and fix it myself. That may result in the removal of my privileges to edit code, but they will know what they are doing. |
|
1690)
Message boards :
Questions and problems :
About the Boinc 7.16.1
(Message 92841)
Posted 18 Sep 2019 by Richard Haselgrove Post: That's odd: after cleaning up the duplicates, I count six visible posts before this one. But the index said two before I came back for this comment. |
|
1691)
Message boards :
Questions and problems :
About the Boinc 7.16.1
(Message 92833)
Posted 17 Sep 2019 by Richard Haselgrove Post: Thanks. I can try to replicate tomorrow, and see if I can find anything in the code that would explain it. And if so, I can report it upstairs. Just before I try to replicate, do you use any command line parameters on either the client or the manager launches? |
|
1692)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92830)
Posted 17 Sep 2019 by Richard Haselgrove Post: I've just come back in after a day out in the late summer fresh air. I can go back out again if need be. |
|
1693)
Message boards :
Questions and problems :
About the Boinc 7.16.1
(Message 92818)
Posted 16 Sep 2019 by Richard Haselgrove Post: I rolled my own a few weeks ago so there may have been changes between when I got mine from git-hub. Don't know where yours came from so it could be an earlier or later build.There have been no effective updates to the v7.16.1 branch since it was made on 30 July. But there is one horrible bug in the Manager code which you need to correct if you built that as well. It's an easy fix - remove one stray comma from a library file - but the side effects are critical. See #3247, #3249 |
|
1694)
Message boards :
Projects :
CPDN URL changing.
(Message 92800)
Posted 13 Sep 2019 by Richard Haselgrove Post: David has now changed his mind as to the cause. A new fix is in place, and should propagate to all in 24 hours or so. |
|
1695)
Message boards :
Questions and problems :
New Windows install no reading app_config.xml
(Message 92797)
Posted 13 Sep 2019 by Richard Haselgrove Post: From the file location, I see you're running Windows. How did you prepare the file? The usual advice is to use the Windows simple text editor 'Notepad', but by default this produces files with the '.txt' extension - so you might have created 'app_config.xml.txt' by mistake. And then Windows hides the .txt extension in File Manager, and simply describes the file as a 'Text Document' instead. To be certain, use Control Panel (Folder Options or File Explorer Options, depending on your version of Windows): on the View tab, uncheck "Hide extensions for klnown file types". That will show you exactly what's going on, and you can edit the file name to remove the extra '.txt' if that is indeed the problem. |
|
1696)
Message boards :
Projects :
CPDN URL changing.
(Message 92789)
Posted 12 Sep 2019 by Richard Haselgrove Post: David Wallom says the team will investigate during office hours tomorrow.and As we are seeing messages about a number of services I am suspecting network equipment problems rather than the underlying CPDN servers.I'm not sure I follow the logic of that last one. |
|
1697)
Message boards :
Projects :
CPDN URL changing.
(Message 92784)
Posted 12 Sep 2019 by Richard Haselgrove Post: I just tried pasting https://www.cpdn.org/ into a browser, and got This site can’t be reached www.cpdn.org took too long to respond.The usual browser rubbish can be ignored - I know I'm online because it's spellchecking as I type! Edit - while I was typing that, the browser did connect - sort of. I got a very crude HTML representation of the front page (no style sheet formatting), and the error message Warning: file_get_contents(https://www.cpdn.org/menu_boinc.xml): failed to open stream: Connection timed out in /home/boinc/cpdnboinc/html/inc/header.inc on line 9 |
|
1698)
Message boards :
Projects :
CPDN URL changing.
(Message 92782)
Posted 12 Sep 2019 by Richard Haselgrove Post: I'd enforce contact with the scheduler first (Manager, Projects tab, update) and scour the Event Log for the proper messages - including 'using old url'. The current CPDN crew are still pretty green where BOINC is concerned: I wouldn't trust them to get a difficult change like that right until I see the results for myself. Following my own advice on this machine just now, I got 12/09/2019 18:53:27 | climateprediction.net | Scheduler request failed: Error 403- but admittedly this machine hasn't followed all the previous changes. |
|
1699)
Message boards :
Projects :
News on Project Outages
(Message 92778)
Posted 12 Sep 2019 by Richard Haselgrove Post: GPUGrid.net is back up, but I needed to retry a lot of backed-off comms. |
|
1700)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92765)
Posted 10 Sep 2019 by Richard Haselgrove Post: so reporting work at Einstein is still borked.See https://einsteinathome.org/content/project-down-4. We're getting the impression that Bernd mopped his brow, packed up his toolbag, and left the room - but forgot to flip the vital switch. |
|
1701)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92763)
Posted 10 Sep 2019 by Richard Haselgrove Post: Now reachable from here, and now exchanging data - reporting old work, issuing new work. |
|
1702)
Message boards :
Questions and problems :
BOINC And windows server 2016
(Message 92727)
Posted 7 Sep 2019 by Richard Haselgrove Post: Thanks for the report. I'll happily pass this upstream to the Windows Installer team, but I'd like a little more information first, please. Looking at https://www.boincstats.com/stats/5/host/breakdown/os/ (Host Operating System breakdown for Einstein@Home), I see multiple versions of Windows Server 2016 attached and (presumably) running - Standard, Datacenter, and Evaluation, in various combinations. But unfortunately, it's hard to assess externally whether they are running "the latest" (by which I presume you mean v7.14.2) version of BOINC, and we don't know whether they used a standard or custom installer. So, could you please post some technical details of how you discovered the MSI incompatibility: in particular, please could you post the text of any error messages and their associated error numbers, if any. Details of your particular server version would also be helpful. |
|
1703)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92713)
Posted 5 Sep 2019 by Richard Haselgrove Post: Grumble. My email provider, British Telecom, subcontracts its email service to Yahoo. BT says Yahoo is experiencing a global outage - but I'm not their customer, so I've got nobody to send a complaint email to. Even if I could send it... So, no email receipts here. Find another way. Later - email is visible again, though my inbox is mercifully quiet. The problem has been noticed: Yahoo email fault hits BT, Sky and TalkTalk customers |
|
1704)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92707)
Posted 4 Sep 2019 by Richard Haselgrove Post: Funny that. The email seems to have gone astray, and my PM inbox is empty. But I've got a little list... ;-) |
|
1705)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92700)
Posted 4 Sep 2019 by Richard Haselgrove Post: And he fixed a minor bug for me over the weekend. That felt like displacement activity, too: sidetracking from the much more important issues which are still backing up. Like the stalled v7.16.1 release - which was something that David himself was champing at the bit for, months ago. |
|
1706)
Message boards :
Questions and problems :
unable to install on Windows 10.
(Message 92669)
Posted 2 Sep 2019 by Richard Haselgrove Post: Now I have it installed, but it will not start. The error is "The application was unable to start correctly (0xc000007b".Ouch. 0xC000007B(from MSDN NTSTATUS Values) I think you need to review whatever steps you took to force it into installing. |
|
1707)
Message boards :
BOINC Manager :
Another BOINC isn't sleeping when computer in use issue.
(Message 92661)
Posted 1 Sep 2019 by Richard Haselgrove Post: The screenshot is working here, but it's on an advertising-supported website - it's possible an ad-blocker or similar might interfere with display. I see a single task 'uploading (suspended - computer is in use)', plus many others either 'Running' or 'Waiting to run'. Most of the running tasks are from Asteroids@home, plus one Rosetta@home: the waiting tasks are Rosetta, World Community Grid, and one Einstein. It looks as if networking is being suspended, but computing isn't. BOINC doesn't have separate 'computer in use' detection mechanisms for the two functions, so I'd zero in on the 'Activity' menu - is the top ('Run') section set to 'always' or 'based on preferences'? |
|
1708)
Message boards :
Questions and problems :
How to move project files to another HD
(Message 92656)
Posted 1 Sep 2019 by Richard Haselgrove Post: This has in fact been discussed many times before. The most recent time I wrote it up was in message 90919. Following those instructions will achieve what you want. |
|
1709)
Message boards :
Questions and problems :
unable to install on Windows 10.
(Message 92648)
Posted 31 Aug 2019 by Richard Haselgrove Post: Is your 'Enterprise' copy of Windows actually running as part of an Enterprise where Group Policy restrictions might have been deployed via a Domain Controller? If so, please explain as much of the policy as you know and are allowed to tell us. |
|
1710)
Message boards :
Questions and problems :
Windows install issues
(Message 92633)
Posted 29 Aug 2019 by Richard Haselgrove Post: What about the question on the cc_config.xml file?cc_config.xml is completely optional, and not created by default. Make one if you feel the need to run something non-standard, once you get the client working. |
|
1711)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92626)
Posted 28 Aug 2019 by Richard Haselgrove Post: Jord discovered, late on a Friday afternoon, that data was being transferred horribly slowly from device to device (phone to tablet? Something like that) and he wanted to free up space before the following morning. Both devices had USB-C sockets, but neither of us had a cable to hand. By the time we'd scratched our heads, compared prices etc., it was nearly 18:30 - which cuts down the 'same day collection' options quite severely. To my surprise, the UK retail chain 'PC-World' had one at a reasonable price, and in stock at the second-nearest branch. So we took a quick hop across the moor to Guiseley, in contraflow to the remains of the evening rush hour. And back in time for dinner. |
|
1712)
Message boards :
Questions and problems :
Daily quota
(Message 92579)
Posted 25 Aug 2019 by Richard Haselgrove Post: They come into play (automatically) when you return a sequence of errored or invalid results. Check your machine, perhaps restart it, consult other users of Universe@Home. In short, try to work out what the problem is, and fix it. You can still be allocated one task per day. If you return that successfully, and it's valid, your quota will gradually be raised as you return successive valid tasks. |
|
1713)
Message boards :
Questions and problems :
need help debugging a problem: Linux 7.16.1
(Message 92573)
Posted 24 Aug 2019 by Richard Haselgrove Post: I'd be prepared to place a small bet that this is a hardware problem, not related to the software version (of either BOINC or SETI) in use. |
|
1714)
Message boards :
Questions and problems :
need help debugging a problem: Linux 7.16.1
(Message 92570)
Posted 24 Aug 2019 by Richard Haselgrove Post: Mine's a Corsair 650W 80+ bronze (according to the invoice), and hasn't had time to degrade - much - yet. The 360W draw is measured: I think that's comfortably below the 80% line (520W) I should be able to sustain without ill effects. But the question I was asking was more about the cable paths and how much power they could sustain. If I'm imagining the concept of a 4-in-1 splitter correctly, that device should be limited to 75W from the motherboard: the connected cards might be limited to 25W each if the slots are wired as x4, or might be limited to half (37.5W) each. Either way, they'll be underpowered compared to what NVidia was expecting. |
|
1715)
Message boards :
Questions and problems :
need help debugging a problem: Linux 7.16.1
(Message 92568)
Posted 24 Aug 2019 by Richard Haselgrove Post: BOINC checks the presence, capabilities, and status of coprocessors at startup. It is not designed as a continuous hardware health monitor: other tools are available for that. As you have discovered, petri33's "Cuda 9.00 special" app for SETI is one such tool, drawing on earlier work contributed by JasonG and Raistmer. The sample stderr you have linked shows the problems go deeper than you describe. It starts with repeated iterations of setiathome_CUDA: CUDA Device 5 specified, checking... Device cannot be used Cuda initialisation FAILED, Initiating Boinc temporary exit (180 secs) setiathome_CUDA: Found 3 CUDA device(s):implying that TWO of your five cards have become inoperable since the task was originally assigned to device 5. One thing to remember is that Petri's CUDA 9 app is designed to drive the cards as hard as possible (well, not quite as hard as the CUDA 10 app I've recently started using, but close). That much strain is highly likely to seek out and expose any weaknesses in the hardware and power supply in use. The PCI Express Power specification allows: 25W (maximum) from PCIe x1 or x4 slots 75W from PCIe x8 or x16 slots 75W from 6-pin auxiliary connectors 150W from 8-pin auxiliary connectors My dual GTX 1660 Ti machine is currently drawing about 360W from the wall, falling to a little over 300W when the CPU is idled. The rest of the machine is pretty minimal (single SSD, no mechanical or optical drives), so the GPUs must be drawing nigh on 150W each. With bus power and an 8-pin connector each, they're well supplied - and have shown no sign of power stress. In the past, you've described using riser cables from < x8 PCIe slots. I don't have access to the full rig specifications, so you'll have to do the power maths yourself - but my instinct would be to do a full power audit: how much do the cards demand under load, and how is each card powered (what combination of auxiliary connector, direct feed from slot, feed from slot via riser cable, feed (if any) from PSU to outboard riser slot). Tot up total draw, total connector capacity, and total supply capability. Make sure that each total is bigger than the previous one. And that's all before we consider cooling and exhaust air impact on neighbouring cards. |
|
1716)
Message boards :
Questions and problems :
Hard drive spinning very fast
(Message 92566)
Posted 24 Aug 2019 by Richard Haselgrove Post: We could also read "hard drive to spin so fast" as a synonym for "hard drive to spin so often" - in other words, lots of activity shown by the hard disk activity LED. In that case, the answer would be 'checkpointing' - writing intermediate result data to disk, so that if processing is interrupted for any reason, it can re re-started from somewhere close to the point previously reached. That can save a tremendous amount of wasted computation on long tasks. |
|
1717)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92527)
Posted 19 Aug 2019 by Richard Haselgrove Post: The sample email Jord sent to me had a hanmail.net email address - possibly Chinese, but registered in Korea according to a whois lookup. Unfortunately, I can't prevent either Chinese or Korean users putting my given name into the 'From' field of a spam email account - we could all do that, although 99.99999% of us wouldn't. I'm sorry for Jord's sake that he seems to have attracted a stalker or spammer who has decided that an email ostensibly from me is more likely to bypass his natural caution, but that's the world we live in. |
|
1718)
Message boards :
Questions and problems :
What happened to "requested" and "granted" credits?
(Message 92517)
Posted 17 Aug 2019 by Richard Haselgrove Post: It was removed from the BOINC templates around 2010. |
|
1719)
Message boards :
Questions and problems :
Offline Repeated Client Execution
(Message 92444)
Posted 9 Aug 2019 by Richard Haselgrove Post: For what you're asking, you don't want to run the BOINC client in offline mode: you want to run the science applications developed and distributed by your chosen project. Those applications are the ones which produce the scientific outputs, which are indeed worth comparing for both speed and precision. The third-party developers for SETI@Home have a bench-test suite - basically a complex Windows batch file, plus some utilities - which allow testers to run comparison tests for arbitrary versions of the science application against arbitrary test or selected datafiles. The script records comparative timings, and the utilities measure the quality of the results (a process similar to project validation). So it can be done, but it would be specific to the project which you have in mind to study. The validation utility, in particular, might need to be re-coded from scratch to mimic the requirements of your chosen project. |
|
1720)
Message boards :
GPUs :
Seems SLI is causing GPU problems
(Message 92441)
Posted 9 Aug 2019 by Richard Haselgrove Post: And please edit the thread title to make it clear to future readers that it was a project software problem, not SLI, that caused your errors. |
|
1721)
Message boards :
BOINC client :
Seems 7.16.1 has been released but not shown on download page
(Message 92434)
Posted 9 Aug 2019 by Richard Haselgrove Post: Effectively, Gianfranco's PPA is "the" bleeding-edge test repository for BOINC under Linux. By adding that PPA, you are giving prior consent to receiving unconfirmed code. Having said that, Gianfranco does build from official release branch code. The v7.16.1 release code branch was forked (without announcement) 10 days ago, but as yet no formal testing processes have been initiated. I installed the same version myself this morning, and I've initiated some conversations with other members of the development team. That's as much as I'm prepared to say until I hear back from the other developers. |
|
1722)
Message boards :
BOINC client :
Cannot find release 7.15.0
(Message 92392)
Posted 6 Aug 2019 by Richard Haselgrove Post: You're in luck - there are still dependencies for VS2005 listed at https://boinc.berkeley.edu/trac/wiki/SourceCodeGit#Windowsbuilddependencies. David A is still building releases with VS2010 to maintain compatibility with Windows XP. Edit - with those old versions, you may need to update some files, especially ca-bundle.crt to maintain secure link compatibility with modern servers like, in particular, SETI. |
|
1723)
Message boards :
BOINC client :
Cannot find release 7.15.0
(Message 92388)
Posted 6 Aug 2019 by Richard Haselgrove Post: Haven't managed to get visual studio installed under WINE yet to see if it is possible to do Windows builds on a Linux box. How recent a version do you need. All the more recent ones are ranked as garbage on the WINE site.I'm building with VS2013 - if you can get hold of it. Microsoft doesn't tend to keep superseded versions hanging around for ever. https://boinc.berkeley.edu/trac/wiki/CompileClient#Windows is pretty good. I've added a couple of gotchas from my own experience. |
|
1724)
Message boards :
BOINC client :
Cannot find release 7.15.0
(Message 92386)
Posted 6 Aug 2019 by Richard Haselgrove Post: Well, after some confusion between openssl and libssl-dev, a lot of purple (warnings) and cyan (also warnings), I think I have a v7.16.1 client - no red! I'm not going to test it until after the SETI outage (when I can flush my - excessive - cache), but watch this space. |
|
1725)
Message boards :
BOINC client :
Cannot find release 7.15.0
(Message 92385)
Posted 6 Aug 2019 by Richard Haselgrove Post: I've checked three different ways now: 1) Fresh clone into a new folder in my Windows build tree 2) Straight ZIP download in Linux 3) Inspect file contents directly on GitHub I've built new clients under Windows: the machines I've deployed those builds on are showing v7.16.1 in the status line and online. I haven't got the Linux client to build yet - going to give that another go today. But I've only ever seen v7.15.0 when downloading/building from master - deliberately or otherwise. |
|
1726)
Message boards :
BOINC client :
Cannot find release 7.15.0
(Message 92382)
Posted 6 Aug 2019 by Richard Haselgrove Post: downloaded the boinc_client zip from github for 7.16.1 (but version.h had 7.15.0)If version.h had 7.15.0, then you didn't download v7.16.1 The version numbers in the copy I downloaded yesterday were all correct: they were updated six days ago by https://github.com/BOINC/boinc/commit/42a488b331e8f8e203691d5cbb036e81b8c8fd13 But thanks for the build instructions: I'll give those a try later today. |
|
1727)
Message boards :
BOINC client :
Cannot find release 7.15.0
(Message 92378)
Posted 5 Aug 2019 by Richard Haselgrove Post: As make runs again, I am just getting more and more impressed at anyone who can manage to build BOINC from source at least on the penguin platform!That is a delight I am reluctantly gathering myself to address. The Windows build instructions, by way of contrast, are a model of clarity, and from memory I only had to work round one single outdatedness. |
|
1728)
Message boards :
BOINC client :
Cannot find release 7.15.0
(Message 92370)
Posted 5 Aug 2019 by Richard Haselgrove Post: Well, up to a point. Until a couple of days ago, my computer list showed several v7.15.0, and one v7.13.0 - most of the 7.15.0 have now been replaced by v7.16.1 so I can get a feel of how the new release code works and report any bugs. All of those odd builds were compiled by me, to develop or test fixes for the central code base. They will all have been different from any other build made from the master branch with different tweaks - you can't just point to a development version number and assume you know everything about it. I've supplied updates to the source code to tweak GPU detection too - but only for NVidia, most recently in #2707 for the RTX range. If there needs to be a similar fix for the "S9100" (Hawaii), let me know and I can submit it. |
|
1729)
Message boards :
BOINC client :
Cannot find release 7.15.0
(Message 92368)
Posted 5 Aug 2019 by Richard Haselgrove Post: The numbering policy is: Even numbers are used for releases Odd numbers are used for development work, not yet ready for release. If you see v7.15.0 out in the wild, it's a private build, made for testing and possibly with modifications not intended for general use. The next release branch is 7.16.1, and that can be selected under 'branches' in GitHub: I've made a private Windows build for testing, but I don't know of anybody else admitting to doing that. |
|
1730)
Message boards :
Questions and problems :
Ubuntu 19.04 no options available
(Message 92352)
Posted 1 Aug 2019 by Richard Haselgrove Post: I'm testing out a Linux Mint 19.1 Tessa box, which also has v7.9.3 as the default repository version. All the options are visible in the dialog there, so this isn't a generic BOINC Manager problem - might be a library issue on the OP's machine? |
|
1731)
Message boards :
Questions and problems :
Ubuntu 19.04 no options available
(Message 92350)
Posted 1 Aug 2019 by Richard Haselgrove Post: If you have never set preferences manually on those machines, they will still be running with web-based preferences. So you can set new values using the global "Computing preferences" page, in your account at any of the projects the machines are attached to. |
|
1732)
Message boards :
Questions and problems :
no running task while computer is not in use
(Message 92340)
Posted 30 Jul 2019 by Richard Haselgrove Post: Anyway I have still the the screenshot from the task list + the renamed old folder if it is needed. Just tell me what to upload and I'll do it.A screenshot of the task list is unlikely to help - unless it shows that all tasks are well past their deadlines, which would account for there being no allocated tasks on the server (as Gary says). Better to make a text copy of the messages in the Event Log, and post that instead. |
|
1733)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92274)
Posted 23 Jul 2019 by Richard Haselgrove Post: Will Richard be granted the opportunity to turn everything off?Interestingly, the front doors of the rack cabinets are secured by combination locks, but the back doors - where the power cables run - aren't. And I saw where Isaac lives. One good yank... But you'll be reassured to hear that I've moved on from San Francicso. Keith Myers treated us to dinner in Orland last night (many thanks, Keith), and we're off to visit the Redwood forests today. |
|
1734)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92255)
Posted 21 Jul 2019 by Richard Haselgrove Post: Grumble - the Amtrak trackside experience. This was truck about 80, of about 145 (didn't count exactly), and it sat there blocking the view up to and above the station roof for a good 10 minutes. 
|
|
1735)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92254)
Posted 21 Jul 2019 by Richard Haselgrove Post: Would it be the Club Quarters Hotel, Central Loop?And the Vagabond Inn in Sacramento has the slowest I've encountered all trip. But that might be just logging in and checking me over - seems to be better now. |
|
1736)
Message boards :
Questions and problems :
Possible to move "leftover" work units into a project directory for processing?
(Message 92244)
Posted 20 Jul 2019 by Richard Haselgrove Post: No. All you see in the .old folder will be the data files. You would need to re-create all the <workunit> and <result> metadata in client_state.xml as well. That would be almost impossible to do without making a mistake which would be even more problematic than the current situation. |
|
1737)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92237)
Posted 19 Jul 2019 by Richard Haselgrove Post: Also a gross misunderstanding as this isn't just anywhere.That is weird: none in Wales, just one in Scotland, a few in England, and all the rest in Northern Ireland. Which side of the fence will they be, when all the dust settles? |
|
1738)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92215)
Posted 18 Jul 2019 by Richard Haselgrove Post: Grumble - Well according to the tracker Richards train arrived over 2 hours late and is still in Denver.Yes to both of those. They do a full train service at Denver, so they can't shorten the 45 minute turnround - especially at high holiday season. when any sleeper vacated has to be turned round and readied for the next customer. Then we needed a new track slot, then the PTC computers stopped communicating with Amtrak, then a customer was taken ill just as we were leaving the station. But we completed the trip, and I'm safely tucked up in my hotel room in Oakland. Meeting Angela and seeing Eric again first thing in the morning. |
|
1739)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92183)
Posted 16 Jul 2019 by Richard Haselgrove Post: Looks like I'm getting the full Amtrak experience - now estimating arrival at 09:00. Which, frankly, I'm putting down as a glory: more time in that forest waterfall of a shower, and to repack properly with the extra 2 hours or whatever. |
|
1740)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92178)
Posted 16 Jul 2019 by Richard Haselgrove Post: Then we're sure he's not on it. :)And as it happens, I wasn't in Denver either. John McClelland from the SETI message boards drove from his home near Colorado Springs to pick me up, and took me to watch a memorable display of...  Hummingbirds at dawn over Pike's Peak, from his patio. And much more besides. Muchas, muchas gracias, John. |
|
1741)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92153)
Posted 14 Jul 2019 by Richard Haselgrove Post: There's water in them thar mountains. I remember hotels by all the different showers I encounter. Grumble - the miserable dribble in a closet in Chicago. Glory - this one felt like a (hot) waterfall in a forest glade. Jord, you'd like it. |
|
1742)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92151)
Posted 14 Jul 2019 by Richard Haselgrove Post: Very useful, shows today's 5 is 2 hours 23 minutes late. Remember in the USA, freight goes first(no matter what they say)Thanks - that'll be really useful. It also says it's going to pick up 45 mins by Denver - cancel that lie-in! I'll be able to go down and watch the fun in the search for breakfast. |
|
1743)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92149)
Posted 14 Jul 2019 by Richard Haselgrove Post: Are you catching Amtrak 5 from Denver?Sure am - Tuesday morning (Amtrak time). Even the guy on the ticket desk warned me to check how late it was running before leaving the hotel. Quick question - is there room in a single-use roomette for a big bag, or should I check it in? Seems I have to decide an hour before departure, when I haven't seen the space. |
|
1744)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92147)
Posted 14 Jul 2019 by Richard Haselgrove Post: Glory: I've moved out west, to Denver. There's S - P - A - C - E - room to move, room for the (far fewer) people to leave gaps. Grumble: It's raining. Proper thunderstorm, right over the city. |
|
1745)
Message boards :
News :
BOINC workshop agenda update
(Message 92142)
Posted 13 Jul 2019 by Richard Haselgrove Post: Greetings from Chicago O'Hare airport airside. Meet the gang from yesterday. 
|
|
1746)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92136)
Posted 11 Jul 2019 by Richard Haselgrove Post: Is it really that difficult to have the roving mike hooked up to a live feed as well as the speaker's mike.Passed to Juan. |
|
1747)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92113)
Posted 9 Jul 2019 by Richard Haselgrove Post: Kevin is trying to nudge presenters to repeat the end-of-talk questions into the microphone, but that doesn't help when people break out into general group discussion. Badges!!! What have I started! |
|
1748)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92110)
Posted 9 Jul 2019 by Richard Haselgrove Post: edit: seti's backBut still not uploading - at least, nothing's getting out of IBM HQ. |
|
1749)
Message boards :
News :
BOINC workshop agenda update
(Message 92105)
Posted 9 Jul 2019 by Richard Haselgrove Post: The live feed should be recorded and available later, for those in different time zones (or busy at work!). The presenter's slides are usually uploaded afterwards, and I'll watch out for the location as we go along. |
|
1750)
Message boards :
News :
BOINC workshop agenda update
(Message 92104)
Posted 9 Jul 2019 by Richard Haselgrove Post: Live stream now active. |
|
1751)
Message boards :
News :
BOINC workshop agenda update
(Message 92103)
Posted 9 Jul 2019 by Richard Haselgrove Post: Workshop is just starting, in housekeeping mode. Once we get into presentations, a live feed should be available at https://www.youtube.com/channel/UCJDOd7pQf1IF8ZxKHpNwung/featured |
|
1752)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92101)
Posted 9 Jul 2019 by Richard Haselgrove Post: We're just starting - domestic housekeeping. Had a brief chat with Eric - he wasn't aware, but is looking into it remotely. I'll ask about the live feed when I get a moment. Try https://www.youtube.com/channel/UCJDOd7pQf1IF8ZxKHpNwung/featured |
|
1753)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92096)
Posted 9 Jul 2019 by Richard Haselgrove Post: The last task on my travelling laptop is failing to upload. Maybe Jeff is working on Bruno? |
|
1754)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92093)
Posted 9 Jul 2019 by Richard Haselgrove Post: And we're down. Must be Jeff this week, because Eric is supposed to be here in Chicago with us. |
|
1755)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92087)
Posted 8 Jul 2019 by Richard Haselgrove Post: Not recording this time, but I think I've been put in charge of the live stream - "we have a plan for recording and livestreaming the workshop". Going to be somewhere on YouTube live. I'll post the url when I get it. |
|
1756)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92084)
Posted 8 Jul 2019 by Richard Haselgrove Post: That is what us Chi-towners call "THE CANYON" or the more common naming of The Financial District.I can see why they call them canyons, but fortunately it's not living up to the 'windy' bit right now. Thank you for providing those long linear sunshades over the main streets, but can you do something about the noise? |
|
1757)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92083)
Posted 8 Jul 2019 by Richard Haselgrove Post: Isn't everyone in Chicago at the BOINC Workshop?Not yet - starts Tuesday, and it's still Monday here. |
|
1758)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92069)
Posted 7 Jul 2019 by Richard Haselgrove Post: Would it be the Club Quarters Hotel, Central Loop?Yes, it is. Did you know that the street it's on (West Adams Street) is the official start of US Route 66? I don't think the hotel is bothered. First image of Chicago, this morning, just along the street:  (note the rest of the wedding party dotted along the brownstone) This is the backdrop they wanted for their wedding photos: 
|
|
1759)
Message boards :
Questions and problems :
rosetta@home tasks don't download
(Message 92064)
Posted 5 Jul 2019 by Richard Haselgrove Post: You should enable the <http_debug> event log flag (Ctrl+Shift+F). That will give you the urls accessed, and the server to-and-fro leading up to the error. |
|
1760)
Message boards :
Questions and problems :
It appears that CPU tasks take way too long to crunch when GPU is active
(Message 92053)
Posted 2 Jul 2019 by Richard Haselgrove Post: Yes but... Any file is just a sequence of bytes: the name is irrelevant, although the windows 'File Type' based on extension saves a lot of time. But if you run a Windows file anywhere else, you are likely to get the error message "This program requires Microsoft Windows" - because the Windows ecosystem is built around a shared library of Dynamic Link Libraries or DLLs to perform common, shared, functions. If those libraries can't be attached by using the expected system calls, the whole structure falls apart. A program could be contrived to run on the bare hardware, sure - but if it requires to do any system-level operations, like writing files to disk, the chances of it succeeding are vanishingly small. |
|
1761)
Message boards :
Questions and problems :
It appears that CPU tasks take way too long to crunch when GPU is active
(Message 92050)
Posted 2 Jul 2019 by Richard Haselgrove Post: If you look at a typical OpenCL application at SETI, it looks something like this: <app_version>
<app_name>setiathome_v8</app_name>
<version_num>822</version_num>
<platform>windows_intelx86</platform>
<api_version>7.5.0</api_version>
<avg_ncpus>0.04</avg_ncpus>
<max_ncpus>0.2</max_ncpus>
<plan_class>opencl_nvidia_SoG</plan_class>
<cmdline></cmdline>
<coproc>
<type>CUDA</type>
<count>1</count>
</coproc>
<file_ref>
<file_name>MB8_win_x86_SSE3_OpenCL_NV_SoG_r3557.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>libfftw3f-3-3-4_x86.dll</file_name>
</file_ref>
<file_ref>
<file_name>MultiBeam_Kernels_r3557.cl</file_name>
</file_ref>
<file_ref>
<file_name>mb_cmdline_win_x86_SSE3_OpenCL_NV_SoG.txt</file_name>
<open_name>mb_cmdline.txt</open_name>
</file_ref>
</app_version>The 'main program' is MB8_win_x86_SSE3_OpenCL_NV_SoG_r3557.exe - that's, in this case, a 32-bit Windows executable, and it runs on the CPU. No way is that going to run under Linux or Mac.The OpenCL bit is MultiBeam_Kernels_r3557.cl - it's only that part which is fully platform-independent. It's written in an intermediate-level code, but expressed in plain text. You can open it with a text editor, and see that it starts with Raistmer's cross-platform notes: //R: here kernels for SETI MultiBeam OpenCL build will be located // This version is using fraction(a * b * b) = fraction((_a + fa) (_b + fb) ^ 2) for enhancing the precision #define M_2PI 6.28318530717958647692f //R: it's single precision constant here ! //#if USE_OPENCL_NV || USE_OPENCL_HD5xxx || USE_OPENCL_INTEL // #define OCL_REAL_LOCAL 1 //#endif #if defined(USE_OPENCL_INTEL) || defined (__APPLE__)&&defined(USE_OPENCL_NV) #pragma OPENCL FP_CONTRACT OFF #endifThat code is compiled on the user's local machine (using a compiler application included in the driver runtime support package) to generate binary code tailored to the exact hardware and software target in use. Einstein embed the OpenCL code into their single-file CPU application, so you can't see the inner workings as clearly as you can at SETI, but it works the same way. |
|
1762)
Message boards :
Questions and problems :
It appears that CPU tasks take way too long to crunch when GPU is active
(Message 92045)
Posted 2 Jul 2019 by Richard Haselgrove Post: OK, let's pick that lot apart, considering: Environment Source code Compiled binaries Data Why cant Einstein or Seti tasks run on whichever GPU is available?That's an environment question: it was decided when SETI switched to the BOINC platform (or possibly later, when BOINC added GPU processing) that a 'task' would be assigned to a particular device, and the result would be expected back from that device. Note the word "assigned": any task is capable of being computed on any device - it just happens to have been sent to yours. The replication copy has been sent to a device on somebody else's computer. It could be the same type of device as yours, or it could be different. I am guessing it is only for verification purposeYes, it's primarily an anti-cheating device. But a weak one, as the successful deployment of rebranding tools confirms. The only real constraint is that the result must be returned from a host with the same HostID as it was issued to. What about resuming from a checkpoint?Happens already. If you have multiple cards of the same type - NVidia GPUs, for example - there is no certainty that a task that was running on one card when you shut down, will resume on the same card on restart. The checkpoint file - being pure data - doesn't mind. one OpenCL application should be able to be used on any OpenCL capable piece of hardware out thereThat certainly applies at the source code level. I think opinion varies at the compiled binary level: some programmers find that minor changes have to be made during compilation, others don't. I believe that you can take Einstein binaries for different GPU types and compare them, and find them identical. |
|
1763)
Message boards :
Questions and problems :
It appears that CPU tasks take way too long to crunch when GPU is active
(Message 92037)
Posted 2 Jul 2019 by Richard Haselgrove Post: It's a difference between the way NVidia and AMD implement OpenCL runtime support. Allegedly. But on the other hand, other people say that: a) NVidia implements OpenCL by calling the underlying CUDA layer b) CUDA has all the most efficient ways of calling the CPU only when needed. The implication would be that NVidia has deliberately crippled OpenCL by failing to implement the full set of inter-layer calling protocols. But I don't have enough first-hand information to say that. |
|
1764)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92032)
Posted 2 Jul 2019 by Richard Haselgrove Post: Nice simple question, I thought. But the answer turns out to be 'yes AND no'.Would it be the Club Quarters Hotel, Central Loop?Some of us don't chill our beer.I've just discovered that my hotel in Chicago next week has what it calls an English pub attached. But no mention of what temperature it sells the beer at. Or even what beer it sells, come to that. It's at the same address, but it seems to be known simply as The Central Loop Hotel these days - that's what I booked it under. Primary attraction - just one bag-dragging block from Monroe station on the Blue Line from O'Hare. |
|
1765)
Message boards :
Questions and problems :
It appears that CPU tasks take way too long to crunch when GPU is active
(Message 92028)
Posted 2 Jul 2019 by Richard Haselgrove Post: You'll find that the CPU is actually doing very little, despite appearing to be busy. It's what's known as a spin-wait loop: the CPU is just asking 'Do you need me yet? Do you need me yet? Do you need me yet?' over and over again. It's one (very inefficient) way of ensuring that it's ready to leap into action at a microsecond's notice when the GPU needs to be told what to do next. The CPU should be able to power-down the floating point unit, the SIMD units, and much else while that's going on. Saves a few watts. |
|
1766)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 92025)
Posted 2 Jul 2019 by Richard Haselgrove Post: People who post a question, but never come back to read the answer. |
|
1767)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92024)
Posted 2 Jul 2019 by Richard Haselgrove Post: IF you are going to Chicago in Summer, it's NOT the raw 87F Temps you have to worry about, it's the HUMIDITY!!! (Could end up feeling like it's 100F!!! (Or close to it.))Well, the people who offered the workshop venue presumably live close by and are accustomed to it. Provided I can still walk three blocks between air-conditioning units without falling over, I should be OK. Failing that, the $10 all-day pass for the water taxi service is looking increasingly attractive. |
|
1768)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 92017)
Posted 2 Jul 2019 by Richard Haselgrove Post: Some of us don't chill our beer.I've just discovered that my hotel in Chicago next week has what it calls an English pub attached. But no mention of what temperature it sells the beer at. Or even what beer it sells, come to that. But with temperatures forecast to reach 87°F, I think I might be ready for a chilled beer. |
|
1769)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 92013)
Posted 1 Jul 2019 by Richard Haselgrove Post: Film? These days it is more likely to be a digital image.I can only assume. Above my wrist was some very complicated engineering with an embedded light source, so the radiographer could exactly line up the (presumed) radiation source with the contours of my arm. And they took care - two separate shots, with my arm at different angles under the light. Under my wrist, a flat plastic box, maybe 5mm thick, with the brand name 'Agfa' embossed. And no wires. |
|
1770)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 92010)
Posted 1 Jul 2019 by Richard Haselgrove Post: Absolutely. There was a lot of calm, professional, efficiency by people who knew exactly what they were doing. And no little technological efficiency, too. The X-ray was deceptively simple: an overhead radiation source in a steerable gantry, my wrist resting on a photographic plate casually placed on an examination couch beside an ordinary chair. Yet it took barely 10 minutes to develop the plate and diagnose what is described as a fracture which can easily be missed. And by the time I'd walked back to A & E for the initial treatment assessment, there was my wrist, up on screen for the paramedic to see. |
|
1771)
Message boards :
Questions and problems :
Re: How to change what drive BOINC stores files on
(Message 92005)
Posted 1 Jul 2019 by Richard Haselgrove Post: Yes, it does. Everything that BOINC needs to store is contained in that single folder and its sub-folders, including all the results so far for a task in progress and completed tasks. The only thing I'd suggest for a project like CPDN is to take it slowly - especially at the first step, "Stop running tasks when exiting the BOINC Manager". CPDN can take a little time to flush everything to disk and shut down the applications in an orderly fashion. |
|
1772)
Message boards :
Questions and problems :
It appears that CPU tasks take way too long to crunch when GPU is active
(Message 92002)
Posted 1 Jul 2019 by Richard Haselgrove Post: It is. But the next question is, 'by how much?' - 2% or 98%? |
|
1773)
Message boards :
Questions and problems :
It appears that CPU tasks take way too long to crunch when GPU is active
(Message 91997)
Posted 1 Jul 2019 by Richard Haselgrove Post: It also depends on how the programming language chosen has been implemented on the hardware platform. OpenCL on NVidia hardware usually needs a lot of CPU support: OpenCL on AMD or iGPU is less likely to (though it does vary according to the problem, the programmer, and the options chosen). |
|
1774)
Message boards :
Projects :
WCG Climate research project: New Beta Test – May 29, 2019
(Message 91995)
Posted 1 Jul 2019 by Richard Haselgrove Post: No, it's available to all - the OP simply copied a version of the url shortened with elipses. Try https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,41609_offset,0#610450 |
|
1775)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 91994)
Posted 1 Jul 2019 by Richard Haselgrove Post: I have to say that I'm incredibly impressed and reassured by how well the system has responded to my little medical incident last week. There's hope for this austere little country yet. I fell in the late afternoon (Wednesday), but didn't damage my ability to walk - so I just completed the errand and went home as normal. On Thursday morning it was clear that my wrist was swollen and difficult to use, so I rang the doctor - really just for advice on how to manage it. But the system kicked in: within an hour I was called back by a triage doctor: two hours later I was seeing a doctor face-to-face (appointment delayed because I didn't want to drive). Three hours later I'd transferred to the hospital (8 miles, two buses): been X-rayed, diagnosed with a Carpal Avulsion, plastered, and booked in for a consultant's appointment the following morning. The consultant cut me free, checked me over, and gave me the splint to support and protect my wrist while it completed healing. And today, when I went to collect my lunch and newspaper as usual, I found all the overgrown weeds which had obscured the path surface had been trimmed back to the fences. The total cost (to me) of the whole episode was £1.20 - two bottles of water for the bus transfers. I hadn't expected to be out and about on the hottest day of the year, and one thing that we've lost during the last century are the public drinking fountains - replaced by safe piped water into our houses. Mind you, they're slowly coming back, as we cut down on single-use plastic. All the medical facilities and treatment were free on the NHS, of course, and all the multiple bus trips were free because I'm over pension age. Long may it continue. |
|
1776)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 91978)
Posted 28 Jun 2019 by Richard Haselgrove Post: Not on any meds, except beer. And not driving for the time being. Good things, free bus passes. Been down to the pharma to collect a reusable ice-pack for the swelling, and some gel rub to go with it. |
|
1777)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 91976)
Posted 28 Jun 2019 by Richard Haselgrove Post: Might have to be quick. The most tedious part of the whole job is the travelling: I didn't think it was wise to drive, and the doctor / hospital are on opposite sides of the principal town, where all buses terminate in the bus station. But they never connect... Had 50 minutes to wait for the hourly service back up to the village. So I popped into the Town Hall to put the accident on the record, and I've already had an email exchange with the Area Rights of Way Officer: last time I reported a problem on that path (collapsed stile), he had the whole thing replaced with a much better one - so I'm claiming that stile as my memorial shrine! Still some daylight left - piccies may follow, or I may just go to the pub instead. Nerves need a little more stabilisation. |
|
1778)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 91974)
Posted 28 Jun 2019 by Richard Haselgrove Post: Ah, that's better. Consultant cut the plaster off, pushed and prodded for a couple of minutes, and decided that I hadn't said 'ouch' loudly enough. So he prescribed a short soft fabric splint, with velcro fastenings, so I can peel it off to take a shower. And scratch that itch. |
|
1779)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 91961)
Posted 27 Jun 2019 by Richard Haselgrove Post: Do you remember that I dragged Jord down to the cafe by the locks on the first day? Well, I did that exact same walk yesterday, as I've probably done a thousand times in the last ten years. Except that yesterday I stepped on a bit of path that wasn't there, and fell flat on my face. Result - a day spent shuttling between doctor and hospital, and a diagnosis of a cracked...  Triquetral bone (that's the bright green one). Tonight I've got an old-fashioned A&E plaster from elbow to hand. Tomorrow I see the consultant, and I'm hoping to get it swapped for something lightweight and adjustable - otherwise three weeks in American heat is going to be unbearable. It's already starting to itch... |
|
1780)
Message boards :
Questions and problems :
bug? computation error on restarting based on time of day
(Message 91867)
Posted 17 Jun 2019 by Richard Haselgrove Post: It might be worth randomly suspending single tasks and allowing them to restart later in their own time. GPU tasks are always removed from GPU memory when suspended for any reason (unlike CPU tasks, which can be left in memory if your preferences permit). That means that on restart, they have to be reloaded from the checkpoint file. Over the years, some projects have from time to time had problems with their checkpointing code. You wouldn't notice any problems when suspending, but you can get a crash when the app tries to read back a bad file. The same thing can happen if your hard disk is flaky. Start with the project that failed on restart, but test other projects too. Only rely on unattended stops/starts when you're certain that all your active projects have reliable checkpoint/restart code. |
|
1781)
Message boards :
GPUs :
Request for Intel iGPU support to be a working feature again.
(Message 91861)
Posted 17 Jun 2019 by Richard Haselgrove Post: As mentioned somewhere in this thread, I updated to version 26.20.100.6890 (Latest) Date: 5/29/2019 (DCH version) from https://downloadcenter.intel.com/download/28816/Intel-Graphics-Windows-10-DCH-Drivers?product=80939, and all my previous problems went away as well. Raistmer is still looking through the test logs I sent him, but I think it might really have been the older drivers this time (he always blames the drivers, and I always blame his coding. We're fine: last time, it turned out to be a compiler optimisation flag, so we were both wrong). My test machine is running at All tasks for computer 8670176: looking good so far. The single inconclusive seems to be a dodgy wingmate with several inconclusives and invalids. Unfortunately, there's no sign of the promised v7.16 yet, but anyone who finds themselves stymied by the DCH extensions list can download the 'artifact' test build from https://ci.appveyor.com/project/BOINC/boinc/builds/25266698/artifacts |
|
1782)
Message boards :
GPUs :
Request for Intel iGPU support to be a working feature again.
(Message 91846)
Posted 15 Jun 2019 by Richard Haselgrove Post: I also refer to a SETI test in that thread: like then, today's test with the SETI app has led to an inconclusive result, so I suspect Intel have tweaked the precision or driver optimisation again. Bernd is aware of the Einstein problem, but too busy to do anything about it: I don't feel like waking Raistmer out of his slumbers before I set off on my travels. So I'm inclined to let this one lie until some cold, boring, winter's evening.Change of plan. The two SETI tasks I ran with the original OEM driver threw up multiple issues, so much so that the first one needed four replications before SETI found a pair which could validate together. That suggests that there may be something more seriously wrong with the current application mix at SETI, and Raistmer and I are going to run some more substantial offline tests. In the meantime, that level of replication is obviously wasteful for the project and all concerned: I suggest that - whilst observed testing is very welcome - you don't put these newest iGPUs into full SETI production until we report back. |
|
1783)
Message boards :
Questions and problems :
GPU detection failed, error 0xc0000005
(Message 91833)
Posted 14 Jun 2019 by Richard Haselgrove Post: 6/6/2019 4:15:21 PM | | [coproc] clGetDeviceInfo failed to get device extensions for device 0Update. We have now established that this error is intrinsic to BOINC, and will occur when the new Intel Windows DCH drivers are installed - whether they are sourced from Microsoft or Intel. BOINC v7.14.x will always fail to detect an iGPU with these drivers, but v7.16.x should work. It is anticipated that testing of the new client should start either over this weekend, or during the week commencing 17 June 2019. |
|
1784)
Message boards :
GPUs :
Request for Intel iGPU support to be a working feature again.
(Message 91830)
Posted 13 Jun 2019 by Richard Haselgrove Post: Yes, that was my reaction at first too. But luckily I had appropriate hardware available, and I plowed through the very limited information provided. And yes, it was a BOINC problem. Not a deliberate removal of a feature (as the thread title seems to imply), but simply the outside world outgrowing us. Anyway, my patch has been accepted. You will both test version 7.16 (expected to be released for testing this weekend), won't you? |
|
1785)
Message boards :
GPUs :
Request for Intel iGPU support to be a working feature again.
(Message 91827)
Posted 13 Jun 2019 by Richard Haselgrove Post: OK, now we're getting somewhere. I've updated to 26.20.100.6890 (Latest) Date: 5/29/2019, and Oblomov still sees OpenCL: Platform Name Intel(R) OpenCL Number of devices 2 Device Name Intel(R) UHD Graphics 620 Device Vendor Intel(R) Corporation Device Vendor ID 0x8086 Device Version OpenCL 2.1 NEO Driver Version 26.20.100.6890 Device OpenCL C Version OpenCL C 2.0 Device Type GPUbut BOINC doesn't. 13/06/2019 19:38:53 | | Starting BOINC client version 7.14.2 for windows_x86_64 13/06/2019 19:38:53 | | log flags: file_xfer, sched_ops, task, cpu_sched, sched_op_debug 13/06/2019 19:38:53 | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8 13/06/2019 19:38:53 | | Data directory: C:\BOINCdata 13/06/2019 19:38:53 | | Running under account Richard Haselgrove 13/06/2019 19:38:53 | | OpenCL CPU: Intel(R) Core(TM) i5-8250U CPU @ 1.60GHz (OpenCL driver vendor: Intel(R) Corporation, driver version 7.6.0.0228, device version OpenCL 2.1 (Build 0)) 13/06/2019 19:38:53 | | No usable GPUs foundThe GPU detection code returned <warning>clGetDeviceInfo failed to get device extensions for device 0</warning> Does that match your diagnostics? Continuing: that test comes a long way down the list - https://github.com/BOINC/boinc/blob/master/client/gpu_opencl.cpp#L844 - it's already found everything else. Oblomov found them: Device Extensions cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_fp16 cl_khr_depth_images cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_icd cl_khr_image2d_from_buffer cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_intel_subgroups cl_intel_required_subgroup_size cl_intel_subgroups_short cl_khr_spir cl_intel_accelerator cl_intel_media_block_io cl_intel_driver_diagnostics cl_intel_device_side_avc_motion_estimation cl_khr_priority_hints cl_khr_throttle_hints cl_khr_create_command_queue cl_khr_fp64 cl_khr_subgroups cl_khr_il_program cl_intel_spirv_device_side_avc_motion_estimation cl_intel_spirv_media_block_io cl_intel_spirv_subgroups cl_khr_spirv_no_integer_wrap_decoration cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_intel_planar_yuv cl_intel_packed_yuv cl_intel_motion_estimation cl_intel_advanced_motion_estimation cl_khr_gl_sharing cl_khr_gl_depth_images cl_khr_gl_event cl_khr_gl_msaa_sharing cl_intel_dx9_media_sharing cl_khr_dx9_media_sharing cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_intel_d3d11_nv12_media_sharing cl_intel_simultaneous_sharingThat's a long list - 1.09 KB (1,117 bytes). I think we might need a bigger buffer... OK, make that https://github.com/BOINC/boinc/blob/master/lib/opencl_boinc.h#L48. But increasing it gives: 13/06/2019 20:31:42 | | Starting BOINC client version 7.15.0 for windows_x86_64 13/06/2019 20:31:42 | | This a development version of BOINC and may not function properly 13/06/2019 20:31:42 | | log flags: file_xfer, sched_ops, task, cpu_sched, sched_op_debug 13/06/2019 20:31:42 | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8 13/06/2019 20:31:42 | | Data directory: C:\BOINCdata 13/06/2019 20:31:42 | | Running under account Richard Haselgrove 13/06/2019 20:31:42 | | OpenCL: Intel GPU 0: Intel(R) UHD Graphics 620 (driver version 26.20.100.6890, device version OpenCL 2.1 NEO, 3166MB, 3166MB available, 211 GFLOPS peak) 13/06/2019 20:31:42 | | OpenCL CPU: Intel(R) Core(TM) i5-8250U CPU @ 1.60GHz (OpenCL driver vendor: Intel(R) Corporation, driver version 7.6.0.0228, device version OpenCL 2.1 (Build 0))I think that earns me a beer - PR to follow later. Oh, all right. PR #3183. NOW beer. Seeya! |
|
1786)
Message boards :
GPUs :
Request for Intel iGPU support to be a working feature again.
(Message 91826)
Posted 13 Jun 2019 by Richard Haselgrove Post: Yes, you're right. "Using current driver" should work. The trouble is that my current driver does work, your current driver doesn't work. So, what's the difference, and is it something for BOINC to fix, Microsoft to fix, or Intel to fix? I've posted my current driver version (25.20.100.6374), but you've said nothing except "since the shift to Windows DCH drivers". Any chance we could have comparable details? Until we work out who broke what, we won't know whose job it is to fix it. Edit - I assume you're somewhere on https://downloadcenter.intel.com/download/28816/Intel-Graphics-Windows-10-DCH-Drivers?product=80939, but that page includes 9 different versions. |
|
1787)
Message boards :
GPUs :
Not getting any work for my Anonymous platform at Einstein@home
(Message 91825)
Posted 13 Jun 2019 by Richard Haselgrove Post: Try doing a command line boinc --helpto see what the start-up options are (I can't find them documented in the Wiki just now) '--fetch_minimal_work' sounds like what you're looking for. |
|
1788)
Message boards :
GPUs :
Request for Intel iGPU support to be a working feature again.
(Message 91797)
Posted 11 Jun 2019 by Richard Haselgrove Post: Have a look at the opening post in another intel gpu opencl thread at Einstein. That's the same UHD 620 that I used in this test. You'll see that I got it to fetch work (needs Beta apps enabled), but not to run it - it stalled after doing the initial data setup, and never reached the first checkpoint. I also refer to a SETI test in that thread: like then, today's test with the SETI app has led to an inconclusive result, so I suspect Intel have tweaked the precision or driver optimisation again. Bernd is aware of the Einstein problem, but too busy to do anything about it: I don't feel like waking Raistmer out of his slumbers before I set off on my travels. So I'm inclined to let this one lie until some cold, boring, winter's evening. |
|
1789)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91795)
Posted 11 Jun 2019 by Richard Haselgrove Post: Well, it's now 05:20 PDT, and nothing's happened y......... |
|
1790)
Message boards :
GPUs :
Request for Intel iGPU support to be a working feature again.
(Message 91793)
Posted 11 Jun 2019 by Richard Haselgrove Post: As you can see in my previous post (now complete), BOINC can see my OEM driver just fine. So... running my secondary monitor with the iGPUYes, that's normally a prerequisite for iGPU computing. I have 'headless' dummy plugs on the motherboard outputs on all my NVidia-equipped desktops. the shift to Windows DCH driversThat's beginning to sound like the nub of the problem. I'll see if I can replicate. Now downloading driver 25.20.100.6374 dated 18 November 2018 from Microsoft. And that's got OpenCL support too: Platform Name Intel(R) OpenCL Number of devices 2 Device Name Intel(R) UHD Graphics 620 Device Vendor Intel(R) Corporation Device Vendor ID 0x8086 Device Version OpenCL 2.1 NEO Driver Version 25.20.100.6374 Device OpenCL C Version OpenCL C 2.0which BOINC can see: 11/06/2019 13:15:21 | | Starting BOINC client version 7.14.2 for windows_x86_64 11/06/2019 13:15:21 | | log flags: file_xfer, sched_ops, task, cpu_sched, sched_op_debug 11/06/2019 13:15:21 | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8 11/06/2019 13:15:21 | | Data directory: C:\BOINCdata 11/06/2019 13:15:21 | | Running under account Richard Haselgrove 11/06/2019 13:15:22 | | OpenCL: Intel GPU 0: Intel(R) UHD Graphics 620 (driver version 25.20.100.6374, device version OpenCL 2.1 NEO, 3166MB, 3166MB available, 211 GFLOPS peak) 11/06/2019 13:15:22 | | OpenCL CPU: Intel(R) Core(TM) i5-8250U CPU @ 1.60GHz (OpenCL driver vendor: Intel(R) Corporation, driver version 7.6.0.716, device version OpenCL 2.1 (Build 716))and use: 11/06/2019 13:21:57 | SETI@home | [cpu_sched] Starting task blc25_2bit_guppi_58406_23898_HIP20842_0094.28629.818.21.44.208.vlar_1 using setiathome_v8 version 820 (opencl_intel_gpu_sah) in slot 4 |
|
1791)
Message boards :
GPUs :
Request for Intel iGPU support to be a working feature again.
(Message 91791)
Posted 11 Jun 2019 by Richard Haselgrove Post: OK, it sounds like it's time to do some testing again. I've got a Windows 10 i5 on the desk in front of me, and BOINC saw the Intel GPU just fine when I bought it - running build 1809. It's recently updated itself to 1903, so worth another look. It's an ultraportable, so I don't usually use the iGPU for crunching (not enough cooling), and I'll have to switch BOINC out of service mode when I get to that stage: but I'll update this post as I go along. Hardware: i5 8250U CPU, Intel UHD 620 Software: Windows 10 Pro, version 1903, build 18362.145 Graphics driver: 23.20.16.4973 dated 28 Feb 2018 (OEM supplied) Well, Oblomov's CLinfo sees the right sort of detail - I'll switch to that machine to post. Number of platforms 1 Platform Name Intel(R) OpenCL Platform Vendor Intel(R) Corporation Platform Version OpenCL 2.1 Platform Profile FULL_PROFILE Platform Extensions cl_intel_dx9_media_sharing cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_d3d11_sharing cl_khr_depth_images cl_khr_dx9_media_sharing cl_khr_fp64 cl_khr_gl_sharing cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_icd cl_khr_image2d_from_buffer cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_spir Platform Host timer resolution 100ns Platform Extensions function suffix INTEL Platform Name Intel(R) OpenCL Number of devices 2 Device Name Intel(R) UHD Graphics 620 Device Vendor Intel(R) Corporation Device Vendor ID 0x8086 Device Version OpenCL 2.1 NEO Driver Version 23.20.16.4973 Device OpenCL C Version OpenCL C 2.1 Device Type GPU ... Device Name Intel(R) Core(TM) i5-8250U CPU @ 1.60GHz Device Vendor Intel(R) Corporation Device Vendor ID 0x8086 Device Version OpenCL 2.1 (Build 611) Driver Version 7.6.0.611 Device OpenCL C Version OpenCL C 2.0 Device Type CPUAnd BOINC sees 11/06/2019 12:49:08 | | Starting BOINC client version 7.14.2 for windows_x86_64 11/06/2019 12:49:08 | | log flags: file_xfer, sched_ops, task, cpu_sched, sched_op_debug 11/06/2019 12:49:08 | | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8 11/06/2019 12:49:08 | | Data directory: C:\BOINCdata 11/06/2019 12:49:08 | | Running under account Richard Haselgrove 11/06/2019 12:49:08 | | OpenCL: Intel GPU 0: Intel(R) UHD Graphics 620 (driver version 23.20.16.4973, device version OpenCL 2.1 NEO, 3166MB, 3166MB available, 211 GFLOPS peak) 11/06/2019 12:49:08 | | OpenCL CPU: Intel(R) Core(TM) i5-8250U CPU @ 1.60GHz (OpenCL driver vendor: Intel(R) Corporation, driver version 7.6.0.611, device version OpenCL 2.1 (Build 611)) 11/06/2019 12:49:08 | | Processor: 8 GenuineIntel Intel(R) Core(TM) i5-8250U CPU @ 1.60GHz [Family 6 Model 142 Stepping 10] 11/06/2019 12:49:08 | | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx tm2 pbe fsgsbase bmi1 smep bmi2 11/06/2019 12:49:08 | | OS: Microsoft Windows 10: Professional x64 Edition, (10.00.18362.00) |
|
1792)
Message boards :
Questions and problems :
Boinc Manager impossible to uninstall
(Message 91781)
Posted 9 Jun 2019 by Richard Haselgrove Post: Even if you didn't keep the original downloaded installer executable, you can download it again - look at any project you attached the laptop to, if you can't remember which version you had installed. Your account/host list will still display that information. All released version are kept in https://boinc.berkeley.edu/dl/, not least for this purpose. |
|
1793)
Message boards :
Questions and problems :
Boinc Manager impossible to uninstall
(Message 91779)
Posted 9 Jun 2019 by Richard Haselgrove Post: The Windows installer is trying to find the instructions for UNinstalling the previous installation. They are included in the boinc.msi file used to INSTALL that previous version. Windows should keep that version of the file lying around for reference, but if it's gone missing - install the exact same version as you installed before, and then uninstall it again. You (as administrator) are installing it, and you (as the same administrator) are uninstalling it: the missing boinc account shouldn't make any difference. |
|
1794)
Message boards :
Questions and problems :
green screen death when using my GPU no message computer then crashes.
(Message 91757)
Posted 7 Jun 2019 by Richard Haselgrove Post: From Wikipedia: Scalable Link Interface (SLI) is a brand name for a multi-GPU technology developed by Nvidia for linking two or more video cards together to produce a single output.If you have two (or sometimes even more) identical NVidia GPUs, you can hook them together to form the equivalent of one big GPU. That doesn't help with the work we do in BOINC: BOINC will still see you as having multiple smaller cards, and will use them independently. It's probably better not to have the SLI link present for our sort of work. |
|
1795)
Message boards :
Questions and problems :
green screen death when using my GPU no message computer then crashes.
(Message 91752)
Posted 7 Jun 2019 by Richard Haselgrove Post: Why that is wedged in there I'm not sure.Advertising. Usually the advertiser's (interpolated) download button is larger, more prominent and more eye-catching than the download product that brought you to the page in the first place. They cause vastly more problems than they ever cure, and you need to exercise care only to download the product you really want. Has anyone successfully used adblockers against this curse? |
|
1796)
Message boards :
GPUs :
GPU usage utility available?
(Message 91735)
Posted 5 Jun 2019 by Richard Haselgrove Post:  (this message board uses BBcode, rather than HTML, and PNG format preserves sharp detail better than JPG) |
|
1797)
Message boards :
GPUs :
GPU usage utility available?
(Message 91733)
Posted 5 Jun 2019 by Richard Haselgrove Post: Thanks - that helps. We can look at Valid tasks for computer 8710401 and see that your GPU is, in fact, returning valid work - most recently at 5 Jun 2019, 20:49:10 UTC (less than an hour ago). It is possible that your machine is working fine when you're not looking at it, but switching itself off (for SETI, at least) when you look at it, simply because it thinks you want to do something more important with it. Your description of 'stop signs' and green arrows confirms that you are using the 'Simple View' of the BOINC Manager program: if you are interested and want to get more involved involved in the SETI project, you can switch to the 'Advanced View' and find more information about what's going on and how you can manage it. But that's not important: you're fine as you are. |
|
1798)
Message boards :
GPUs :
GPU usage utility available?
(Message 91729)
Posted 5 Jun 2019 by Richard Haselgrove Post: Also, look in your event log for a message like the followingPlease post the whole of the first 30 lines, not just selected highlights. You might leave out the parts that seem unimportant you, but are meaningful to us. |
|
1799)
Message boards :
GPUs :
GPU usage utility available?
(Message 91723)
Posted 5 Jun 2019 by Richard Haselgrove Post: How do you know that the GPUs aren't being used? What exactly are you seeing, or not seeing, that tells you "no go"? The most useful information to give potential helpers is the first 30 lines or so of the Event Log, after a fresh system restart. A lot to get our collective teeth into in there. |
|
1800)
Message boards :
Questions and problems :
green screen death when using my GPU no message computer then crashes.
(Message 91703)
Posted 4 Jun 2019 by Richard Haselgrove Post: how do i know or how can I tell ….. if my fans are spinning on my NVIDIA Quadro K4200 (4095MB) driver: 391.25 ???Cut a narrow strip of stiff paper, and push it gently through the ventilation slots in the cover while the machine is running. If the fan's spinning, you'll hear the blades hit the paper, and you ears will give you an idea of how fast it's running. If all you hear is the sound of silence, either you've picked the wrong ventilation slot, or the fan has stopped. |
|
1801)
Message boards :
GPUs :
GPU usage utility available?
(Message 91700)
Posted 4 Jun 2019 by Richard Haselgrove Post: Actually, the Windows 10 Task Manager has quite a good GPU usage monitor, but you have to dig a bit to find it. The GPUs are listed on the 'Performance' tab, but when you first open it the four mini-graphs all show nothing happening: they refer mostly to video processing activity, which isn't relevant for Distributed Computing. Each graph has a 'picker' - drop-down list at top left - which can display different values. I think the names are different for each manufacturer's products, and I don't have an active W10 GPU to check with, but if you search through the 10 or so possibilities, you should find a useful display. |
|
1802)
Message boards :
Questions and problems :
green screen death when using my GPU no message computer then crashes.
(Message 91687)
Posted 2 Jun 2019 by Richard Haselgrove Post: "4 to 5 minutes" sounds like the timescale for a thermal problem - dusty, stuck, or failed fan, perhaps. Take the lid off and give it a careful, but thorough, inspection and hoovering. BSODs usually display a 'STOP' number, with sometimes an English translation as well. Those might be helpful. |
|
1803)
Message boards :
Questions and problems :
Where is the SETI gpu application source located?
(Message 91673)
Posted 30 May 2019 by Richard Haselgrove Post: My current SaH NVidia SoG build identifies itself as r3584. Tracing that back leads me to https://setisvn.ssl.berkeley.edu/trac/browser/branches/sah_v7_opt/AKv8 |
|
1804)
Message boards :
Questions and problems :
Problem suspending tasks
(Message 91665)
Posted 29 May 2019 by Richard Haselgrove Post: Or set 'No new tasks' for any project which might be in danger of over-fetching. The project whose tasks are being suspended won't be at risk of fetching - work requests are inhibited for any project which has even one task suspended. The problem with 'Suspend network activity' is that it also prevents result files being uploaded and completed tasks being reported. |
|
1805)
Message boards :
Questions and problems :
Problem suspending tasks
(Message 91663)
Posted 29 May 2019 by Richard Haselgrove Post: This problem is already one of the identified issues in the backlog of improvements waiting for developer time. Race condition when suspending tasks My personal opinion (stated in that issue, but unverified) is that the 'race condition' in the title occurs because of the difference in speeds between: a) the Manager sending each successive 'single task' RPC b) the Client processing all parts of the suspend instruction If a single running task is suspended, a message is sent to the project application to close and (potentially, depending on preference settings) remove itself from memory. The key question in your case might be how quickly SRBase apps can stop, rather than how quickly they can start. Personally, if I need to do something like this, I use the same workround as you: suspend all unstarted tasks as a separate block action: only suspend the running task when there's nothing available to take its place. |
|
1806)
Message boards :
Questions and problems :
resource "share' not working as expected when set to zero
(Message 91639)
Posted 26 May 2019 by Richard Haselgrove Post: RS is not a global preference, it is project specific by its nature and as far as I know it's not transmitted to other projects.My mistake - ignore. Not enough coffee error. But forgive me a slightly related question, where do projects get the lists of other projects users participate in?From BOINC Combined Statistics - https://boinc.netsoft-online.com/. It's one of the generalised stats sites (though not one commonly mentioned by users). An 'include' call at https://github.com/BOINC/boinc/blob/master/html/inc/user.inc#L32 gets that block (complete) from netsoft, so the project you're viewing doesn't actually 'know' anything about your other projects: the page is assembled on the fly. That's why the figures ("last public export") are always a bit below the internal ones for the current project. You're right, it definitely happens. I foundI'll have a look through the client code to see if I can find that "reset if all zero" bug that floyd describedI hope you find something. I sure looked after the incident I described and I looked again today, still don't see how it is possible but it happened. Unfortunately I can't remember details but they must be important. 26/05/2019 16:27:33 | | All projects have zero resource share; setting to 100in the event log after a test, which led me to https://github.com/BOINC/boinc/blob/master/client/cs_statefile.cpp#L544. In turn, that code was added on 28 January 2010, just in time for version 6.10.32, which in turn has "The long awaited for backup project mechanism has now been implemented. Projects with a resource share of 0 are considered backup projects." (change log) I think the two must be both intentional and related, although the checkin note says simply "client: fix my last checkin". A good point. Given the deliberate coding, I think we should probably leave it alone, although there is some untidiness. Even after the 'setting to 100' appeared in the event log, the project page in BOINC Manager still showed zero across the board: after I plugged the network cable back in and did an update on all projects, my normal values appeared for all projects, all venues. So it's non-destructive.we can lose that now that zero has been ascribed an active meaningYes when it is only about changing a meaningless value. But on a second thought, maybe parts of the code rely on the sum of shares being non-zero, for example when calculating fractions. I think the best we can do is to add a note to the documentation to say "When setting backup projects, leave yourself at least one leg to stand on". ;-)But for the time being (until I find the smoking gun in the code), I'm keeping an open mindNo gun, sadly not even smoke. But I heard a bang and someone fell over. Hate to look like a fool. Edit - added to List of project preferences |
|
1807)
Message boards :
Questions and problems :
resource "share' not working as expected when set to zero
(Message 91635)
Posted 26 May 2019 by Richard Haselgrove Post: A corollary to that: in theory, Resource Share can only be set on a project preferences website or via an Account Manager (BeemerBiker mentions BAM!, but other AMs are available). RS can't be set via BOINC Manager or boinccmd - the only local control would be via editing the raw data files. Not every project updates their server code in sync with BOINC releases - many of them leave it untouched for months or years. For a long time, there were reports that specific projects wouldn't allow the special 0 value to be entered via their websites. What happens if a user specifies 0 via an AM, for a project which can't handle it? BOINC global preferences - including RS - are supposed to be propagated from project to project during client contacts: would a non-updated project substitute unwanted values during that process? Also, Einstein@Home discovered that for many years a bug in the server code had been dropping elements from the final venue settings during propagation: that also disrupted things. I'll have a look through the client code to see if I can find that "reset if all zero" bug that floyd described: we can lose that now that zero has been ascribed an active meaning. But for the time being (until I find the smoking gun in the code), I'm keeping an open mind whether it's in the client, an outdated server, or the AM route. |
|
1808)
Message boards :
Questions and problems :
Cant seem to find 7.14 for linux
(Message 91632)
Posted 25 May 2019 by Richard Haselgrove Post: Gianfranco Costamagna's ppa (also known as LocutusOfBorg) has been consistently reliable and up-to-date in the past. He's one of the package managers I included in my notification list when I took a stint at release management. |
|
1809)
Message boards :
Questions and problems :
Cant seem to find 7.14 for linux
(Message 91623)
Posted 25 May 2019 by Richard Haselgrove Post: I think I know what's happened, but I need to talk to both the people concerned before I can be sure. That should happen on Thursday (holidays permitting) - I'll post again when I have some firm information. |
|
1810)
Message boards :
Questions and problems :
How to force BOINC to use the slowest GPU ?
(Message 91621)
Posted 25 May 2019 by Richard Haselgrove Post: You can configure BOINC to do that. Full details are in the Client configuration page of the user manual. First, create a configuration file cc_config.xml The easiest way to do this is to make a temporary modification to the 'Event log options...' - Ctrl+Shift+F from BOINC Manager. Change something, save it, then change it back again and save it again. That creates a complete configuration file pre-populated with all the standard options. Then, add an extra line to it, using a simple text editor like Notepad. From the manual: <ignore_nvidia_dev>N</ignore_nvidia_dev>You want to follow the example exactly, because the 'best' GPU - the noisy one you want to disable - should be device 0. Put the new line in the <options> section, in alphabetical order. [It'll work anywhere, but keep the file organised - it'll be much easier to remember what you've done in the future] Save the file, and restart BOINC. Check the first few lines in the event log on restart to ensure that the change has worked properly. |
|
1811)
Message boards :
Questions and problems :
How to terminate a project communication request
(Message 91617)
Posted 24 May 2019 by Richard Haselgrove Post: There's no way to explicitly cancel an update request. There's a timeout parameter - default 5 minutes - which you can turn down if it happens often. If something seems to have got stuck, I find the easiest way of tickling it back into life is to suspend network activity (Activity menu) and re-enable it a few seconds later. |
|
1812)
Message boards :
Questions and problems :
Manager opens when restarting computer
(Message 91597)
Posted 21 May 2019 by Richard Haselgrove Post: ... depending (I am guessing) on whether the boinctray program is running...Wrong guess. boinctray.exe is the 'user active' (mouse, keyboard, or sundry HID activity) detection utility. It has no visible interface components. |
|
1813)
Message boards :
Questions and problems :
Manager opens when restarting computer
(Message 91596)
Posted 21 May 2019 by Richard Haselgrove Post: 'Run minimised in tray' was never a user option. It was, and is still, the default startup option created by the Windows BOINC installer: it does this by creating the command line "D:\BOINC\boincmgr.exe" /a /s(or whatever your chosen installation location was) Starting visible on-screen is usually the result of creating a secondary startup command manually in one of the other Windows startup locations. |
|
1814)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91589)
Posted 21 May 2019 by Richard Haselgrove Post: I wonder what that bodes?A full West Coast breakfast? |
|
1815)
Message boards :
BOINC client :
Boincmgr auto-launches boinc client
(Message 91587)
Posted 21 May 2019 by Richard Haselgrove Post: Shouldn't this be made optional and configurable? The author of this thread has asked, reasonably and appropriately, that the client should not start when the manager is launched: other authors in other threads have been frustrated and annoyed when they have not been able to start the client from the manager. We (upstream BOINC development) have provided two controls to allow user choice: the default is for the manager to start the client if not running, but you can * Deselect "Run the client?" on the Options --> Other options... dialog (affects future launches only) * Use the command line switch '--no-daemon' when launching the manager. These may be a bit clunky, and may be badly documented, but at least they leave the final choice in the hands of the end user. This was the clear majority view in my consultation at Opinions requested from home Linux users. Note that my consultation was aimed at home users only: my observations here would be superfluous if Fedora / RHEL / CentOS and their derivatives are only ever installed under corporate management environments. |
|
1816)
Message boards :
GPUs :
Is there a way in Linux Mint 19.1 to delay boinc start at boot?
(Message 91561)
Posted 19 May 2019 by Richard Haselgrove Post: I see one possible point of confusion: BeemerBiker is showing us logs relating to an old, development version of BOINC (v7.9.3 - note the 'odd' middle integer), whereas Germano will be seeking to build onwards and upwards from the current release version 7.14.2 |
|
1817)
Message boards :
Questions and problems :
Need delay in Ubuntu else client does not see GPUs.
(Message 91550)
Posted 19 May 2019 by Richard Haselgrove Post: This was asked three days ago in the GPU area: Is there a way in Linux Mint 19.1 to delay boinc start at boot? I've asked the developers to look into it in #3150: could you nip over to the other thread and answer Germano's questions, please? He's a member of the development group who specialises in packaging BOINC for Linux distros. |
|
1818)
Message boards :
Questions and problems :
Need help in retrieving the sah_v7_opt svn repository
(Message 91520)
Posted 16 May 2019 by Richard Haselgrove Post: SVN update finally completed while I was busy in a meeting. Looks like I have the whole of C:\Sources\setisvn\branches\sah_v7_opt, showing a clean checkout (there are errors in other parts of the tree, which needn't concern us here). The upload of Petri's code was r4000, by Raistmer on 06 February 2019 - no updates since then. 38MB compressed, if you want it. |
|
1819)
Message boards :
Questions and problems :
Need help in retrieving the sah_v7_opt svn repository
(Message 91519)
Posted 16 May 2019 by Richard Haselgrove Post: I'm using Windows Chrome, and following the 'Web interface' link on the 'porting' page gives me this warning: Your connection is not privateBut there is an 'advanced' button, offering This server could not prove that it is setisvn.ssl.berkeley.edu; its security certificate is not trusted by your computer's operating system. This may be caused by a misconfiguration or an attacker intercepting your connection.and the option to proceed anyway. I proceeded, because I know that I'm not about to upload my credit crad details into any repository, SETI or otherwise. Can you access it via the command line with svn checkout?Again, because I run Windows, I'm using Tortoise SVN GUI. I used the 'update' command from my previous working copy 15 minutes ago, and it's still running. But I've already got a new branch C:\Sources\setisvn\branches\sah_v7_opt\PetriR_0.97 - it's going through other admin stuff currently, rather than source code. Why has Petri got folders full of server stuff like assimilator / db / splitter / validate in a v7_opt branch? |
|
1820)
Message boards :
Questions and problems :
Need help in retrieving the sah_v7_opt svn repository
(Message 91512)
Posted 15 May 2019 by Richard Haselgrove Post: I was able to reach https://setisvn.ssl.berkeley.edu/trac/browser/branches/sah_v7_optwithout problems, and it displays as 'source:branches/sah_v7_opt@4031' - confirming checking 4031 is the current head. Note that I am currently accessing the http:// insecure version (I had to persuade my browser to bypass the security check). The checkout command might work better with http, too, despite what the porting page says. |
|
1821)
Message boards :
GPUs :
radeon R7 240 -- ubuntu 18.04 -- boinc not computing on it
(Message 91508)
Posted 15 May 2019 by Richard Haselgrove Post: Unfortunately, BOINC conforms to the traditional definition of insanity: when something goes wrong, it keeps trying to do the same thing in the hope that the outcome will be different. Until someone can work out why that 'Creating Command Queue' call is failing (and having the OpenCL driver version number will help), I think we're stuck. |
|
1822)
Message boards :
GPUs :
radeon R7 240 -- ubuntu 18.04 -- boinc not computing on it
(Message 91504)
Posted 15 May 2019 by Richard Haselgrove Post: OK, we have a smoking gun: ERROR: OpenCL kernel/call 'Creating Command Queue. (clCreateCommandQueue)' call failed (-6) in file ../../src/GPU_lock.cpp near line 1359.That's with the stock SETI@home v8 v8.22 (opencl_ati5_nocal) x86_64-pc-linux-gnuapplication - one for the Linux/ATI gurus to ponder over. I'll have a look round the rest of the tasks for SETI host 8715400, but unfortunately the '100 restarts' limit (all with the same outcome) uses up the available 64 KB stderr space, so we lose the start of the error report. |
|
1823)
Message boards :
GPUs :
radeon R7 240 -- ubuntu 18.04 -- boinc not computing on it
(Message 91502)
Posted 15 May 2019 by Richard Haselgrove Post: vlar_0 absent on output file.That looks like a very incomplete copy of a line which probably started out looking like 15/05/2019 09:37:27 | SETI@home | Computation for task blc33_2bit_guppi_58406_23240_HIP20842_0092.3027.409.21.44.0.vlar_1 finishedPlease give us more detail like that - it helps us to look up and find out more about what's going wrong. From the full name, we can reach task 7678888327 and take it from there. |
|
1824)
Message boards :
GPUs :
Is there a way in Linux Mint 19.1 to delay boinc start at boot?
(Message 91469)
Posted 13 May 2019 by Richard Haselgrove Post: I started typing 'Linux delay serv...' into Google, knowing the BOINC runs as a service under Linux. It's a common enough question for Google to autocomplete it as linux delay service startup on bootIf you paste that into the search engine of your choice, you should get some good hints. I'm not a Linux person, but I think that I've seen this question often enough to propose that we ought to have the delay built in to the systemd we ship out with the repository ppas (if I'm not muddling two different bits of terminology there). If you find a working solution among the search results, please post it here, or if a Linux specialist reading here would like to take up the challenge, we could maybe fix this at source. |
|
1825)
Message boards :
The Lounge :
acknowledgement letter
(Message 91461)
Posted 10 May 2019 by Richard Haselgrove Post: There are a lot of SETI research groups around the world. Several, including SETI@Home, are grouped under the Berkeley SETI Research Center. The SETI Institute, although close by San Francisco and Berkeley, is not directly connected to the Berkeley initiatives, although they are both active in the same worldwide research effort. Some years ago, NASA gathered together a much wider group of interested researchers from around the world. The proceedings of their week-long conference can be seen at https://nai.nasa.gov/seminars/featured-seminar-channels/conferences-and-workshops/2014/3/16/search-for-life-beyond-the-solar-system-conference/ You can see that both Jill Tarter (SETI Institute) and Eric Korpela (SETI Programs at the University of California, Berkeley) were relegated to a footnote session on the final afternoon. |
|
1826)
Message boards :
Questions and problems :
Screen name.
(Message 91447)
Posted 9 May 2019 by Richard Haselgrove Post: The only part of "BOINC Central" that has any need to consider the implications of GDPR are these forum.For the record - not true. "BOINC Central" provides and maintains all the backroom tools that give effect to user choices, while at the same time maintaining the integrity of the database which tracks workunits, tasks, time, credit and so on. 'Users' and 'Tasks' are tightly integrated: Einstein issues certificates to the users who discover new pulsars - so they need to know who ran what. And many other examples: GPUGrid issues badges for contribution levels to each separate published research paper. GDPR requires that user data like that can be deleted on request. History, integrity, and the 'right to be forgotten' required a delicate juggling act, involving (IIRC) replacing user and computer ID numbers with dummy values when the user withdraws consent. |
|
1827)
Message boards :
Questions and problems :
Download hang - World Community Grid
(Message 91444)
Posted 8 May 2019 by Richard Haselgrove Post: Use the <http_debug> Event Log option (Ctrl+Shift+F from BOINC Manager) to find out exactly what sort of "transient HTTP error" is being reported. Ask again if you can't make any sense of the detailed messages. |
|
1828)
Message boards :
Questions and problems :
GPU detection failed, error 0xc0000005
(Message 91442)
Posted 8 May 2019 by Richard Haselgrove Post: Any error code like that is pretty much guaranteed to be a Microsoft Windows NT status code, and very hard for anyone except a programmer to track down. 0xc0000005 is STATUS_ACCESS_VIOLATIONWith a history like yours, it's possible that left-over code from the nvidia GPU is trying, but failing, to get information about the card you disabled. I'd suggest that you run a utility like Display Driver Uninstaller to clean out the old, unwanted, code, reboot, and then run the Windows 10 updater to replace the intel driver that you do still want. |
|
1829)
Message boards :
Questions and problems :
Old reliable host can't connect to E@H
(Message 91400)
Posted 5 May 2019 by Richard Haselgrove Post: Yes, the log would have been conclusive, but never mind. The security certificate bundle that BOINC uses for https links was updated around v7.10, so the older version could have had that problem. But 5/4/2019 4:36:04 PM | | Project communication failed: attempting access to reference siteindicates that google.com could be contacted, not as stated. So there's still a query there. |
|
1830)
Message boards :
BOINC client :
What does busy mean in work_fetch event log entry, also rr_sim doing something strange with time slice.
(Message 91392)
Posted 5 May 2019 by Richard Haselgrove Post: I think it's to do with https://github.com/BOINC/boinc/blob/master/client/work_fetch.h#L153: // estimate the time a resource will be saturatedHigh-priority jobs are those in danger of missing their deadlines. They are run as 'Earliest deadline first' in an effort to return them as quickly as possible, jumping the normal resource-share queue. |
|
1831)
Message boards :
Questions and problems :
Old reliable host can't connect to E@H
(Message 91384)
Posted 5 May 2019 by Richard Haselgrove Post: When I paste that url into a browser, I get a perfectly good reply from https://einsteinathome.org/get_project_config.php so there's some redirecting going on. Try pasting that url into the address box on the attach page. |
|
1832)
Message boards :
Questions and problems :
Old reliable host can't connect to E@H
(Message 91381)
Posted 4 May 2019 by Richard Haselgrove Post: Einstein is pretty hot on respecting the new GDPR European privacy laws. You're asking (I think) to reconnect to an old account: you shouldn't trigger a 'new account' message. (They might be refusing - still - to accept new accounts via BOINC Manager, unless you're using the latest client which can show you the 'terms of use' page and record your acceptance). It would help if you could describe exactly how you're trying to make this re-connection: they might have a problem after all the privacy upheavals, but we can't get them to fix it if we don't know exactly what you're doing. |
|
1833)
Message boards :
Questions and problems :
CPU usage claims 100% but CPU temperature doesn't reflect this?
(Message 91364)
Posted 3 May 2019 by Richard Haselgrove Post: You could be using 100% of a tiny integer NOP loop, and no memory, while waiting for something to happen. Or you might be using 100% of the AVX SIMD power of the floating point unit, while shovelling 64-bit floats in and out of RAM. They both show as 100%. |
|
1834)
Message boards :
Projects :
CPDN: Tasks using a lot of memory.
(Message 91360)
Posted 3 May 2019 by Richard Haselgrove Post: I'll add David Wallom's comment from this morning: We are definitely in two camps, one with average CPU times of ~ 900s and one between 4000s - 9000s. With the extremely high memory load I would envisage the difference is caused by memory and bus type not just processors.and move this thread to the Projects area for future reference - CPDN users may wish to set max_concurrent via app_config.xml for this task type. |
|
1835)
Message boards :
Questions and problems :
Screen name.
(Message 91351)
Posted 2 May 2019 by Richard Haselgrove Post: I asked on the contributors' call, but we haven't got anything else to suggest. The GDPR is new, complex, law, and will only be fully defined when the first few test cases have been determined by a Court. Until then, amateur interpretations which turn out to be wrong might establish the wrong sort of liability... TACC identifies itself on its website as an offshoot of the University of Texas: universities in general have well established legal offices in their administrative structures. Might be best to ask for help there. |
|
1836)
Message boards :
Questions and problems :
Queue size (preferences) not being used on SETI - why?
(Message 91349)
Posted 2 May 2019 by Richard Haselgrove Post: The basic principal is is that the client does all the work requesting. How the settings reach the client depends on what you've done. If you make any changes locally, then what you've set in the client takes precedence - as your lower left screenshot makes clear ("Using local preferences"). I'd recommend using the extended Event Log messages. 'Sched_op_debug' is quiet enough to be used permanently, and gives you a better idea what you're asking for and what you got in reply. The brutal, but most effective, is 'work_fetch_debug'. It's too verbose to use permanently, but you may need it for this one. Set it, apply, wait 10 seconds, unset it, and save. Then go see what you caught, and post it here for advice. |
|
1837)
Message boards :
Questions and problems :
Multiple boinc clients are shown in "Startup Task" page
(Message 91340)
Posted 2 May 2019 by Richard Haselgrove Post: You're over-engineering things. BOINC has it's own control in the Options menu in BOINC Manager: Other options --> Run Manager at login? Uncheck that - job done. Task Manager shouldn't be used or necessary unless things have got seriously screwed up. Some of those entries will be for supporting tools most people have never heard of, like boinctray.exe |
|
1838)
Message boards :
Questions and problems :
Screen name.
(Message 91321)
Posted 1 May 2019 by Richard Haselgrove Post: I'm slightly surprised that TACC are out of the loop on this, because they were one of the co-sponsors with David Anderson of his new Science United proposal. Legal advice on GDPR was a bit of a last-minute scramble, and multiple sources were used. The lead people for BOINC were Kevin Reed of WCG (outside the EU, but advised by IBM); Oliver Behnke of Einstein (inside EU, at least the German part); and Laurence Field of CERN (straddling the EU border!). But I'm not sure if their accumulated wisdom has been gathered together into an easy reference document. I should be speaking with at least two of those people tomorrow evening, and I'll ask. It might also be worth mentioning to TACC that there is a relatively new provision of a 'server stable' release version of the BOINC server code, currently accessible by cloning the 'Server_release/1.0/1.0.4' tag in the BOINC github repository. The server release was last updated on 6 March 2019, so it should include all the GDPR-related changes from last year. |
|
1839)
Message boards :
Questions and problems :
Screen name.
(Message 91318)
Posted 1 May 2019 by Richard Haselgrove Post: That was the reference, yes. I was never a real fan, but I was around when it was broadcast and saw some of the episodes. |
|
1840)
Message boards :
Questions and problems :
Screen name.
(Message 91316)
Posted 1 May 2019 by Richard Haselgrove Post: A lot of work was done around this time last year to make the core BOINC code GDPR-compliant. The main problems were to do with getting informed user consent for the storage and display of their personal information; informed consent for the transfer of that personal information to third parties (external stats sites); and implementing the 'right to be forgotten' (ability to delete accounts). I don't remember giving you a random name being part of the process, let alone taking away your ability to change it unilaterally. Just be grateful that you are not a number. |
|
1841)
Message boards :
Questions and problems :
Screen name.
(Message 91314)
Posted 1 May 2019 by Richard Haselgrove Post: Since Texas isn't a member of the European Union, I suspect their lawyers have failed to fully comprehend the GDPR. |
|
1842)
Message boards :
Questions and problems :
Why am being forced into two consecutive 24 delays with no work returned yet
(Message 91312)
Posted 1 May 2019 by Richard Haselgrove Post: You returned the 'disk limit exceeded' errors at 30 Apr 2019 3:58:40 UTC, and requested new work at 1 May 2019 0:44:29 UTC - that's less than 24 hours. I'm not sure exactly when Einstein resets it's 'daily' clock in relation to your time zone, but it might be worth a manual update when you wake up in the morning. (In this context, it's the Einstein server which is setting the delays) |
|
1843)
Message boards :
Questions and problems :
Why am being forced into two consecutive 24 delays with no work returned yet
(Message 91297)
Posted 30 Apr 2019 by Richard Haselgrove Post: Fingers crossed. It might be wise to mark those library files as <executable/> as well - they certainly contain binary code which is going to be executed. |
|
1844)
Message boards :
Server programs :
mariadb index limitaion (ERROR 1071 (42000))
(Message 91293)
Posted 30 Apr 2019 by Richard Haselgrove Post: It depends what you're planning to use the database for. If it's for private, in-house usage only, and you can control exactly what's written into it, then you could be fine. But if it's facing an international public audience, then you have to use a character set which can handle everything which is being thrown at it, currently up to 4-byte emojis. I was sent a PM here yesterday, which appeared to contain ðŸ‘�: it turned out to be a simple 👍. You've probably seen some curious versions of your (accented) name over the years. Further discussion of the character sets used in an international setting at #2117. |
|
1845)
Message boards :
Questions and problems :
Why am being forced into two consecutive 24 delays with no work returned yet
(Message 91289)
Posted 30 Apr 2019 by Richard Haselgrove Post: Problems that require <copy_file/> to solve also arise when multiple different versions exist and have to be stored under different names in the filing system, but programmers are told to build their apps expecting a simple, generic, name. It's such a bloomin' nuisance that it has its own special name under Windows - 'DLL Hell'. NVidia were guilty of it in the earliest days - recycling the generic cudart.dll and cufft.dll over several (incompatible) generations. Under Windows, they've learned their lesson, and now use a strongly versioned name for each new release. It can be necessary to do a <copy_file/> with rename (<file_name> / <open_name>) to unambiguously resolve the confusion, but let's hope not. It's another reason to use Dependency Walker, to find out exactly which version of the filename has been embedded in the binary. Far more robust than using a Hex editor for the same purpose... |
|
1846)
Message boards :
Projects :
CPDN: Tasks using a lot of memory.
(Message 91287)
Posted 30 Apr 2019 by Richard Haselgrove Post: Since it's Dave, I'm assuming CPDN. [Mod hat on] I'll leave it here for a day or two to catch the attention of general readers - see if they can help. Then I'll move it into the Projects area with a CPDN tag in the title, for future reference. [Mod hat off] |
|
1847)
Message boards :
Server programs :
mariadb index limitaion (ERROR 1071 (42000))
(Message 91280)
Posted 30 Apr 2019 by Richard Haselgrove Post: There's no definitive fix, but it's a known problem - some discussion and suggestions in #2140. Also an external reference to StackOveflow: https://stackoverflow.com/questions/1814532/1071-specified-key-was-too-long-max-key-length-is-767-bytes |
|
1848)
Message boards :
Questions and problems :
Why am being forced into two consecutive 24 delays with no work returned yet
(Message 91275)
Posted 30 Apr 2019 by Richard Haselgrove Post: The Windows and Linux processes are different. BOINC started by using Linux conventions, so you may be in luck, but I remember having to introduce David Anderson to https://docs.microsoft.com/en-us/windows/desktop/dlls/dynamic-link-library-search-order#search-order-for-desktop-applications when the Linux techniques failed under Windows. You may have to reverse the process. |
|
1849)
Message boards :
Questions and problems :
Why am being forced into two consecutive 24 delays with no work returned yet
(Message 91274)
Posted 30 Apr 2019 by Richard Haselgrove Post: You mean lines like 29-Apr-2019 19:51:46 [Einstein@Home] Aborting task p2030.20170414.G44.77-01.19.C.b0s0g0.00000_1646_1: exceeded disk limit: 127.11MB > 19.07MBI think that will be an individual limit on the task - you wouldn't have a limit below 20 meg on any global or system disk usage. Looking on my machine, I can see <workunit>
<name>p2030.20170414.G44.61-02.33.N.b5s0g0.00000_1438</name>
<app_name>einsteinbinary_BRP4</app_name>
<version_num>134</version_num>
<rsc_fpops_est>17500000000000.000000</rsc_fpops_est>
<rsc_fpops_bound>350000000000000.000000</rsc_fpops_bound>
<rsc_memory_bound>260000000.000000</rsc_memory_bound>
<rsc_disk_bound>20000000.000000</rsc_disk_bound>
...That last line of 20,000,000 is the culprit - it translates to 19.07 binary MiBWhy such big disk usage? My guess is that you've put a <copy_file/> on that big cuFFT library, so the whole darn thing is copied into the slot folder and counts towards disk usage. Try just allowing BOINC to create a softlink as usual (remove <copy_file/>, leave the rest of app_info unchanged) and run a test task. I think Linux should be able to follow softlinks, but I don't know for sure where libraries are concerned. If it doesn't work, you'll have to find a way of installing the FFT library in such a way that Linux finds it by whatever it uses as a 'PATH' equivalent. My Beta task from ten years ago actually succeeded where I'd expected it to fail, because I'd installed NVidia's developer SDK and the system had a way of finding the library in that package. Drivers don't install FFT support, but the developer tools do - but it has to be exactly the right version, and it changes with every CUDA version release. |
|
1850)
Message boards :
Questions and problems :
Why am being forced into two consecutive 24 delays with no work returned yet
(Message 91272)
Posted 29 Apr 2019 by Richard Haselgrove Post: When I was preparing the AIstub files under Windows, I used SysInternals' Dependency Walker to see what libraries were needed: I don't know the name of the equivalent tool under Linux, but there must be one. The results of not checking can still be seen at SETI Beta message 39386 |
|
1851)
Message boards :
Questions and problems :
Suspending computation (computer is in use) function ignores external mouse and keyboard
(Message 91270)
Posted 29 Apr 2019 by Richard Haselgrove Post: Added to an ongoing discussion in #3044 concerning static versus dynamic detection of HIDs (Human Interface Devices]. |
|
1852)
Message boards :
Questions and problems :
Why am being forced into two consecutive 24 delays with no work returned yet
(Message 91266)
Posted 29 Apr 2019 by Richard Haselgrove Post: By default, the daily quota reduces to a minimum of one task per day - that gives you an escape route. I imagine the same applies to Einstein, although they have heavily customised their code. Is this the machine you are running under Anonymous Platform (app_info.xml)? If so, ensure that any extra file you put in the project directory is properly declared and referenced - otherwise it may not be available when the app is actually run from the slot directory. |
|
1853)
Message boards :
Questions and problems :
Why am being forced into two consecutive 24 delays with no work returned yet
(Message 91263)
Posted 29 Apr 2019 by Richard Haselgrove Post: Einstein says you've exceeded your daily quota: 2019-04-29 01:11:56.3583 [PID=29900] [send] stopping work search - daily quota exceeded (24>=11) 2019-04-29 01:11:56.3622 [PID=29900] Sending reply to [HOST#12775352]: 0 results, delay req 85052.00It's "until tomorrow", not quite the full 24 Your quota is low because you've errored every one of the 43 attempts so far: try to use the remaining time to fix the error, then you can start increasing the quota each time you return a successful completion. Edit - you're missing a library: ../../projects/einstein.phys.uwm.edu/einsteinbinary_cuda64: error while loading shared libraries: libcufft.so.8.0: cannot open shared object file: No such file or directory |
|
1854)
Message boards :
BOINC client :
7.14.2 and 7.12.1 both fail to get work units on very fast systems
(Message 91255)
Posted 29 Apr 2019 by Richard Haselgrove Post: When I look at https://milkyway.cs.rpi.edu/milkyway/server_status.php it says at the bottom Task data as of 29 Apr 2019, 10:41:21 UTC. That's about 20 minutes ago. A LOT can change in 20 minutes, so it now showing to have 11683 tasks RTS means that they were that number at 10:41:21 UTC, not now. Aside from that, even if that number is a constant 11K+, you still don't know how many tasks were in the feeder at the given moment your client asked for work. If there were none, or way fewer than you're asking for, it won't give you work.Also, two other things are not included in their (BOINC standard) template SSP: Number of tasks available per app - NBody or separation. It wouldn't help if most of the 11683 were NBody. Number of tasks reported per hour - with a maximum runtime of 43 seconds, that could be huge. (22377 recent computers all doing a task every 40 seconds is over 2 million tasks reported every hour - you get an idea how much is actually happening under the hood) |
|
1855)
Message boards :
BOINC client :
7.14.2 and 7.12.1 both fail to get work units on very fast systems
(Message 91254)
Posted 29 Apr 2019 by Richard Haselgrove Post: Pure coincidence on the reply - I'd been downstairs doing something else, and just happened to come back to the computer and look for messages as you posted. Since then, I've been out to the shops and back. Timing matters. Which may also have a bearing on this problem. With the Milkyway tasks finishing so quickly, anyone who gets the 'no work' reply and goes into backoff woll clear the backoff via a completion and will ask again - the delay being solely governed by the 91 second server delay. It's possible that you're seeing a whole group of hosts in lockstep - all asking (and asking again) at the same time. The client backoffs were deliberately designed with a randomisation factor to avoid that problem, but there's nothing random about the server delay. The advantage of the 'update' click could be as simple as that slight extra randomness in the timing. Oh, wait. The 'zero sent' continues even after the cache is empty and there's no completed work either to clear the backoff, or needing to be reported? That destroys both our hypotheses! |
|
1856)
Message boards :
BOINC client :
7.14.2 and 7.12.1 both fail to get work units on very fast systems
(Message 91250)
Posted 29 Apr 2019 by Richard Haselgrove Post: See my following post. It's the difference between warehouse storage and front desk pickup. You see the numbers in the warehouse, and sure - there's plenty of work back there. But you don't see what's on the front desk - the project doesn't show that to you, because it changes several times a second. All that clicking 'update' does is to send an immediate request (if you need work) - it doesn't change the state of the front desk pickup supplies. You stand exactly the same chance of success as if BOINC had asked automatically. |
|
1857)
Message boards :
BOINC client :
7.14.2 and 7.12.1 both fail to get work units on very fast systems
(Message 91247)
Posted 28 Apr 2019 by Richard Haselgrove Post: The TV cliffhanger is nicely set up for the feature-length series closer next week. Meanwhile, back at BOINC... Quickly, before I fall into bed. The BOINC structure is uniform across projects, with minor local tweaks. There are two numbers to consider - and please pass these on to Jake for consideration. The first is the total number of workunits created by what are generically known as workunit generators, and are familiar to SETIzens as 'splitters'. At SETI, this number hovers around the 600,000 level, and is subject to some hysteresis - it takes time to turn off the generators when the RTS buffer hits peak, and they aren't called back into service until it falls to trough. The SETI generator has a typical wavelength of around an hour: I don't have a ready knowledge of the Milkyway figures, or even whether there's an equivalent of the Haveland graphs. The RTS buffer is stored in databases and disk files. The second number relates to tasks held in fast cache memory by a process known as the 'feeder'. That number's probably measured in hundreds, and has a cycle time measured in seconds. When you request work, it's the tasks in the feeder cache which are scanned for suitability: you need that fast cache response time. "No tasks allocated" equates either to "feeder empty" or "no suitable tasks in feeder". It's the old trade counter vs warehouse problem: Are these in stock? Yes Can I pick one up? No Why not? Because I'd have to send Joe over to the warehouse first, and he's on lunch My suspicion, reading this thread, is that Jake is talking about the workunit generators and RTS: I don't think he's reached the page in the manual about the feeder yet. Perhaps one of you could point him to https://boinc.berkeley.edu/trac/wiki/BackendPrograms#feeder, but tell him not to blindly wind up <feeder_query_size>N</feeder_query_size>to something obscene: both timing (the scheduler has to search the list) and size (it has to fit into memory without paging) are critical. And so to bed. |
|
1858)
Message boards :
BOINC client :
7.14.2 and 7.12.1 both fail to get work units on very fast systems
(Message 91242)
Posted 28 Apr 2019 by Richard Haselgrove Post: OK, I can manage the next bit now. The next request was 15823 Milkyway@Home 4/28/2019 10:14:06 AM Scheduler request completed: got 0 new tasksThat's effectively 91 seconds after the one in my last post. In this mode, the client will set a backoff every time it gets a got 0 new tasks response, and will clear it every time one of your previously cached tasks finishes computation. Those are both known design features, like 'em or not. At the moment, you're completing cached work much more often than once every 91 seconds, so you'll keep asking at every possible occasion. But as soon as you run dry, there's nothing to clear the backoffs (no more work finishing), and they'll build up. The only way to cure this one is to persuade the project to provide more tasks. A possible mitigation would be to set a cron job to trigger boinccmd to issue a project update every five minutes, but you might get very unpopular with the server administrators. They obviously aren't set up to satisfy your voracious appetite. |
|
1859)
Message boards :
BOINC client :
7.14.2 and 7.12.1 both fail to get work units on very fast systems
(Message 91241)
Posted 28 Apr 2019 by Richard Haselgrove Post: Richard is the expert in decrypting the work fetch debug output. Everything appears normal for intervals. The work requests look normal. What I don't understand is why you are getting backoffs for 10 and 5 minutes directly after the scheduler acknowledges receipt of reported work. That is coming from the scheduler and not from your host or client. Normally the scheduler backs off if there are issues in contacting the servers or the client has issues downloading work and the client can't acknowledge correct reception of the sent tasks. Have you looked at the Transfers tab in the Manager after you have requested work and see if you have task downloads in backoff?No need. Look at that event log again, without the clutter: 15402 Milkyway@Home 4/28/2019 10:12:30 AM Scheduler request completed: got 0 new tasksThe project requested 91 seconds. The backoff was done by the client, as a normal reaction to the lack of available work. And if no work was assigned by the server, there'll be no files to download, and nothing will show in the transfers tab. Sorry, I've had a busy weekend showing visitors round Yorkshire. They've moved on to London now, but I found myself surprisingly tired (and I've got a watercooler appointment with the TV later tonight). I should be back to normal tomorrow, and I'll try to look through the rest of the log before your morning starts. |
|
1860)
Message boards :
BOINC client :
7.14.2 and 7.12.1 both fail to get work units on very fast systems
(Message 91234)
Posted 28 Apr 2019 by Richard Haselgrove Post: Yes, <work_fetch_debug> is a bit of a blunderbuss. Best to set it once, wait until it's done just one cycle, and then unset it again while you pick over the pieces. [That's why I got them to put an 'apply' button on the dialog :-)] But it is powerful - if you could fillet out that one complete cycle from [work_fetch] ------- start work fetch state ------- to [work_fetch] ------- end work fetch state ------- and post it here, we could take a look. Might contain some clues. |
|
1861)
Message boards :
BOINC client :
7.14.2 and 7.12.1 both fail to get work units on very fast systems
(Message 91232)
Posted 28 Apr 2019 by Richard Haselgrove Post: I think we need to understand a little bit better where this delay is coming from. First, every time you contact a project, the project itself asks you to pause a little while before asking again - SETI for 303 seconds, Milkyway for 91 seconds. That shows in the Event Log, and it's best to have regard to it, but it obviously isn't the problem here. Most of the time, it's your machine which does the asking for work, when it thinks it needs it - and that depends on your cache settings. It's usually best to ask 'little and often', but that does depend on whether the project regularly has work to send you. I'd always suggests setting the <sched_op_debug> Event Log flag (which you can do from the Manager, Options menu) - it gives you the very basic stuff, like 28/04/2019 07:09:43 | SETI@home | [sched_op] CPU work request: 0.00 seconds; 0.00 devices 28/04/2019 07:09:43 | SETI@home | [sched_op] NVIDIA GPU work request: 4332.36 seconds; 0.00 devices 28/04/2019 07:09:43 | SETI@home | [sched_op] Intel GPU work request: 0.00 seconds; 0.00 devices- I usually like to request about an hour of work, once an hour, but we can tweak that later. The problems like you're describing more usually arise when you wait and ask for a lot of work at once, and happen to hit a moment when the project hasn't got any - that tends to be when BOINC decides not to bother asking for a while, and fetch from a backup project instead. But we need to understand the overall picture before we can be sure of that. Got to go out now, but the artifacts Keith mentioned are at https://ci.appveyor.com/project/BOINC/boinc/builds/23992763/artifacts. I'll dig out some notes on how to use them later. Late edit - here are the notes I wrote last time for using those artifacts. You'll need the middle one of the three downloadable links, labelled 'win-client'. |
|
1862)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 91227)
Posted 27 Apr 2019 by Richard Haselgrove Post: And this morning my first HOT shower in a week! What a pleasure!The mystery of the shower may have been solved! This morning, for the first time in four visits, Uli couldn't get any hot water out of it either - even though the bathroom was still covered in steam from the shower I'd had five minutes earlier. Turns out you have to be bold with it - turn the tap on full, 180° turn. That wakes the boiler up. |
|
1863)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91195)
Posted 24 Apr 2019 by Richard Haselgrove Post: Or even the hail. |
|
1864)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91183)
Posted 23 Apr 2019 by Richard Haselgrove Post: Trying to remember a very forgettable thing, the longest classic outage. IIRC it was nearly a month long, but I can't remember the exact length. Anyone have better memory?Didn't we have a long one when thieves stole the data cable linking the lab to campus? |
|
1865)
Message boards :
The Lounge :
BOINC Politics thread
(Message 91171)
Posted 23 Apr 2019 by Richard Haselgrove Post: There I was, hoping that Trump and the Queen could gang up on Theresa May and persuade her to resign... :-) |
|
1866)
Message boards :
The Lounge :
That traveling Dutchman is coming back to Britain and is no longer going to Germany
(Message 91161)
Posted 21 Apr 2019 by Richard Haselgrove Post: You'll be glad you missed this on your last visit to Richard's part of the country. Ilkley Moor fire: Crews battling 'intense' moorland blazeHaven't been up there to investigate yet - it's centred on the other side of the moor from me - but we often see small columns of smoke from up there. Bradford Council has banned game shooting on the public parts of the moor, but the private parts are actively managed for, I think, grouse, and the old heather is burned off in late winter to encourage new ground cover for the young chicks in spring. As usual, the media is getting over excited about their 'dramatic' pictures: Notre Dame it ain't. A few years ago (2006), we had Aftermath of moor fire will be felt for years, with supposedly centuries of peat below the heather destroyed for ever. It was back to normal within a couple of years, with a bit of normal winter rainfall and regrowth. The fire crews will have a bit of extra overtime because of some silly picnicker, and then it'll be forgotten when the regular news cycle picks up again. Check LBA current conditions for a more realistic view. |
|
1867)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91140)
Posted 18 Apr 2019 by Richard Haselgrove Post: We've just finished the developer conference call, and we reached a consensus that the proposal to remove the menus items from Linux will be shelved for now - the menus remain unchanged for general users. It's possible that they might still be removed from the service installs made from repository distributions, but that would be selective and wouldn't impact home users building from source. Later, we'll try to find time to investigate the whole Linux situation - User / Service modes, and starting / stopping the client in each mode - in more depth. Your comments here will be most useful, but in the meantime, please considered the first phase of this consultation closed. Thanks for your input. |
|
1868)
Message boards :
BOINC Manager :
Limit Work By CPU Core Temperature
(Message 91108)
Posted 16 Apr 2019 by Richard Haselgrove Post: Take a look at TThrottle, which has been written independently to work alongside BOINC and do what you suggest. |
|
1869)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91098)
Posted 15 Apr 2019 by Richard Haselgrove Post: Cogent reasons for something like 'delayed start' have been expressed at SETI - in particular, to allow time to ensure that video drivers, and hence GPU runtime components, have initialised and are ready for use. It's on my ToDo list to pass those back to the developers in the context of this service discussion, but I want to get the fine-tuning of the available controls across first. |
|
1870)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91096)
Posted 15 Apr 2019 by Richard Haselgrove Post: I think that 'active users' more often use the 'shut down connected client' or 'stop service' tools prior to carrying out configuration, maintenance, or upgrade changes which can only be carried out when the client is inactive. For computers which are shut down at certain times of day for energy conservation reasons (either overnight when the user is asleep, or during the heat of the day where air conditioning is in use), some long-running project applications are felt to be sensitive to abrupt closedowns and are better handled with a deliberate BOINC 'stop' before the computer shutdown. Les runs one of those projects - note how his troubleshooting/maintenance routine involves a triple action - suspend, exit, reboot. That comes from experience with his favoured project.They can also suspend all projects indefinitely, without limit of time - which might be almost as bad as stopping the service.Isn't that just what people try to achieve when they stop the service? Or do they also want to remove the applications from memory? Could't that be done without actually stopping the client? In that case it wouldn't be a problem to simply resume. |
|
1871)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91095)
Posted 15 Apr 2019 by Richard Haselgrove Post: I'm not immediately aware of any formal documentation on the tray icon, but I'll have a look later. Also, I'm a Windows user, so not all features may be identical: but from memory, the tray icon actions are: double-click: restores a minimised Manager to its previous on-screen position. right-click: context menu including suspend, snooze, about, exit. hover (no click): tooltip with current activity status. no action: alert for new arrivals in the 'notices' tab. The tray icon is displayed by the main BOINC Manager application, not (as is frequently misunderstood) by the separate windows 'boinctray.exe' applet which monitors mouse and keyboard activity to detect when the computer is "in use" (as BOINC defines it). |
|
1872)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91092)
Posted 15 Apr 2019 by Richard Haselgrove Post: If something goes wrong and everything locks up, (not often), I Suspend models and BOINC, Exit BOINC, and then reboot.That's the bit where your experience overlaps into this enquiry. When you say 'Exit BOINC', can you confirm whether your intention is to exit the Manager, the service, or both? What control do you use to effect the Exit? |
|
1873)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91087)
Posted 14 Apr 2019 by Richard Haselgrove Post: Richard, thanks for the link to the SETI discussion.The discussion has moved on since you read it, and we've had quite a discussion about root and sudo. I'm clearer now, and I've taken the liberty of reposting your ITMOTB link. I think it's worth reposting part of my latest thoughts here too. I think the proposal which led to this discussion came from the Debian stable, which I think counts as "real *nix"? The originator's actual words are: |
|
1874)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91078)
Posted 13 Apr 2019 by Richard Haselgrove Post: Yes, Gary has identified the correct precursor that led to this discussion (and - when this one got off to a slow start - a parallel discussion at SETI). It seemed to me that the proposed change - removing, for example, the tools to stop a running client - came from a contributor who was firmly in the 'Linux as multiuser' camp, who had failed to consider the wider environment. Computer science is evolving all the time. When the Linpack top-500 list was first compiled in June 1993, the top-10 page contained Cray supercomputers with 16 cores, and NEC machines with just four cores. The current (November 2018) top 10 still contains Crays, but the smallest has 387,872 Intel Xeon cores and an unknown number of Tesla P100 GPUs. I think I'd call that a cluster, rather than a computer. Similarly, operating systems keep evolving. I think that it's wrong to characterise Linux as multiuser and Windows as single user: Gary has made the case for the single-user mode of operation with Linux, but Windows is multi-user, too. Windows is commonly installed in corporate settings where individual nodes run under group policies centrally set by a Windows server running as a Domain Controller. That even allows features like hot-desking - an individual user with access rights controlled by an administrator can log on to any node on the system, and find that their installed programs, data folders, and even desktop appearance follow them from place to place. But they still wouldn't have the power to install BOINC on any of them. An IT support worker with the (jealously guarded!) master password would have to do that for them. I'm always amused when a public display screen, ticket machine, or bank ATM is stuck on a Windows XP boot error message. I think that BOINC should be as flexible as possible, and that contributors from one particular historical background shouldn't mould it too rigidly into their particular world-view. But I started this thread to gather alternative opinions, aware that I shouldn't be too rigid about my own world-view, either. |
|
1875)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91075)
Posted 12 Apr 2019 by Richard Haselgrove Post: It would be a nightmare for a BOINC developer to maintain all those possibilities.Would it be possible for BOINC (centrally) to define a name for a script to be run, and for the package maintainers to define the contents of the named script for their distro? |
|
1876)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91074)
Posted 12 Apr 2019 by Richard Haselgrove Post: Therefore the ability to stop the Client should be removed. There are clear alternatives to just removing the option and I now wonder why they where never mentioned in the github PR or issue.Here, here. That's why I got involved, and why we're having this discussion. My feeling is that the advocates of the current issue / PR had a particular (and narrow) view of 'the typical Linux user' - probably guided by the Linux users in their personal workspace - and proposed solutions that seemed natural for that particular stereotypical user. |
|
1877)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91072)
Posted 12 Apr 2019 by Richard Haselgrove Post: ...there is a different "UAC" system (visual sudo) in every distribution...Users say they only know of gksudo, and it was deprecated in Ubuntu from 18.04 Ubuntu forums suggest sudo -H Ideas? Edit - https://bugs.launchpad.net/ubuntu/+source/umit/+bug/1740618 |
|
1878)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91071)
Posted 12 Apr 2019 by Richard Haselgrove Post: But surely that's something the package preparer (repo manager) can configure for the distro they're in charge of? That's where this whole subject originated. |
|
1879)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91069)
Posted 12 Apr 2019 by Richard Haselgrove Post: @ Christian, I think things have moved on a little since the situation you describe. I used to run BOINC in service mode under Windows XP, but all my recent installations have been in user mode because of the Windows 7 (and later) restrictions on GPU drivers. So today, as part of this enquiry, I installed BOINC in service mode on a newish, cleanish machine running Windows 10 Pro v1809. I took the plain vanilla installer for v7.14.2 from the download page here. Everything worked as I remembered, and all the stop/start/snooze/suspend/resume controls worked as normal. Except for one. I could shut down the connected client immediately, but when I tried to restart the Client from the Manager, I had to acknowledge a Windows UAC dialog confirming that I wanted to make changes to the system. I think that highlights that all communications between the Client and the Manager (information display in one direction, command and control in the other direction) are handled by network RPC protocols: these bypass all the user account and security sandboxes, and explain why you can manage a remote machine in exactly the same way as you manage your local machine (the only possible block is at the firewall level). Starting the client in service mode, under a different user security regime, is the only time that the Manager has to work with operating system security protocols - and I think that applies equally to Windows and Linux. |
|
1880)
Message boards :
BOINC Manager :
Opinions requested from home Linux users
(Message 91056)
Posted 11 Apr 2019 by Richard Haselgrove Post: I'm a user of BOINC under Windows. I'm familiar with the the separate Client and Manager components; the difference between Service mode (runs when no user is logged in) and User mode (required under Windows for GPU computing, because of driver restrictions); how to stop, pause, snooze and restart the client; and much else. All of these apply to the Linux version of BOINC as well. I think the only significant change to what I've just said is that GPUs can be used in Service mode under Linux. Because Windows is controlled centrally by Microsoft, there are few active versions in circulation at any one time, and it is relatively easy for the BOINC developers to manage 'one size fits all' in the Windows desktop environment. But Linux is available in multiple different flavours, and with multiple alternative desktop environments (or so I understand - I don't pretend to know the details). For this reason, BOINC distribution has been largely delegated to the distribution managers who maintain the software installation packages. By default, these distributions of BOINC are installed in Service mode. There is a move to make the Linux version of BOINC effectively 'Service Mode Only', and to remove certain controls which a Windows would expect to find in their user-mode interface: the ability to snooze the client from the system tray icon, for example, and to 'Shut down the connected client' (local or remote) from the main BOINC Manager menu. I'm in regular contact with BOINC developers, but I'm not competent to debate the merits or de-merits of this particular proposal. Hence this consultation thread. Please add your comments, and perhaps indicate your particular Linux flavour, and your level of experience with it (advanced, intermediate, beginner, or refugee from Windows...) |
|
1881)
Message boards :
Questions and problems :
Presence of "Ghosts Tasks" prevents Downloads ?
(Message 91054)
Posted 11 Apr 2019 by Richard Haselgrove Post: The original statement you queried was "Checking presence of 572 project files". Note the word 'files'. These are data files stored on the hard disk of your computer by the project - they include the application programs used to run the tasks, graphics used in Simple View, the buttons in advanced view to link to different pages on the project website. They will also include any data files used for any of the tasks, but there isn't a 1-to-1 correlation between 'files' and 'tasks'. |
|
1882)
Message boards :
Questions and problems :
Presence of "Ghosts Tasks" prevents Downloads ?
(Message 91052)
Posted 11 Apr 2019 by Richard Haselgrove Post: I'd also say that those messages don't look like standard BOINC messages. WCG uses different server code from the rest of BOINC, so it's probably better to ask on the WCG forums as well. |
|
1883)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91041)
Posted 10 Apr 2019 by Richard Haselgrove Post: Well, I saw that can't be found message too and it loaded instantly indicating no entry. About 5 seconds later if found the Beta webpage. Took me another few seconds to retry main and it was back. So transient here and I'm using Google public DNS, not the flaky ISP's DNS.Seems you've hit the nail on the head: Start: 2019-04-10 12:32:00 PDT End: n/a Severity: Minor Performance Issue Status: Identified The network team has received reports of DNS lookup issues which may impact off campus users attempting to access campus web sites and other applications. The team is currently investigating. CMR: INC0847343http://systemstatus.berkeley.edu/incident/detail?i=20508 |
|
1884)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91037)
Posted 10 Apr 2019 by Richard Haselgrove Post: Well, your list is more up to date.It's only the dates that were updated, the last time I gave it a thorough test - I didn't need to update the IP addresses. Just added the web server for completeness. |
|
1885)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91032)
Posted 10 Apr 2019 by Richard Haselgrove Post: My current list is # 208.68.240.118 setiboincdata.ssl.berkeley.edu # upload server Aug 2017 # 208.68.240.110 boinc2.ssl.berkeley.edu # download server Aug 2017 # 208.68.240.126 setiboinc.ssl.berkeley.edu # scheduler Aug 2017 # 208.68.240.127 boinc2.ssl.berkeley.edu # download server Aug 2017 # 208.68.240.110 setiathome.berkeley.edu # website Aug 2017You only want to un-comment one of the two download servers at a time. |
|
1886)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91028)
Posted 10 Apr 2019 by Richard Haselgrove Post: Another party today?No - both the website and the operational servers are working fine from here. If that message means that your machine can't work out what IP address to use ("Can't resolve hostname" in the Event log), then it's a local DNS failure at your ISP and can be overcome by activating the hosts file. If the IP address is being resolved (check with <http_debug>), then it's a routing problem between your machine and Berkeley - not a lot you can do about that except contact the ISP and complain. |
|
1887)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 91007)
Posted 9 Apr 2019 by Richard Haselgrove Post: It'll be a long outrage if you achieve the full moonlight howl. There's just the thinnest sliver of a New Moon showing here... |
|
1888)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90997)
Posted 9 Apr 2019 by Richard Haselgrove Post: I see an Aricebo file has been put in the splitter. I hope I get some of those WUs. Its been a while since we've had Aricebo data.02ap19aa - this month. I think they arrive automatically when recording has finished, whether we're ready for them or not. |
|
1889)
Message boards :
Questions and problems :
Where to set cache location
(Message 90919)
Posted 3 Apr 2019 by Richard Haselgrove Post: All BOINC's data is stored in a single data folder (including subfolders). There is no run-time mechanism for changing this folder location, but it can be set at installation time (which may be where you remember setting it previously). If you set the wrong location during your last installation, proceed as follows. Locate your current data folder. By default, it will be at C:\ProgramData\BOINC\, but check that it's there. The current location is shown in the BOINC Event Log at startup. Un-install your current BOINC version. This removes the programs and some settings, but does not remove your existing data. Locate your BOINC data folder, and move the entire thing - folder, files, subfolders - to your preferred new location. Re-install BOINC from your original download file (or fetch a new copy if you've deleted it - doesn't matter). Pause on the third installer screen, which will display the current installation paths. Click on the 'Advanced' button in the lower-right corner. Click on the second 'Change...' button (for Data directory) and pick the new location you moved the data folder to. You can choose a new location for the program files, too, but it's not essential. Check the other settings if needed, and complete the installation as normal. When you restart BOINC, you should pick up exactly where you left off. |
|
1890)
Message boards :
Questions and problems :
Scheduler problem - not receiving more than 1 AMD GPU task at a time
(Message 90911)
Posted 3 Apr 2019 by Richard Haselgrove Post: Welcome back. There is indeed some new code to test - #3076 - but note that this is primarily aimed at finishing off the max_concurrent problem, not the 'recovering from massive FLOPS error' problem. |
|
1891)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90898)
Posted 2 Apr 2019 by Richard Haselgrove Post: Looks like the project is down again. A double outrage since the earlier one was too short.Yup, same here. And it's specifically SETI - not my usual problem with the router rebooting. As this post proves. |
|
1892)
Message boards :
Projects :
There are NO health/biology projects for Raspberri Pi systems
(Message 90864)
Posted 1 Apr 2019 by Richard Haselgrove Post: Here is the Linpack performance of the latest Pi 3+ in comparison to older versionsUnfortunately, I can't find any results for Linpack tests on my NVidia GPUs. BOINC has the theoretical figure as '4087 GFLOPS peak', which we know is ludicrous in benchmark terms, but I'd guess we might be in the low hundreds of GigaFLOPS. Very roughly, that's a thousand times faster than even the best Pi. My 20 hour job would take 20,000 hours (if the Pi could handle the memory needed): that's over 2 years. We'd all be dead before they found a cure. |
|
1893)
Message boards :
Projects :
There are NO health/biology projects for Raspberri Pi systems
(Message 90859)
Posted 1 Apr 2019 by Richard Haselgrove Post: There's also the question of what computational power the projects need for health/biology research. One of my NVidia GTX 970 GPUs - still a reasonably powerful device - is closing in on completion of a GPUGrid health/biology task - after 20 hours of solid computing. I hate to think how long a Raspberri Pi would have taken for that task - certainly longer than the project researchers are willing to wait. |
|
1894)
Message boards :
Questions and problems :
When trying to run boinscr on Linux I get "chdir:No such file or directory"
(Message 90841)
Posted 31 Mar 2019 by Richard Haselgrove Post: Would that command not be 'boincscr'? |
|
1895)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 90820)
Posted 27 Mar 2019 by Richard Haselgrove Post: Big grumble, no glory. Very disappointed to report that MPs call for end to Keighley snooker clubs' men-only rules comes from my patch. As the old saying goes, "You can always tell a Yorkshireman, but you can't tell him much." At least the local MP seems to be on the ball. |
|
1896)
Message boards :
Questions and problems :
Reached a limit on tasks in progress
(Message 90789)
Posted 25 Mar 2019 by Richard Haselgrove Post: What does the Event Log message say, exactly? If it says something like 19-Feb-2019 09:20:02 [SETI@home] This computer has reached a limit on tasks in progressthen it's a limit which has been set on the project's servers - SETI@home, in this case. |
|
1897)
Message boards :
Questions and problems :
Can I run the windows client with a corporate network certificate?
(Message 90783)
Posted 25 Mar 2019 by Richard Haselgrove Post: BOINC Manager doesn't need to contact the internet, and doesn't use certificates. The only thing that BOINC Manager can do is to launch your default browser, and the browser will use its own security settings. The BOINC client does need to contact the internet. The BOINC client (boinc.exe) would have been installed, by default, in 'C:\Program Files\BOINC\'. In that same folder, you will also find a file 'ca-bundle.crt'. I'm told that this contains multiple certificates in plain-text format. You should be able to append (or prepend) your corporate certificates to the bundle. Note that the BOINC-supplied bundle will be replaced with each new client installation, so you will need to repeat the operation after every update. |
|
1898)
Message boards :
BOINC Manager :
BOINCfast.exe
(Message 90779)
Posted 24 Mar 2019 by Richard Haselgrove Post: There's also remote control of BOINC on your mother's computer from yours.Controlling BOINC remotely |
|
1899)
Message boards :
Questions and problems :
Scheduler problem - not receiving more than 1 AMD GPU task at a time
(Message 90766)
Posted 21 Mar 2019 by Richard Haselgrove Post: We held another of the developer conference calls today. David Anderson was not present (this one was at 13:00 UTC, convenient for Europe but very poorly timed for California), but sent in a written report. He said he's ready to start work on a new client release, but knows that work fetch issues have to be addressed first. I was asked to liaise with David on the matters which need attention. I've opened a new issue to consolidate the state of play: #3065. Please add any relevant comments, either here or in the issue. |
|
1900)
Message boards :
Server programs :
Terminating jobs early
(Message 90707)
Posted 16 Mar 2019 by Richard Haselgrove Post: I don't think so, but that one's a bit above my pay-grade: my experience is more as a client-user than as a server-administrator. I would say that cancelling test jobs in a closed test environment has very few downsides, but is something to be avoided in a live environment if at all possible. Sometimes the BOINC system cancels jobs issued to clients if they haven't started running yet - say if a late return from a wingmate completes validation on the workunit. That's not a problem - nothing is wasted. But if you cancel a job which has already started running, then the time spent so far IS wasted, and your volunteer users might start to feel that their efforts aren't being valued, and start to drift away. |
|
1901)
Message boards :
Server programs :
Terminating jobs early
(Message 90704)
Posted 16 Mar 2019 by Richard Haselgrove Post: Try https://boinc.berkeley.edu/trac/wiki/CancelJobs |
|
1902)
Message boards :
Server programs :
Frequency of getting new tasks on local grid
(Message 90703)
Posted 16 Mar 2019 by Richard Haselgrove Post: Yes, in the client control section of the project config file on the server. https://boinc.berkeley.edu/trac/wiki/ProjectOptions#client-control <next_rpc_delay>x</next_rpc_delay> |
|
1903)
Message boards :
Server programs :
Identifying computers OS
(Message 90687)
Posted 13 Mar 2019 by Richard Haselgrove Post: The server receives, parses, stores, and displays information passed to it by the BOINC client running on the remote machine in question. All that information (and the only information that the server has access to) is passed in the file 'sched_request_climateprediction.net.xml'. That file is re-written every time the client is preparing to make contact with the server for whatever reason, but the most recent version is always accessible on the client machine for reference. Here is a sample of the relevant section from one of my genuine Windows machines. <host_info>
<timezone>0</timezone>
<domain_name>redacted</domain_name>
<ip_addr>redacted</ip_addr>
<host_cpid>823018b1231e10f28da302ec7a92c01a</host_cpid>
<p_ncpus>4</p_ncpus>
<p_vendor>GenuineIntel</p_vendor>
<p_model>Intel(R) Core(TM) i5-4690 CPU @ 3.50GHz [Family 6 Model 60 Stepping 3]</p_model>
<p_features>fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx smx tm2 pbe fsgsbase bmi1 smep bmi2</p_features>
<p_fpops>4742537574.411942</p_fpops>
<p_iops>16819291849.400866</p_iops>
<p_membw>333333333.333333</p_membw>
<p_calculated>1549370695.497608</p_calculated>
<p_vm_extensions_disabled>0</p_vm_extensions_disabled>
<m_nbytes>8444608512.000000</m_nbytes>
<m_cache>262144.000000</m_cache>
<m_swap>16887267328.000000</m_swap>
<d_total>1000185266176.000000</d_total>
<d_free>974493057024.000000</d_free>
<os_name>Microsoft Windows 7</os_name>
<os_version>Professional x64 Edition, Service Pack 1, (06.01.7601.00)</os_version>
<n_usable_coprocs>3</n_usable_coprocs>
<wsl_available>0</wsl_available>
<host_info>You could compare that with the equivalent taken from a BOINC client running under WINE, and see if you could find any distinctive and diagnostic difference.
|
|
1904)
Message boards :
Questions and problems :
more "phantom" AMD gpus: got 2x as many rx560 as I really have
(Message 90686)
Posted 13 Mar 2019 by Richard Haselgrove Post: The Einstein project has a supported, official, route (via project preferences) to enable parallel processing of two or more tasks on any GPU. Two tasks per card is often a sweet spot, but it does vary by card. Search for posts on this subject by archae86. Parallel processing is not necessarily advantageous at every project. If it works for you at Einstein, it would be wiser to use the supported route, rather than try to emulate the 'phantom GPU' (multiple drivers) route being discussed in this thread. |
|
1905)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90662)
Posted 12 Mar 2019 by Richard Haselgrove Post: 2 hours 15 minutes, but it's down now. |
|
1906)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90639)
Posted 10 Mar 2019 by Richard Haselgrove Post: You escaped just in time. It's snowing here now - huge fluffy flakes. |
|
1907)
Message boards :
Questions and problems :
Account info for new computer
(Message 90635)
Posted 9 Mar 2019 by Richard Haselgrove Post: It is not possible to change the HostID in that way unless the two hosts are attached to the same user account (email address). I think the enquirer said that the original email account and password had been forgotten. If you still have the previous computer which had been attached using the migration tool, or even just the hard disk from that computer, all is not lost. Finish off the current work on the new machine (set 'No New Tasks') - that won't be transferable either. Then, use BOINC Manager to 'Remove' the computer from the SETI@Home project - leave BOINC installed as it is. Then, find the file account_setiathome.berkeley.edu.xml on the old hard disk - it will be at the top level of the 'BOINC data directory', but somebody else will have to help you find that on a Mac. Use a USB stick or suchlike to copy it to the equivalent place on the new machine, and you should have your old account back next time you start BOINC. |
|
1908)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90629)
Posted 8 Mar 2019 by Richard Haselgrove Post: And now I've been entrusted with his laundry.Laundry delivered to the sleeping Jord - washed, dried and folded. I must have been a Dhobi wallah in a previous life. |
|
1909)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90624)
Posted 8 Mar 2019 by Richard Haselgrove Post: And now I've been entrusted with his laundry. Let's hope my washing machine hasn't joined the same plumbers' union as the shower. |
|
1910)
Message boards :
Documentation :
Translation of Boinc Home Page
(Message 90623)
Posted 8 Mar 2019 by Richard Haselgrove Post: I suspect this may be partly due to the move of the template (.po) files from the main BOINC github repository to the separate BOINC-site repo in #3014 a month ago. The Italian .po file has a copyright date of 2017, and there have been many changes to the home page since then - it is possible that some strings have not yet been made translatable. Edit - it looks as if https://github.com/BOINC/boinc-site/blob/master/BOINC-Web.pot is up-to-date (automatically generated). But I don't know how you would access that in Transifex/ |
|
1911)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90622)
Posted 8 Mar 2019 by Richard Haselgrove Post: I do hope your shower works for visitors as well. Richard says his shower works, but for me it only does cold, very cold and ice cold. This while the faucet on the sink does give hot water. So I suspect sabotage. 🙄😒😤😟🤪😵😠I think it must be something he said. My shower has just welcomed me with a perfectly adequate supply of hot water, the same as it does every morning. Maybe it understands Dutch better than I do? Or perhaps I have inadvertently allowed Brexit-supporting plumbing into my house - that would be more worrying. I don't remember it having this problem with Americans. |
|
1912)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90599)
Posted 6 Mar 2019 by Richard Haselgrove Post: All of that should be available from Listen - Breakthrough Initiatives, but I've never found it there and I haven't got time to look now - Jord is hungry!......25% of Parkes observing time for five years commencing in 2016OCT. |
|
1913)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90595)
Posted 6 Mar 2019 by Richard Haselgrove Post: From Parkes: 2018OCT Observing Schedule Release Notes CASS has recently been approached by NASA with a request to track the Voyager 2 spacecraft with the Parkes Telescope on a regular basis for the next 3-4 months 2019APR Observing Schedule Release Notes At the start of the semester the telescope will be fitted with the 20cm multi-beam receiver (MB20) and the Ultra Wideband Low (UWL) receiver. This will be the receiver fit out until the VLBI run at the end of the semester Source: Australia Telescope National Facility: Schedules & Rosters |
|
1914)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90593)
Posted 6 Mar 2019 by Richard Haselgrove Post: . . So those of us saving resources could .....Donate them to SETI so they can hire more staff? :) |
|
1915)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90577)
Posted 6 Mar 2019 by Richard Haselgrove Post: It's had a relapse. |
|
1916)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90514)
Posted 5 Mar 2019 by Richard Haselgrove Post: Mostly sunny so far, only had a bit of rain down that hill into town. Richard has taken me on a long walk, I suspect he wants to dump me somewhere, but too bad for him, I have navigation on my phone. 😁I tried to sell him off to the lady who runs the cafe by the locks, but no - he followed me home anyway. |
|
1917)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90504)
Posted 5 Mar 2019 by Richard Haselgrove Post: Pleased to confirm that Jord has safely arrived in Yorkshire, to the typical British welcome of sunshine and showers - often both at the same time. |
|
1918)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90489)
Posted 4 Mar 2019 by Richard Haselgrove Post: Well, I've had a quick look round the local treescape, and it looks very much like it did before the storm. The daffodils are (mostly) standing proud, and nodding gently in the breeze. I think we're good to go. |
|
1919)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90477)
Posted 3 Mar 2019 by Richard Haselgrove Post: All of mine are back too, and some machines have even picked up some new work - fortunately, all the machines which had a completely idle GPU. |
|
1920)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90403)
Posted 2 Mar 2019 by Richard Haselgrove Post: I like crunching for GPUGrid.https://www.gpugrid.netNote for anyone tempted to switch from SETI to GPUGrid - things are different there. In particular, time moves much quicker. Their deadline is five days, but they like to have the work back in under 24 hours. And the tasks are much bigger - on my cards, they usually take 12 hours, but at the moment they're slipping in a few which last 20 hours. You do not want a big cache over there! |
|
1921)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 90402)
Posted 2 Mar 2019 by Richard Haselgrove Post: Are those what we used to call BTUs, or British Thermal Units? |
|
1922)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 90394)
Posted 2 Mar 2019 by Richard Haselgrove Post: According to OVO/Boost yes.It's the industry standard - applies to everyone. https://www.gov.uk/guidance/gas-meter-readings-and-bill-calculation |
|
1923)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 90391)
Posted 2 Mar 2019 by Richard Haselgrove Post: Your gas uaage is measured in kWh?No, it's measured in cubic metres (or, if you have a very old meter, cubic feet). But it's billed in kWh. That's what those 60 different supply companies - OK, 50 different companies, at least ten have gone bust - do. They turn volume into power by multiplying by 39.8 and dividing by 3.6: it's not clear how many of them get it right. |
|
1924)
Message boards :
Questions and problems :
Update GPU.
(Message 90374)
Posted 2 Mar 2019 by Richard Haselgrove Post: Depends if you're willing to spend a little time and effort, as well as money. You will find that there's a huge range of modern GPUs. At the top end, they will complete your scientific research much more quickly: at the bottom end, a card of similar speed to your current ones will use significantly less electricity. Before you purchase anything, make sure it will work in your current computers. Manufacturers are discontinuing driver support for older operating systems like Windows XP: your existing cards will go on working, but a new card will probably require a new driver. Check all the small print. |
|
1925)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90357)
Posted 28 Feb 2019 by Richard Haselgrove Post: Well, we have a local stream, which powered the local economy for several centuries. But yes, I do have a washing machine - a washer-drier, even, so you don't have to expose your smalls to the neighbours. And until the modern economy finally collapses, I have an electricity supply to power it. |
|
1926)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90344)
Posted 27 Feb 2019 by Richard Haselgrove Post: And it was about http://seti.berkeley.edu - the generic 'all searches' site, not SETI@Home.When I searched for "does Seti use drupal" I got hits for posts asking for help in moving from drupal 6 to drupal 7.That post was 7 years ago. Hard to say at this point. Shrugs ... |
|
1927)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90338)
Posted 27 Feb 2019 by Richard Haselgrove Post: I got a couple of pages to refresh, including a Server Status page showing everything green at 20:30 UTC. But it had gone down again before I could click 'retry uploads'. Off to the pub for a cold one. Well, probably a luke-warm one in this weather. |
|
1928)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90328)
Posted 27 Feb 2019 by Richard Haselgrove Post: Echo of slow grinding sound as we drag our anchors across the bed of the Atlantic. Maybe heading off to rejoin our old tax havens in the Caribbean. |
|
1929)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90324)
Posted 27 Feb 2019 by Richard Haselgrove Post: Then back I'll probably take the Chunnel. Never been in such a long claustrophobic environment before. :)You can always practice on the London Tube. If you stay for three weeks, you'll be getting perilously close to Brexit! For some reason, Eurostar - that's the foot-passenger service, capital to capital, not your car shuttle - has been reporting a reduced timetable 'for operational reasons', for more than a week now. I wonder if they're preparing emergency doors for Brexit. And if so, which side will the door-knob be on? |
|
1930)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90321)
Posted 27 Feb 2019 by Richard Haselgrove Post: If dry, we walk. If wet, we go by car :-) |
|
1931)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90319)
Posted 27 Feb 2019 by Richard Haselgrove Post: What to believe, the Met Office or the BBC? :)That's the glory of the free market competition system. If you don't like one set of weather, you're free to choose a different one. Mind you, the BBC did choose the cheapest option when it was forced to ditch the state-sponsored service and into the competitive market. Maybe you get what they pay for? |
|
1932)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90317)
Posted 27 Feb 2019 by Richard Haselgrove Post: https://www.bbc.co.uk/weather/bd20 
|
|
1933)
Message boards :
Questions and problems :
Scheduler problem - not receiving more than 1 AMD GPU task at a time
(Message 90266)
Posted 26 Feb 2019 by Richard Haselgrove Post: I thought I was the one being a pain asking about all the problems that I'm having!You'd be surprised who's been emailing who. I was personally asked by Bruce Allen (Director not only of Einstein, but also of the Institute in Hannover that hosts Einstein) to add that agenda item to the project call agenda and speak to it - and by his secretary to write my own minutes afterwards. And I think you've uncovered another one. Work fetch is currently (I think) determined by Client Scheduling October 2010 - Estimated credit: they decided it was too difficult to maintain resource share according to real credit, and so they used REC instead. So I pulled out these two tables. Bill: RAC REC Einstein@Home 18,906.01 142,297,886.589 MilkyWay 4,143.10 216,419.883 SETI@home 6,462.70 605,565,081.292 OpenCL: AMD/ATI GPU 0: AMD Radeon(TM) Vega 8 Graphics (43,980,464 GFLOPS peak) Richard: RAC REC Einstein@Home 8,119.18 8,322.576 GPUGRID 391,764.22 303,164.317 NumberFields@home 3,222.80 1,834.406 SETI@home 10,627.91 367,015.648 OpenCL: NVIDIA GPU 0: GeForce GTX 970 (4,087 GFLOPS peak) OpenCL: NVIDIA GPU 1: GeForce GTX 750 Ti (1,639 GFLOPS peak) OpenCL: Intel GPU 0: Intel(R) HD Graphics 4600 (192 GFLOPS peak)They're not directly comparable, but between them, they teach us a lot. Bill: I know you can't run MilkyWay at the moment because of the bug, so your MilkyWay figures (both RAC and REC) are currently low. I left out SETI Beta for the same reason - you only ran a few test tasks, so the averages didn't have time to stabilise. Richard: this is my daily driver, and the work pattern has been steady for months, if not years. I run: Einstein in the Intel GPU GPUGrid on the GTX 970 SETI on the GTX 750 Ti and (rarely) on the GTX 970 NumberFields on the CPU Comments: Bill's REC has been blown out of the water by the GFlops error. I don't think we'd thought of that before. If that's the only GPU in the system (as yours is), it shouldn't make much difference - but it would unbalance the work fetch calculations between CPU, other GPUs, and the Ryzen GPU component. In my case, the balance between GPUGrid and SETI is most interesting. Discount RAC - GPUGrid give credit away like popcorn - but REC for SETI should be half what it is. REC isn't taking account of the actual device speeds. It's interesting how close RAC and REC are for Einstein and GPUGrid (that says something about the policies of those two projects). I think NF@Home is another slight overpayer, but SETI is, as we all know, way out of line. So, now I've got to write that up in a place and in a way that it will be taken seriously. We know that David is under pressure to review work fetch because of the changes he made when fixing the first part of Keith's problem: he won't want to do that twice, but I don't think he'll want to add this to that work. And no other developer has yet come forward to say that they have the capacity to review this whole area of code. Bummer. |
|
1934)
Message boards :
Questions and problems :
Scheduler problem - not receiving more than 1 AMD GPU task at a time
(Message 90250)
Posted 25 Feb 2019 by Richard Haselgrove Post: You missed #3006 ;-) Yes, I know all about those - I've been making myself a pain in the arse all over BOINC for the last three weeks. See summary at Agenda item 4 of MINUTES: BOINC PROJECTS CALL #4. Bill knows all about it as well - he's been testing both client patches, and Eric's server patch at SETI Beta. We're in a bit of a hiatus waiting for Laurence to update the Server Stable branch. |
|
1935)
Message boards :
Questions and problems :
Scheduler problem - not receiving more than 1 AMD GPU task at a time
(Message 90248)
Posted 25 Feb 2019 by Richard Haselgrove Post: Well, we used to have a safety net: if (wacky_dcf(p)) {
// if project's DCF is too big or small,
// its completion time estimates are useless; just ask for 1 second
//
req_secs = 1;That was back in the day when we used DCF, before CreditNew. I wonder what wacky_dcf() does now.
|
|
1936)
Message boards :
Questions and problems :
Scheduler problem - not receiving more than 1 AMD GPU task at a time
(Message 90247)
Posted 25 Feb 2019 by Richard Haselgrove Post: I'm not sure what speed estimate the Client uses during the work fetch calculation: it might be using 59.63 GFLOPS from the server APR, or it might be using 43,980,464 GFLOPS from its own calculation. Call it 60 GF to be optimistic. I think that gives a maximum SETI runtime of about 40 minutes, or 36 tasks per day. There is no point whatsoever in requesting more than 3 days work, because you'll never get more than 100 tasks from SETI. My personal suggestion would be to try a setting of 2.5 days + 0.05 days: if you get things working cleanly, that should result in BOINC turning in a completed task and requesting a new one roughly once an hour. Now that BOINC is reporting tasks no more than an hour after they complete, there's no point in going for a big 'additional' number. I don't know what would happen if BOINC is paying attention to the absurd driver speed during work fetch. It might - I rather hope it would - say "that's impossible - there must be something wrong", and drop down to a safe level. If I get time, I might try to look through the code. But just to check it out, I'd turn the work fetch right down, just to see what happens. I run 0.25 days + 0.05 days for SETI (except on Tuesdays). That chugs along quite happily if you have a good internet connection and backup projects. |
|
1937)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90245)
Posted 25 Feb 2019 by Richard Haselgrove Post: Cambridge to Chelmsford looks like about an hour and a half by express coach, changing at Stansted Airport. Depends on your personal taste how that sounds. Cambridge to Oxford (for CPDN) will take about 10-20 years, until Chris Grayling reopens the railway line.... |
|
1938)
Message boards :
Questions and problems :
Limit boinc_master bandwidth on Mac OS
(Message 90239)
Posted 24 Feb 2019 by Richard Haselgrove Post: Are you sure that the server is "remote"? BOINC consists of two parts, the Manager and the Client. Both run on your own computer, and communicate by TCP/IP on address 'localhost' (127.0.0.1), port 31416. Check your networking tool to see if this is the traffic it is seeing. |
|
1939)
Message boards :
Projects :
News on Project Outages
(Message 90222)
Posted 22 Feb 2019 by Richard Haselgrove Post: Sawn Kwang has posted in the Technical News area. It appears they had two separate problems: a power outage and a cooling failure. On 2019-02-20, at about 1930 UTC there was a power outage at UWM. The E@H Web site front-end went down when the power shut off, but power has been restored.Then, there's a second post about networking: Re the Server Status page: It looks like the server status page is not working; it says everything is down. This is probably due to the networking at UWM is not fully operational yet after the power outage and data-center migration.I think that's our problem with uploads too. Either the network still hasn't been properly configured for the new IP addresses and routing, or we're still waiting for new DNS settings to propagate. We can't do anything about either. |
|
1940)
Message boards :
Projects :
News on Project Outages
(Message 90215)
Posted 22 Feb 2019 by Richard Haselgrove Post: My BRPS tasks are also GPU, and the uploads are sent in the first instance to einstein4.aei.uni-hannover.de But the ultimate failure, as it has been several times in recent weeks, is HTTP/1.1 504 Gateway Time-out: that's an onwards transmission to another server, with might be either in Hannover or in Milwaukee, Wisconsin. |
|
1941)
Message boards :
Projects :
News on Project Outages
(Message 90212)
Posted 22 Feb 2019 by Richard Haselgrove Post: I run the Binary Radio Pulsar Search application. Tasks were downloading, uploading, and reporting OK yesterday while the website was down. But uploads failed around the time that Jim1348 reported that his uploads were stuck, and they haven't moved since. |
|
1942)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90166)
Posted 17 Feb 2019 by Richard Haselgrove Post: You'll be pleased to hear that my house was built somewhere around 1850, when builders still built from (very tough) natural materials, and were in tune with the local geography. So, it's not on the flood plain at the bottom of the hill. It's not in the teeth of the gale at the top of the hill. It's not facing the prevailing winds from the West. It did have a good clear view of last year's Beast from the East approaching up the valley - but with the dry air from that direction, the snowdrifts were easily managed. So, bring on your worst! |
|
1943)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90162)
Posted 17 Feb 2019 by Richard Haselgrove Post: It's now listed as an untrusted, aggressive, advertiser from Clacton-on-Sea, Essex. |
|
1944)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 90143)
Posted 15 Feb 2019 by Richard Haselgrove Post: Kudos to British Telecom: Hello,Edit - my actual speed is 54.93 Mbps already. I wonder how much they'll overshoot by this time? |
|
1945)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90142)
Posted 15 Feb 2019 by Richard Haselgrove Post: Weather forecasts are tricky things. Sometimes "northern and western parts" means Scotland and Wales, sometimes it means here. Actually, I think 'here' has a micro-climate all of its own, which doesn't appear in any forecast. I'll try to work out which deity is in charge - possibly Rombald the Giant, though I'm not sure if he's been elevated yet. |
|
1946)
Message boards :
Questions and problems :
Slow scheduling
(Message 90140)
Posted 15 Feb 2019 by Richard Haselgrove Post: 1) Asteroids@home, with a WU name identical for GPU and CPU, has been that way for years and my complaint will be likely ignored. I'll isolate it to a single machine with the GPU it needs or a separate BOINC directory or VM.The identical name (by itself) shouldn't be a problem. The CPU and GPU tasks will need different different binary code to run them, and they should be distinguishable by Plan Class (shown in brackets in the 'Application' column in BOINC Manager, Advanced view - probably for the GPU tasks only). Provided the separate plan class for GPU apps is in place, BOINC should track the estimated run times for the two different processes independently. [I thought that Einstein was the only GPU-project still failing to do that, because they deliberately opted out of CreditNew, for understandable reasons] |
|
1947)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90120)
Posted 13 Feb 2019 by Richard Haselgrove Post: I must find my tuit and get my RPi up and running........I've got a round one somewhere. I can bring it to the Dutchman's central meet if you want to borrow it. |
|
1948)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90104)
Posted 13 Feb 2019 by Richard Haselgrove Post: Morning all. Party still in full swing, I see? |
|
1949)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 90070)
Posted 12 Feb 2019 by Richard Haselgrove Post: Is this the first time that a feature request has progressed from idea to implementation within the course of a single SETI maintenance session, during which can't search the message boards? #3017 |
|
1950)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 90055)
Posted 12 Feb 2019 by Richard Haselgrove Post: You might think that the Razorbill's departure might have been scheduled for the convenience of booked passengers. Or does the Razorbill need high tide, while the passengers need low tide? Six hours in the souvenir shop to match the two up? |
|
1951)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 90045)
Posted 12 Feb 2019 by Richard Haselgrove Post: Further down, it explains that T.O.M. was actually Hiram Percy Maxim, who Wiki identifies as He was the son of Sir Hiram Maxim, inventor of the Maxim Machine gun.which was "an instrument of torture and discipline", if ever there was one. |
|
1952)
Message boards :
BOINC client :
Problem building client with VS2017
(Message 90034)
Posted 12 Feb 2019 by Richard Haselgrove Post: That sounds very plausible. OpenCL is supposed to be a universal, platform independent, programming language. I've seen Windows 10 install an NVidia driver from Microsoft, which had CUDA support but not OpenCL. Later, I let the same machine download an Intel GPU driver from Microsoft, which came with OpenCL. Lo and behold, when I put the NVidia card back in the machine, that had OpenCL support too, using the Intel tools. There are BOINC users who advocate using 'clean install' and a driver removal utility at every update: I think that's overkill, but worth having available as a standby in case problems like this crop up. |
|
1953)
Message boards :
Questions and problems :
Rosetta hogging resources
(Message 90018)
Posted 11 Feb 2019 by Richard Haselgrove Post: Isn't that the mechanism designed for 'non compute intensive' tasks? (Like the old Quake Catcher application, which would simply sit in the background doing very little, waiting for the earth to move) |
|
1954)
Message boards :
BOINC client :
Problem building client with VS2017
(Message 90017)
Posted 11 Feb 2019 by Richard Haselgrove Post: ... and everyone else uses VS2013.Including the CI ('continuous integration') automated testing which takes place within minutes after every code change. If you build on VS2013, and you see that the AppVeyor check has passed on the branch you're interested in, you know you can concentrate on the logic and not worry about syntax errors. |
|
1955)
Message boards :
Questions and problems :
Rosetta hogging resources
(Message 90008)
Posted 11 Feb 2019 by Richard Haselgrove Post: There are many projects that force down work and flood the cache, ignoring their resource share.Projects servers can't actually 'force down work' - the firewall and NAT (Network Address Translation) in your router would prevent that. Project servers can only communicate with your computer in the form of a reply to a request made by your computer. I'd be interested to see the request/reply pair for an event like this - the <sched_op_debug> event log flag is useful for this: 11/02/2019 08:45:02 | SETI@home | Sending scheduler request: To report completed tasks. 11/02/2019 08:45:02 | SETI@home | Reporting 5 completed tasks 11/02/2019 08:45:02 | SETI@home | Requesting new tasks for NVIDIA GPU 11/02/2019 08:45:02 | SETI@home | [sched_op] NVIDIA GPU work request: 7683.10 seconds; 0.00 devices 11/02/2019 08:45:06 | SETI@home | Scheduler request completed: got 8 new tasks 11/02/2019 08:45:06 | SETI@home | [sched_op] estimated total NVIDIA GPU task duration: 7907 secondsNote that the amounts requested and received are expressed in seconds - duration requested and runtime estimated. They should match. |
|
1956)
Message boards :
BOINC client :
Problem building client with VS2017
(Message 90004)
Posted 11 Feb 2019 by Richard Haselgrove Post: I've successfully built with VS2013, but I haven't tried VS2017. I think you would have to update the whole vs2013 solution file - Microsoft probably has an import wizard which will break the back of that process, but you may still need to tweak some lines manually. You would also need to assemble a Windows build dependencies tree for VS2017 (that gives you curl, openssl, wxwidgets, and zlib - there may be more). I don't know how much the format changes between versions. |
|
1957)
Message boards :
GPUs :
2nd GPU installed, but not being used
(Message 89986)
Posted 10 Feb 2019 by Richard Haselgrove Post: The simplest way these days - now you've inadvertently given away that you're running Windows - is to make, apply, and then unmake and save a trivial change to one of the Event Log Options (Ctrl+Shift+F, from BOINC Manager). That will create a complete, working, cc_config.xml file with your current settings as defaults. Then, edit that file with a plain text editor like Notepad, and change 0 to 1 in the relevant line. As Jord says, for this particular value a restart of BOINC is necessary: for many of the others 'Read config files' from the Manager is sufficient. |
|
1958)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 89953)
Posted 7 Feb 2019 by Richard Haselgrove Post: That's a good point. All the modern talk is about the customer switching, but it happens the other way too - and nobody talks about that. I once had a bank account, and my custom was 'bought' twice by new owners without my consent. I started with Girobank, but found myself taken over by first Alliance & Leicester, then later by Santander. Nobody asked my permission - I was just a chattel. |
|
1959)
Message boards :
Projects :
Einstein@Home: Can't create account / Captcha refuses to verify
(Message 89949)
Posted 7 Feb 2019 by Richard Haselgrove Post: Another post from Shawn: The issue should be fixed. If it appears again, please inform us. |
|
1960)
Message boards :
Projects :
Einstein@Home: Can't create account / Captcha refuses to verify
(Message 89936)
Posted 6 Feb 2019 by Richard Haselgrove Post: Reply from Shawn Kwang, Einstein Administrator: We're aware of the problem. And I think we have a fix, but it may not have be deployed yet. Hopefully this will be fixed soon. |
|
1961)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89935)
Posted 6 Feb 2019 by Richard Haselgrove Post: I've compared notes with Eric and Jeff in the SETI lab. According to my logs, the upload problems we were suffering yesterday cleared at 02:08 UTC this morning. They saw the same message as "At 6:10pm [PST] on Tuesday the 5th..." - which is the same time, within a couple of minutes. So it seems that Bruno's woes were indeed related to the Campus IST issue, and were solved by replacing the failed router. |
|
1962)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 89926)
Posted 6 Feb 2019 by Richard Haselgrove Post: And Swindon...On the positive side: at least I won't have 5 roundabouts to navigate off the boat, but 7... {facepalm} :-DIf you don't want to navigate too many roundabouts, stay clear of Milton Keynes! |
|
1963)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 89920)
Posted 6 Feb 2019 by Richard Haselgrove Post: LOL. So long as I'm not expected to do the same. |
|
1964)
Message boards :
Projects :
Einstein@Home: Can't create account / Captcha refuses to verify
(Message 89918)
Posted 6 Feb 2019 by Richard Haselgrove Post: Done. I've passed on your comment: you should at least be able to read replies at https://einsteinathome.org/content/bug-report-captcha-problem-new-account If you need to reply to any questions before you get your account open, post them here and I'll pass them on. |
|
1965)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 89916)
Posted 6 Feb 2019 by Richard Haselgrove Post: On the positive side: at least I won't have 5 roundabouts to navigate off the boat, but 7... {facepalm} :-DI don't think it'll be too bad. Once you're past the first two and out of the ferryport, you'll be on the main waterfront drag, and you simply head due West for the next 70 miles, as the road gradually gets bigger and turns into the M62. Or you could take the A1079 past Beverly and York - I don't know that road myself, but it should be OK. |
|
1966)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89908)
Posted 5 Feb 2019 by Richard Haselgrove Post: I think the upload problem is entirely down to Bruno, and there's been no sign of anyone kicking his backside yet. Still slow and patchy upload service here.Not sure if the temporary fix is working well, the UL are still very slow and with a lot of retries.But we can reach the campus - this message board is on a berkeley.edu domain. If we can, why shouldn't they? Edit - that blew it. Now this site |
|
1967)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89902)
Posted 5 Feb 2019 by Richard Haselgrove Post: But we can reach the campus - this message board is on a berkeley.edu domain. If we can, why shouldn't they? |
|
1968)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89896)
Posted 5 Feb 2019 by Richard Haselgrove Post: The outage starts and the DL are now happening, very slowly, but happening (about few KBps only)I've had no problems at all with downloads - maybe that's an internet routing problem? But uploads have been requiring multiple retries - I hope they give Bruno a good scrub behind the ears while they have the chance. |
|
1969)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 89878)
Posted 1 Feb 2019 by Richard Haselgrove Post: In the old days I could have named 10 stores that would have an 8 port gig switch in stock.I'm surprised by that. My local trade counter has 50 models listed still, and has four models in stock to pick up as soon as I could get there. (Though they won't be much use to you on this side of the Atlantic.) What has changed is the price. My first 100 Mbit switch was only four ports, and cost over £100: these 8 port 1 gig ones start at £9.99. But I've never heard of 'Tenda' as a manufacturer. |
|
1970)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 89868)
Posted 30 Jan 2019 by Richard Haselgrove Post: To inject a balancing note: a few days before Christmas, I was trying to use an airline website to book a complicated transatlantic flight (arriving and departing at different cities, plus connecting flights). The return connection wasn't showing up, although the individual flights were showing. I rang the number shown on the website, and explained my problem to the person who answered the phone. He tried the same booking, saw and confirmed the problem. He filed a report to tech support, and told me that the overnight maintenance shift would fix it: I should be able to complete the booking the following morning. I tried again, and the reported problem had indeed been fixed (I hit a second, different, snag, so had to go through the process again - but that was as quick and easy). Number of menu redirects: 0 Total length of phone call: under 7 minutes. This was calling the UK number of a European, but not British, airline. |
|
1971)
Message boards :
Questions and problems :
Problems with remote client control
(Message 89862)
Posted 30 Jan 2019 by Richard Haselgrove Post: I've been told that in my client version the log doesn't go in stdoutdae.txt(that doesn't exist) anymore but in a system log, I can find boinc log messages in /var/log/messages that look like this:OK, useful check - we can cross off some of the obvious possibilities. BOINC does create stderr logs as well as stdout, but I doubt there will be anything of use. You can add extra reporting to the stdoutdae log by creating a cc_config.xml file and setting logging flags: the only one which might be useful here is <gui_rpc_debug>. That would at least indicate whether the remote request is reaching the client. You should also double-check whether a firewall could be implicated, just to be on the safe side. |
|
1972)
Message boards :
Questions and problems :
Problems with remote client control
(Message 89859)
Posted 30 Jan 2019 by Richard Haselgrove Post: Purely for testing purposes while you try to sort this out, are you able to verify independently that the BOINC client is actually running on the headless machine? Probably 90% of the "can't connect to client" problems we investigate turn out to be "client wasn't running, so there's nothing to connect to". If the client is actually running, the next stage would be to try to locate the persistent Event Log file (stdoutdae) on the headless machine, and see if that contains any clues. |
|
1973)
Message boards :
The Lounge :
BOINC Politics thread
(Message 89822)
Posted 29 Jan 2019 by Richard Haselgrove Post: The snow front has reached Yorkshire, and England is turning into Narnia while the White Witch is still on her feet. |
|
1974)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 89821)
Posted 29 Jan 2019 by Richard Haselgrove Post: Yes, driving into Bradford is easy, driving out of Bradford is easy, but driving through Bradford is a pain in the fundament. (Bradford sits in a bowl in the hills: 'into' means downhill, 'out of' means uphill, but picking the right hill is tricky) I'll send you some suggested routes. |
|
1975)
Message boards :
BOINC Manager :
Make project list cpu/os sensitive
(Message 89791)
Posted 25 Jan 2019 by Richard Haselgrove Post: SETI? (SaH v8 only, not Astropulse) |
|
1976)
Message boards :
BOINC Manager :
Make project list cpu/os sensitive
(Message 89789)
Posted 25 Jan 2019 by Richard Haselgrove Post: There's one remaining problem in that list, in that LHC is declared - falsely - to support ATI gpus (because of an unfortunate naming clash). We cleaned up the BOINC Manager 'attach' wizard for the major platforms - Linux, Mac, Windows - recently, so that v7.14.2 displays the same information correctly: it didn't before. But I don't think that work has been completed for Android yet, so maybe Android users aren't getting that project compatibility information on their devices? |
|
1977)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89755)
Posted 22 Jan 2019 by Richard Haselgrove Post: David Anderson wrote:Does that imply they weren't keeping backups before?Jord: |
|
1978)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 89751)
Posted 22 Jan 2019 by Richard Haselgrove Post: I remember Lichfield (just). Handover point between parents and grandparents, between Manchester and Oxford. Age probably about 9 or 10. |
|
1979)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 89748)
Posted 22 Jan 2019 by Richard Haselgrove Post: How middle would Coventry be?I suggested Stratford (not yours) or Warwick, so we're thinking the same ball park. |
|
1980)
Message boards :
The Lounge :
The traveling Dutchman came to Great Britain in March 2019
(Message 89738)
Posted 22 Jan 2019 by Richard Haselgrove Post: Not to mention meet up with each other! |
|
1981)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89728)
Posted 22 Jan 2019 by Richard Haselgrove Post: Main scheduler is already down, but I snuck a post onto a message board. |
|
1982)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89719)
Posted 22 Jan 2019 by Richard Haselgrove Post: We had a sprinkle, and some still visible at the top of the hill. All gone from my half-way ledge. |
|
1983)
Message boards :
Questions and problems :
Fedora 29: no GPU although CUDA is installed
(Message 89696)
Posted 19 Jan 2019 by Richard Haselgrove Post: It depends what tool you use for your 'remote session'. If you use Microsoft's Remote Desktop, it loads a cut-down GPU driver which can't be used by BOINC. If you use {any, if not all} of the third-party remote access tools, they use the installed standard driver, and BOINC can continue to use the GPU. |
|
1984)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 89675)
Posted 17 Jan 2019 by Richard Haselgrove Post: Thanks for that. I'll be avoiding London though, probably a marvelous city but not to drive in.Except in a completely mud-covered Land Rover - preferably 15 years old. You don't care. Even taxi drivers respect that. |
|
1985)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 89673)
Posted 17 Jan 2019 by Richard Haselgrove Post: Grumbles over @ Einstein, uploads have been stuck for well over 6 hrs, no glory there.A repeat of a previously known and understood problem. It might clear in the next few minutes, if Bernd has set the script to run every 12 hours: if he's set it for 24 hours, he'll get an email from me first thing tomorrow morning (UTC). |
|
1986)
Message boards :
The Lounge :
BOINC Politics thread
(Message 89632)
Posted 15 Jan 2019 by Richard Haselgrove Post: And a motion of no confidence in the Government tabled tonight, to be debated and voted on tomorrow. Under normal circumstances, that would lead to a General Election next month. But these are not normal circumstances. |
|
1987)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89622)
Posted 15 Jan 2019 by Richard Haselgrove Post: I'd have a go at Yar-rice, with perhaps a bit of throat clearing before the first 'r', for the second part. But I'll pass on the first. |
|
1988)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89618)
Posted 15 Jan 2019 by Richard Haselgrove Post: It just went downJust as I was trying to answer Jord's question. Jord wrote: I can take FRB being Fast Radio Burst, but anyone got an idea what the GJ is in blc34_2bit_guppi_58405_85972_GJ687_0028.4001.818.21.44.244.vlar_0 ? That's the sort of question one would like to think could be answered on the Breakthrough Listen website, but I haven't found it yet. |
|
1989)
Message boards :
News :
Client version 7.14.2 released
(Message 89589)
Posted 14 Jan 2019 by Richard Haselgrove Post: OK, we can solve that. First, some background. BOINC consists of two parts: The Manager (boincmgr.exe - which you can see and interact with) - this is for 'command and control' only. I doesn't do any work by itself. The Client (boinc.exe) - this part actually ensures that scientific work is done. When you logon to Windows, both parts should be started automatically. You should be able to see one copy of each in the Task Manager list of active processes. BOINC prevents you starting a second copy of either. This probably explains why double-clicking has no effect. Now the tricky bit. When you click on the red 'X' in the top-right hand corner of the Manager, that doesn't actually stop it running. Instead, it minimises to an icon in the 'notification area' at the bottom-right of the screen, next to the clock. And Windows has made it increasingly difficult to find / see / restore these icons. Windows 10 makes it particularly hard. But you can overcome that. https://winaero.com/blog/always-show-tray-icons-windows-10/ shows you multiple ways. You can show icons for all running programs, or if that's too much for your particular setup, you can turn them on and off individually. Find BOINC (boincmgr.exe) in that list, and turn icon display on for it. That makes life much easier. If you want to restore the full Manager window to see how it's getting on, simply double-click on the small icon in the notification area. |
|
1990)
Message boards :
News :
Client version 7.14.2 released
(Message 89586)
Posted 14 Jan 2019 by Richard Haselgrove Post: I'm using Windows 10 Pro.Then you should be getting messages about the availability of v7.14.2, not as you typed. Try going directly to the download page on this website: https://boinc.berkeley.edu/download.php if that url is different from the one your message is directing you to, please tell us what the faulty one is (and exactly where you're seeing it), so that we can investigate further. |
|
1991)
Message boards :
BOINC Manager :
BOINC notifications
(Message 89577)
Posted 13 Jan 2019 by Richard Haselgrove Post: The control that Ageless described prevents the balloon alerts (which some might find intrusive), but it does not prevent the information from sitting silently in BOINC Manager for you to refer to, or ignore, at your discretion. |
|
1992)
Message boards :
News :
Client version 7.14.2 released
(Message 89573)
Posted 13 Jan 2019 by Richard Haselgrove Post: Which operating system are you using? (Android, Linux, Mac OS X, Windows) |
|
1993)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89550)
Posted 11 Jan 2019 by Richard Haselgrove Post: There we went again. At this rate, we're going to have to set up a BOINC lounge at SETI... |
|
1994)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 89545)
Posted 10 Jan 2019 by Richard Haselgrove Post: Reminds me of the idiot boss from XP days who used his desktop (taskbar) clock as a holiday planner. As Boss, he had to have an administrator account (of course), so all the orders he processed for the rest of the day were dated a month or few into the future.Haven't used doze this century or not on my own machines at any rate. Would blocking the M$ domain(s) in the router be a way of stopping automagical updates?It is W7, so it can be shut off, but there is an idiot who insists on surfing the web off the machine, so it does need updates. We eventually bought him a Windows Domain Controller, and locked everybody's clocks to NTP. |
|
1995)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89529)
Posted 8 Jan 2019 by Richard Haselgrove Post: Decent hand-pulled and locally brewed ale did it for me. But I've got a bottle of malt whisky handy in case anybody posts anything as stupid as #2954 again. |
|
1996)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89524)
Posted 8 Jan 2019 by Richard Haselgrove Post: They haven't said which database - of three. They'd hardly bother with an announcement for Astropulse Science - they could just hold the tapes and assimilators for a day or two while we carried on working. I'd always assumed defrag happened automatically on the BOINC database every week. Which leaves SaH science - and I'd be surprised if they could do that in mere hours. Time for me to head out for some party juice, methinks. |
|
1997)
Message boards :
The Lounge :
BOINC Politics thread
(Message 89512)
Posted 8 Jan 2019 by Richard Haselgrove Post: Private providers of public services, part 2. Sexual assault forensics centres failing some victims That's enough for one day - Ed. |
|
1998)
Message boards :
The Lounge :
BOINC Politics thread
(Message 89511)
Posted 8 Jan 2019 by Richard Haselgrove Post: Another triumph for the free market. Energy firm ceases trading after ban Read the story. They were banned for crap customer service, which is odd - customer service (billing, accepting payments, issuing refunds) is all that they do. The actual gas and electricity is delivered direct to people's homes by wholesalers, through an infrastructure created while the industries were nationalised. So, yet another carcass for the lawyers to pick over. Presumably the directors will walk away scot free ("Limited Liability", and all that). According to the broadcast version of this story, there have been nine failures recently - I didn't catch exactly when they started counting. So that's nine sets of programmers, nine website designers, nine call centre teams, nine advertising agencies, nine compliance teams, nine regulators offices, all working themselves out of a job. And for "customer choice"?
|
|
1999)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89509)
Posted 8 Jan 2019 by Richard Haselgrove Post: Ah, thank you. Time to wind the clock. It had gained about one and a quarter seconds when I checked this morning. |
|
2000)
Message boards :
Questions and problems :
Can't get project screensavers to run
(Message 89494)
Posted 6 Jan 2019 by Richard Haselgrove Post: Microsoft are apparently phasing out support for screensavers.And modern TFT flat screens don't need 'saving' in the way that old-fashioned CRT monitors did. |
|
2001)
Message boards :
The Lounge :
BOINC Politics thread
(Message 89484)
Posted 5 Jan 2019 by Richard Haselgrove Post: None.The Guardian has done some digging. Brexit ferry firm: the lawyer, the submariner and the shotgun-toting former veal exporter Not reassuring. |
|
2002)
Message boards :
BOINC client :
RPI and working location.
(Message 89476)
Posted 4 Jan 2019 by Richard Haselgrove Post: That web site makes assumptions about BOINC installation locations which may not be accurate for all installations, or for all versions of BOINC. The default working location for most distro installs was changed from /var/lib/boinc-client to /var/lib/boinc w.e.f. version 7.10 Perhaps somebody could add a comment. #2945 |
|
2003)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89435)
Posted 2 Jan 2019 by Richard Haselgrove Post: That indeed is a fine timepiece RichardWell, you can thank my uncle for my ability to show it to you today. He found it for sale in an antiques shop 30 or 40 years ago, and snapped it up as an impulse buy. He's looked after it since then, and decided to 'pre-bequeath' it to me while he was still mobile enough to come north and supervise its re-assembly. I drove down to London to collect it (just squeezing everything into one car load, no spare seats) and they followed by train. I believe the style of dial - with the emphasis on the single clear minute hand - is called an 'observatory' dial, specifically designed for taking precise measurements: it would be interesting to hear Eric's comments on that. And the mechanism is also very precise, with a temperature-compensated pendulum. It's called a 'Regulator' - the master clock for a large household or institution, from which all the smaller clocks are set. I'll try and get it started on the mark tomorrow morning, and Jord can tell you how much it's gained or lost when he visits in March. |
|
2004)
Message boards :
The Lounge :
BOINC Politics thread
(Message 89428)
Posted 2 Jan 2019 by Richard Haselgrove Post: You really could not make it up!See how many of these chancers you've ever heard of before. https://beta.companieshouse.gov.uk/company/10709921/officers |
|
2005)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89426)
Posted 2 Jan 2019 by Richard Haselgrove Post: Don't hoover any longer and I will in March. :)Yes, you forgot to do the cornice round the lower plinth last time. But at least the hoover's handy. 
|
|
2006)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89423)
Posted 2 Jan 2019 by Richard Haselgrove Post: Fake. Not enough dust. ;-)Well, it was built for dusty houses like mine, so the mechanism is housed in a plain but dustproof wooden hood. You should have seen how much dust I hoovered off that once I get it down to eye level! |
|
2007)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89417)
Posted 2 Jan 2019 by Richard Haselgrove Post: That's a beautiful clock, Richard.It's even more impressive internally, though hard to photograph without moving and dismantling.  A very simple mechanism, but the precision of the machining is breathtaking. |
|
2008)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89415)
Posted 2 Jan 2019 by Richard Haselgrove Post: Grumble, grumble. I've been given this lovely clock:  Proper Victorian mechanical engineering, about 150 years old. Runs perfectly, but needs winding every seven (or at a pinch, eight) days. I usually maintain it on maintenance day. Which yesterday wasn't... And I wound it a bit early last Tuesday, because there were other things happening. So it stopped. And it stopped with the minute hand blocking the winding keyhole. So I had to move it out of the way. And now it doesn't line up at the top of the minute*. And now I'll have to stop it on a good timing mark, and start it again 10 hours and 48 minutes later. Modern kids with their GPS synchronised smart phones just don't know what time means. * The dial is a full 12" (30 cm) across, so you can actually see the minute hand move at each 1-second tick or tock. Alignment matters! |
|
2009)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89412)
Posted 2 Jan 2019 by Richard Haselgrove Post: Hold on tight. Scheduler is down, but the boards are still running - as rough as a boulder field. Summat's up. |
|
2010)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 89319)
Posted 24 Dec 2018 by Richard Haselgrove Post: I thought but without any real evidence that the computer restart would assign the number 1 to the Main monitor - that's the way they've ended up on my machines. If not, we'll probably have to enquire what adaptor(s) is/are driving each monitor. It's possible that Windows is assigning number 1 to whatever monitor is on the first detected GPU. |
|
2011)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 89317)
Posted 24 Dec 2018 by Richard Haselgrove Post: I've used multiple monitors since Windows 98, so here goes. Some of this is theory, because I don't want to mess up my monitors now I've got them the way I like, but I am typing on Windows 7, and I have the monitor control window open on the other screen, so we should be good to go. Right-click on an unoccupied area of open desktop, and choose 'Screen Resolution' (third from bottom) In monitor display box, click 'Detect'. This is usually the first thing to try when things go screwy - if you boot the machine with a monitor disconnected, that monitor may not show, but reconnecting it, then 'Detect', usually restores normality. Select the monitor you wish to make your main display, and tick the 'Make this my main display'. Click the 'Apply' button. The taskbar should move immediately to the screen you chose, but any desktop icons may still be all over the place. Do a machine restart, while keeping your fingers crossed on both hands. See how it comes back up. If the monitors are right, start tidying up the icons. If not, come back and have another grumble. |
|
2012)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89311)
Posted 22 Dec 2018 by Richard Haselgrove Post: That's actually what I'm wearing mostly these days - should be good enough except on the steeper bits of the Cow & Calf. I get mine from Trek & Trail, one of the outlets in Salts Mill - the World Heritage site and Hockney gallery about 5 miles down the valley. |
|
2013)
Message boards :
The Lounge :
These are no longer with us, may they have peace.
(Message 89309)
Posted 22 Dec 2018 by Richard Haselgrove Post: Paddy Ashdown, 77, first leader of the merged Liberal Democrat party in British politics. |
|
2014)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89307)
Posted 22 Dec 2018 by Richard Haselgrove Post: Indeed. Boots, more like - but not wellies. |
|
2015)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89305)
Posted 22 Dec 2018 by Richard Haselgrove Post: Oi - you didn't mention walking across the moor! Yes, we can do that - I'm already half-way up the side of https://www.youtube.com/watch?v=qWXwqEGdWLc - but you'll have to bring your own weather. And your own 'at. |
|
2016)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89292)
Posted 19 Dec 2018 by Richard Haselgrove Post: Now you've gone too far the other way! I drink 'ale' in the formal, British, sense of that word - more specifically, 'bitter', which is served at cellar temperature to enhance the taste.That alehouse feels all wrong to those of us who live in Celsius/Centigrade regions of the world. Ale at blood temperature? No thanks.Not a vampire, okay. So it would be 3 degrees for you. The same pub does sell several 'lager' beers, all chilled to within an inch of their lives. I don't know whether that's because drinkers demand something to take the taste away, or because brewers know that they couldn't sell the stuff if the customers could taste it. I've had a PM overnight going into more detail than you'd ever want to know, but without sources so I can't claim it is authoritative. It does suggest 50–55°F (10–13°C) for my bitter, which sounds about right. |
|
2017)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89281)
Posted 18 Dec 2018 by Richard Haselgrove Post: That alehouse feels all wrong to those of us who live in Celsius/Centigrade regions of the world. Ale at blood temperature? No thanks. |
|
2018)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89269)
Posted 18 Dec 2018 by Richard Haselgrove Post: Judging by the varying start times (sometimes very early, Pacific time), I always assumed it needed a human being to press the 'stop' button, at the very least. But they may then be able to roll over in bed and go back to sleep while the automation takes over... Edit - and less than two minutes later, down she goes. 05:16, Berkeley time. |
|
2019)
Message boards :
Projects :
GPUGrid
(Message 89261)
Posted 17 Dec 2018 by Richard Haselgrove Post: <max_stderr_file_size>0</max_stderr_file_size>Yes, confirmed. It all seems to happen in lib/diagnostics.cpp: I see static double stderr_file_size = 0; static double max_stderr_file_size = 2048*1024; static double stdout_file_size = 0; static double max_stdout_file_size = 2048*1024;but I had real difficulty following how those variables are set or used this morning. We ought to move this conversation out of the projects area if it's going to continue. |
|
2020)
Message boards :
Android :
Connecting to a different BOINC account manager
(Message 89255)
Posted 17 Dec 2018 by Richard Haselgrove Post: At the moment, volunteer developers are having a blitz on bringing the Android code up-to-date. At least five areas of code are waiting for testing and approval before being made available to the public in finished form: two more are still being worked on. All will have to be merged and tested as a complete system, before going out as a public Beta. And sorry, no, we don't maintain an ETA for work as complex as that. |
|
2021)
Message boards :
Projects :
GPUGrid
(Message 89249)
Posted 17 Dec 2018 by Richard Haselgrove Post: In general, if something is written in the docs, it can be trusted. Occasionally, things get out of whack (something gets changed in code without updating the docs, fo example): if something like that comes to our attention, either Jord or I will try to correct it. If nothing is written in the docs, it is not safe to make any assumptions: consistency is not guaranteed, however desirable. I've had a search through the code, and I can't find anywhere where the specific case of max_stdout_file_size=0 is handled. My best guess is that a new file would be created every time BOINC is restarted, and just one single run would be kept as the 'old' file. That's probably not what you wanted. |
|
2022)
Message boards :
Projects :
GPUGrid
(Message 89243)
Posted 16 Dec 2018 by Richard Haselgrove Post: I'll compare your dropbox cc_config.xml with mine more thoroughly after dinner.Can't see any significant differences - not at this time of night, anyway. |
|
2023)
Message boards :
Projects :
GPUGrid
(Message 89233)
Posted 16 Dec 2018 by Richard Haselgrove Post: If you're seeing, specifically, timeouts - you could try a couple of config options: <dont_contact_ref_site>1</dont_contact_ref_site>Cuts out the 'internet access' check (doesn't pester Google). I guess we tend to know whether the internet is up without BOINC telling us... <http_transfer_timeout>60</http_transfer_timeout>I think this controls scheduler requests as well. Default is 300 seconds - I think 60 is plenty on a decent connection (if it ain't happened by then, it ain't going to happen). But with GPUGrid, I was getting a more proactive 'Couldn't connect to server' before the timeout. I'll compare your dropbox cc_config.xml with mine more thoroughly after dinner. |
|
2024)
Message boards :
Projects :
GPUGrid
(Message 89215)
Posted 14 Dec 2018 by Richard Haselgrove Post: Not even in stdoutdae.txt? I try not to over-fill my logs (just a couple of fairly quiet debug options), and I retain 50 MB - so my 'old' file currently takes me back to mid-July. |
|
2025)
Message boards :
Projects :
GPUGrid
(Message 89200)
Posted 14 Dec 2018 by Richard Haselgrove Post: BOINC is not designed to handle multiple overlapping scheduler requests to different projects. Looking at my overnight logs (GPUGrid was indeed down), I see BOINC waiting 22 or 23 seconds for the reply, and trying at most two consecutive times before going into backoff: 13/12/2018 23:00:43 | GPUGRID | Sending scheduler request: Requested by project. 13/12/2018 23:01:05 | GPUGRID | Scheduler request failed: Couldn't connect to server 13/12/2018 23:01:05 | GPUGRID | Sending scheduler request: Requested by project. 13/12/2018 23:01:28 | GPUGRID | [sched_op] Deferring communication for 00:01:43 13/12/2018 23:01:28 | GPUGRID | [sched_op] Reason: Scheduler request failedSo there's a maximum delay of 45 seconds before another project gets a chance: if you can complete 100 SETI tasks in 45 seconds, you're on your own. This is perhaps specific to GPUGrid because it's one of the few projects which requests a scheduler contact every hour (as shown in the log above), so I don't mind discussing it in this thread for the time being. But, on the other hand, my BOINC clients run typically for a month at a time: you mention yours was starting up. Could that, perhaps, be because you had shut it down to run an unsupported external tool like a rescheduler? Was BOINC trying to do something different at startup from the normal scheduled checkin? As always, supply the full context with supporting evidence if you want to lodge a feature request for development. |
|
2026)
Message boards :
Questions and problems :
Detecting situations where, "Scheduler RPC deferred for xx:yy:zz" has been issued.
(Message 89199)
Posted 14 Dec 2018 by Richard Haselgrove Post: That's a pity - I can't submit that code as it stands, then. The duplicate definition warning is a doddle - I can use unique names. I suspect that the Windows build system (VS2013) has warnings suppressed to avoid frightening the horses. 'Not declared in scope' is more annoying, because I lifted the code from a recognised source. It would be interesting to know if the code builds with unique names but leaving the maths alone. The official way to fix it would be to add #include <math.h>in the list of includes at the top of the file. There is an alternative time formatting function where I can print HH:MM:SS by using format %T - but that requires an intermediate buffer to hold the output, and the whole app crashed when I tried it... |
|
2027)
Message boards :
Questions and problems :
Detecting situations where, "Scheduler RPC deferred for xx:yy:zz" has been issued.
(Message 89189)
Posted 12 Dec 2018 by Richard Haselgrove Post: This is why I'm happy to build 7.14.2 and have it ready to go. I'm just thinking out loud here - I have no concerns about sticking a 7.14.2 boinccmd in a 7.2.42 host and seeing what happens. I can easily download a task or two and test on the 1min deferral. If the 7.14.2 boinccmd chokes, I'll just reverse the change. So I'm not looking for you to do any more or provide further answers unless you know for certain there would be problems with running a 7.14.2 boinccmd in the older BOINC versions. I'm quite happy to test that. If there are problems, I'll probably just continue on with the current setup and slowly upgrade all the 7.6.33 machines to 7.14.2, by which time maybe there might be a kernel/amdgpu module/OpenCL libs combination that will finally be usable with SI GPUs and allow me to upgrade the balance of the fleet.There's absolutely no reason not to try. If I were you, I'd make a backup copy of the working boinccmd binary on any machine you want to test, and then replace it with the boinccmd from your your newly-built (and hacked) v7.14.2 There are only three possibilities: It crashes, because of a different library dependency between the build machine and the older Linux on the test machine. It runs, but produces dodgy output (either because of the older BOINC client - unlikely - or because I've written Windows-only code). It works. If (1) - revert to backup! |
|
2028)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89187)
Posted 11 Dec 2018 by Richard Haselgrove Post: Edit 2: Main also now.The boards are up, but none of my hosts have reported yet ('Scheduler request failed: Couldn't connect to server'). By my definition, the outrage continues. Edit - one (small) batch got through, but now it's 'Scheduler request failed: HTTP service unavailable'. Button abuse time... |
|
2029)
Message boards :
Questions and problems :
Detecting situations where, "Scheduler RPC deferred for xx:yy:zz" has been issued.
(Message 89186)
Posted 11 Dec 2018 by Richard Haselgrove Post: Could well be. I'll start backtracking to find where that number might be set - whether it's the executables or the most recent batch of datafiles. But not tonight. |
|
2030)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89175)
Posted 11 Dec 2018 by Richard Haselgrove Post: At Einstein, it's just the export. SHAWN KWANG, Administrator wrote: I want to remind people that in one week, our stats exports will change and honor the option Do you consent to exporting your data to BOINC statistics aggregation Web sites?GDPR is a European Union law, so Einstein - having an operating base in Germany - is being especially cautious. The same code is being made available to all BOINC projects, but projects based outside the EU may take a more relaxed attitude. |
|
2031)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 89167)
Posted 11 Dec 2018 by Richard Haselgrove Post: No, you can't lose your credits. You can lose the display of your credits on websites outside Einstein's direct ownership, unless you give renewed explicit permission for your private data to be shared with external sites. As required by law. |
|
2032)
Message boards :
Questions and problems :
Detecting situations where, "Scheduler RPC deferred for xx:yy:zz" has been issued.
(Message 89162)
Posted 11 Dec 2018 by Richard Haselgrove Post: I think this might be good enough: name: Einstein@HomeFind the equivalent of https://github.com/BOINC/boinc/blob/master/lib/gui_rpc_client_print.cpp#L81 in your working copy of the code - that's the function void PROJECT::print() in the library file /lib/gui_rpc_client_print.cpp. Look for the two lines time_t foo = (time_t)last_rpc_time;
printf(" last RPC: %s\n", ctime(&foo));and replace them with time_t foo = (time_t)last_rpc_time;
printf(" last RPC: %s", ctime(&foo)); //ctime supplies its own /n
if (min_rpc_time > dtime()) {
time_t foo = (time_t)min_rpc_time - (time_t)dtime();
printf(" Scheduler RPC deferred for: %02.0f:%02.0f:%02.0f\n", floor(foo/3600.0), floor(fmod(foo,3600.0)/60.0), fmod(foo,60.0));
}That's a small change to clean up the blank line, and (conditionally) show the deferral in HH:MM:SS, like BOINC Manager. Build a new boinccmd, and enjoy.Thanks to stackoverflow |
|
2033)
Message boards :
Questions and problems :
Detecting situations where, "Scheduler RPC deferred for xx:yy:zz" has been issued.
(Message 89157)
Posted 10 Dec 2018 by Richard Haselgrove Post: Having a second bite at this. The good news is - for Einstein, I'm getting last RPC: Mon Dec 10 11:26:15 2018 RPC deferred until: Mon Dec 10 13:24:26 2018There's been no contact with the scheduler for over an hour because of 10/12/2018 12:46:26 | Einstein@Home | Sending scheduler request: Requested by user. 10/12/2018 12:46:26 | Einstein@Home | Reporting 8 completed tasks 10/12/2018 12:46:26 | Einstein@Home | Requesting new tasks for Intel GPU 10/12/2018 12:46:27 | Einstein@Home | Scheduler request failed: Peer certificate cannot be authenticated with given CA certificatesso the client has invented this backoff by itself - and it shows up. That's one of my questions answered. And I've given notice of "Proposed work on boinccmd output" in #2904 |
|
2034)
Message boards :
Questions and problems :
Detecting situations where, "Scheduler RPC deferred for xx:yy:zz" has been issued.
(Message 89153)
Posted 9 Dec 2018 by Richard Haselgrove Post: FWIW: name: Einstein@Home master URL: http://einstein.phys.uwm.edu/ user_name: Richard Haselgrove team_name: Raccoon Lovers resource share: 100.000000 user_total_credit: 112876275.710468 user_expavg_credit: 275403.015925 host_total_credit: 12692614.418247 host_expavg_credit: 8478.457242 nrpc_failures: 0 master_fetch_failures: 2 master fetch pending: no scheduler RPC pending: no trickle upload pending: no attached via Account Manager: no ended: no suspended via GUI: no don't request more work: no disk usage: 0.000000 last RPC: Sun Dec 09 16:56:10 2018 RPC deferred until: Sun Dec 09 16:57:10 2018 project files downloaded: 0.000000 GUI URL: name: Common questions description: Read the Einstein@Home Frequently Asked Question list URL: https://einsteinathome.org/faq |
|
2035)
Message boards :
Questions and problems :
Detecting situations where, "Scheduler RPC deferred for xx:yy:zz" has been issued.
(Message 89152)
Posted 9 Dec 2018 by Richard Haselgrove Post: As always when doing this sort of work: AAAARRRRGGGGHHHH that bastard C++ language again First, the easy bit: what you see in --get_project_status is written by https://github.com/BOINC/boinc/blob/master/lib/gui_rpc_client_print.cpp#L81 Sure enough, "last RPC:" is printed on line 104, and there's no reference to "RPC deferred until:". There's a perfectly good variable 'min_rpc_time', and we can display that the same way as 'last_rpc_time' - that works nicely. But both are written as absolute times since Unix year dot. What we're interested in is the difference between 'until' and now: how much longer have we got to wait? And when is now? The client uses gstate.now, the Manager uses dtime(), and boinccmd uses ... I haven't the f'ing foggiest idea. Everything I throw at it is off by billions of seconds. But if someone could write me a nice formatted "RPC deferred for: HH:MM:SS" function using those values, it would be a piece of piss to drop it in. They all seem to be declared as doubles, but some of them might contain tm structures. When I tried using a nice simple difftime, it crashed boinccmd. Oh, but I did find a bug. You see a blank line after the 'last RPC' line because ctime supplies its own /n. You can have that one for free. Why is project_files_downloaded (time) always zero? More coffee, Jeeves. |
|
2036)
Message boards :
BOINC client :
Edit Resource Share per Computer?
(Message 89151)
Posted 9 Dec 2018 by Richard Haselgrove Post: Make that four. The third one is 'school', but there's also a case 0 of 'default' or '---'. |
|
2037)
Message boards :
Questions and problems :
Detecting situations where, "Scheduler RPC deferred for xx:yy:zz" has been issued.
(Message 89150)
Posted 9 Dec 2018 by Richard Haselgrove Post: Now there's a proper question to get the juices flowing! Some background to get us started. BOINC has several components. The key one is the 'client' or 'daemon': all the really important stuff happens inside here, but you can't see it - it has no user interface. Instead, there are several alternative stand-alone interface modules which communicate with the client and with the outside world. There's BOINC Manager, of course, and boinccmd - and also independently-written tools like BoincTasks and BoincView. All of these rely on a common communications standard called an RPC (Remote Procedure Call). The principle is that the interface tool says 'Send me all you've got on project status' (or whatever), and gets a great wodge of XML in return. Then, it's up to the interface module to pluck the interesting or important bits out of the wodge and present them to the user in its own particular way. In development terms, this process tends to be driven by BOINC Manager. Somebody wants to display new information, so they add a display in the Manager, and perforce they have to add the underlying data to the RPC. But there's been no pressure to keep every interface tool up-to-date with new features. I suspect that we'll find that the backoff time is in the XML sent from the client to boinccmd, but the code to transfer it from XML to output simply hasn't been updated. boinccmd is pretty simple, so I'll have a go at that this afternoon: since you already have the build tools to make an operational version of boinccmd, you can update your working copy. We will have to double-check when this particular element was added to the client and Manager, in case you need to update the sender as well as the receiver. "Who would be the best person to ask?", indeed. BOINC these days is maintained by a group of volunteer developers, and I operate on the fringes of that - I've even had a few minor patches accepted by the core group. So, adding extra code is a matter of opening an issue or pull request on Github, and seeing whether it sinks or swims. But, I'm not a great fan of that process. It lacks any overarching design stage before we start lobbing code into the pot. There have been too many cases recently where somebody has had a great idea and coded it, only to stir up a howl of protest from people who rely on the old way of doing things - some of my work has been to integrate the best of the old and the new, and the measure of my success has been that absolutely no-one has noticed or commented - it just works the way you want it to, whichever side of the fence you're on. I have a slight worry that this may be one of those cases. Some sysadmins may have an existing investment in scripts which rely on boinccmd outputting in a rigidly familiar format - exactly X lines in a get_project_status report, with my data on line Y, for example. It's not good practice - the XML 'search and report if found' technique is better - but it's quick. And dirty. And common. So, we need to have a meta-discussion about the design, before we start adding extra data in the public source. I can try and feed that into the group. Going back to your original use-case: there are many reasons for scheduler communications to be deferred. Some are dictated by the project - in Einstein's case, delay 60 seconds between any two contacts, and wait an hour if the request was during maintenance. Others are made up by the client, and are for variable times: they tend to be triggered (initially) by communications failures, but also by project task shortages ("don't bother asking for work if there wasn't any last time"). You perhaps need to investigate whether all of these trigger the 'Scheduler RPC deferred for' line in the Manager property page - project-requested backoffs certainly do, but I don't know about the internally-generated ones. That's enough for this cup of coffee - I'll dig further into the code later. |
|
2038)
Message boards :
Questions and problems :
gpu stopped processing when 2nd user signed in
(Message 89145)
Posted 8 Dec 2018 by Richard Haselgrove Post: Found some work on shared user logs under fast user switching - v7.3, March 2014: [boinc_alpha] v7.3.8 - 'all users' installation testing But nothing on GPUs yet. |
|
2039)
Message boards :
Questions and problems :
gpu stopped processing when 2nd user signed in
(Message 89143)
Posted 8 Dec 2018 by Richard Haselgrove Post: No, the drivers are the same - IIRC the problem is that a program started by the first user can't use drivers running under the second user's account. I did some work with Rom Walton on this - that says how long ago it was. I think the best the second user can do is to shutdown BOINC and start it again in the second account. Then you have to repeat the process when the users swap again. That's dredged up from a very old memory - I'll go and have a delve in the archives again. |
|
2040)
Message boards :
BOINC Manager :
Usage limits for "in use" versus not
(Message 89138)
Posted 7 Dec 2018 by Richard Haselgrove Post: Totally agree. #41 |
|
2041)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89071)
Posted 28 Nov 2018 by Richard Haselgrove Post: I'm still picking away slowly at this one. The original problem at Einstein - and one which I've seen over the years at other projects - is not being able to remove a project which is throwing up bad messages in the event log (bad url or duplicate attachment, for example). The simple answer now seems to be definitive, and could be put into an FAQ. Stop using an Account Manager In BOINC Manager, switch to Advanced View (required step). Open Tools menu Select "Stop using [name of account manager]" You'll be given a confirmation dialog explaining the consequences, but nothing frightening will happen: it simply says If you stop using [name of account manager], you'll keep all your current projects, but you'll have to manage projects manually.And it works. Tasks will continue running, you can use NNT to prevent downloading new ones, and you (immediately) gain the power to remove unwanted or badly-configured projects. |
|
2042)
Message boards :
Questions and problems :
7.14.2 does not suspend running tasks when CPU usage limit is lowered
(Message 89038)
Posted 26 Nov 2018 by Richard Haselgrove Post: There's a Robert Petersen at SETI who owns host 8620508. That machine does have 12 CPU cores - but it also has both an NVidia GPU and an Intel GPU. All three devices show tasks in the current task list. If that's you, could you confirm that all three of the current tasks are running on the CPU only, or could there, perhaps, be one running on each device? If two tasks are running on GPUs, try suspending GPU computation, and see if that brings the count down to one - and if it stops the fans running? It's hard to tell from the technical specifications whether those devices are running in a full-size, well-ventilated computer case - or are all crammed into a workstation-class laptop. Cooling the latter is always problematic, but a full case with a properly-specified cooling system should be near silent even at full power. |
|
2043)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89035)
Posted 26 Nov 2018 by Richard Haselgrove Post: It seems that I may have over-worried about spamming project user tables. I appear to be: Science United user b356c73a at Acoustics@home Science United user 0c7077f2 at LHC@home Science United user c205d69d at Amicable Numbers Science United user 353e0849 at PrimeGrid Science United user fd9dddbc at CPDN and Anonymous at Einstein@Home None of those seem to have changed since yesterday. Now to find a clean way out, given that SU keeps overwriting my 'No new tasks' settings. (Apologies to Einstein and CPDN - I had to resort to the abort button for some of theirs) |
|
2044)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89033)
Posted 26 Nov 2018 by Richard Haselgrove Post: The question is - is SU going to change the way it works? :P I think I might re-apply the clothespeg to my nose and give my SU test account another spin around the block - this time watching out for those 'random username' attach/detach events. If I get a different username every couple of hours, that would certainly be worth reporting, and I could also shoehorn it into the next developer call. |
|
2045)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89031)
Posted 26 Nov 2018 by Richard Haselgrove Post: I'm pondering that. I certainly want to feed it in to the core BOINC development team (if such a thing exists), but just at the moment you're aware of the 'office politics' problem - I'm not sure how best to get it into the system. Look, for example, at David Anderson's comment of 22 January 2016 in #1455 - though that was before SU, of course. Suggestions for how to avoid it languishing in the backlog welcome. |
|
2046)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89025)
Posted 25 Nov 2018 by Richard Haselgrove Post: Science United testing complete. Took longer than I expected - not only did it keep attaching new projects, but it succeeded in detaching one project (PrimeGrid), reattaching, and fetching new work - all while preserving the 'No New Work' flag. That's not nice. And if it uses a new randomised account name for each re-attach (hard to tell), that's a lot of spamming of project user tables. Another point: while attached to an Account Manager, you're expected to use the AM to do all project attachments and similar manoeuvres via the AM website. But if you happen to discover BOINC Manager and open it, the website address isn't visible anywhere - just the normal range of project buttons. The only clues that you've joined an AM are the disabled 'Remove' button, and some changes to the Tools menu. I don't think that's enough. I looked at Gridcoin, but they're recommending BAM! as their account manager, so I'm done here. Phew. |
|
2047)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89024)
Posted 25 Nov 2018 by Richard Haselgrove Post: To answer those two, first read (very forensically) the developer Wiki on Account Managers I linked earlier (here it is again). The section Account manager RPCs talks about 'a get_project_config.php file', but that's for your client to attach to the Account Manager. It's nothing to do with verifying the project url. Lower down, in the 'action' section, read projects: A list of currently attached projects. For each project:(my emphasis) Because neither the AM itself, nor the client when told to attach to a project, does a second get_project_config to confirm the url before attaching, it is CRITICAL that this is accurate. Account manager administrators aren't warned about this, or about the precise technical meaning of a MASTER url. All the problems so far arise from outdated information in this field. BAM, to give Willy due credit, got both CPDN and Einstein right. Because BAM supports a wider range of projects than the master list supplied by BOINC, I suspect Willy maintains his own list manually, and is very precise in his actions - a man after my own heart. Others are more slapdash. As to CPDN not sending work - frankly, I'm glad it didn't. I'm using a test BOINC account and data folder for this, and I wouldn't want to be stuck with a 4-month simulation until the Spring! Given all the problems Oxford has simply running a stable set of servers with stable BOINC code, I expect they simply didn't have any work to send at the critical moment. BTW, I'd avoid Grid Republic if I were you. It tries to install an ancient version of BOINC (v6.8.44, dated 22 August 2012), and doesn't give you any choice about where to install it. But it does pick up and try to re-use the folder locations of your regular BOINC installation. I'm an experienced Windows user and developer (I've even run Windows Servers with Active Directory security and user account management), and it took me four or five attempts to get the installer to run past the various error messages. And the guidance link for what to put in the account/password boxes takes you direct to Facebook. So far, Science United seems to have the right project data for both CPDN and Einstein - though I'm slightly surprised to find Primegrid classified as Physics (or possibly Earth Sciences). And it keeps re-attaching the same projects, and over-riding my 'no new tasks' setting. |
|
2048)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89016)
Posted 24 Nov 2018 by Richard Haselgrove Post: That's OK - I'm using my own time, as a follow-up to things I partially learned while acting as release manager for v7.10 And here's a nice little log snippet: 24/11/2018 17:34:00 | | Contacting account manager at http://www.gridrepublic.org/ 24/11/2018 17:34:02 | | Account manager contact succeeded 24/11/2018 17:34:02 | | Attaching to http://climateprediction.net/ 24/11/2018 17:34:04 | http://climateprediction.net/ | sched RPC pending: Project initialization 24/11/2018 17:34:04 | http://climateprediction.net/ | [sched_op] Starting scheduler request 24/11/2018 17:34:04 | http://climateprediction.net/ | [sched_op] Fetching master file 24/11/2018 17:34:09 | http://climateprediction.net/ | [sched_op] Got master file; parsing 24/11/2018 17:34:09 | http://climateprediction.net/ | [sched_op] Found 1 scheduler URLs in master file 24/11/2018 17:34:09 | http://climateprediction.net/ | Master file download succeeded 24/11/2018 17:34:14 | http://climateprediction.net/ | sched RPC pending: Project initialization 24/11/2018 17:34:14 | http://climateprediction.net/ | [sched_op] Starting scheduler request 24/11/2018 17:34:14 | http://climateprediction.net/ | Sending scheduler request: Project initialization. 24/11/2018 17:34:14 | http://climateprediction.net/ | Requesting new tasks for CPU and NVIDIA GPU 24/11/2018 17:34:14 | http://climateprediction.net/ | [sched_op] CPU work request: 1.00 seconds; 0.00 devices 24/11/2018 17:34:14 | http://climateprediction.net/ | [sched_op] NVIDIA GPU work request: 1.00 seconds; 0.00 devices 24/11/2018 17:34:16 | climateprediction.net | Scheduler request completed: got 0 new tasks 24/11/2018 17:34:16 | climateprediction.net | [sched_op] Server version 707 24/11/2018 17:34:16 | climateprediction.net | This project is using an old URL. When convenient, remove the project, then add https://climateprediction.net/ https://www.climateprediction.net/get_project_config.php returns <master_url>https://climateprediction.net/</master_url>but acct_mgr_reply.xml contains <url>http://climateprediction.net/</url> That little difference between http and https is enough to trigger the error message - and in this case, the epithet 'old' is probably correct. Trying a manual attach (not under AM control) produced 24/11/2018 17:49:50 | climateprediction.net | Already attached to a project named climateprediction.net (possibly with wrong URL)The 'manual attach' account (with the correct url) can be removed through BOINC Manager, but the GR account (with the bad url) can't - 'Remove' button is disabled. I think this is the original Einstein state that prompted this thread. Shutting down BOINC and restarting it reduced the project count to 1, but as often as not it deleted the wrong account. And 'Synchronising' with the Account Manager (that's a specific reserved word for AM interactions) re-creates the account with the bad url. For an account created via an AM, client_state.xml contains the line <attached_via_acct_mgr/>(about 40 lines into the <project> specification, in this case). Removing this line - something NOT recommended for inexperienced users - allows the project to be removed using the standard BOINC Manager button. And 'Stop using Grid Republic..." (i.e. doing away with AM operation, and returning to the native BOINC mode of operation we're familiar with) should stop it coming back. ------------------------------------ I've tested BAM! as well, and that sent out the correct Master URL for both Climate Prediction and Einstein. Grid Republic (the website) couldn't contact Einstein today (probably used a bad url :P), so wouldn't allow me to test attachment to Einstein. Now for another break, and then I'll see how Science United and Gridcoin respond to the same AM stress tests. |
|
2049)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89006)
Posted 23 Nov 2018 by Richard Haselgrove Post: OK, since Account Managers aren't well understood by the helpdesk regulars (certainly not by me), I'll park these here for reference. https://boinc.berkeley.edu/trac/wiki/AccountManagement https://boinc.berkeley.edu/wiki/Account_managers |
|
2050)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 89004)
Posted 23 Nov 2018 by Richard Haselgrove Post: OK, I'm slowly going to get back up to speed after a period of distraction. In no particular order: 1) Confirmed that the message says 'old' (for url), when it really means 'different' or 'unexpected'. https://github.com/BOINC/boinc/blob/f14d96d2be2e5a290177f7b5391b0802fbd3756c/client/cs_scheduler.cpp#L617 2) Despite the urgency (red letter alert) of the message, you can ignore it, and things will just burble along as usual. I've got one like that. 3) For people who have got themselves into a mess and want to sort themselves out, the 'correct' way to identify the 'correct' url is to stick 'get_project_config.php' into the address bar of a browser after the url of the project website. Thus, https://einsteinathome.org/get_project_config.php returns (among other guff) <master_url>http://einstein.phys.uwm.edu/</master_url>That one's a keeper. After another sleep, I'll have a go at 4) Can I prove that some AM out there is attaching using a url that doesn't comply with (3) above? 5) If so, can I recreate the 'attached twice to the same project' error? 6) And if I succeed, can I work out a procedure for cleaning up the mess? |
|
2051)
Message boards :
Questions and problems :
How do I stop BOINC screen saver from starting?
(Message 89001)
Posted 23 Nov 2018 by Richard Haselgrove Post: Use your normal operating system screensaver management tools. Choose a different screensaver, or 'none'. If you're using Windows, the BOINC installer gives you the option (on the 'advanced' settings page) to install the BOINC screensaver module, or not, as you wish. |
|
2052)
Message boards :
Questions and problems :
Account Managers and Project URLs
(Message 88991)
Posted 22 Nov 2018 by Richard Haselgrove Post: Christian wrote an even fuller explanation of the problem in #2642. I suspect the user problem is not helped (though without checking it at this time of night) by BOINC using the error message "using an 'old' url" when it simply means 'different'. |
|
2053)
Message boards :
The Lounge :
These are no longer with us, may they have peace.
(Message 88989)
Posted 22 Nov 2018 by Richard Haselgrove Post: Roy Bailey - Folk singer and Professor of Sociology. Time to roll home. |
|
2054)
Message boards :
Questions and problems :
Proper usage of command-line started boinc client with boinc manager
(Message 88984)
Posted 22 Nov 2018 by Richard Haselgrove Post: You can't launch 'few copies' without making a deliberate effort to use the options provided. But if you do want to run multiple copies (as we've discussed several times on these boards), you can both run multiple clients, and control them using multiple managers. It just requires a little bit of reading, a little bit of understanding, and if necessary questions can be asked. (and it still requires fewer command line parameters than the SETI SoG application :P) |
|
2055)
Message boards :
Questions and problems :
Proper usage of command-line started boinc client with boinc manager
(Message 88981)
Posted 22 Nov 2018 by Richard Haselgrove Post: No. Port 31416 is hard-wired as the default. https://github.com/BOINC/boinc/blob/f14d96d2be2e5a290177f7b5391b0802fbd3756c/lib/common_defs.h#L24 |
|
2056)
Message boards :
Questions and problems :
Proper usage of command-line started boinc client with boinc manager
(Message 88933)
Posted 20 Nov 2018 by Richard Haselgrove Post: OK, conference call over. Here are a working pair of command lines - first, for the client, and then for a Manager to manage it. D:\BOINC\boinc.exe --allow_multiple_clients --redirectio --detach_console --gui_rpc_port 31418 --dir D:\BOINCdata2 start D:\BOINC\boincmgr.exe /m /n 127.0.0.1 /g 31418 /p password These are designed to run a secondary BOINC instance (e.g. for testing) on a machine where a primary client is continuing to run. If you will only ever have one client running at a time, you can lose 'allow_multiple_clients', '/m', and the non-standard port numbers. My own BOINC program files are in D:\BOINC\ - that's a user choice, so you'll have to substitute your own. Both my client directories specify the same 8-letter gui_rpc_auth.cfg, as you see here (I don't need BOINC GUI RPCs to pass my perimeter security). The standard manager startup relies on the DATADIR registry key to find gui_rpc_auth.cfg, so if you don't specify /p, it may fail to pick it up or get the wrong one. Since all comms are over TCP/IP networking, each client/manager pair needs to be constrained to use unique ports and matching passwords so the traffic doesn't get muddled. You are recommended to use BOINC v7.14.2 for this work - there was a bug which meant you had to start the client with boinc.exe rather than just boinc, and in fixing that bug another bug was introduced. I think both bugs are fixed in 7.14.2 |
|
2057)
Message boards :
Questions and problems :
Proper usage of command-line started boinc client with boinc manager
(Message 88927)
Posted 20 Nov 2018 by Richard Haselgrove Post: Will post a working reply when the projects conference call has finished - tied to the wrong desk at present. |
|
2058)
Message boards :
BOINC Manager :
Suggestion for "activity" drop down menu
(Message 88907)
Posted 20 Nov 2018 by Richard Haselgrove Post: That's what the right-click 'snooze' menu item on the notification area icon is for. |
|
2059)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88856)
Posted 14 Nov 2018 by Richard Haselgrove Post: It's been a long time since I had that particular flashboack, and I'm sorry to say I primarily associate it with hot, third-world holidays - Yalta (Crimea), Cairo, Morocco are possible candidates. But the memory's getting quite hazy now. |
|
2060)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88853)
Posted 14 Nov 2018 by Richard Haselgrove Post: Ah, Richard, welcome to the Los Angeles Basin smog. 5000 to 10000 foot mountains surround the basin trapping the stuff in, unless we get Santa Ana winds which start wildfires.Yes, been there, smelled that. I was aged 5 at the time, and for many years (decades, even) after leaving, catching a whiff of whatever it was you Americans put in your gasoline back then instantaneously transported me back there. |
|
2061)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88836)
Posted 13 Nov 2018 by Richard Haselgrove Post: No smoke from wildfires round here (except when they clear the heather from the grouse moors), so they're planning to generate their own. I've just been down to the public environmental consultation for the waste incinerator they're planning to build about a mile upwind of me:  I don't think anyone had visualised it in three dimensions until I got there. |
|
2062)
Message boards :
News :
Client version 7.14.2 released
(Message 88822)
Posted 12 Nov 2018 by Richard Haselgrove Post: It's an additional offering, not a replacement. WCG now have their own wrapping paper for delivery, but it's the same client underneath. |
|
2063)
Message boards :
Documentation :
Docker password
(Message 88805)
Posted 12 Nov 2018 by Richard Haselgrove Post: The command is actually "git clone https://..." That works without needing a password, using my normal Tortoise GUI (Windows) interface to Git. |
|
2064)
Message boards :
Questions and problems :
7.14.2 does not suspend running tasks when CPU usage limit is lowered
(Message 88793)
Posted 9 Nov 2018 by Richard Haselgrove Post: "Switch between tasks every N minutes" isn't an instruction, it's a permission. The actual decision will be made on priority (the successor to the old debt mechanism, which divvied up resource share according to time run (only)). The new 'priority' mechanism attempts to make resource share follow the credit that would be awarded, if credit was awarded fairly by all projects according to the cobblestone definition. Which it isn't, so don't even think about it. If the priorities have shifted enough that another project needs to take its turn, the switch will be made when one of the now-lower priority projects has reached its TSI (or a multiple thereof). https://boinc.berkeley.edu/trac/wiki/ClientSchedOctTen |
|
2065)
Message boards :
Questions and problems :
7.14.2 does not suspend running tasks when CPU usage limit is lowered
(Message 88783)
Posted 7 Nov 2018 by Richard Haselgrove Post: Which tasks, from which projects? What does <checkpoint_debug> say? |
|
2066)
Message boards :
Questions and problems :
7.14.2 does not suspend running tasks when CPU usage limit is lowered
(Message 88780)
Posted 7 Nov 2018 by Richard Haselgrove Post: You can take your pick of whatever version you want from https://boinc.berkeley.edu/dl/?C=M;O=D (that link displays newest first) The numbers are {version}, {major revision}, {minor revision} Major revision: odd numbers are development/test versions (expect bugs): even numbers are getting close to release. Minor revision: bugfixing towards final release. The smaller the revision number, the more bugs. Speaking for Windows only: every installer removes the existing program files from the current version, and deploys the new files. When reverting to an earlier version, the installer often reports that the installation failed to complete cleanly. If that happens, simply re-run the installer for the version you're trying to test: it will repair whatever problem it's complaining about and finish the job. In general, any version will work with your existing projects, applications, and tasks. But there are some boundaries where structures were changed, and you should avoid crossing these. The most recent showstopper like that occurred early in the v6.13 testing phase, in preparation for the release of version 7. Don't revert to v6.12 or earlier without anticipating major problems. |
|
2067)
Message boards :
Questions and problems :
Repeatedly running the same job
(Message 88778)
Posted 7 Nov 2018 by Richard Haselgrove Post: Yes. The volunteer developers for the SETI@Home project have a sophisticated bench testing tool which allows chosen tasks to be run repeatedly, using different versions of the SETI scientific application and different configurations. The tool monitors and reports both the run time of each test, and the accuracy of the results generated compared to a reference result. The version I know only runs on Windows: it's basically a batch file, with a few small command-line apps to handle things like the accuracy test. A tool like that would only work for the specific project it was designed for, so you should ask at SETI if you're interested. At the moment the project is going through a fairly mature, settled, phase, so there's not much development work going on: but there is preliminary work active at the moment in preparation for bringing a new data source online, so the applications may need to be tweaked again soon, and that should bring the developers out of hibernation. |
|
2068)
Message boards :
Questions and problems :
7.14.2 does not suspend running tasks when CPU usage limit is lowered
(Message 88748)
Posted 6 Nov 2018 by Richard Haselgrove Post: In his other thread on CPU time, Paul is asking about wrapper apps. Wrapper apps typically don't checkpoint, and BOINC won't suspend a task that hasn't checkpointed - that's been the case for years, and is not new for v7.14.2 Paul, please consider that a possible change in work mix might be causing this when you reply to Richie's questions: I'll reply to the CPU time issue later, but I have a busy day first. |
|
2069)
Message boards :
Projects :
Tn-Grid is not in the project's list
(Message 88743)
Posted 5 Nov 2018 by Richard Haselgrove Post: The web site still says, just now, We would also like to thank the members of the BOINC.Italy community who are helping us in this testing phase.(my emphasis) In general, I don't think projects still in testing are admitted to the project list, so it may be this statement which is holding things up. The other things on the page (sponsorship by National Research Council, published scientific papers) look good. I suggest you advise them to remove or re-phrase the testing statement - "thanks to ... who helped us during the testing phase", perhaps? - and then to write to the boinc_admin email shown on the email list page. It's above the level of what we can do directly here. |
|
2070)
Message boards :
Questions and problems :
comma in user name
(Message 88739)
Posted 4 Nov 2018 by Richard Haselgrove Post: They didn't like that - they reopened #594 (which has been sitting on the stocks for 11 years) instead. |
|
2071)
Message boards :
Questions and problems :
comma in user name
(Message 88738)
Posted 4 Nov 2018 by Richard Haselgrove Post: Thanks for the bug report. Please advise him and the other users that the user name is not needed, and can be deleted, to allow the message to be sent to its destination via the user ID number alone. Reported upstream: #2795 |
|
2072)
Message boards :
BOINC client :
How does BOINC get WU CPU time
(Message 88736)
Posted 4 Nov 2018 by Richard Haselgrove Post: Interesting question. Let's start with the simpler case, where a project writes their own executable file for use in a BOINC environment. What they do is to include API calls to BOINC-supplied library code, and it's the BOINC library (linked in to the final application) which performs the timing functions. The BOINC library also provides the messaging facilities which enable the BOINC client to control the science application, and to read back progress and timing data while it is running and on completion. In the case of wrapper apps, the BOINC library functions are included in the wrapper application, not the final science app. The whole purpose of the wrapper, of course, is to enable a project to use a scientific tool where the source code can't be modified to add the API calls. So I think that BOINC can probably only take exact measurements from the wrapper, and use those as proxies for the science app itself: assuming that it starts when it's told to, stops when it's told to, and runs (? at 100%) in between. You could test that: with a traditional app with the embedded BOINC library, measured CPU time will decrease relative to elapsed time if you over-commit the CPU. If I'm right about measurements only being taken from the wrapper, you won't be able to see any effect on CPU time from over-commitment. |
|
2073)
Message boards :
Questions and problems :
VirtualBox 6 now in beta
(Message 88677)
Posted 30 Oct 2018 by Richard Haselgrove Post: Why not link VirtualBox 6.0 Beta 1 released |
|
2074)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88650)
Posted 27 Oct 2018 by Richard Haselgrove Post: On a Saturday? You know something we don't? ;-) |
|
2075)
Message boards :
Questions and problems :
Displayed time to complete a project.
(Message 88646)
Posted 27 Oct 2018 by Richard Haselgrove Post: I am running BOINCTasks and I have an option what columns to show.That option also exists in the standard Manager shipped with BOINC. Options --> Select columns. |
|
2076)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 88633)
Posted 26 Oct 2018 by Richard Haselgrove Post: First train of the day was on-time arriving in Birmingham, so was the one I wanted to catch, so I missed it (frequently the first one runs a minute or so early, and the second a few minutes late....)Read the small print of the Passengers Charter. Train companies are fined if they wait for passengers making what we quaintly used to call 'connections', and rewarded for leaving passengers on the platform so they can spend money in the retail franchises. It's all about competition now: the train coming in is competing against the train going out, and they're both competing against the stationmaster for your business. #JoinedUpThinking |
|
2077)
Message boards :
The Lounge :
BOINC Politics thread
(Message 88622)
Posted 24 Oct 2018 by Richard Haselgrove Post: As I said in the SETI politics thread yesterday, decisions being made by a single individual without oversight by congress fits my personal definition of a dictatorship.A Florida man accused of groping a woman aboard a Southwest Airlines flight over the weekend later defended his lewd actions to police, saying “the President of the United States says it’s ok to grab women by their private parts,” according to court documents.Don't remember hearing about that one getting past congress but then not everything gets reported over here. |
|
2078)
Message boards :
Questions and problems :
Task error "finish file present too long"
(Message 88571)
Posted 22 Oct 2018 by Richard Haselgrove Post: I think we're talking about two different things. "finish file present too long" is a 10 second limit hard coded into the BOINC client: when the science is all over and done, the science application should write a file to say that it's completed and BOINC can clean up and report it: then the science application should shut itself down and get out of the way so that BOINC can move on to the next job. See https://github.com/BOINC/boinc/blob/f14d96d2be2e5a290177f7b5391b0802fbd3756c/client/app_control.cpp#L127 It's hard to see how changing a batch (data) spec could change how quickly an application shuts down. Dave is more likely to be thinking of EXIT_TIME_LIMIT_EXCEEDED, which is a different beast entirely: that one can certainly be controlled by a project by correcting <rsc_fpops_bound>. |
|
2079)
Message boards :
BOINC Manager :
Any way to set the Default back to Save and Not Cancel
(Message 88567)
Posted 21 Oct 2018 by Richard Haselgrove Post: Our pleasure :-) But it's more subtle than that. Cancel will be the default for people who haven't saved anything before: Save will be the default for people who've already got something saved. |
|
2080)
Message boards :
Questions and problems :
Projects run without my permission
(Message 88565)
Posted 21 Oct 2018 by Richard Haselgrove Post: The sequence is vital. 1) Abort 2) Update 3) Reset (but see below) Update doesn't request work: it reports the full state of the project, including telling the server that you have aborted the tasks. You won't get those same tasks ever again (fingers crossed!). Only if you are short of work (by your current global cache settings) will it request work. In your most recent log, you did not abort the resent tasks LATeah0104S_1028.0_0_0.0_2223312_1 LATeah0104S_1028.0_0_0.0_2224330_0 LATeah0042F_1512.0_735174_0.0_1 and you reset the project before you updated it to tell the server that you (intended to but hadn't) aborted them. Look at your task list on the project server to see how many are currently listed as 'in progress'. If there are only three, then a single Resend / Abort / Update cycle will achieve what you want - don't bother with reset (unless to want to recover the disk space) - at the moment, you're just wasting bandwidth downloading everything again and again. |
|
2081)
Message boards :
Questions and problems :
Broken link/missing files for joining Science United?
(Message 88562)
Posted 21 Oct 2018 by Richard Haselgrove Post: Is there a forum for Science United?https://scienceunited.org/forum_index.php Best of luck with that one. |
|
2082)
Message boards :
Questions and problems :
Projects run without my permission
(Message 88561)
Posted 21 Oct 2018 by Richard Haselgrove Post: If it happens again, I suggest you
Resetting the project by itself doesn't tell the server that you don't want to see the tasks ever again: it's intended to let you make a clean start and then carry on. |
|
2083)
Message boards :
Questions and problems :
Not enought space for LHC
(Message 88553)
Posted 20 Oct 2018 by Richard Haselgrove Post: The only way I know to do it is to run a second instance of BOINC, only attached to LHC, and give that second instance a separate data folder on the HDD via the command line. It's a solution which is documented and has been discussed on these boards periodically, but it isn't trivial to set up. |
|
2084)
Message boards :
Questions and problems :
Projects run without my permission
(Message 88547)
Posted 20 Oct 2018 by Richard Haselgrove Post: Logs are kept through restarts, in the files stdoutdae.txt and stdoutdae.old - both in your BOINC data directory, wherever you chose to locate that. The location will be listed right at the top of the restarted log. |
|
2085)
Message boards :
Questions and problems :
Projects run without my permission
(Message 88544)
Posted 20 Oct 2018 by Richard Haselgrove Post: On the Einstein question: it's been the case for a long time (not new in v7.12) that BOINC will collect a "Resent lost task" from a project even when 'no new tasks' is selected: *new* is interpreted very strictly in that case. I'd be interested to sent what your event log shows for those Einstein tasks, before you categorise it as a bug. |
|
2086)
Message boards :
Questions and problems :
Projects run without my permission
(Message 88543)
Posted 20 Oct 2018 by Richard Haselgrove Post: I rather agree, but there is some logic behind it. The basic program which runs on your computer is still BOINC, and that won't change. Science United is an optional new layer which sits on top of BOINC, and behaves as an Account Manager (they have existed for some time, but this is a new one). The behaviour described by MarekK in the opening post sounds to me like the way Science United is designed to be used: you choose - on the Science United website - an area of science that you are interested in supporting, and the Account Manager function translates that into project names and attaches them for you. One advantage would be that if a new project comes along (and convinces SU that it's fit, proper, and worthwhile), then it can be attached automatically to the existing pool of volunteers in that area of science, without having to do their own recruitment drive. We'll see how that works. |
|
2087)
Message boards :
Questions and problems :
How to solve libcurl3 dependency in Boinc Manager
(Message 88529)
Posted 19 Oct 2018 by Richard Haselgrove Post: Yes, the major distros all have native packages - and for those that haven't, Gianfranco's (LocutusOfBorg) PPA has good reports. As of the conference call yesterday, one of the package maintainers was complaining that they hadn't even been notified that the 7.14 code was ready for testing, let alone release - so you might be stuck back at 7.12 for a few days. That one's not bad, but we've fixed up a few loose ends, so 7.14 should be better. Keith also posted this problem at SETI, and some of the guys there are getting it to work with a curl34 PPA. |
|
2088)
Message boards :
Questions and problems :
How to solve libcurl3 dependency in Boinc Manager
(Message 88524)
Posted 19 Oct 2018 by Richard Haselgrove Post: We need to compare and contrast the build version and installation method between Keith's experience (failed) and Dave's experience (working). Details, please. |
|
2089)
Message boards :
Questions and problems :
How to solve libcurl3 dependency in Boinc Manager
(Message 88521)
Posted 19 Oct 2018 by Richard Haselgrove Post: What do you mean the one from your libcurl3-less distribution?I believe that there have been attempts made to popularise BOINC in 'static-linkage' format - in other words, with all library files included in a huge download, instead of relying on the host operating system. These builds aren't supplied by, or supported by, the core BOINC developers. But Linux is open source, so packages may be available through Linux user groups. |
|
2090)
Message boards :
Questions and problems :
How to solve libcurl3 dependency in Boinc Manager
(Message 88518)
Posted 19 Oct 2018 by Richard Haselgrove Post: Thinking about it, the BOINC client also has a dependency on libcurl3, so it won't be running, and even remote Managers will have no chance of connecting to it. Curl itself reports that the current version is 7.61.1, released on 5 September 2018 - the next version release is due on 31 October 2018. I have no idea, and can find no explanation, how those numbers map onto the libcurl3 and libcurl4 terminology - perhaps a developer can help us. I've written to the development mailing list. |
|
2091)
Message boards :
BOINC Manager :
IS there a BOINC Manager version that doesn't need libcurl3?
(Message 88517)
Posted 19 Oct 2018 by Richard Haselgrove Post: Answered in you other thread https://boinc.berkeley.edu/forum_thread.php?id=12677 Locking this duplicate to keep the conversation in a single location. |
|
2092)
Message boards :
Questions and problems :
How to solve libcurl3 dependency in Boinc Manager
(Message 88516)
Posted 19 Oct 2018 by Richard Haselgrove Post: No. BOINC Manager communicates with the BOINC client over a network connection, so that it can work equally with local and remote clients. And libcurl provides the network underpinning for BOINC. I can ask the developers/maintainers of BOINC for Linux if they are aware of this removal of libcurl3: I have to say that I'm surprised. Your quickest solution will be to use a remote BOINC Manager (either an older Linux, or Windows) to control the client on the affected machine: on Linux, the client is usually started automatically with the operating system, and can run without management until something needs to be changed. |
|
2093)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88478)
Posted 16 Oct 2018 by Richard Haselgrove Post: Soon :-)in comparison to the age of the universe :-) |
|
2094)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88474)
Posted 16 Oct 2018 by Richard Haselgrove Post: Thanks, that looks like an excellent start. Cheers. |
|
2095)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88472)
Posted 16 Oct 2018 by Richard Haselgrove Post: Have I missed anything? Been a busy day, and I could murder a pint of proper British hand-pull. |
|
2096)
Message boards :
The Lounge :
BOINC Politics thread
(Message 88465)
Posted 16 Oct 2018 by Richard Haselgrove Post: I wonder whether the descendants of the original 'plantation' of (protestant) Scots in Ulster - like the DUP - will make a mad dash for 'Remainer' Scotland? |
|
2097)
Message boards :
BOINC Manager :
English display in French version
(Message 88458)
Posted 15 Oct 2018 by Richard Haselgrove Post: Since we call a simple wxMessageBox within the safety wrapper, we could also follow https://docs.wxwidgets.org/3.0/classwx_message_dialog.html and the Public Member Functions like SetOKLabel and related. There seem to be only the most basic of state labels: Yes, No, OK, Cancel. If we put those into the safety wrapper, and our own translations into the existing files, we might be able to fix this with minimal coding and no extra files. |
|
2098)
Message boards :
BOINC Manager :
English display in French version
(Message 88454)
Posted 15 Oct 2018 by Richard Haselgrove Post: I was the last person to work on that page, and the event shown seems to be https://github.com/BOINC/boinc/blob/master/clientgui/ProjectInfoPage.cpp#L851 There's no BOINC code for that procedure call, so I assume we're relying on the default handlers provided by wxWidgets - I don't know how (if ever) we could shoehorn translations into it. |
|
2099)
Message boards :
Questions and problems :
Use ONLY GPU??
(Message 88429)
Posted 12 Oct 2018 by Richard Haselgrove Post: http://boinc.bakerlab.org/rosetta/prefs.php?subset=project is where you want to be. You need to click 'edit preferences' (at foot of list) to deselect CPU - it worked for me. Though I would have expected to have seen similar checkboxes for deselecting the different types of supported GPU, too. Edit - I don't see any sign that Rosetta supports GPU crunching at either https://boinc.bakerlab.org/rosetta/apps.php or https://boinc.berkeley.edu/projects.php |
|
2100)
Message boards :
Questions and problems :
Use ONLY GPU??
(Message 88427)
Posted 12 Oct 2018 by Richard Haselgrove Post: Assuming you are attached to a project or projects which has applications for your type of GPU: Log into your account at the website for each of those projects. At each one, visit your project preferences page: edit your preferences to deselect 'Use CPU' at those projects. That will prevent any new CPU work being requested or allocated: it is regarded as courteous to complete and report any CPU work already issued. Do remember that all GPU tasks require some fractional CPU time in support: the fraction needed varies wildly from project to project, application to application, GPU (type) to GPU (type). |
|
2101)
Message boards :
BOINC Manager :
New behaviour with 7.14
(Message 88421)
Posted 11 Oct 2018 by Richard Haselgrove Post: Just look at the caption on the top command button. As usual, it shows what the response will be when clicked. 'Show active tasks' - currently showing 'ready to start' tasks too, will change to show running tasks only. 'Show all tasks' - currently showing running and pre-empted ('waiting to run') tasks only, will change to show cached and ready to report tasks too. |
|
2102)
Message boards :
BOINC Manager :
New behaviour with 7.14
(Message 88418)
Posted 11 Oct 2018 by Richard Haselgrove Post: Did you by any chance click the 'Show active tasks' button by mistake in a mouse-by? That would have the effect you describe, and it's exactly the same in v7.14 as it always has been. |
|
2103)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 88355)
Posted 8 Oct 2018 by Richard Haselgrove Post: I use a couple of DVI KVM switches by ATEN. The newer 8-port seems better at spoofing the connected monitor, but the older 4-port doesn't (and runs very hot, and eats wall-warts). They do have a 4K display port range - https://www.aten.com/gb/en/products/kvm/desktop-kvm-switches/?f3071%5B%5D=9 - but they won't be cheap. I haven't dared look at the price, and remember you'd need oodles of cables. The biggest problem I have is that I have to set both KVMs so that both screens are connected to any machine I need to reboot. If I reboot without - especially - the older KVM pointing to a live monitor, it'll never register above 1024x768 until I reboot again: the native resolution of that one is 1600x1200. If I reboot without a monitor connected, all the desktop icons shuffle themselves around on whatever screen it can find (possibly even the fake dummy plug in my Intel GPU output). It can take a while to sort everything out if there's a power outage and all the machines reboot at once. Once everything is up and running, I leave the computers pumping out the full signal (no monitor sleep), and switch off the monitor power buttons. |
|
2104)
Message boards :
Questions and problems :
Unrecognized tag in cc_config.xml
(Message 88327)
Posted 6 Oct 2018 by Richard Haselgrove Post: I think the BOINC code is designed only to run the WSL detection tests under 64-bit versions of Windows - that will rule out all but a tiny minority of XP users. It does run under Windows 7/64 (my preferred OS), but it - obviously - always finds that WSL is not installed, and sets the 'don't use' tag, so that projects can set their servers not to send WSL work to hosts incapable of processing it. (it actually uses a slightly different tag for that purpose - <WSL_available> 1|0 at the bottom of <host info> in the sched_request) |
|
2105)
Message boards :
Questions and problems :
Unrecognized tag in cc_config.xml
(Message 88324)
Posted 6 Oct 2018 by Richard Haselgrove Post: It's a small, harmless, bug. Don't bother to try to fix it - it'll just come back again. WSL stands for Windows System for Linux - a sort of reverse WINE. It's a new feature in Windows 10 only. One of the projects was experimenting with it, so one of our volunteer developers added support code so it could be used like VBox by other projects too. Unfortunately, he missed one of the places where it needed handling: that was fixed within a couple of days, but sod's law dictated that v7.12.1 was prepared for release between the bug and the fix, and the fix wasn't carried over (I say with feeling, because I wrote the fix). But you've reminded me to check that the fix works in v7.14.0, and it does. So one way to eliminate the message is to join the v7.14.0 testing programme. |
|
2106)
Message boards :
BOINC client :
meaning of Credit here in BOINC Client
(Message 88314)
Posted 5 Oct 2018 by Richard Haselgrove Post: The speed of the core client makes no difference to credit. The speed of the project science applications does. |
|
2107)
Message boards :
Questions and problems :
Scheduler request failed: Error 408
(Message 88310)
Posted 4 Oct 2018 by Richard Haselgrove Post: Where are you seeing this error number? it isn't in the list of BOINC codes. I'm wondering whether it's an http error: 408 REQUEST TIMEOUT You might need to get Andy, David or Sarah to check the server logs for your IP address. |
|
2108)
Message boards :
BOINC client :
Ask about BOINC Client Release
(Message 88305)
Posted 4 Oct 2018 by Richard Haselgrove Post: Link on front page under 'Learn more'. User manual |
|
2109)
Message boards :
Questions and problems :
Exclusive apps cannot be specified with absolute path
(Message 88269)
Posted 29 Sep 2018 by Richard Haselgrove Post: As far as I know, BOINC runs at 23 and should give way politely.BOINC itself probably isn't your problem. It uses very few computer resurces - 14 minutes of CPU time in the 11 days since this computer was last rebooted. Resource contention is more likely to be caused by the project science apps, which run at up to 100% CPU usage given half a chance. But they should, also, run at very low priority and give way politely, as you say. It might be worth probing their behaviour next time the problem re-surfaces. |
|
2110)
Message boards :
Questions and problems :
Limit Resource Share for a single project on a single computer?
(Message 88257)
Posted 28 Sep 2018 by Richard Haselgrove Post: As jglrogujgv said, you have four independent sets of website parameters at your disposal. Are you using them all already? |
|
2111)
Message boards :
Projects :
News on Project Outages
(Message 88247)
Posted 27 Sep 2018 by Richard Haselgrove Post: Independently of the previous item, Andy Bowery wrote: We will be taking the CPDN project offline tomorrow [Friday 28 September] in order to perform some backend work. The project will go offline at 8:45am (UK time). The project will be restored by the end of the working day. |
|
2112)
Message boards :
Projects :
News on Project Outages
(Message 88241)
Posted 27 Sep 2018 by Richard Haselgrove Post: CPDN again. The security certificate for their BOINC website - which also handles their scheduler, which issues new work and accepts completed work - has failed to auto-renew. There is also a delay on manual renewal. Staff are aware and will keep trying again until the certificate has been renewed - hopefully this afternoon. Until then, expect difficulty reporting completed work and requesting new work. You will also face security warnings if you try to access www.cpdn.org. Edit - the certificate has been successfully renewed and https://www.cpdn.org/cpdnboinc/ is working again. |
|
2113)
Message boards :
Questions and problems :
Broken link/missing files for joining Science United?
(Message 88230)
Posted 26 Sep 2018 by Richard Haselgrove Post: The same thing happens if you open the Science United website and try any of the internal 'join' links. There's also a report that existing users can't access their account options, with the same error message quoted. I think it's their problem, not ours. |
|
2114)
Message boards :
Questions and problems :
Screensaver blank macos Mojave 10.14
(Message 88194)
Posted 25 Sep 2018 by Richard Haselgrove Post: To anticipate the developer's next questions, Which version of BOINC were you running? Which project's screensaver did you expect to see? |
|
2115)
Message boards :
Questions and problems :
Screensaver blank macos Mojave 10.14
(Message 88186)
Posted 25 Sep 2018 by Richard Haselgrove Post: This morning, the lead Mac developer for BOINC circulated this memo to everyone involved in developing BOINC code and BOINC project applications for Mac: Apple has just released OS 10.14 (Mojave) and Xcode 10. Both of these will require significant changes to the BOINC source code and also to BOINC project applications. Until we have made these changes, developers should use only Xcode 9.4.1 or earlier.I'll pass on your specific observation so it can be looked at alongside the other changes needed, but please be aware it may take some time to solve. |
|
2116)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88164)
Posted 25 Sep 2018 by Richard Haselgrove Post: Well, I bet that took all of us by surprise! |
|
2117)
Message boards :
Projects :
News on Project Outages
(Message 88129)
Posted 19 Sep 2018 by Richard Haselgrove Post: David Wallom wrote: The project is back up though we have been experiencing some unreliability which is unexpected so will be investigating over the next week or so. |
|
2118)
Message boards :
Questions and problems :
on last UBUNTU 18.4.1 cannot set max CPU time to be used
(Message 88127)
Posted 19 Sep 2018 by Richard Haselgrove Post: If you want to hand edit any file, edit global_prefs_override.xml, and then the max_ncpus_pct value. Do know it's percentages that are used here.That's what setting preferences via the BOINC Manager interface does: it creates an override file (which is optional and may not be present on your machine), and thus detaches you from the alternative method of setting preferences via a project web site. Check the Event Log after startup to see whether the line 18-Sep-2018 17:42:36 [---] Reading preferences override fileis present. If it is, edit it: if not, edit your global preference set on a project web site. As Dave says, it's usually better to change the first preference ("Use at most nnn% of the CPUs"), rather than fiddling with CPU time. |
|
2119)
Message boards :
News :
Web site changes
(Message 88053)
Posted 13 Sep 2018 by Richard Haselgrove Post: I assume there will be a new or updated version of Boinc manager and client because the current versions help dropdown menus point to non-existent urls. Also the same applies to Boinctasks.The menus and help buttons are working again from BOINC Manager (the target url has been replaced). I'm not in a position to test BoincTasks at the moment - could somebody do the honours, please? |
|
2120)
Message boards :
Questions and problems :
I can not remove a project (Rosetta)
(Message 88033)
Posted 11 Sep 2018 by Richard Haselgrove Post: Yes, I suppose so. It's a bit of a sledgehammer approach, though - removing all logs and statistics too, possibly losing favoured ID numbers. Some people (including me!) don't like doing that, so I'd hold it back as a last resort and try more forensic approaches first. |
|
2121)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88031)
Posted 11 Sep 2018 by Richard Haselgrove Post: Yes, because SETI is back up again. That must be a record. |
|
2122)
Message boards :
Questions and problems :
I can not remove a project (Rosetta)
(Message 88029)
Posted 11 Sep 2018 by Richard Haselgrove Post: Unfortunately, no. Uninstalling BOINC only removes BOINC's programs - it deliberately leaves user data, such as project files, intact. |
|
2123)
Message boards :
Questions and problems :
I can not remove a project (Rosetta)
(Message 88027)
Posted 11 Sep 2018 by Richard Haselgrove Post: Yes. At this point, I'd suggest: Stop BOINC (strictly speaking, this means the BOINC Client, nothing to do with the Manager. A full exit from BOINC Manager should do it.) Delete the file you have (correctly) identified Start BOINC Manager again - which will in turn re-start the BOINC Client. I've never run an account manager myself, so I'm hazy on details: I have heard reports that sometimes the rogue project returns after that process. If it does, come back here and we'll have another look. |
|
2124)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 88023)
Posted 11 Sep 2018 by Richard Haselgrove Post: Evidently not Jeff's Tuesday. |
|
2125)
Message boards :
Questions and problems :
Way to adjust/disable project backoff
(Message 88020)
Posted 11 Sep 2018 by Richard Haselgrove Post: There is also <maintenance_delay>nseconds</maintenance_delay>If you often have short outages (below the default 3600 seconds, or 1 hour), that would allow work to restart quicker. But be aware of the downside: if you have a longer outage, that will allow the clients to make many more failed attempts to contact the server, and the random client backoff mentioned by Jord will be more likely to reach the maximum 4 hours. Bookmark Project configuration for a long list of tuning options. |
|
2126)
Message boards :
Questions and problems :
Help links are broken in Boinc Manager
(Message 88017)
Posted 11 Sep 2018 by Richard Haselgrove Post: On my machines, the links that BOINC Manager creates are exactly the same as always, and the default browser is invoked normally. The problem, which is ongoing, is as stated in the error message: The requested URL /manager_links.php was not found on this server.A necessary file has been moved or deleted on this server. |
|
2127)
Message boards :
Questions and problems :
I can not remove a project (Rosetta)
(Message 88012)
Posted 10 Sep 2018 by Richard Haselgrove Post: It sounds as if you originally added Rosetta via an Account Manager, such as BAM! If that's the case, use the account manager to detach it again. |
|
2128)
Message boards :
Documentation :
Wiki Update Request: Leiden Clasical - Closing down
(Message 88001)
Posted 9 Sep 2018 by Richard Haselgrove Post: https://boinc.berkeley.edu/wiki/Leiden_ClassicalYou are referring to a very old page ("last modified on 17 February 2009, at 14:51") in a part of the documentation that is no longer actively maintained - I don't think I even know how I would have found it without your link. We can try and find the links that would lead to it (which would need to be deleted too), but it may take some time. Leiden Classical no longer appears on the current list of BOINC projects |
|
2129)
Message boards :
GPUs :
Two GPUs in a PC displayed wrongly
(Message 87997)
Posted 9 Sep 2018 by Richard Haselgrove Post: This is a known weakness in the BOINC support for GPUs as general purpose compute devices. Although the BOINC client running on your computer is fully capable of identifying each device separately (look at the first few lines of the Event Log after restarting BOINC), the design decision taken around the year 2008 was to treat all devices as identical when reporting them to the BOINC project servers. By 2014, it was acknowledged - see 'A Brief History of BOINC', slide 62, in the 2014 BOINC Workshop report - that there were "Reflections on software: things we need to change - Coprocessor model". But since then, development funding for BOINC has dried up, and I see nothing to suggest that this deficiency will be addressed in the near future. |
|
2130)
Message boards :
News :
Web site changes
(Message 87989)
Posted 9 Sep 2018 by Richard Haselgrove Post: It's explained, and discussed, on this message board in the thread NSF funds new model for BOINC. There was a fuller explanation, plus links to the actual NSF funding proposal, behind the 'details' link in the opening post, but unfortunately 'tbd.php' (the working name before Science United was chosen) is another of the files which has gone missing from this website.I re-posted the NSF funding proposal links on another message board, but I've discovered this morning that the actual PDF files are also missing, not just the links to them. Posting an alternative link for my own reference, as much as anything else. Collaborative Research: SI2-SSI: Expanding Volunteer Computing |
|
2131)
Message boards :
News :
Web site changes
(Message 87987)
Posted 9 Sep 2018 by Richard Haselgrove Post: you can't login to any of your project accounts, as SU creates anonymous accounts.Ooops. Sorry about the misinformation. So there's no way to report errors to project administrators, or ask for help with a technical problem? Or for the project to tell you what you've achieved? That's a - backwards step, I'd say. And no way to communicate in your native language with the members of your national team? (I did see that teams are disabled at SU) |
|
2132)
Message boards :
News :
Web site changes
(Message 87983)
Posted 8 Sep 2018 by Richard Haselgrove Post: Science United functions as an 'Account Manager' like BAM! or Grid Republic - indeed, you'll see those three names listed together if you follow 'Use account manager...' from the Tools menu in your current BOINC Manager. If you follow the Science United link on the new front page here (not the big green 'Join' button), you'll see on their home page: Science United lets you help scientific research projects by giving them computing power. These projects do research in astronomy, physics, biomedicine, mathematics, and environmental science; you can pick the areas you want to support.So, you would be running the same BOINC software, with access to exactly the same community resources, as you are running now. The only difference would be that you would choose the area(s) of science that you wish to contribute to, rather than selecting projects by name - and I have to say that I find some of the project names already in use meaningless to the point of distraction. |
|
2133)
Message boards :
Questions and problems :
No Transfers?
(Message 87968)
Posted 7 Sep 2018 by Richard Haselgrove Post: I'm afraid you have mis-understood the normal processes which occur when you reach "Computation for task ... finished". Let me try to explain - first in the general case for the majority of science projects, and then separately for the special case of Milkyway@Home. Most projects write the scientific results of their computations into one or more data files. These files can vary in size from a few tens of bytes, up to a hundred megabytes or more, depending on the project. Once BOINC has finished computation, it will immediately start to upload any data files created to the project server. Obviously, a 100MB file will take some time to transfer, and these are the files you will see on the transfers tab in BOINC Manager: they will also be listed in the event log, shortly after the 'completed' and 'starting' messages. Only after the final data file has been uploaded and acknowledged by the project can you consider that your scientific contribution is fully complete. It's at this point that the status displayed by BOINC Manager, which will have passed through 'Running' and 'Uploading', will finally switch to 'Ready to report'. By that time, it's already too late to catch any activity on the transfers tab. 'Reporting' the completed task is a separate operation, which involves updating a number of database entries at the project to keep the administration and record-keeping tidy. It's more efficient and saves on the server workload if multiple reports can be batched together and dealt with at the same time. That's why you sometimes see 'Ready to report' status showing for an extended period, but BOINC normally reports them all once the oldest has been waiting for an hour or more. That's the end of the general procedure applying to all projects, but as I said at the beginning, Milkyway is different. They don't write their scientific data into a separate file, but instead include it in the administrative 'report' stage (which is otherwise more concerned with timing). For that reason, there are no Milkyway file uploads listed in your event log, and nothing appears on your transfer tab (although you would see a download transfer if you were watching either of those places the next time new work is allocated by the project). Everything I've written here applies to all versions of BOINC: there's nothing special about Windows 10. The same would apply if you were running Linux or OS X, or Windows 7 or Windows XP. ---------- Welcome to BOINC. No doubt other questions will come to mind as you find your way around - feel free to come back here and ask them as you come across them. |
|
2134)
Message boards :
News :
Web site changes
(Message 87964)
Posted 7 Sep 2018 by Richard Haselgrove Post: Having participated (by invitation) as a community representative on the BOINC Working Group last year, I can confirm that I have heard nothing (over many months of weekly conference calls) to suggest that anyone has any intention to discontinue the current BOINC Client / BOINC Manager / attach to project mode of working. I did mention our current short-term problems with this website during the developer conference call yesterday: if they haven't been resolved by the time of the next scheduled developer call on 20 September, I will be suggesting much more robustly that we may need to consider recruiting a new webmaster. |
|
2135)
Message boards :
News :
Web site changes
(Message 87961)
Posted 7 Sep 2018 by Richard Haselgrove Post: I wonder if this SU idea has something to do with money. Is the funding for BOINC finished? Perhaps you are trying to gain new funding by just changing the name and a few fundamental differences and call it a new idea that deserves funding.Yes, that is indeed the case - at least in part. It's explained, and discussed, on this message board in the thread NSF funds new model for BOINC. There was a fuller explanation, plus links to the actual NSF funding proposal, behind the 'details' link in the opening post, but unfortunately 'tbd.php' (the working name before Science United was chosen) is another of the files which has gone missing from this website. You probably also need to refer to New governance model [for BOINC]. The links in the opening post are outdated, but Jord summarises the funding position in the second post. Note the dates: central funding for BOINC ceased in 2015, over three years ago. |
|
2136)
Message boards :
News :
Web site changes
(Message 87949)
Posted 6 Sep 2018 by Richard Haselgrove Post: More importantly, will there be a permanent redirect so that existing links to previous solutions or to resource threads like FAQs still function? The historical record will be seriously weakened unless reply and quote links can be followed. |
|
2137)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87934)
Posted 4 Sep 2018 by Richard Haselgrove Post: I like the pub (distraction) idea though. CHEERS!Been there, done that, got back home again. I think I'll just head up to bed - keep some WUs warm for me in the morning. |
|
2138)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87903)
Posted 3 Sep 2018 by Richard Haselgrove Post: The Kittyman posted to say that Eric had agreed to give up part of his holiday to work from home and try to fix today's workflow problems. |
|
2139)
Message boards :
Questions and problems :
BOINC BOINC 7.2.42 can't connect to local host
(Message 87899)
Posted 3 Sep 2018 by Richard Haselgrove Post: Yes, that would explain the symptoms. You cannot connect to local host - the client - if the client isn't running. The next job would be to locate any crash dump file, (boinc itself might have written error messages into stdoutdae.txt or stderrdae.txt), and try to interpret them. |
|
2140)
Message boards :
BOINC client :
Authorization Failure -155
(Message 87877)
Posted 29 Aug 2018 by Richard Haselgrove Post: I suggest you put a password into gui_rpc_auth.cfg (why do you have two of them, btw?). Anything will do - mine has eight letters, 'password'. Then, give that password to boinccmd via --passwd password, along with everything else. |
|
2141)
Message boards :
BOINC Manager :
Creating Custom Boinc client (boing gui manager) Installer for windows
(Message 87874)
Posted 29 Aug 2018 by Richard Haselgrove Post: Installshield packages up all the required components into a boinc.msi file, and it's Microsoft's Windows Installer which performs both the installation and the uninstallation. It leaves traces of its work in the Windows registry. These are taken from a Windows 7 registry after a BOINC v7.12.1/x64 installation. [HKEY_CURRENT_USER\Software\Microsoft\Installer\Products\E86DD6DAB39F51D4AB35BFD20DC7868A\SourceList] [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Installer\UserData\S-1-5-21-1446909043-1926004228-2671199157-1001\Products\E86DD6DAB39F51D4AB35BFD20DC7868A\InstallProperties] [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Uninstall\{AD6DD68E-F93B-4D15-BA53-FB2DD07C68A8}]I suspect that your problem arises from the GUID strings: "InstallSource"="C:\\Windows\\Downloaded Installations\\{EB573524-9B18-4288-8ACA-7D0A716094A7}\\" [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Uninstall\{AD6DD68E-F93B-4D15-BA53-FB2DD07C68A8}] (the binary UninstallString shown above is the same GUID {AD6DD68E-F93B-4D15-BA53-FB2DD07C68A8} shown in the key name) These GUIDs will probably need to be generated afresh with every alteration of the msi file contents: you'll see {AD6DD68E-F93B-4D15-BA53-FB2DD07C68A8} as a modification to win_build/installerv2/BOINCx64.ism in https://github.com/BOINC/boinc/commit/bd12338dbd29083daa5a4b022592ca31ff68cd98. I can only suggest you search for references to uninstall GUIDs in the InstallShield documentation. |
|
2142)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87867)
Posted 28 Aug 2018 by Richard Haselgrove Post: ... the message boards are MIA ATM.I've reported tasks too, but the whole data-driven web site seems to be down - WOW didn't pick up an update, although work was reported well before the top of the hour. |
|
2143)
Message boards :
Server programs :
Limit number of jobs in progress to be 2 at most.
(Message 87859)
Posted 28 Aug 2018 by Richard Haselgrove Post: To be honest, I haven't a clue. I'm just a reader of documentation - I've never even seen a live BOINC server, let alone operated one. I would be surprised if you have to re-create the project. I would expect you might have to stop and then re-start the servers daemons after editing the config file: you might need to cancel any previously created workunits and create new ones. Trial and error, as always. |
|
2144)
Message boards :
Server programs :
Limit number of jobs in progress to be 2 at most.
(Message 87852)
Posted 28 Aug 2018 by Richard Haselgrove Post: You probably want to have a look at Job Limits. One that catches my eye is <max_ncpus>N</max_ncpus>If that isn't adequate, try a config_aux.xml file, described in the following section. |
|
2145)
Message boards :
Questions and problems :
boinccmd: No usable Gpu Found - boingmgr: Gpu found
(Message 87850)
Posted 28 Aug 2018 by Richard Haselgrove Post: Somewhat related. Does logging to std{out,err}dae.txt work when you use --detach_console? It doesn't appear to be working on Windows 10 but I'm curious if it used to work.As noted in the GPU area thread, I have a batch file set up to invoke {use,test} a secondary boinc client instance. Both stdoutdae and stderrdae are correctly updated in the secondary data directory, but I do have "--redirectio --detach_console" (both entries, in that order) on my command line. This is Windows 7. |
|
2146)
Message boards :
Questions and problems :
Boinc is unable to connect to the core client
(Message 87825)
Posted 27 Aug 2018 by Richard Haselgrove Post: Thanks for quick help!By which, I assume that boinc.exe isn't? First quick check: Open BOINC Manager Switch to 'Advanced' view, if you're not using that already From the 'Options' menu, select 'Other options..." See whether 'Run the client?' is checked. It should be. If it wasn't, check it and try again. If still no joy, open Windows Explorer (the file management tool). Navigate to C:\Users\[YourLoginName]\AppData\Roaming\BOINCLook for the file stderrdae.txtLet us know whether it exists, how big it is, and when it was last datestamped. If it does exist, open it, and look for the last (most recent) occurrence of BOINC Windows Runtime Debugger Version xxx Dump Timestamp : xx/xx/xx yy:yy:yy- that should be today. Below that, look for "*** Dump of thread ID...". Copy from there to the end, and post it here. |
|
2147)
Message boards :
Questions and problems :
boinccmd: No usable Gpu Found - boingmgr: Gpu found
(Message 87820)
Posted 27 Aug 2018 by Richard Haselgrove Post: I think this is specifically a Windows issue, because of the comment at https://github.com/BOINC/boinc/blob/master/client/main.cpp#L447: // This bit of silliness is required to properly detach when run from within a commandBut I don't think it's specific to any single version of Windows. The check for usable GPUs happens under the control of yet a third invocation of the program, and under the second bug (the current BOINCs) the location of the program to be re-invoked isn't being passed round correctly. |
|
2148)
Message boards :
Questions and problems :
Boinc is unable to connect to the core client
(Message 87814)
Posted 27 Aug 2018 by Richard Haselgrove Post: We may need to go through multiple steps here. First, we need to be certain that the BOINC Client (the normally-silent working part of BOINC) is actually running: it is normally started automatically by BOINC Manager. Open the Windows Task Manager - the easiest way is to right-click on the task bar across the bottom of your screen - and look at the 'processes' tab. You should see two separate entries: boinc.exe (described as 'BOINC client') boincmgr.exe (described as 'BOINC Manager for Windows') If the Manager is listed, but the client is not, then the client may have crashed, and we'll have to go searching for evidence. But let us know what you see first. |
|
2149)
Message boards :
Questions and problems :
boinccmd: No usable Gpu Found - boingmgr: Gpu found
(Message 87813)
Posted 27 Aug 2018 by Richard Haselgrove Post: There is a known problem in this area, which is being worked on at the moment. If you allow BOINC Manager to start the client (the standard, recommended, way), GPUs should be detected as normal. Any problems in this area, especially with Windows 10, are usually to do with Microsoft updates installing cut-down drivers, with crucial parts - CUDA, OpenCL - missing. Staring the boinc client from the command line should be possible, but is subject to problems requiring workrounds. Older versions of BOINC - up to and including v7.8.3 - had a bug which caused GPU detection to fail if you typed "c:\BOINCPATH\boinc". The workround for this bug is to type "c:\BOINCPATH\boinc.exe", with the filename extension explicitly included. This bug was fixed from v7.10.2, but a replacement bug was introduced during the fix. The problem in the later BOINCs occurs if the command line is invoked from a batch file in some other location than c:\BOINCPATH\. The workround here is to place the commands you want in a batch file, and place that batch file in c:\BOINCPATH\. Then, you can run the batch file from a shortcut in another part of the system, and it should work. This is the first time I've read about anybody trying to start the client directly from the registry, so I can't be certain that "batch file in the program directory" will work - I haven't tested it myself - but it's worth a try. Please let us know how you get on. Further discussions on these related bugs can be found in the GPU area - Has BOINC 7.12.1 broken NVIDIA detection? - and in Github issue #2657. |
|
2150)
Message boards :
Questions and problems :
Unable to post a message in a thread
(Message 87802)
Posted 26 Aug 2018 by Richard Haselgrove Post: There is a limit on the number of links per post here, because multiple links are a characteristic of some forms of spamming. I don't know what it is just now, but toning things down a bit would help. WCG uses different forum code, so the anti-spam measures there would be different, too. |
|
2151)
Message boards :
Questions and problems :
Can't delete Boinc
(Message 87795)
Posted 25 Aug 2018 by Richard Haselgrove Post: I don't think we've ever had a confirmed case of BOINC being distributed in viral form, but it's always a possibility, and you're right to question it. https://csgrid.org/csg/ or "Citizen Science Grid" is a legitimate BOINC project associated with the University of North Dakota, and they do use 'exact' in some of their application names, so it could be genuine. But even so, it should be your free choice to install it, and to remove it. You don't mention which version of Windows you're running, so it's hard to give exact advice, but I'd suggest you try: 1) Look to see if there is any reference to BOINC in the list of installed programs. Look in the Windows Control Panel for 'Programs and Features', and follow the instructions for uninstalling a program. 2) Open the Windows Task Manager (right-click on the task bar running across the bottom of your screen). Select any programs with 'BOINC' in their name, and then click the 'End process' button to the bottom-right of the window. Do the same with any programs with 'Exact' in their name. Then see if the folder can be deleted. 3) Restart the computer in 'Safe Mode' (press F8 repeatedly as soon as the boot process starts). Then see if the folder can be deleted. Those three suggestions are in increasing order of drasticness. The best would be suggestion (1), if you can, because that should remove any lingering attempts to start the program again, as well as removing the programs themselves. Come back and ask again if you need more help. |
|
2152)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 87790)
Posted 25 Aug 2018 by Richard Haselgrove Post: You can go even lower than that - £4.32 for one designed for travelling with a short lead that won't get tangled up. But buy from a local shop - silly to pay half as much again for postage. Trust HU-4440P 4 Port USB 2.0 Mini Hub |
|
2153)
Message boards :
GPUs :
Has BOINC 7.12.1 broken NVIDIA detection?
(Message 87780)
Posted 24 Aug 2018 by Richard Haselgrove Post: OK, after other commitments this morning, I've been able to devise a definitive test. Starting the second client with the batch file on my desktop (i.e. in a different location from the BOINC program files), I get these displays with process explorer. With BOINC v7.8.3:  That was the one that wouldn't detect GPUs unless you started it with boinc.exe on the command line, instead of just boinc. Otherwise, it worked. With BOINC v7.12.1:  Notice that the command line starts with just "boinc.exe". That's hard-wired now, but in the process we've lost the previous "D:\BOINC\..." - that's what's causing the problems in this thread. Instead of a getting a fix, we just shifted the deckchairs around a bit. That's enough for me to submit a technical report - I'll edit in the issue number when I've written it. Submitted: #2657 |
|
2154)
Message boards :
GPUs :
Has BOINC 7.12.1 broken NVIDIA detection?
(Message 87776)
Posted 23 Aug 2018 by Richard Haselgrove Post: In that case, it could be a side effect of the same problem that mmonnin and I were just discussing. If BoincTasks doesn't set the working directory explicitly, it could trigger the same problem as our second instance batch file launchers. That makes it more important to fix the client, because BoincTasks is a popular tool. |
|
2155)
Message boards :
GPUs :
Has BOINC 7.12.1 broken NVIDIA detection?
(Message 87772)
Posted 23 Aug 2018 by Richard Haselgrove Post: Indeed. Ironically, it was my own report (but picked up from an email on boinc_alpha) that led to the last change - #2029. Still poking at it. And I think I've found a workround. As I think I've posted before, I start my second client instance with a batch file: D:\BOINC\boinc.exe --allow_multiple_clients --redirectio --detach_console --gui_rpc_port 31418 --dir D:\BOINCdata2That was placed on my desktop. Note that my BOINC program folder is on hard drive D:, whereas my desktop is on SSD C: Putting the batch file directly into D:\BOINC\, and starting it from a shortcut on the desktop, restores GPU detection here - could you test, please? It'll be a working directory problem. Now, how to code that in C++? |
|
2156)
Message boards :
GPUs :
Has BOINC 7.12.1 broken NVIDIA detection?
(Message 87770)
Posted 23 Aug 2018 by Richard Haselgrove Post: OK, thanks. My test machine is also Win7, so it'll be a fair comparison of NV / AMD. Edit - and the same problem occurs under v7.12.1 - why? I'm seeing 23-Aug-2018 18:14:07: CreateProcess failed: 'The system cannot find the file specified. (0x2)'in the stderrdae.txt file in the second data directory. There was some work in this area prior to v7.10.2 - there was a complaint about problems starting BOINC at the command line without specifying the name as boinc.exe But I have boinc.exe in my batch file - that's not (exactly) it. But it gives us a place to look. |
|
2157)
Message boards :
Questions and problems :
Users of grcpool beware!
(Message 87768)
Posted 23 Aug 2018 by Richard Haselgrove Post: I didn't know about the auto message part but I know Amicable #s does ban bad hosts pretty quickly. Some others have systems to reduces their quota eventually getting down to -1.BOINC has an automatic mechanism which should bring the quota down to (+) 1 - though there is some debate as to how effective it is. A value of -1 is considered a blacklisting of the host: it is normally only applied manually, as at CPDN. |
|
2158)
Message boards :
GPUs :
Has BOINC 7.12.1 broken NVIDIA detection?
(Message 87767)
Posted 23 Aug 2018 by Richard Haselgrove Post: If this is a 2nd client then GPU detection broke on 2nd clients in 7.10.Since I was responsible for releasing v7.10, I'd be interested in learning more about that. I don't think I've seen it reported before. (I do have a machine with a second client installed for testing, and with dual NVidia cards, so I'll take a look myself. What should I be looking for?) |
|
2159)
Message boards :
GPUs :
Has BOINC 7.12.1 broken NVIDIA detection?
(Message 87762)
Posted 22 Aug 2018 by Richard Haselgrove Post: BOINC itself is still detecting NVidia just fine, but as Bernie says it's most likely the incomplete drivers from a Windows 10 update that's causing the problem. |
|
2160)
Message boards :
BOINC Manager :
Any way to set the Default back to Save and Not Cancel
(Message 87752)
Posted 21 Aug 2018 by Richard Haselgrove Post: There was a change made in the Manager released alongside client version 7.10.2: that change is continued into the Manager released with client 7.12.1 Anything older than that (like 7.8.3 and before) will work the old way. There will only be inconsistencies if you update different machines at different times. |
|
2161)
Message boards :
Questions and problems :
Is the "Daily schedules -> Network" feature in Computing preferences broken?
(Message 87730)
Posted 20 Aug 2018 by Richard Haselgrove Post: Did you also select 'Network activity based on preferences' on the Activity menu in BOINC Manager? If you had set 'Network activity always', that would cancel out your time-of-day preferences. |
|
2162)
Message boards :
BOINC Manager :
Any way to set the Default back to Save and Not Cancel
(Message 87679)
Posted 16 Aug 2018 by Richard Haselgrove Post: Yes. Merge pull request #2585 into the source code in time for the next client release. |
|
2163)
Message boards :
Questions and problems :
Find my BOINC CPID
(Message 87658)
Posted 15 Aug 2018 by Richard Haselgrove Post: A surprisingly interesting question. There are two values which might be called into use as 'CPID'. One describes an individual computer (or 'host', as we tend to refer to it in BOINC circles), and the other describes you, the computer's owner. You can find both of them in the file called 'client_state.xml', which you can find in your BOINC data directory. On this machine, I'm finding: <host_cpid> 823018b1231e10f28da302ec7a92c01a (right at the top - line 5 - under <host_info> <cross_project_id> d7965dba1521bf2508ec781b0b43bd23 <external_cpid> 68aa4b6077c3fe48975e530f6ad94ca5The second two occur at intervals throughout the file, within each new <project> section. Both <cross_project_id> and <external_cpid> are identical throughout the file, in whichever <project> section they appear - they are truly cross-project. But in several projects, <external_cpid> is blank. Which is odd, because that's the one which is used to identify me uniquely at the sites which collect statistics from all BOINC projects (such as BOINCstats) I have no idea why there are three different values, instead of two. And I also have no idea which of the two (or three) values GridCoin wants you to use - they will have to explain that for themselves. In BOINC circles, and thus for this message board, <external_cpid> is the most important one. That depends - critically - on the user using the same e-mail address when making accounts at each separate project. It also depends on a new computer being attached to a project where the user already has an existing active computer. If those two conditions are satisfied, the new computer will quickly learn the true <external_cpid> for the user's account. |
|
2164)
Message boards :
Server programs :
Aborting Task elapsed time exceede
(Message 87645)
Posted 14 Aug 2018 by Richard Haselgrove Post: All task deadlines are set, held, and tested as absolute Unix time values in UTC. Your Kararchi clients should also be set to UTC, but with a time zone correction so that most times - including those used by BOINC - are displayed in a format that corresponds to local clocks. None of that should affect the decisions about whether a task has, or has not, exceeded the allowed time. So, you've seen delay_bound of 604800 - that's 7*24*3600, so a week. No problem there. But <rsc_fpops_bound>12e12</rsc_fpops_bound>
<rsc_fpops_est>14e14</rsc_fpops_est>is bonkers. You're saying that you will kill jobs ('bound') two orders of magnitude before you expect them to finish ('est').rsc_fpops_est is really a crucial value. You really need to run some test jobs on a machine with known speed (typically ~1 Gigaflop for tasks run on CPUs, much higher for GPUs). Then set rsc_fpops_est to match. And rsc_fpops_bound - the failsafe that kills tasks if they get into an endless loop - should always be set higher than you expect the tasks to run - usually by a factor of 10, perhaps 100 if you're finding runtime hard to estimate. |
|
2165)
Message boards :
Questions and problems :
How to finish a work unit without pausing?
(Message 87643)
Posted 14 Aug 2018 by Richard Haselgrove Post: Most BOINC tasks, from most projects, 'checkpoint' periodically. The default is every 60 seconds, and the restart from a checkpoint is quick and easy - so in the vast majority of cases, nobody is concerned if BOINC switches to another project once an hour. If Rosetta is an outlier in this respect, then it needs special treatment. Delaying task switch is a good one, up to and beyond the normal run time of the task type/project in question. Another good one is to check that 'leave tasks in memory when suspended' is active, through web preferences or BOINC Manager (depending on the preferred management style in operation). But it's possible that not all projects benefit from that setting. |
|
2166)
Message boards :
Server programs :
Aborting Task elapsed time exceede
(Message 87622)
Posted 14 Aug 2018 by Richard Haselgrove Post: But what does <no_delay_bound> will do ?It's not a documented value, so probably nothing except log an error message. You need a limit, and the default value of a week should be enough to get started. |
|
2167)
Message boards :
Server programs :
Aborting Task elapsed time exceede
(Message 87620)
Posted 13 Aug 2018 by Richard Haselgrove Post: Client and server clocks not properly synchronised? Look on the client machine(s) what is actually being received and stored in client_state.xml for <flops> and <rsc_fpops_bound>, and work out for yourself whether it makes sense. |
|
2168)
Message boards :
BOINC client :
Authorization Failure -155
(Message 87619)
Posted 13 Aug 2018 by Richard Haselgrove Post: As already discussed in this thread, this is NOT an authorisation failure with the account manager. Instead, it is a failure between boinccmd and boinc (the BOINC client), both running locally on your own machine. Find the gui_rpg_auth.cfg password for your local BOINC client, and pass it with --passwd on the command line for boinccmd. |
|
2169)
Message boards :
Questions and problems :
Can't detect VB notice.
(Message 87613)
Posted 13 Aug 2018 by Richard Haselgrove Post: It might be worth peeking to see if BOINC is listed in https://appdb.winehq.org/, with or without any warnings. I built a debug version of the Manager recently, with symbols to make that backtrace more useful. I'll dig out the reference. message 87207 |
|
2170)
Message boards :
Questions and problems :
Can't detect VB notice.
(Message 87609)
Posted 13 Aug 2018 by Richard Haselgrove Post: Won't that be solved by uninstalling the 64bit VBox and installing the 32bit version?I'm not sure that mixed-mode VBox installation is possible. Windows hosts x86/AMD64 on https://www.virtualbox.org/wiki/Download_Old_Builds_5_2 leads to a single download file, not a choice. I might see what I can dig out of it. Edit - close, but no cigar. Running the VBox installer with --extract gets you two separate msi files for 32 and 64. But running the wrong one gives you 
|
|
2171)
Message boards :
Questions and problems :
Can't detect VB notice.
(Message 87608)
Posted 13 Aug 2018 by Richard Haselgrove Post: Edit: It seems to install but hangs when I try and run the 64bit version.Which bit hangs - Manager or Client? Remember the common problem of the Manager opening, but sticks at "can't communicate with client". Which, as often as not, means the client has crashed immediately. |
|
2172)
Message boards :
Questions and problems :
Running boinc tasks after sleep
(Message 87604)
Posted 13 Aug 2018 by Richard Haselgrove Post: If you open BOINC Manager after the machine awakens from sleep, and call up the Event Log (Ctrl+Shift+E), does it show any messages related to the sleep and the subsequent awaken? If there are any, perhaps you could post them here. |
|
2173)
Message boards :
Questions and problems :
Can't detect VB notice.
(Message 87603)
Posted 13 Aug 2018 by Richard Haselgrove Post: We had a fairly intense discussion about the VBox notice in development circles earlier this year. My view is (was) that because it only appears when the 'bitness' of BOINC and Windows are mis-matched, it wasn't a significant problem, because there was never a need to mis-match them, and a mistake should be flagged up. If you think you've found a genuine need for a cross-install, we should investigate that first - and if we can't solve that one, then I'd accept that we should re-visit the VBox notice. |
|
2174)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 87600)
Posted 13 Aug 2018 by Richard Haselgrove Post: I see that nobody - even in the consumer entertainment business - has a customer service department that works over the weekend. |
|
2175)
Message boards :
Questions and problems :
Can't detect VB notice.
(Message 87599)
Posted 13 Aug 2018 by Richard Haselgrove Post: If you're running a 64-bit emulated Windows as WINE, you should be running a 64-bit version of BOINC too. This message should only appear when a 32-bit version of BOINC is running under a 64-bit version of Windows. Download 64-bit BOINC installer (manually from download_all if needed), uninstall current, install replacement. Ah, you've tried that. Maybe we need to debug the 64-bit install problems? |
|
2176)
Message boards :
Questions and problems :
Scheduling 2 GPUs
(Message 87569)
Posted 10 Aug 2018 by Richard Haselgrove Post: I think <ignore_nvidia_dev> is one of those cc_config items which can't be changed dynamically while the client is running. You would have to use boinccmd --quit and then restart the client - that can be done from the command line too, but it's best to allow 10-15 seconds for all processes to clean up and shut down before the restart. Apart from that, your idea sounds feasible, and is the only way to achieve what you want until we finally get #41 implemented. Both running GPU tasks will be stopped when the client is stopped, and only one will be restarted in 'day' mode. The second will sit 'waiting to run' until a GPU becomes available, and then be restarted in preference to a new task starting. [As it happens, I'm writing this on a computer where GPU0 is driving two high-res screens, and I've been doing some programming work all morning. I'd almost forgotten that GPU0 has also been running an intensive GPUGrid task all the time. If you haven't experimented recently to see if BOINC can co-exist with your foreground work, you might consider doing so.] |
|
2177)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 87549)
Posted 9 Aug 2018 by Richard Haselgrove Post: I watched one of those once, for a business customer I was doing contract support work for. Customer orders existing broadband upgrade, from 500 Kbits to 2 Mbits (it was a long time ago) British Telecom office generates two work orders: a) cease existing service b) provide new service BT Engineering generates six job tickets: 1) Tell computer to reject existing login 2) Send engineer to premises to collect existing router (we refused to let him in) 3) Send engineer to exchange to remove jumper from line 4) Tell computer to accept new login 5) Send engineer to premises to deliver new router (identical) 6) Send engineer to exchange to attach new jumper to line in roughly that order. The six steps were spread over about a fortnight, and despite hanging on to the old router for as long as possible, we were without service for two separate periods totalling about a week. They got one of my better complaint letters.... @Jord - I'll pass on your apologies at the conference call. |
|
2178)
Message boards :
Projects :
CPDN Offline Again (second act)
(Message 87543)
Posted 9 Aug 2018 by Richard Haselgrove Post: It's Thursday, so they're down for a backup again. They really ought to borrow SETI's standing outage script! |
|
2179)
Message boards :
The Lounge :
Single Board Computers And Their Use To computational Crunching
(Message 87542)
Posted 9 Aug 2018 by Richard Haselgrove Post: My only caveat would be that some projects don't have tasks available for the arm processor. CPDN, my main project is one, I can't remember which others.Those which support Linux on ARM are marked on the Projects page. The information isn't yet shown in the 'Attach to Project' wizard in BOINC Manager, but I'm working on it. |
|
2180)
Message boards :
Questions and problems :
Screen Saver not rely a Screen Saver.
(Message 87539)
Posted 9 Aug 2018 by Richard Haselgrove Post: Yes. I was the sole programmer for a UK niche product called 'FunderFinder', which enjoyed success in its target market from the early 1990s to the mid 2000s. I wrote it for MSDOS, and converted it for Windows 3.0 when that was released (and maintained it through Windows XP). Although I studied mainframe computing to Diploma standard in the early 1970s, I left the computing field for a completely different career until microcomputers became available: my programming in the desktop market is entirely self-taught. In my retirement, I am slowly learning the open source programming techniques used in BOINC, so that I can make minor contributions to the project. So yes, it can be done - through I haven't reached the screensaver page in the manual. |
|
2181)
Message boards :
Questions and problems :
Screen Saver not rely a Screen Saver.
(Message 87527)
Posted 8 Aug 2018 by Richard Haselgrove Post: The developer is the person who happened to be around at the time, and he or she developed for the platform and operating system that were around at the time. That operating system has died: so might the developer, for all you or I know. What is needed now is a new developer for the new operating system. It might be you. |
|
2182)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87517)
Posted 8 Aug 2018 by Richard Haselgrove Post: Even a humble MRI scanner might be enough... |
|
2183)
Message boards :
Server programs :
Run tasks by earliest deadline
(Message 87515)
Posted 8 Aug 2018 by Richard Haselgrove Post: I found the original question ambiguous. It could be a client-side question, but it's unusual for someone to go right down to the bottom of the list and pick the server area for those. He could just as easily have been referring to Prioritizing jobs in a multi-user project. That's server documentation, and the 'users' in this case are the research scientists. In server land, the task deadline is known as the 'delay_bound', and is set in the input template. I don't know enough about user accounts and input templates to comment about how the delay bound and the job priority could be made to interact dynamically. |
|
2184)
Message boards :
Questions and problems :
can the "TdrLevel" check be switched off?
(Message 87514)
Posted 8 Aug 2018 by Richard Haselgrove Post: I'm always worried when I read "somebody told me...". Some very strange people have written some very strange things on the internet. If you're making low level changes in an area you're not previously familiar with, always refer to the original source documentation: TDR Registry Keys. It's a DWORD. BOINC doesn't concern itself with stuff like TDR - it leaves that to the operating system. Windows manages the drivers. If you break the driver, BOINC can't use it. If you fix the driver, BOINC uses what it's given. |
|
2185)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87490)
Posted 7 Aug 2018 by Richard Haselgrove Post: That was a short summer holiday. It's down now. Another guess is that they were planning something long and difficult... |
|
2186)
Message boards :
News :
BOINC Workshop 2018
(Message 87484)
Posted 7 Aug 2018 by Richard Haselgrove Post: Be careful what you ask for! The reply to my question was 'Thanks for asking. Here are the keys to the Wiki, here's a box of slides. Help yourself.' So, several mis-steps later... WorkShop18! |
|
2187)
Message boards :
News :
BOINC Workshop 2018
(Message 87480)
Posted 6 Aug 2018 by Richard Haselgrove Post: They are most commonly added to the notice calling the workshop in Boinc Events, but no sign yet. With all the changes to BOINC recently, and with new participants preferring GitHub and/or GoogleDocs, I'm not sure what the plan is this year. I'll ask around. (Edit - autoreply: David Wallom of CPDN - our host in Oxford - is on annual leave until next week, and very sensibly hasn't taken his email with him.) |
|
2188)
Message boards :
Questions and problems :
Task view STATUS s update questions
(Message 87479)
Posted 6 Aug 2018 by Richard Haselgrove Post: Descriptions like "Running (2 CPUs)" are written by the local BOINC client running locally on your machine, in response to instructions present - either in the <app_version> sent by the server, in app_info.xml, or in app_config.xml. The last two files are not usually present, but are optional files you can write yourself. I've never heard of a project requiring two CPUs for a single task running on a GPU. Some projects have multi-threaded CPU applications available - usually intended for 4-core or 8-core CPUs - but they are intended for CPU use only: any GPU present would be scheduled independently. Without having more detail, and the names of the project(s?) involved, it's hard to generalise. |
|
2189)
Message boards :
Projects :
CPDN Offline Again (second act)
(Message 87431)
Posted 2 Aug 2018 by Richard Haselgrove Post: The database is back up at 15:24 UK time. That gives us a reasonable benchmark of 2 hours 30 minutes for future outages. |
|
2190)
Message boards :
Projects :
CPDN Offline Again (second act)
(Message 87430)
Posted 2 Aug 2018 by Richard Haselgrove Post: Posted at 13:14 UK time today: We will be taking the project offline today at 13:00 (UK time) in order to take a database backup. (Edit - to be fair, Andy sent it out 30 minutes before the outage started, but it spent 45 minutes stuck in the Oxford University email handling system) |
|
2191)
Message boards :
Questions and problems :
Activity - Run Always
(Message 87429)
Posted 2 Aug 2018 by Richard Haselgrove Post: there is no automated way to do so.It would be possible using the boinccmd command-line tool: --set_run_mode {always | auto | never} [ duration ]You could write a script to change the setting on all the machines in your workgroup (using Controlling BOINC remotely), and launch it using the Windows Task Scheduler. But that's taking you off into the heavy-haulage backwoods of BOINC - you may not want to go quite that far yet. |
|
2192)
Message boards :
Questions and problems :
Activity - Run Always
(Message 87425)
Posted 1 Aug 2018 by Richard Haselgrove Post: BOINC Manager --> Advanced view --> Activity menu --> select 'always' in all three sections. |
|
2193)
Message boards :
Questions and problems :
Unable to install BOINC Update
(Message 87422)
Posted 1 Aug 2018 by Richard Haselgrove Post: This is happening because your computer has lost track of the previous installer file, which is needed to remove the old version of BOINC prior to installing the new one. There is a long FAQ on the subject: the simplest way is usually to re-install the version that you were using before, which puts everything back in the expected place so that the new installer can find it. |
|
2194)
Message boards :
News :
BOINC Workshop 2018
(Message 87400)
Posted 31 Jul 2018 by Richard Haselgrove Post: Make that 150 years - I've linked the Wiki article. It was named after the Rev Dr Francis Randolph, the 18th-century cleric who was a chief benefactor of the Ashmolean - so very appropriate to mention it in this context. |
|
2195)
Message boards :
Questions and problems :
boinccmd --get_simple_gui_info
(Message 87398)
Posted 31 Jul 2018 by Richard Haselgrove Post: Values are in seconds, or down to (meaningless) microseconds after the decimal point. 29047 seconds is 484 minutes, just over 8 hours. Better? |
|
2196)
Message boards :
News :
BOINC Workshop 2018
(Message 87395)
Posted 31 Jul 2018 by Richard Haselgrove Post: I once stayed in a hotel across the street from the Ashmolean...The Randolph - the place to see and be seen in Oxford, for at least the last 50 years to my personal knowledge? You must have had a generous boss! |
|
2197)
Message boards :
Questions and problems :
Installation suspended
(Message 87393)
Posted 31 Jul 2018 by Richard Haselgrove Post: I doubt it was the act of installation which was suspended - more likely, the running of SETI once installation was complete. That is a user preference, under your control. Go to http://setiathome.berkeley.edu/prefs.php?subset=global, and uncheck the second and third checkboxes under 'When to suspend'. Edit - forgot you have to switch to the editable version of the page first. Green button, 'Edit preferences', at the bottom of the section. |
|
2198)
Message boards :
News :
BOINC Workshop 2018
(Message 87392)
Posted 31 Jul 2018 by Richard Haselgrove Post: And left it in the pub tonight. Bother.I've received a PM anxious about the welfare of my umbrella. Fear not. It made it out of the Turf, came up to the rooftop terrace bar (where we failed to see the lunar eclipse - cloud cover), and on to the Lamb & Flag - where it rested overnight, overlooked during our somewhat abrupt departure. And from where I retrieved it the following day, still hung up in the corner of the snug where I'd left it. That says one of two things - either the cleaning team in the L&F is as slack as hell, or (and probably more likely), Oxford pubs are used to absent-minded professors leaving umbrellas, research papers, government secrets and so on behind them, and allow a reasonable time for retrieval before recycling them to the next tourist, student, or spy that needs them. It gave me an excuse to wile away a couple of hours in the Ashmolean before the pubs opened, anyway. |
|
2199)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87384)
Posted 31 Jul 2018 by Richard Haselgrove Post: is their any way to change the default high lighted button in the options, back to 'save' instead of cancel? :-)Yes, that's written and approved. Just waiting to agree the wording in a confirmation dialog I added at the same time. #2585. |
|
2200)
Message boards :
Questions and problems :
BOINC 7.12 not opening TCP listener on Win 10 x64 Pro
(Message 87378)
Posted 30 Jul 2018 by Richard Haselgrove Post: I have Windows 10 machine updated to 1803 and WSL detection seems to be working fine. This is my daily driver so I'm not exactly enthusiastic about doing a factory reset to test WSL installed before FCU :(If it's critical, I can wipe that test install and start again - I prefer to run the W7 boot for normal work. But give me a few days to finish recovering from the workshop first... |
|
2201)
Message boards :
Questions and problems :
BOINC 7.12 not opening TCP listener on Win 10 x64 Pro
(Message 87374)
Posted 30 Jul 2018 by Richard Haselgrove Post: Well, this is my startup (bare test machine) for v7.12.1 with WSL - ran Windows update to get the July cumulative update before running. 30/07/2018 18:32:48 | | cc_config.xml not found - using defaultsJust missing an OpenCL driver because I haven't run the machine with the intel_gpu active yet. |
|
2202)
Message boards :
Questions and problems :
BOINC 7.12 not opening TCP listener on Win 10 x64 Pro
(Message 87369)
Posted 29 Jul 2018 by Richard Haselgrove Post: Would you be willing to run a debug build or two? I think that's the fastest way to figure out what's going on.Yes, I can compile and run - though the machine which is Windows-10-capable usually runs Windows 7. What are you suspecting may be the problem? (I have FCU loaded already, so reverting would be a faff) |
|
2203)
Message boards :
Questions and problems :
Problem install BOINC on Linux-Ubuntu 18.04
(Message 87345)
Posted 27 Jul 2018 by Richard Haselgrove Post: There's an active discussion of Authorization Failure -155 in the BOINC client area. We need to look at this - there were some changes in v7.10 which may be causing problems. |
|
2204)
Message boards :
News :
BOINC Workshop 2018
(Message 87344)
Posted 27 Jul 2018 by Richard Haselgrove Post: I packed an umbrella...And left it in the pub tonight. Bother. I know which pub and where I left it - I'll have to go back in the morning. And so to bed. |
|
2205)
Message boards :
BOINC client :
Authorization Failure -155
(Message 87326)
Posted 26 Jul 2018 by Richard Haselgrove Post: Further evidence: C:\BOINC>C:\BOINC\boinccmd.exe --passwd rubbish --project_attach http://setiathome.berkeley.edu/ my_account_key Authorization failure: -155I'm using Windows, where registry data is used to manage boinccmd <--> client authentication 99.9% of the time. I deliberately broke that, and triggered the failure message. I suspect the other users in this thread are running Linux, where identifying the data folder and hence reading gui_rpc_auth.cfg may be more problematic. |
|
2206)
Message boards :
BOINC client :
Authorization Failure -155
(Message 87325)
Posted 26 Jul 2018 by Richard Haselgrove Post: I'm not sure we're looking in the right place. Isn't "Authorization failure: -155" more concerned with whether boinccmd is authorized to issue commands to the local client, rather than whether the user account is recognised by the project? I need to think about and research that. Edit - confirmed: I get messages like this C:\BOINC>C:\BOINC\boinccmd.exe --project_attach http://setiathome.berkeley.edu/ my_account_key 26-Jul-2018 12:56:12: GUI RPC error: Already attached to project Operation failed: already attached to project C:\BOINC>C:\BOINC\boinccmd.exe --project_attach http://www.enigmaathome.net/ my_account_keywith, in the second case, an event log message 26/07/2018 12:58:27 | Enigma@Home | Invalid or missing account key. To fix, remove and add this project .The auth failure -155 means you need to pass the --passwd on the boinccmd command line - see Boinccmd tool: "Otherwise you need to supply (as password) the string stored in the file gui_rpc_auth.cfg in the client's data directory. If you run boinccmd remotely you also need to configure the client to accept remote control." |
|
2207)
Message boards :
News :
BOINC Workshop 2018
(Message 87320)
Posted 26 Jul 2018 by Richard Haselgrove Post: Thanks Richard, have posted link to audio on CPDN boards. I trust this is OK.Andy says yes :-) |
|
2208)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87317)
Posted 26 Jul 2018 by Richard Haselgrove Post: Just a thought over breakfast, before I head into town - has anyone suggested that you run sfc? System File Checker |
|
2209)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87313)
Posted 25 Jul 2018 by Richard Haselgrove Post: Perfect timing! The conference reaches the hackfest stage tomorrow - oops, today in this part of the world - and we'll have a room full of developers. I'll see if someone will take a look. Your crash dump does look complete, with the extra debug trace information, so they'll have a good start to work from. Fingers crossed. |
|
2210)
Message boards :
News :
BOINC Workshop 2018
(Message 87305)
Posted 25 Jul 2018 by Richard Haselgrove Post: Final reminder: the recordings are available at BOINC Workshop 2018 - talks audio The speakers in sequence start from message 87248. |
|
2211)
Message boards :
News :
BOINC Workshop 2018
(Message 87304)
Posted 25 Jul 2018 by Richard Haselgrove Post: CM - Proof of UserID ownership That's the last of the main speakers. Next session is small group discussions, which I won't record. So that probably concludes this workshop feed. Thank you for your interest. |
|
2212)
Message boards :
News :
BOINC Workshop 2018
(Message 87301)
Posted 25 Jul 2018 by Richard Haselgrove Post: Kevin Reed - BOINC: migration to community model |
|
2213)
Message boards :
News :
BOINC Workshop 2018
(Message 87300)
Posted 25 Jul 2018 by Richard Haselgrove Post: What temperatures are expected on Friday? Oh, only 27C? But with chance of thundery showers. Good luck.I packed an umbrella... |
|
2214)
Message boards :
News :
BOINC Workshop 2018
(Message 87298)
Posted 25 Jul 2018 by Richard Haselgrove Post: We're back from lunch. While the next session records, I've been asked to make two public service announcements, primarily of interest to people within close travel distance to Oxford UK. 1) Dinner tonight. 18:30 this evening at the Chiang Mai Kitchen, Kemp Hall Passage, 130A High Street, Oxford, OX1 4DH. All welcome. 2) Community building & discussion. A walk (pub crawl?!) through the environs of Oxford. Meet at the Oxford e-Research Centre (workshop venue), 09:30 Friday for 10:00 start. Up to six miles, bus back, timing will be 'dynamic'. Again, all welcome. Further details on the extended workshop agenda. |
|
2215)
Message boards :
News :
BOINC Workshop 2018
(Message 87295)
Posted 25 Jul 2018 by Richard Haselgrove Post: GDPR discussion and interpretation Slides at https://tinyurl.com/yad2m7uv |
|
2216)
Message boards :
News :
BOINC Workshop 2018
(Message 87294)
Posted 25 Jul 2018 by Richard Haselgrove Post: Regrouping for GDPR. This is a pre-recorded video which will be available in the workshop proceedings, so I'll give my recorder a rest until we reach the discussion stage. |
|
2217)
Message boards :
News :
BOINC Workshop 2018
(Message 87292)
Posted 25 Jul 2018 by Richard Haselgrove Post: Keith Uplinger - unit testing code Break - we ALL need more coffee. |
|
2218)
Message boards :
News :
BOINC Workshop 2018
(Message 87291)
Posted 25 Jul 2018 by Richard Haselgrove Post: Christian Beer - Software testing (Software testing and Continuous Integration mechanism review) |
|
2219)
Message boards :
News :
BOINC Workshop 2018
(Message 87289)
Posted 25 Jul 2018 by Richard Haselgrove Post: Client releases (Kevin Reed and Richard Haselgrove) |
|
2220)
Message boards :
News :
BOINC Workshop 2018
(Message 87288)
Posted 25 Jul 2018 by Richard Haselgrove Post: Kevin Reed - server release process |
|
2221)
Message boards :
News :
BOINC Workshop 2018
(Message 87287)
Posted 25 Jul 2018 by Richard Haselgrove Post: Marius Millea - BOINC server using Docker (technical - sorry, I can't show the live demo) |
|
2222)
Message boards :
News :
BOINC Workshop 2018
(Message 87286)
Posted 25 Jul 2018 by Richard Haselgrove Post: And here we go again. Recordings will appear on the same link as yesterday: BOINC Workshop 2018 - talks audio Just a brief intro to day 2 to start with. And this is how we expect the rest of the day to run: Day 2 schedule |
|
2223)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87284)
Posted 24 Jul 2018 by Richard Haselgrove Post: All catering is done by gender neutral bots.Unlike the Oxford pub I left about 40 minutes ago. At 22:50, a bot visited our table to say 'last orders: bar closes in 10 minutes'. I didn't need another, but had a recent refill in front of me. By 23:10, the story was 'Please leave the pub now - we're closing', and by 23:15 it was 'You're the last one in here -off you go'. Meanwhile another member of staff was smiling and saying 'don't worry, take your time'. WTF happened to drinking up time? My less confident BOINC workshoppers left at the first warning, so I was left to track down the night bus by myself. Maybe I should have held back on the anecdotes about the early days of computing? :-) |
|
2224)
Message boards :
News :
BOINC Workshop 2018
(Message 87276)
Posted 24 Jul 2018 by Richard Haselgrove Post: David Anderson - BOINC We're moving to more general discussion for the last 15 minutes, I'll record it, but not bother to upload it if we ramble too much[*]. Tomorrow is more centrally devoted to BOINC itself, rather than the projects. I'll have to ask the participants again if it's OK to record them. * Not really anything useful, except which pub we're adjourning to - the Bear in Blue Boar Street, for anyone local). |
|
2225)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87274)
Posted 24 Jul 2018 by Richard Haselgrove Post: It may be hot where you live but Seti just went down.Just in time for David's live demo of the new auto-attach feature! |
|
2226)
Message boards :
News :
BOINC Workshop 2018
(Message 87273)
Posted 24 Jul 2018 by Richard Haselgrove Post: Ilia Chernov - Inst Applied Math |
|
2227)
Message boards :
News :
BOINC Workshop 2018
(Message 87270)
Posted 24 Jul 2018 by Richard Haselgrove Post: Patrick Schoefer - SETI.Germany (about team competitions like their Pentathlon) |
|
2228)
Message boards :
News :
BOINC Workshop 2018
(Message 87269)
Posted 24 Jul 2018 by Richard Haselgrove Post: ... will be made available by the workshop administrators after it's all over. I've suggested to David Wallom that he should try to associate the slides with these recordings in the official report. |
|
2229)
Message boards :
News :
BOINC Workshop 2018
(Message 87265)
Posted 24 Jul 2018 by Richard Haselgrove Post: The Innovation Grid (Macauley Lynch, Richard Phillips) |
|
2230)
Message boards :
News :
BOINC Workshop 2018
(Message 87263)
Posted 24 Jul 2018 by Richard Haselgrove Post: Team Blocklink AG (Jan Schotta, Lorenz Langgartner, Maximillian Weigel) |
|
2231)
Message boards :
News :
BOINC Workshop 2018
(Message 87262)
Posted 24 Jul 2018 by Richard Haselgrove Post: In response to WCG. Why don't they share the project's source code to others be able to contribute, maybe further optimize or port to other platforms?As in my last answer, it isn't their code: they would at the very least need permission from the providing researchers - but that's not insurmountable. A possibly more significant problem is covered in the talk: WCG is run by IBM, and IBM is careful to protect its reputation. So, for example, even the researcher's code is security-audited before acceptance. But I'll try to get one of the WCG team to reply further. OK, I asked Juan in the coffee break. WCG tweak the submitted code to work with BOINC (checkpoints etc.), but they are very careful to avoid touching the scientific algorithms. Most of the research code that is submitted to them has been in use for several years, and the researchers are comfortable with its accuracy and reliability. Any 'improvements' would need not only the security vetting, but also a scientific re-audit to be acceptable in a research context. |
|
2232)
Message boards :
News :
BOINC Workshop 2018
(Message 87259)
Posted 24 Jul 2018 by Richard Haselgrove Post: CM - Gridcoin |
|
2233)
Message boards :
News :
BOINC Workshop 2018
(Message 87258)
Posted 24 Jul 2018 by Richard Haselgrove Post: Just been out for a brief wander round the shadier bits of Oxford - phew, it's hot out there. The building the climate guys live in (and which we're borrowing for the week) has a tall atrium, good airflow, and north-facing windows: it's very tolerable by comparison. Maybe they knew something in advance before they moved in? Listen to David's talk: he had some good factoids. If it's still this hot on Thursday, the UK National Grid will have to reduce electricity supply by 20%, because the wires will go into thermal runaway. And we're running at 97% load already... While we wait to reconvene - that GPU question. World Community Grid manage the distribution of work to volunteers, but they don't originate the research themselves. That is submitted by partner organistations around the world, and the partners provide the original software they want to be run. WCG can tweak it, but they don't go to the extent of re-writing it for another platform - or for GPUs. If the research partners provide GPU software, WCG can and have distributed it through BOINC - but none of their current projects use GPUs. Not all scientific problems can be designed to run as many simple parallel threads at the same time (which is what a GPU provides): researchers won't write for GPUs unless they have a problem which can be solved that way. |
|
2234)
Message boards :
News :
BOINC Workshop 2018
(Message 87256)
Posted 24 Jul 2018 by Richard Haselgrove Post: Steven Clark - Nanohub And we break for lunch - back in 1 hour. |
|
2235)
Message boards :
News :
BOINC Workshop 2018
(Message 87255)
Posted 24 Jul 2018 by Richard Haselgrove Post: Matt Blumberg - Grid Republic and Charity Engine |
|
2236)
Message boards :
News :
BOINC Workshop 2018
(Message 87254)
Posted 24 Jul 2018 by Richard Haselgrove Post: Next talk uploaded - Juan Hindo for World Communtiy Grid I asked your question about GPUs - you'll hear her answer, but I might be able to post a brief summary in the next break. |
|
2237)
Message boards :
News :
BOINC Workshop 2018
(Message 87252)
Posted 24 Jul 2018 by Richard Haselgrove Post: Third talk up - Bernd Machenschalk for Einstein@Home We're taking a break - I need coffee! |
|
2238)
Message boards :
News :
BOINC Workshop 2018
(Message 87251)
Posted 24 Jul 2018 by Richard Haselgrove Post: Sorry, my fault - I was still setting up and I missed the introductory words. But I think I got the full body of the main talk. |
|
2239)
Message boards :
News :
BOINC Workshop 2018
(Message 87249)
Posted 24 Jul 2018 by Richard Haselgrove Post: Second talk uploaded - Ivan Reid for LHC Anyone been able to playback? |
|
2240)
Message boards :
News :
BOINC Workshop 2018
(Message 87248)
Posted 24 Jul 2018 by Richard Haselgrove Post: First talk uploaded - David Wallom for CPDN (our host) Talks are at BOINC Workshop 2018 - talks audio Could somebody please confirm accessibility? |
|
2241)
Message boards :
News :
BOINC Workshop 2018
(Message 87244)
Posted 24 Jul 2018 by Richard Haselgrove Post: Yes, I'm logged in and recording. First talk has just started: I'll upload talks as soon as I can after each one finishes. There will be a google drive link for downloading. If anyone wants to ask a question, you can try posting it here and I'll try to ask it at an appropriate moment: it will be hard to ask individual speakers in real time. I'll try to ask the specific questions about WCG and Folding when we get to those speakers. |
|
2242)
Message boards :
The Lounge :
Needful Things
(Message 87236)
Posted 23 Jul 2018 by Richard Haselgrove Post: Turns out the room has a fan, but it's nowhere near enough. It's been a long time since I felt the urge to have a cold shower before bed, but I think that might have done the trick - while bringing back unwelcome memories of boarding school. Throwing all the bedding except half a sheet onto the floor will help, too. |
|
2243)
Message boards :
Questions and problems :
How do I remove LHC@home from my BOINC manager?
(Message 87232)
Posted 23 Jul 2018 by Richard Haselgrove Post: As I said before, uninstalling the BOINC program and cleaning the BOINC registry will make no difference (except that you won't be able to run ANY BOINC project). Everything to do with accounts and projects is stored in the boinc data folder. |
|
2244)
Message boards :
News :
BOINC Workshop 2018
(Message 87231)
Posted 23 Jul 2018 by Richard Haselgrove Post: There is no live feed from the venue this year. I am planning to attend, and will attempt to make audio recordings of the principal speakers and sessions. Like last year, I will try to upload them as soon as each talk is completed, but that depends on compatible network facilities being available at the venue - I won't know that until I get there tomorrow morning. |
|
2245)
Message boards :
The Lounge :
Needful Things
(Message 87224)
Posted 22 Jul 2018 by Richard Haselgrove Post: Bring a fan, Richard.That's a thought. I've got a fan for the bedroom, though I haven't needed to use it for the last couple of (slightly cooler) weeks. It's easier to sling it into the back of the car than it would be as hand baggage on Jet2. Though I may get funny looks on the bus from my B&B to the conference venue... |
|
2246)
Message boards :
Questions and problems :
How do I remove LHC@home from my BOINC manager?
(Message 87213)
Posted 22 Jul 2018 by Richard Haselgrove Post: The only files you absolutely need to delete/edit are in the data folder: account_lhcathomeclassic.cern.ch_sixtrack.xml (your LHC file file may be slightly different, depending on when and how you attached. But it will start 'account_' and be recognisable as an LHC file). Delete the file. client_state.xml: remove the entire section starting <project>(again, the name may vary slightly) down to the start of the next project. If you're not running ant tasks at the moment, the final line will probably be </app_version>. Delete the entire 'lhcathomeclassic.cern.ch_sixtrack' subfolder in the projects folder. And that's it. No project data is stored in the registry, and uninstalling the program leaves the data untouched (by design) - no point in wasting your time working along those lines. You might want to wait until someone with knowledge of account manager files comes along, in case there's another file that might need attention. |
|
2247)
Message boards :
Questions and problems :
How do I remove LHC@home from my BOINC manager?
(Message 87209)
Posted 22 Jul 2018 by Richard Haselgrove Post: Did you originally attach to LHC by using an account manager like BAM! ? If so, you'll have to remove it the same way. That's the only reason I know for the 'remove' button being grayed out. You can also set 'No New Tasks' for LHC while you sort it out. |
|
2248)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87207)
Posted 21 Jul 2018 by Richard Haselgrove Post: Because I need to check my GoogleDrive ready for the conference, I've popped up a debug copy of BOINC Manager v7.12.1 for Windows, if anyone wants to give it a try. This one (64-bit only) was compiled with VS2013, and the runtime DLLs - debug versions where appropriate - and the pdb file are all included in the zip: it should run if you just unpack these files into a separate folder and run boincmgr.exe from there. You can only have one copy of the Manager running at a time, so exit your standard manager before using this one (leave the client running so you've got something to test with!). Otherwise, use the /m switch on the command line to allow multiple copies. Google Drive link |
|
2249)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87206)
Posted 21 Jul 2018 by Richard Haselgrove Post: To amplify that previous thought. BOINC says: 21/07/2018 12:38:49 | SETI@home | update requested by user 21/07/2018 12:38:53 | SETI@home | sched RPC pending: Requested by user 21/07/2018 12:38:53 | SETI@home | Sending scheduler request: Requested by user. 21/07/2018 12:38:54 | SETI@home | [error] No start tag in scheduler reply 21/07/2018 12:38:54 | SETI@home | [sched_op] Deferring communication for 00:01:04 21/07/2018 12:38:54 | SETI@home | [sched_op] Reason: can't parse scheduler reply My router says:  My eyeballs say that a telecoms cable is unplugged ;-) But the notices tab didn't crash the manager. |
|
2250)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87205)
Posted 21 Jul 2018 by Richard Haselgrove Post: I very much doubt that Avast could do (or prevent) anything at installation time that would cause this problem: it would be more likely to prevent whole chunks of BOINC (client, manager) from being installed or from functioning at all. But you're right to suggest it could be implicated. Modern anti-virus products don't just scan files after they've landed on your hard disk: they also monitor internet traffic, and prevent things they don't like from reaching their destination. Perhaps one of the projects is sending a notice that Avast doesn't like, and the manager crashes when it tries to parse Avast's human-friendly warning that the download has been blocked? I've had BOINC complain (but not crash) that it can't parse the human/browser friendly message that my router sends to tell me that my internet connection is down. I've been searching through Github to see what Manager crashes have been addressed in the past. I found #2165 BOINC Manager crashes when language is Spanish (faulty translation format) #1210 BOINC Manager crash - Statistics Tab - Clicking "All projects (sum)" while a project hasn't initialized #2557 Mac: Patch wxWidgets 3.1.0 to fix BOINC Manager crash on OS 10.6. #1304 Sometimes Images in Notices can crash BOINC Manager #1311 BOINC manager crashes while attaching to project under Greek windows OS 824b3e7 Manager: don't crash if GUI RPC returns empty reply ... though it's not clear how this could happen. All of these have been fixed, but they're food for thought. In particular, the last one might be similar to my hypothetical Avast problem. It might be worth disabling Avast's "Internet Shield" (or whatever they call it) component before clicking the notices tab, and seeing if it still crashes. |
|
2251)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87196)
Posted 20 Jul 2018 by Richard Haselgrove Post: Debug builds aren't usually made unless - like now - someone wants to track down a specific problem. I can make a debug manager at v7.12.1, if you're game? It would be built with VS2013 - the release builds are kept at VS2010 for Windows XP compatibility - so would need a couple of updated DLLs as well. I think the developer would prefer to see the callstack trace from current code, rather than an earlier version. |
|
2252)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87193)
Posted 20 Jul 2018 by Richard Haselgrove Post: Well, I've tried a few of my Windows 7/64 machines, and they all display notices properly - no crashing. To track this down may require running a debug version of the Manager to show exactly where the problem occurs. Ideally, the debug version should be compiled on the same machine as the release version, but that may be a problem - the developer concerned is flying to Europe (to the UK, as it happens) this weekend for a conference. That may delay things. |
|
2253)
Message boards :
BOINC client :
Fail to add project
(Message 87190)
Posted 20 Jul 2018 by Richard Haselgrove Post: That is the standard configuration for BOINC. If project communication fails, BOINC tries to access a 'high availability server' (by default, google.com, but configurable) to try and determine whether it's a project server problem or a general comms problem. Nothing to worry about. |
|
2254)
Message boards :
Projects :
CPDN Offline Again (second act)
(Message 87178)
Posted 19 Jul 2018 by Richard Haselgrove Post: Down again?Yes. It was another planned and scheduled outage to take a database backup. Edit - backup is now complete and project should be back online. |
|
2255)
Message boards :
Questions and problems :
BOINC manager closes on selecting 'Notices' tab
(Message 87172)
Posted 19 Jul 2018 by Richard Haselgrove Post: Although you are only running two projects at the moment, your computer is attached to several more, and the client will attempt to fetch notices from those projects too. I would, first of all, suspect POGS, as the project has closed down and the old server may be responding in unexpected ways. To help us work out what the problem might be, I'd appreciate it if you could back up your account_pogs.theskynet.org.xml file (in the BOINC data directory), and then 'Remove' the POGS project. If that makes the problem go away, we can use the backup file to test further. If the problem continues, we could try other projects the same way, but I think there's less likely to be a problem with the live projects. |
|
2256)
Message boards :
BOINC client :
Fail to add project
(Message 87169)
Posted 19 Jul 2018 by Richard Haselgrove Post: When working with a client, you can often get useful information by enabling extra debug logging information. If you're using the GUI BOINC Manager, switch to 'Advanced View', and from the Options menu select 'Event Log options' (or Ctrl+Shift+F). If you're using the command line, make a cc_config.xml file, and read it in with boinccmd --read_cc_config. Either way, your initial problem seems to be with communication with your server. Use the <http_debug> log flag for comms problems. Go through the 'attach project' process again, and examine the log. The client should have requested the configuration file by issuing an RPC for 'get_project_config.php'. The debug log should indicate what went wrong: if it doesn't, try opening that file yourself in a web browser on the machine the client is running on. My guess is that it's a DNS or firewall issue. |
|
2257)
Message boards :
Questions and problems :
How to ignore CPU in one BOINC instance?
(Message 87141)
Posted 17 Jul 2018 by Richard Haselgrove Post: Correct. Each BOINC instance is independent, and neither knows anything about the other. If the other one is using CPU resources, they would count as 'non-boinc' loading, and could cause this instance to stop running if your preferences were set that way (it sounds like they're not). If both are running CPU tasks, the tasks will run slower - same as they would on any overloaded machine. You MIGHT get a small bonus - if you're running the multi-threaded tasks, they always have to pause some threads periodically so that the data can be synchronised between threads and the despatcher can work out what to do next. That makes MT tasks slightly inefficient, and the more threads are in use, the more inefficient they become: I think some people have commented on that. The app running in your second boinc instance will take advantage of those pauses, and vice-versa. |
|
2258)
Message boards :
Questions and problems :
How to ignore CPU in one BOINC instance?
(Message 87136)
Posted 17 Jul 2018 by Richard Haselgrove Post: And GPUGrid is using such old server software that you can't search for individual task names and see what host processed them. Looking at the host details page for one of my machines, the url is https://www.gpugrid.net/show_host_detail.php?hostid=43404 If you could look up the hostid for the 'other' instance, and type that into the address bar, what happens? |
|
2259)
Message boards :
Questions and problems :
How to ignore CPU in one BOINC instance?
(Message 87133)
Posted 17 Jul 2018 by Richard Haselgrove Post: The normal way would be to set up different host venues on the project server, set the usage preferences there, and set the different HostIDs to use those separate preferences. Is there still a problem setting preferences at GPUGrid? |
|
2260)
Message boards :
Promotion :
Move project to github and post to hackernews and reddit
(Message 87108)
Posted 14 Jul 2018 by Richard Haselgrove Post: It's on Github: https://github.com/BOINC/boinc There are links on the home page of this website. |
|
2261)
Message boards :
Questions and problems :
Resource Share, what if?
(Message 87097)
Posted 13 Jul 2018 by Richard Haselgrove Post: I tend to use 'no new tasks' rather than 'suspend' - many of them are projects which I have attached to briefly to run one or two test tasks to explore a problem, but not felt attracted into a long-term relationship. If I did find an unexpected task running, I would read the Event Log in detail to try to understand the circumstances, and report a bug if I found one. It's often easiest to do that by opening the event log archive file 'stdoutdae.txt', and using the 'find' facility in your text editor. The full life-cycle for a task looks like 11-Jul-2018 21:16:03 [SETI@home] [sched_op] Starting scheduler requestTo track down what happened, it's important to find that original 'Sending scheduler request: To fetch work.' that triggered the download. It's a common misconception that projects "send" work: they can't. They only respond to requests for work (if the project initiated the call, it wouldn't get through your firewall). Look back over what you were doing around the time that task was downloaded. |
|
2262)
Message boards :
Questions and problems :
Resource Share, what if?
(Message 87095)
Posted 13 Jul 2018 by Richard Haselgrove Post: It's still there, waiting for you to use it, but operationally it won't play any part. 13/07/2018 08:56:46 | | [work_fetch] --- state for CPU ---That's a fetch share of 1 (==100%) for just one project in each resource section. The same would apply to runtime allocation. |
|
2263)
Message boards :
Questions and problems :
Resource Share, what if?
(Message 87090)
Posted 12 Jul 2018 by Richard Haselgrove Post: Where did you look? Some of the Event Log options - like <Work Fetch debug> - give second-by-second data about the share, for each resource, taking into account all the current backoffs, suspensions, preferences, etc. |
|
2264)
Message boards :
Questions and problems :
Resource Share, what if?
(Message 87084)
Posted 12 Jul 2018 by Richard Haselgrove Post: Yes. No. Edit - that was a bit blunt ;-) A resource share of zero has a special, different, meaning: it makes that project a 'backup' project - work is only fetched, one task at a time, if no work is available from any other project and the computer is at risk of becoming idle within three minutes. Only use it with that meaning in mind - so it makes no sense when the project is suspended anyway. |
|
2265)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87070)
Posted 11 Jul 2018 by Richard Haselgrove Post: Final whistle :-)Not quite final. 1-1 at the end of normal time - they get to play 30 minutes overtime. |
|
2266)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 87067)
Posted 11 Jul 2018 by Richard Haselgrove Post: What's at 2100?Either it's all over, or it isn't. |
|
2267)
Message boards :
Projects :
CPDN Offline Again (second act)
(Message 86982)
Posted 10 Jul 2018 by Richard Haselgrove Post: I think they've got it! By George they've got it! You can now connect to CPDN using https://www.climateprediction.net/ Buy quick, while stocks last. |
|
2268)
Message boards :
Projects :
CPDN Offline Again (second act)
(Message 86980)
Posted 10 Jul 2018 by Richard Haselgrove Post: Well, I was going to gatecrash that meeting and I was working on preparing a workflow document explaining last week's problem. But in testing it, I find there's progress. https://www.climateprediction.net/ is now configured to process PHP, so https://www.climateprediction.net/get_project_config.php works in a broswer and will be usable for attachment. But they've got the same SSL configuration error on https://www.climateprediction.net/ as they had on https://www.cpdn.org/cpdnboinc/ last week, so the BOINC client can't connect to it. They've fixed that one once, they can fix it on the other server the same way. I just wish it didn't take several days for the lightbulb to come on over every baby step. |
|
2269)
Message boards :
Projects :
CPDN Offline Again (second act)
(Message 86975)
Posted 10 Jul 2018 by Richard Haselgrove Post: Did you get through the next stage, 'fetching master file', correctly? Never mind, still failing here. 10/07/2018 11:49:45 | climateprediction.net | [sched_op] Fetching master file 10/07/2018 11:49:46 | climateprediction.net | [sched_op] Reason: Master file request failedI've sent them the full SSL log pinpointing the error, but I don't think anyone is listening any more. No-one who knows how to drive a server, that is. |
|
2270)
Message boards :
Projects :
CPDN Offline Again (second act)
(Message 86973)
Posted 10 Jul 2018 by Richard Haselgrove Post: Don't bother, I've just done that and reported back. CPDN uses two separate servers (well, lots of them, really, but only two concern us here) Everything concerned with BOINC (attaching, checking your account, help desk etc.) goes through https://www.cpdn.org/cpdnboinc/ This is the address you should use for attaching. Everything to do with climate (the science of predicting it etc.) goes through https://www.climateprediction.net/ That site is working for browsers, but doesn't work for BOINC. Just at the moment, BOINC needs to get one tiny piece of information from it after attaching, but a certificate problem is preventing that final step. So, if you are attached - stay that way. Don't change anything. If you need to attach, await further news. |
|
2271)
Message boards :
Questions and problems :
Compiling for headless use?
(Message 86941)
Posted 8 Jul 2018 by Richard Haselgrove Post: It's encouraged, but we do ask you to participate in bug-hunting and catching problems. The bleeding edge is where the mistakes are made and (hopefully) caught. |
|
2272)
Message boards :
BOINC client :
boinccmd update
(Message 86931)
Posted 8 Jul 2018 by Richard Haselgrove Post: I think the answer is 'very rarely'. From memory, it's something like 'after 10 consecutive failures to contact the scheduler'. |
|
2273)
Message boards :
BOINC client :
boinccmd update
(Message 86929)
Posted 8 Jul 2018 by Richard Haselgrove Post: boinccmd gives an instruction to the boinc client - boinc.exe or just boinc, depending on your operating system - to carry out your request. Exactly the same thing happens if you use the graphical interface 'BOINC Manager'. To update a project, the client writes a file 'sched_request_<url>.xml' - you can see the most recent one in your boinc data directory. It contains information about your account with that project, information about your computer, any work which has been completed, and any request for new work that you might need. The client uses a process known as an 'RPC' or Remote Procedure Call to transfer this file to the scheduler component on the project's server. In return, it expects to receive a sched_reply file, which can again be seen in your data folder. Hope that gets you started. |
|
2274)
Message boards :
Questions and problems :
What's wrong with the way deadlines work using BOINC?
(Message 86903)
Posted 5 Jul 2018 by Richard Haselgrove Post: In the early days of the internet, when CPUs were slow and GPUs didn't exist, 10 days wasn't much work to cache, and people working away from home or on the road might be away from communications for days at a time. Even now, some trades - submarine crew come to mind - may be unable to make personal use of the internet for months. |
|
2275)
Message boards :
Questions and problems :
What's wrong with the way deadlines work using BOINC?
(Message 86899)
Posted 5 Jul 2018 by Richard Haselgrove Post: It is certainly possible for the project administrators, scientists, or researchers - the people, not the abstract project - to make a mistake and create tasks with a longer runtime than the assigned deadline (we were talking to one of them just the other day). If your client receives such a task, it will do its best to complete the work as quickly is possible, running that task in 'Earliest Deadline First' mode and postponing other jobs. The client will automatically abort a task which has been cached, and work hasn't even started on it by the time the deadline is reached. If work has already started on it, but is unfinished by the time the deadline is reached, a warning will be printed in the Event Log at startup (these scenarios typically apply if the user switches the machine off and takes a holiday). I thought that project servers (schedulers) wouldn't send tasks if it was 'infeasible' (the word used) for the client to finish it before the deadline. But different projects have schedulers from different revisions, and I don't know the age of the Primegrid scheduler. |
|
2276)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 86891)
Posted 4 Jul 2018 by Richard Haselgrove Post: We had an example of that just the other day. I spent two days writing a new pop-up for BOINC, and within 20 minutes of me posting the final thing, Jord pointed out I'd used the wrong word (no hard feelings - fortunately, I was still around to fix it). Don't ever expect a programmer to write coherent english while their head is buzzing with "why the f... can't I access usingLocalPrefs from DlgAdvPreferences.cpp?". When I did programming for real and for money, I had the great good fortune that the project leader knew that, and hired both a programmer and a professional author - separate people - to make the words match the deeds. |
|
2277)
Message boards :
Questions and problems :
CANCEL as default on options change
(Message 86840)
Posted 2 Jul 2018 by Richard Haselgrove Post: Argh. Feel free to comment on #2585 Sorry about that. |
|
2278)
Message boards :
Questions and problems :
CANCEL as default on options change
(Message 86838)
Posted 2 Jul 2018 by Richard Haselgrove Post: We're going to need to wait to v7.14 anyway, but I'd like to get it into the system. Last call... 
|
|
2279)
Message boards :
Questions and problems :
CANCEL as default on options change
(Message 86836)
Posted 2 Jul 2018 by Richard Haselgrove Post: OK. Unfortunately my walk in the sun was replaced by an emergency call-out, but I'm back now. Of the two suggestions, I'm inclined to pick PDW's, but you've got about 10 minutes to convince me otherwise while I brew coffee and get back in the saddle. |
|
2280)
Message boards :
Questions and problems :
CANCEL as default on options change
(Message 86828)
Posted 2 Jul 2018 by Richard Haselgrove Post: Happy to accept improvements, but can we get our thinking caps on quickly, please? I'm ready for a final test before commit and PR, and I'd like to push it up before the Americans start tinkering with the codebase again... Edit - test successful, I must have found them all. I'll go out and enjoy the sunshine for an hour or two. Fortunately, changing the wording is the simplest possible update, and nothing can possibly go wrong, go wrong, go... |
|
2281)
Message boards :
Questions and problems :
CANCEL as default on options change
(Message 86826)
Posted 2 Jul 2018 by Richard Haselgrove Post: Thanks Gary. I hate C++ even more after this morning's session - why can't you access a variable between modules, only functions? - but I've eliminated an annoying delay and - TaDa - I think it works.  Now all I have to do is hope I've reverted all the botched attempts that failed and make a clean copy of the ones which didn't. I may be some time... PS Message boxes are very boring things in GUIs - no fun emphasis allowed - and wxGetApp().SafeMessageBox() seems to be even more boring than most. |
|
2282)
Message boards :
Projects :
Distributed Hardware Evolution
(Message 86822)
Posted 2 Jul 2018 by Richard Haselgrove Post: Hmm.. never came across that wiki!.. perhaps it's only accessible from another.. not on the main list (https://boinc.berkeley.edu/trac/wiki/ProjectMain).It is, actually. In the section https://boinc.berkeley.edu/trac/wiki/ProjectMain#Submittingandhandlingjobs, about five lines down: "Submitting jobs locally". There's a huge amount of (pretty dense) detail in there: it'll be worth getting familiar with it before you go live, in case users suggest you implement features you haven't found yet. Or did you mean BackendState? Now you mention it, I can't find that link either. |
|
2283)
Message boards :
Questions and problems :
CANCEL as default on options change
(Message 86813)
Posted 1 Jul 2018 by Richard Haselgrove Post: God, how I hate that bastard offspring of a programming language they call C++, but I think I'm nearly there. I prefer more verbose confirmations:  But BOINC developers prefer short, terse, ones:  Apart from the deliberate mistake (now fixed - only saw it when I previewed here), what do you think? |
|
2284)
Message boards :
Questions and problems :
Multiple BOINC projects; individual daily schedules??
(Message 86803)
Posted 1 Jul 2018 by Richard Haselgrove Post: If you want to go the multiple clients route, they can all use the same data directory.Are you really, really sure about that? Each client will be working with its own client_state.xml, and they'll all have the same name. |
|
2285)
Message boards :
Questions and problems :
Multiple BOINC projects; individual daily schedules??
(Message 86796)
Posted 30 Jun 2018 by Richard Haselgrove Post: OK, this is what I set up some time ago. I'm exclusively Windows, so both my clients are Windows: I don't know anything about how WINE works, so you're on your own - but this may help get you started. The basic BOINC installation is entirely as normal. I choose to put the programs in D:\BOINC, and the data in D:\BOINCdata, but that's entirely arbitrary. The only change in this primary BOINC installation is that I have the line <allow_multiple_clients>1</allow_multiple_clients>in the options section of cc_config.xml in that primary data folder. The line "Config: allow multiple clients" appears in the event log at startup. I have two batch files on my desktop. They are Start client #2.bat D:\BOINC\boinc.exe --allow_multiple_clients --redirectio --detach_console --gui_rpc_port 31418 --dir D:\BOINCdata2 Start Manager #2.bat start D:\BOINC\boincmgr.exe /m /n 127.0.0.1 /g 31418 /p password Now to see if they both still work :-))) Yay! 30/06/2018 21:28:36 | | Starting BOINC client version 7.10.2 for windows_x86_64That looks like a data directory I haven't used in a while! Note that the --dir specified for starting the client has to be separate from your primary data folder. I'm trying to find where I set "don't use GPUs" in this client - it's not obvious. The primary client is happily using all three GPUs in the machine, for three different projects. |
|
2286)
Message boards :
Projects :
Distributed Hardware Evolution
(Message 86790)
Posted 30 Jun 2018 by Richard Haselgrove Post: I aborted the task after the first trickle - http://dhep.ga/boinc/result.php?resultid=154497 You're writing an awful lot to stderr_txt - it's already overflowed after the first 15 minutes of your 328 hour run. BOINC only reports the last 64KB - I hope there was nothing important at the beginning. |
|
2287)
Message boards :
Projects :
Distributed Hardware Evolution
(Message 86789)
Posted 30 Jun 2018 by Richard Haselgrove Post: OK, I grabbed a task, and here's your proof. 30/06/2018 19:36:59 | | [cpu_sched_debug] schedule_cpus(): startAnd in BOINC Manager:  That's 14 days of work, to be completed in 7 days. Immediate EDF, as surmised - the deadline is impossible on any machine (and this is my newest, fastest, i5) Please be prepared to consider factual evidence like this when you you wander into an environment where people really do know what they're talking about. |
|
2288)
Message boards :
Questions and problems :
Multiple BOINC projects; individual daily schedules??
(Message 86788)
Posted 30 Jun 2018 by Richard Haselgrove Post: Setting separate locations won't help you with this one, I'm afraid. Using two different clients would do it, though. You'd need to make sure that they didn't both try to use the same resource at the same time (which would work, but slow everything down). It's a little tricky to set up (especially setting up two different Managers to control the two clients), but it can all be done - I think I have a crib sheet somewhere. |
|
2289)
Message boards :
Projects :
Distributed Hardware Evolution
(Message 86787)
Posted 30 Jun 2018 by Richard Haselgrove Post: Hi Richard yes in that case DHE tasks would get the lowest priority compared to other BOINC projects as they have the latest deadline...Not necessarily so. The question BOINC asks is: Can this task be completed by its own deadline? That depends on so many external factors: how 'big' the task is (fpops_est); how fast the machine is; what proportion of the time BOINC is active; resource share of your project; resource share of other projects; amount of work cached; etc.; etc. You cannot know what the scheduling priority of your tasks might be on any given computer unless you know the complete state of every project being run on that machine: as BOINC does. |
|
2290)
Message boards :
Questions and problems :
Not detecting GPU on Linux
(Message 86778)
Posted 30 Jun 2018 by Richard Haselgrove Post: You might try https://wiki.debian.org/NvidiaGraphicsDrivers? |
|
2291)
Message boards :
Questions and problems :
CANCEL as default on options change
(Message 86776)
Posted 30 Jun 2018 by Richard Haselgrove Post: OK, having let that marinade for a couple of hours, how about this suggestion? If the user is currently using website preferences, the 'cancel' button is the default. If the user is already using local preferences, the 'save' button is the default. If a website user actively clicks their non-default save button, a confirmation message box asks them if they really want to switch mode, as suggested in the original #601. Existing local preference users wouldn't see the confirmation. |
|
2292)
Message boards :
Questions and problems :
Not detecting GPU on Linux
(Message 86775)
Posted 30 Jun 2018 by Richard Haselgrove Post: I'm not a Linux specialist, but this is what I've picked up from message boards over the years: The Linux kernel and the video driver are tightly bound together. If you update Linux (by itself), that linkage gets broken. The video driver - especially the special bits of the driver we use for computing - stops working. The result is: you have to (re-) install the video driver after each Linux update, to re-establish the link between them. |
|
2293)
Message boards :
Questions and problems :
CANCEL as default on options change
(Message 86773)
Posted 30 Jun 2018 by Richard Haselgrove Post: It was actually a conscious, deliberate, change: the previous design caused problems for arguably more people than the new one. You can set your computing preferences for BOINC in two different ways: either via the web sites of the projects you're attached to, or via the preferences dialog in BOINC Manager. When you first install BOINC, it will be using the values from a web site. If you simply look at the current values using the Manager dialog - even without changing anything at all - and then exit via 'OK' (it wasn't even 'save'), you switch to controlling BOINC locally, and stop paying any attention to subsequent website changes. It isn't obvious that this change has happened (although the words are there on the screen), and people less experienced than you pop up on message boards wondering why their website preferences aren't working. The problem is compounded by well-meaning message board helpers who answer questions about preferences by reference to the Manager only (because that's their preferred method) without mentioning the by-product of needing to switch permanently to local control. As it happens, I also change the preferences on three machines up to six times a day each (three pairs of 'change and change back'), but I got used to the new layout during testing and I don't find it a problem. My 'use case' is to balance two projects, one of which requires fast turnround and hence short caches, the other of which suffers lengthy outages and works better with long caches. If the cache length could be set per project..... I'd be interested to hear why you feel the need to change your machine settings so often: if we could start a dialogue, we might be able to propose a change which addresses the needs of both new and experienced users. |
|
2294)
Message boards :
Projects :
Distributed Hardware Evolution
(Message 86770)
Posted 29 Jun 2018 by Richard Haselgrove Post: I was referring to my previous post - my attempt to analyse which manifestation of 'priority' each of you were referring to. I think Michael is referring to process priority on the CPU (aka 'niceness' in Linux). |
|
2295)
Message boards :
Questions and problems :
BOINC 7.8.3 is not picking up the avx, avx2 flags in Ubuntu 16.04.3
(Message 86764)
Posted 29 Jun 2018 by Richard Haselgrove Post: Thanks - I think the boinccmd is sufficient to say "no fault found" within the BOINC ecosystem. Time to toss it back into Fred's lap, I'm afraid, with the results from what we've found here. |
|
2296)
Message boards :
Questions and problems :
BOINC 7.8.3 is not picking up the avx, avx2 flags in Ubuntu 16.04.3
(Message 86762)
Posted 29 Jun 2018 by Richard Haselgrove Post: Thanks Juha - everything is there, including avx2. I suspect that the difference between your i5-6200U and Jim's i7-3770 may be genuine - I have an i7-3770K running Windows, and that reports avx but not avx2 as well. Would you be able to guide Jim to the new location for stdoutdae - it *might* be /var/lib/boinc following the systemd changes in v7.10.2 |
|
2297)
Message boards :
Questions and problems :
BOINC 7.8.3 is not picking up the avx, avx2 flags in Ubuntu 16.04.3
(Message 86759)
Posted 29 Jun 2018 by Richard Haselgrove Post: And that, of course, contains the ones you said were missing: ssse3 sse4_1 sse4_2 avx (but no avx2). We really have to work out whether the problem is the BOINC client not sending the messages, or BOINCtasks screwing up when receiving them: different fixers needed ;-) We need that intermediate display from BOINC - does boinccmd --get_messages help? Or could you use your Windows machine manager to attach remotely to the Linux client? |
|
2298)
Message boards :
Questions and problems :
BOINC 7.8.3 is not picking up the avx, avx2 flags in Ubuntu 16.04.3
(Message 86757)
Posted 29 Jun 2018 by Richard Haselgrove Post: I don't think I can use that BOINC log - it's even more incomplete than the last one, there should be a second line. Can you please post the log the message log output from BOINC itself (either via BOINC Manager, or from stdoutdae.txt), to eliminate the possibility of artefacts being introduced by BOINC Tasks? It should reach as far as arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pcjudging by your previous examples, and it's important that we show the gaps where pdpe1gb and tsc_known_freq should be. |
|
2299)
Message boards :
Questions and problems :
BOINC 7.8.3 is not picking up the avx, avx2 flags in Ubuntu 16.04.3
(Message 86752)
Posted 29 Jun 2018 by Richard Haselgrove Post: OK, getting closer - but it's a different answer. Jim, you posted three event log headers at BT on 13 June, and a computer properties listing on 14 June. Lining those up, I get fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pc fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb invpcid_single pti tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx rdseed adx smap clflushopt intel_pt xsaveopt xsavec xgetbv1 xsaves ibpb ibrs stibp dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_eppBefore I added spaces for alignment, the top line was 236 characters - so not buffer limited. I'd suggest that BOINC's processor detection routine is asking fewer questions than the modern Linux kernel does, but we need more confirmation before I open a Github issue - ideally a comparison pair (BOINC and native) like I've just posted, but taken from BOINC v7.10.2 |
|
2300)
Message boards :
Questions and problems :
BOINC 7.8.3 is not picking up the avx, avx2 flags in Ubuntu 16.04.3
(Message 86749)
Posted 29 Jun 2018 by Richard Haselgrove Post: OK, Fred has looked into it at some length, and has concluded that it is a buffer limitation. Can BOINC be modified to fix that?Please advise Fred that the buffer size was increased from 256 bytes to 1024 bytes some 7 years ago, precisely because the latest generation of processors blow past 256 bytesThat's from https://github.com/BOINC/boinc/commit/9298b7ff6eec1cd56ec8d287b565f667435fe98f We're going to have to do more digging to find out where the truncation is happening, but I'm sure a proposal to fix it would be accepted. |
|
2301)
Message boards :
Projects :
Distributed Hardware Evolution
(Message 86746)
Posted 29 Jun 2018 by Richard Haselgrove Post: I see from Michael's email to boinc_projects that (rsc_fpops are very high at the moment and will be brought down to something more sensible)A high <rsc_fpops_est> will be interpreted by the client as a very long runtime estimate: that will invoke the second form of (scheduling) priority more-or-less whatever deadline is set. I think that completes the explanation. |
|
2302)
Message boards :
Projects :
Distributed Hardware Evolution
(Message 86740)
Posted 28 Jun 2018 by Richard Haselgrove Post: As a minor intervention in this discussion, the word 'priority' is tricky and means different things to different people in different contexts. I think Michael is referring to process priority on the CPU (aka 'niceness' in Linux). I think mmonnin is referring to BOINC scheduling priority (aka 'Earliest Deadline First'). If I'm right, both of you could also be right at the same time. |
|
2303)
Message boards :
Questions and problems :
Does WINE misreport CPU speed to BOINC?
(Message 86735)
Posted 28 Jun 2018 by Richard Haselgrove Post: In general, I think that CPDN is one of the projects that doesn't maintain speed statistics on the server, but instead relies on 'benchmarks' and a single DCF value. The benchmark value for your CPU (there's no equivalent for GPUs) is calculated periodically by code embedded in the BOINC client. One interesting comparison would be the reported benchmark figures for the same machine, by the same (numeric) version of BOINC, once as a native Linux build, and again as a Windows build running under WINE. The other variable to consider is the relative efficiency of the CPDN science applications - the same version of one particular program, as compiled separately for Linux and Windows. I'd suspect that in general scientific laboratory staff are better at configuring Linux compilers than the Windows equivalent, but there may be cases where labs have hired in Windows talent because they know it supports the greatest number of volunteers. |
|
2304)
Message boards :
Questions and problems :
Does WINE misreport CPU speed to BOINC?
(Message 86732)
Posted 28 Jun 2018 by Richard Haselgrove Post: We would need to know which project, and which task types, you're asking about: there's probably somebody here with knowledge of most of them who could help explain it. Do be aware that for most projects, the bulk of the runtime estimation work is done on the project servers, not on your computer. When you start running a new application type, the servers pay very little attention to the speed estimates supplied by your computer: instead, they start accumulating statistics on how well your computer performs on the new task. Once you have 'completed' 11 tasks of the new type - meaning that you have processed the computations successfully, returned the results, and had them accepted as valid by the server - the runtime estimates for newly issued work will be based of your reported experience so far: that would be a better point to make the comparison. |
|
2305)
Message boards :
Projects :
CPDN project offline again
(Message 86694)
Posted 26 Jun 2018 by Richard Haselgrove Post: I'm now getting a new message: 26/06/2018 17:01:28 | climateprediction.net | Scheduler request failed: Peer certificate cannot be authenticated with given CA certificates I'll report that one upstairs too. |
|
2306)
Message boards :
Projects :
CPDN project offline again
(Message 86689)
Posted 26 Jun 2018 by Richard Haselgrove Post: Investigation suggests that the scheduler is declared to require https access, but the scheduling server doesn't have an SSL certificate yet. Andy says he's working on that this afternoon - just one more push! |
|
2307)
Message boards :
Questions and problems :
How to stop using GPU?
(Message 86688)
Posted 26 Jun 2018 by Richard Haselgrove Post: 'Lost' tasks - i.e. caused by detaching without reporting first - are indeed resent even against project preferences - that's known, even if undesirable. But once the tasks have been aborted and reported (via 'update project'), they should go away for good. |
|
2308)
Message boards :
Questions and problems :
How to stop using GPU?
(Message 86676)
Posted 26 Jun 2018 by Richard Haselgrove Post: Before you go, could you please check what the 'properties' dialog shows for your Einstein project in BOINC Manager? Mine looks like this:  That should show the effects of setting preferences on the Einstein website, and as you can see, I choose not to use either CPU or NVidia GPU for that project - it's using my intel_gpu instead. BOINC as a whole honours those limits: my client doesn't request the unwanted work, and Einstein doesn't send it. I'd like to work out where it's going wrong on your system: maybe we have a bug somewhere that needs fixing, or maybe our documentation is unclear. |
|
2309)
Message boards :
Questions and problems :
How to stop using GPU?
(Message 86672)
Posted 25 Jun 2018 by Richard Haselgrove Post: If you set the project preferences to 'off', then no new tasks should be fetched for the GPUs. But that doesn't solve the problem of any tasks that may already have been downloaded and might be running now, or queued up to run later. Using 'Snooze GPU' may allow you to let the GPU cool down for a bit, or run those tasks during the cooler part of the day. |
|
2310)
Message boards :
BOINC Manager :
Today, BOINC Manager notice event date June 1 wrong?
(Message 86644)
Posted 21 Jun 2018 by Richard Haselgrove Post: No, the notice refers to a broadcast at the beginning of this month - it has probably been visible for some time, even if you've only just noticed it. Visit the SETI@Home website for discussion of where the broadcast can be watched online - they appear to have been particularly strict in their geo-location rules. |
|
2311)
Message boards :
GPUs :
BOINC 7.10.2 - Windows 7 - OpenCL GPU Detection
(Message 86508)
Posted 9 Jun 2018 by Richard Haselgrove Post: There is a pre-compiled clinfo.zip available for download. There was no specific change in the GPU detection routines in v7.10.2 |
|
2312)
Message boards :
GPUs :
Intel GPU and CPU at the same time
(Message 86507)
Posted 9 Jun 2018 by Richard Haselgrove Post: Were you using exactly the same application (not just an application from the same project) in all cases? Whether or not an iGPU (or indeed any GPU) uses a full CPU core to support OpenCL is usually determined by the compiler and toolchain used to prepare the app, and sometimes by the developer's choice of settings. |
|
2313)
Message boards :
Questions and problems :
Help needed for BOINC please
(Message 86497)
Posted 8 Jun 2018 by Richard Haselgrove Post: Sorry - even retirees take a lunch break. May I ask which is the best metric to look at on the PCs to check it is working correctlya nd processing work? There are so many tabs and buttons would like to know what is best to look atIn the first instance, it's probably better not to look at the PCs at all. For all standard BOINC projects, the project has a web page for 'Your account', and within that, a page for 'Your computers'. Here's one of mine. There's a lot of detail - which can be useful for troubleshooting - but nothing private which can't be allowed into the public domain. You, as account holder, would be able to see more private details of your own machines. You can quickly run your eye down the last column, and see that every machine has contacted the server today - except the last one, which failed on 19 May. If you want to delve more deeply, you can click 'details' or 'tasks' for each computer: you'll see them by name, but only by ID number for mine. World Community Grid doesn't use exactly the same display code as everyone else, so their page will look and display differently. I don't run that particular project, so I can't describe it exactly, but I'm sure someone else here can do so - and I'm sure the information will be accessible, somehow or other. Later on, as you gain experience (and as the distributed computing bug bites deeper!), you could look out for a utility called 'BOINCtasks' (there used to be a sticky link at the top of this forum - I must look into that). That would allow you to look at the status of all your local machines from a central workstation. Tab - Projects - reset Projects - what does this button actually do? Will it delete all processing we have ever done? Or will it kick start BOINC if it has stopped working?That button is more like a big red 'emergency stop' button, or a sledgehammer. It discards anything your own computer has received from the project {*} and might be working on - without, if my memory serves me right - even having the courtesy of telling the project that somebody else will have to take your place. Don't use it except in desperation. You've discovered the Event Log - most problems can be diagnosed from there, possibly by adding extra detail (Ctrl+Shift+F). Post here again if you need help decoding it. [*] 'reset' doesn't delete the records of work you have completed in the past and reported back to the project server. Edit - I meant to add to my previous answer: you also have a setting 06/06/2018 20:01:32 | | Suspending network activity - computer is in use Assuming you have a decent network connection (and no corporate policy restricting internet access during the working day), I'd switch that off too - your science will get back to the project quicker, and you won't notice it going. That switch probably dates back to the days of dial-up modems. |
|
2314)
Message boards :
Questions and problems :
Help needed for BOINC please
(Message 86495)
Posted 8 Jun 2018 by Richard Haselgrove Post: There doesn't seem to be a lot wrong with that, except - as you said originally - that work wasn't being completed. You might need to think about some of your preferences - user controlled settings. Your log has two significant lines: 06/06/2018 20:01:32 | | Running under account User1 06/06/2018 20:01:32 | | don't compute while active If these are business machines, they are probably logged off or switched off overnight. If they are busy during the working day with the users active, that doesn't leave much time in the day for BOINC to work in. BOINC will stop as soon as the user logs off, and BOINC will stop as soon as the user touches mouse or keyboard. For the daytime problem: you need to allow computation while the user is active. Test this out using one machine at first. BOINC is designed to run its programs at the lowest possible priority, so they should keep out of the way and have no discernable effect on the user's foreground tasks. If you do feel that the foreground tasks do slow down or show signs of stuttering when BOINC is active, you could try reducing the number of CPU cores the machines are allowed to use:  Once you find a setting that you and your users are comfortable with, deploy it to the other machines as well. Unfortunately, there's (still) no way to set BOINC to use all cores when the user is away on their lunch break, but reduce to 2 or 3 when they come back and start working again. For the nighttime problem: it's good business practice for users to log off when they've finished for the day, but if you're willing to meet the energy cost, you could let the machines 'run for science' overnight. To do this, set BOINC to run 'in service mode' - that will involve uninstalling the current version, and then re-installing the same one with the 'service mode' checkbox selected (you'll have to go into the 'advanced' page of the installer to change the default setting). Your downloaded work - including tasks in progress - won't be affected by that change: they'll pick up again afterwards right where they left off. Some BOINC projects can make use of your 'HD Graphics 530' video components, but World Community Grid isn't one of them. The video chips can't be used by BOINC in service mode, so it'll make no difference to you, but others reading this thread might need the warning. |
|
2315)
Message boards :
Questions and problems :
Screensaver
(Message 86477)
Posted 7 Jun 2018 by Richard Haselgrove Post: Another problem is that when I click on "Options/Read Config File" there is no indication that anything is happening or has happened. (The same with "Save" Config")You can see both outcomes in the Event Log. |
|
2316)
Message boards :
Projects :
CPDN project offline again
(Message 86429)
Posted 2 Jun 2018 by Richard Haselgrove Post: I can think of a way, but I'd better send you a PM about it. |
|
2317)
Message boards :
BOINC client :
Boinc overestimating time for WU
(Message 86418)
Posted 1 Jun 2018 by Richard Haselgrove Post: What do you mean "in another client"? Are you somehow running two instances of the Boinc manager?No, he said 'client' and he meant client - they're different programs. How you manage them is a separate - and potentially difficult - question. |
|
2318)
Message boards :
News :
Client version 7.10.2 released
(Message 86320)
Posted 26 May 2018 by Richard Haselgrove Post: We're going to have to keep an eye on this one. I think (but I can't prove it) that immediately after 1803, BOINC Manager didn't appear in that list. Since then, there's been one more monthly rollup (but not a full version upgrade), and now it does. There has been discussion about this issue in the BOINC developer community, and support from at least one of the 'senior' developers for setting the visibility switch via the installer when a new version of BOINC is deployed. Has anybody come across documentation about how that value is memorised by Windows? (My guess would be a GUID registry value). And given that Windows itself can only change it when the application is running, could it be set at installation? Does anyone have any example of any other applications which set their own visibility? Later: I've been reading through MSDN, specifically https://msdn.microsoft.com/en-us/library/dd378460(v=vs.85).aspx#notification https://msdn.microsoft.com/en-us/library/ee330740(v=vs.85).aspx#install_icon The latter contains Each icon in the notification area can be identified in two ways:Our icon (in code) doesn't seem to have a GUID - the handle seems to be retrieved at runtime. That should make it compatible with Vista, but unknown at installation time. Um. |
|
2319)
Message boards :
GPUs :
Intel GPU and CPU at the same time
(Message 86314)
Posted 25 May 2018 by Richard Haselgrove Post: Einstein is a particularly curious case. The intel_gpu app does, as you say, slow down dramatically if all CPU cores are loaded with BOINC applications. But you can bypass that by using a program like Process Lasso to peg the iGPU app - NOT any of the others - to real-time process priority. I find there's a tiny stutter every 11 minutes or so, as one task finishes and the next starts up, but no other detrimental effect on the usability of the machine. However, you may find that the CPU apps - especially, heavily optimised floating point apps like SETI and perhaps Einstein (untested) - run slower in that configuration. Lighter weight, and primarily integer, CPU apps suffer less. Real-time process priority is NOT recommended for general use, but in this case it helps. Explore with care. |
|
2320)
Message boards :
News :
Client version 7.10.2 released
(Message 86313)
Posted 25 May 2018 by Richard Haselgrove Post: Windows 10 pro, version 1803I thought that might be the case. Microsoft has been playing around with the display of icons in the notification area, and version 1803 seems to be the worst case yet. Even the up-arrow to 'Show hidden icons' doesn't show the BOINC Manager icon any more - but it's running. All is not lost. This site seems to have a comprehensive set of options: https://winaero.com/blog/always-show-tray-icons-windows-10/ I've tested them under 1803, and they still work. Perhaps the neatest is pressing the Win + R keys on the keyboard to open the 'Run' command dialog, and pasting shell:::{05d7b0f4-2121-4eff-bf6b-ed3f69b894d9}into the box: that opens the Windows 7 style customisation page, and you can set the BOINC Manager behaviour to 'Show icon and notifications'. BOINC Manager does have to be running for it to show in the customisation list.I had my machine set to show icons before the 1803 update, so I haven't been able to test every part of the process: perhaps you could let us know how you get on. |
|
2321)
Message boards :
BOINC client :
Boinc overestimating time for WU
(Message 86305)
Posted 24 May 2018 by Richard Haselgrove Post: I suspect that the main problem may be caused by running two different Einstein applications on the same machine. That needs a history lesson. The very first versions of BOINC only ran CPU applications, and most projects only had one type of application. The BOINC client on your machine kept track of the real-world speed of those single-application projects by means of a single (one per project) value called 'DCF', or Duration Correction Factor. Then GPUs came along, and multiple-application projects, and it all fell apart - the single DCF value couldn't adjust estimates for two different applications simultaneously. Instead, as part of the CreditNew release in 2010, DCF was replaced by APR (Average Processing Rate) tracked on the server, which can adjust for an arbitrary number of applications and devices at the same time. The Einstein project - specifically - didn't accept the CreditNew design, and so didn't adopt the integrated Runtime Estimation tools either, and never developed their own replacement. If you look at the Project properties for the Einstein project in BOINC Manager, you'll probably see a line like Duration correction factor 0.5865(that's my intel_gpu running Einstein on this machine). SETI and every other project will either hide DCF because it's redundant, or show exactly 1.0000 as the value. You'll probably be able to see DCF dropping every time the AMD finishes a task, and bobbing back up again every time the iGPU finishes one. The only workrounds I can think of for this are: 1) run different projects on the two different kinds of GPU (and the CPU, for that matter) 2) wrap the current Einstein applications up in an app_info.xml file (see Anonymous platform), and declare your own speed rating (<flops> value) for each app. But that's tricky. Apart from that, the only true solution involves work by the Einstein developers. A lot of work. Edit - also read Gary Roberts' comment recently at Einstein. |
|
2322)
Message boards :
News :
Client version 7.10.2 released
(Message 86297)
Posted 23 May 2018 by Richard Haselgrove Post: I have had nothing but trouble since downloading 7.10.2. Boinc runs, but when I leave the boinc window, I can't get back to it except by using the download file to repair the software.Operating system, and version thereof? |
|
2323)
Message boards :
Questions and problems :
GPU unused, unstable CPU consumption
(Message 86239)
Posted 18 May 2018 by Richard Haselgrove Post: Both options are available. You can set preferences in either of two different places: either in the global preferences web page of any of the projects you're attached to, or locally from the Tools menu in BOINC Manager. Pay attention to the settings you use, and be consistent: local settings override web-site settings, so you can't mix them. With that out of the way, the ones you are interested in are at the top of the page in both places, and the (current) wording should be the same: Use at most 100 % of the CPUsSome projects may have older versions of the website code, so check the wording carefully. It sounds as if you might have chosen the second of the options to change (throttle by time). With GPUs especially, I would set that to 100%, and change the first option (number of CPUs) instead. Note that works as an integer number of cores used: your 70% would permit 2 of 4 cores to be used, or 5 of 8 cores. |
|
2324)
Message boards :
Questions and problems :
Quick question: How to reorder tasks in received time in BOINC manager?
(Message 86218)
Posted 15 May 2018 by Richard Haselgrove Post: Depends on the operating system. In Windows, it's a registry hack: [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager\Tasks] "SortColumn"=dword:ffffffff You have to shut down the Manager (you can leave the client running), and either set that value, or simply delete the SortColumn key - BOINC Manager will recreate it with the null value next time it runs. Yes, I'd like to have a "remove filter/sort" - like Microsoft Access - button too. |
|
2325)
Message boards :
News :
BOINC Workshop 2018
(Message 86211)
Posted 14 May 2018 by Richard Haselgrove Post: https://downforeveryoneorjustme.com/oerc.ox.ac.uk says it's down for everyone. I think the CPDN project shares the same underlying hardware, and they've been having problems recently. The project thread CPDN project offline again is likely to get more frequent updates than here. |
|
2326)
Message boards :
News :
PMC meeting minutes posted
(Message 86209)
Posted 14 May 2018 by Richard Haselgrove Post: Just looked in a secondary email account, and found notification for more minutes. Looks like they have ended up in http://boinc.berkeley.edu/trac/wiki/BoincPmcPage - February, two from April, and a tentative forward schedule. |
|
2327)
Message boards :
Questions and problems :
Are there any current issues with BOINC and Ubuntu 18.04 LTS release?
(Message 86185)
Posted 11 May 2018 by Richard Haselgrove Post: We are in the final stages of testing BOINC v7.10.2 as a release candidate. You are more than welcome to test it, but you would need to adjust your viewpoint and get used to running BOINC in service mode from a repository distribution - quite a lot will be different (not least, file locations, ownership, and permissions). I believe you're used to running applications under Anonymous Platform, so that change will be significant for you. On the plus side, this release contains a fix for TBar's long-standing complaint about formatting errors for the timestamps in the Event Log: #2420. Please ask him to thank Juha Sointusalo for that. |
|
2328)
Message boards :
BOINC client :
GPU stopped due to quad core task
(Message 86173)
Posted 9 May 2018 by Richard Haselgrove Post: Jord is absolutely right, and I've done that with the Milkway Nbody tasks. They can run on any number of threads that you care to set for them, but do keep the numbers you use for avg_ncpus and nthreads equal. You may find that any existing work you have already downloaded before making the change still runs using the old settings. Have patience - when you have downloaded new work after the change, and starting running it, it should behave as you want. |
|
2329)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 86137)
Posted 8 May 2018 by Richard Haselgrove Post: Then the party is on in the fun and sun capital of the world, the lower left coast, where the palm trees sway in the gentle breezeLike we had for our public holiday yesterday. Most surprisingly. |
|
2330)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 86131)
Posted 8 May 2018 by Richard Haselgrove Post: Let's hope it's not a long one, then. We have darkening skies, and a 97% chance of rain in the next hour. And we're sending it in your direction... |
|
2331)
Message boards :
Projects :
CPDN project offline again
(Message 86130)
Posted 8 May 2018 by Richard Haselgrove Post: No = the people who need to know will know already. But if you want to find out for yourself why you can't report results, you can turn on <http_debug> for the Event Log, and read the header messages to and from the scheduler. |
|
2332)
Message boards :
Projects :
CPDN project offline again
(Message 86125)
Posted 8 May 2018 by Richard Haselgrove Post: And again: As you might have noticed that the project went down again over the weekend. Again this was due to the main application machine on the underlying VMware/GPFS infrastructure. We have redirected the project services to the backup server (this is currently on HTTP only), please let me know if you have problems connecting. |
|
2333)
Message boards :
Questions and problems :
One email address can be used to register two or more accounts on different projects, bug or feature?
(Message 86113)
Posted 6 May 2018 by Richard Haselgrove Post: Each project is a completely separate entity. Email addresses can be used (and re-used) at more than one project, but the email address is only the unique key within each project. You can use the same email address, and even the same password, at two different projects, but there's no transfer between them. If you change your password at project A, you can still log in with the old password at project B - until you change that password separately and independently. The CPID is a hash generated from the email address, and is shared between projects. It's only used for sharing statistics: you can't use a CPID to join or log into a new project. |
|
2334)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 86110)
Posted 5 May 2018 by Richard Haselgrove Post: That certainly would explain what you're seeing. I use D:\BOINC for programs, and D:\BOINCdata for data, and so far it hasn't been a problem for me. The only difference between us is that you have one folder as a sub-folder of the other, but I don't see why that should suddenly have become a problem. You mentioned that you used the Win x64 installer for 7.10.2 Can you confirm that you also used the x64 version of 7.8.3? And did you explicitly uninstall the old version yourself, or just let BOINC handle it for you? Just trying to get a picture of exactly what steps you went through. Oh, and are you using a service install or an ordinary user install? |
|
2335)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 86105)
Posted 5 May 2018 by Richard Haselgrove Post: I've used both 7.9 and 7.10 installers multiple times during this development cycle, and I've never had that problem. Would you mind trying one more time, first making a note of where your current BOINC Data directory is located, and then looking at the installer screen so see what the new installation folder will be? That might help us understand *why* BOINC can't find it when it starts up. |
|
2336)
Message boards :
Questions and problems :
Manager lost Project(s) and can not add new ones
(Message 86096)
Posted 4 May 2018 by Richard Haselgrove Post: I think you must have subscribed to a testing/development repository. We have been working on a new release, and versions 7.9.n have been available recently for testing. We have now moved on to a release candidate 7.10.2 build: if this passes testing, it should become available in the stable production repositories. |
|
2337)
Message boards :
Projects :
CPDN project offline again
(Message 86088)
Posted 3 May 2018 by Richard Haselgrove Post: And now Andy Bowery writes again: Just to let you know that the project is now back online, rather than running from the backup project server, we are running from the main project server. The infrastructure still remains at risk due to ongoing instabilities in the main VMware/GPFS infrastructure. |
|
2338)
Message boards :
Questions and problems :
Projects continue to process even after exiting BOINC Manager
(Message 86087)
Posted 3 May 2018 by Richard Haselgrove Post: If by 'X out', you mean the red cross in the top-right corner, this is by design. The cross removes the Manager window from your screen, so can use the space for something else, but doesn't fully close it: instead, it's minimised to the 'notification area' or 'system tray' at the bottom right (by the clock). BOINC itself does very little processing, and can safely be left running while you work. It's the scientific research applications that BOINC manages which can sometimes get in the way and slow things down - although the effect is designed to be minimal. If you need to stop the science applications from running, there are various options you can use. The default settings are to suspend computation when the computer is 'in use': that means when activity is detected with the mouse or keyboard. So, it doesn't work if you're just passively watching a film or something. It is possible to set BOINC to stop processing when it detects that a specified application is running on the computer: that can be useful if you play computer games or have an intensive business application. And finally, you can exit BOINC completely (closing the science applications as you do so) by selecting 'Exit BOINC Manager' from the File menu at the top left of BOINC Manager. This is probably closest to your original idea, but have a look round the other options and ask again if you have a specific need. |
|
2339)
Message boards :
BOINC client :
Upgraded Boinc to 7.10.1 and Manager can't connect to localhost anymore
(Message 86078)
Posted 2 May 2018 by Richard Haselgrove Post: I've sent out the release / rebuild instruction for v7.10.2 - hopefully that will resolve all the confusion. |
|
2340)
Message boards :
Questions and problems :
Manager lost Project(s) and can not add new ones
(Message 86074)
Posted 2 May 2018 by Richard Haselgrove Post: [slightly off topic] In order to receive email notifications about new messages on this or any BOINC project message board, you have to have two things in place. 1) You have to 'subcribe' (button above this message list) to each individual thread that you wish to follow 2) You have to choose an appropriate notification frequency in your account's community preferences If both of those settings are in place, and anyone is still not receiving notifications as expected, please let us know and we'll ask the powers-that-be to investigate. |
|
2341)
Message boards :
Projects :
CPDN project offline again
(Message 86072)
Posted 2 May 2018 by Richard Haselgrove Post: Andy Bowery wrote: The CPDN project is now offline. I will be taking a database dump of the database in order to resurrect the master-slave relationship on the two database servers in the project. In order to do this I need a database where no transactions are taking place. Once this is complete, it is likely we will start the project from the backup project server, rather than the main project server, due to ongoing instabilities in the main GPFS infrastructure. |
|
2342)
Message boards :
BOINC client :
Upgraded Boinc to 7.10.1 and Manager can't connect to localhost anymore
(Message 86071)
Posted 2 May 2018 by Richard Haselgrove Post: There is a new BOINC version available now (still 7.10.1 ... but it's a new one). It fixed this problem. Client is able to run and Manager is able to connect to it.Can you give me particulars of when and where this new package was released - which distros have access to it? I'll update the sticky warning message I put in Questions and problems yesterday. |
|
2343)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 86066)
Posted 1 May 2018 by Richard Haselgrove Post: Now that it looks like that problem might be fixed.IST have declared 'problem over': Description: UPDATE: Tuesday, May 1, 2018 2:31pm – The firewall and network continue to remain stable. This issue is now resolved at this time.Keep an eye on it when SETI unleashes the outrage recovery, please? (by that time I'll probably be asleep). Let them know if it isn't. |
|
2344)
Message boards :
Questions and problems :
WARNING - Debian and Ubuntu update [broken] fixed
(Message 86053)
Posted 1 May 2018 by Richard Haselgrove Post: UPDATE The problem described below has been fixed, and a fresh update to v7.10.1 should restore normal operation. There will soon be a version 7.10.2 to eliminate the confusion. The new version 7.10.1 (only) of BOINC doesn't currently run on Debian, Ubuntu and derivatives. This requires additional work by the Linux package maintainers, and we don't currently have a completion date. Please avopid updating for the time being: we will make a new version 7.10.2 available as soon as possible after the problem has been fixed. In the meantime, please revert to an earlier version, and watch for announcements here. |
|
2345)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 86052)
Posted 1 May 2018 by Richard Haselgrove Post: And they seem to be making slow, but steady progress. Uploads have been horribly sticky all morning, but seem now to be working a lot better. And I can teach this board to post! The good news is that once I've cleared the uploads, getting new work - AND downloading it - seems to be relatively easy. Looking forward to hearing the postmortem report. |
|
2346)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 86046)
Posted 30 Apr 2018 by Richard Haselgrove Post: Not what IST say - there may be more to come. Description: UPDATE: Monday, April 30, 2018 2:20pm – This continues to be a sporadic ongoing issue and the network team is working to resolve the issue. There is no ETA at this time. |
|
2347)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 86043)
Posted 30 Apr 2018 by Richard Haselgrove Post: Update to Gary's link: Description: UPDATE: Monday, April 30, 2018 10:17am – IST staff is aware of instability to the Palto Alto Firewall in the Earl Warren Data Center and are troubleshooting to determine the cause and work toward a resolution.And Eric has made a News item: The UC data center switched over to a new firewall this morning. Since then packets into and out of the data center have been suffering drops. The Data Center staff is debugging the problem, we'll probably be dropping packets until it's resolved.Further updates can be accessed at http://ucbsystems.org/category/active/unscheduled-outage/ |
|
2348)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 86034)
Posted 30 Apr 2018 by Richard Haselgrove Post: The transatlantic cable, its endpoints, and the associated routers? Edit - the kittyman says it's happening stateside as well, so scrub that one. |
|
2349)
Message boards :
BOINC client :
Upgraded Boinc to 7.10.1 and Manager can't connect to localhost anymore
(Message 86016)
Posted 27 Apr 2018 by Richard Haselgrove Post: There is a known problem with the source code for v7.10.1 - one of the bug-fixes relies on a precursor patch, and I missed that when calling the release 'ready'. I did issue a recall, but evidently not quickly enough. We'll get a v7.10.2 out as soon as I can find someone to transfer the missing patch: unfortunately, it's a complicated one I don't feel confident enough to tackle myself. Edit - BOINC consists of two parts, the Client and the Manager. You'll have updated both, not just the Manager. It would be helpful to know whether your problem is a) Both client and manager are running, but they can't communicate b) The client ISN'T running, so there's nothing for the manager to connect to Are you able to use operating system tools to check the status of the client, independently of the Manager connection? If the client fails to start, there might be error messages in the log files which could help us. |
|
2350)
Message boards :
Questions and problems :
http 401 unauthorized
(Message 85984)
Posted 24 Apr 2018 by Richard Haselgrove Post: I'm not 100% sure about this, but I think that for World Community Grid you have to open an account first via their website, and only then attach to it through BOINC. That would be different from most other projects, where new account creation can be done as part of the same process as attaching to the project. If you haven't already, create an account at https://www.worldcommunitygrid.org/join.action#signup, and see if that gets you past the 'attach to project' problems. |
|
2351)
Message boards :
Documentation :
Cobblestone / BOINC credit / CPU time / GFLOPs equivalency documentation discrepancy
(Message 85974)
Posted 23 Apr 2018 by Richard Haselgrove Post: Sorry, I could have answered this from memory yesterday, but it's proved hard to locate the documentary evidence that you have the right to request. Your second statement is the one officially recognised by BOINC: 1 BOINC credit equals 1/200 day CPU time @ 1 GFLOPs/secondThere are two complementary calculations: "How many credits should I get for a day's processing on a 1 GFLOP device?" Answer = 200 "How many Fpop-days did I compute in order to earn 1 BOINC credit?" Answer = 1/200 The second version is actually easier to track down an authoritative source for. A number of places - most notably, the home page of this web site, but also a number of the statistics sites like https://boincstats.com/en/stats/-1/project/detail - will quote you the total processing power of the BOINC platform, or a subset of it like a single project. The value today is stated as 14.879 PetaFLOPS (this site), or 20,880.857 TeraFLOPS (BOINCstats). I think BOINCstats archives a wider range of data from more projects than this site does. The source code for this web site is at https://github.com/BOINC/boinc-site/blob/master/index.php#L65: $petaflops = number_format($credit_day/200000000, 3);- which is where the magic 200 that you're trying to track down can be found. Similarly, the BOINCstats page I linked is showing Recent average credit RAC 4,176,171,453 Average floating point operations per second 20,880,857.3 GigaFLOPS / 20,880.857 TeraFLOPSIt's no coincidence that the first figure is exactly 200 (less several orders of magnitude) larger than the second figure - fpops is calculated from credit, not the other way round. How did the documentation discrepancies come about, I hear you ask? When BOINC and this web site was first set up, the original figure was 100, and this will have been copied into the earliest sources. But in the summer of 2010, it was adjudged that a mistake had been made, and the true definition should have used 200 all along. This is the evidence which is hard to track down, but I found a smoking gun in SVN revision 21962 of Jul 15, 2010. My memory suggests that there was more detail in an email - perhaps on the development or projects email lists - around that time: I'll have another look now that I have an exact date. With regard to the conflicting documentation sources you've uncovered, I'd suggest that BOINC's own Wikis, and the English-language Wikipedia, are more likely to be reliable: projects, Account Managers, and translations are secondary sources which may not have been updated (yet) to reflect the 2010 redefinition. At least with a Wiki, you can always check the dates on the page history. |
|
2352)
Message boards :
Questions and problems :
Curious event or happening.
(Message 85961)
Posted 22 Apr 2018 by Richard Haselgrove Post: The eventlog held in stdoutdae.txt (BOINC data directory) is continuous across reboots. |
|
2353)
Message boards :
Questions and problems :
Curious event or happening.
(Message 85959)
Posted 22 Apr 2018 by Richard Haselgrove Post: Have you looked at the Event Log? |
|
2354)
Message boards :
Questions and problems :
BOINC - Virus False Positive?
(Message 85955)
Posted 21 Apr 2018 by Richard Haselgrove Post: It's not only scans that you need to exclude. Modern anti-virus products also monitor files as they are downloaded over the internet - suspect files end up in quarantine, rather than in the intended folders. |
|
2355)
Message boards :
Questions and problems :
BOINC - Virus False Positive?
(Message 85951)
Posted 21 Apr 2018 by Richard Haselgrove Post: Anything in a slots\n location belongs to a science project, not to BOINC itself. restart.idx is a file which we've been discussing at the GPUGrid project, which is the project this particular file belongs to. Have a read of BOINC Trojan - which should definitely have had a question mark on the end. The consensus is that this is a false positive by the AV provider. |
|
2356)
Message boards :
Questions and problems :
Setting up headless quad nVidia Ubuntu rig
(Message 85935)
Posted 19 Apr 2018 by Richard Haselgrove Post: Read Controlling BOINC remotely. |
|
2357)
Message boards :
Projects :
CPDN project going offline this afternoon
(Message 85897)
Posted 16 Apr 2018 by Richard Haselgrove Post: Only have finished tasks. Reporting for my other project causes cpdn back off times to increase. Is there a way to stop BOINC manager "contacting" the server until cpdn reporting is working?Sadly, no. One was proposed and even coded, but rejected by the custodians of the central codebase. Don't worry about it - they'll all be reported in one connection once the servers are back up, and the differential timings will help the servers cope with the mad scramble to create and distribute new work. |
|
2358)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85849)
Posted 12 Apr 2018 by Richard Haselgrove Post: Only commercial (shops and offices) is done to that rigid timetable. I think apartments and houses are usually paid monthly. |
|
2359)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85847)
Posted 12 Apr 2018 by Richard Haselgrove Post: OK, I'm curious. When is Tax Day in Britain. Sounds like it might be the same as U.S. April 15th other thereabouts and not landing on a weekend.The old English quarter days (also observed in Wales and the Channel Islands) are Lady Day (25 March) Midsummer Day (24 June) Michaelmas (29 September) Christmas (25 December) The financial year end was Lady Day. In 1752, Britain switched from using the Julian to the Gregorian calendar (rather belatedly, it has to be admitted). So the human calendar shifted by 11 days relative to solar time. To keep the tax take consistent, the financial year end was shifted by the same 11 days, so the year ends on 5 April and restarts on 6 April. Unfortunately, commercial landlords also collected rent (in advance) on quarter days, and have continued to do so (Why is rent paid on quarter days???). So, commercial landlords pocketed an extra 11 days rent in 1752, and have kept it ever since. I don't actually have to send in my tax papers until 31 January next year, but all the data was complete last week and I can do it any time now. Or not. |
|
2360)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85838)
Posted 12 Apr 2018 by Richard Haselgrove Post: Anybody with Fermi or Kepler based cards got any Arecibo Vlars on the gpu yet?Yes, got some on my GTX 470. I'll let you know what it feels like when I try to do my income tax return on that machine at the weekend. |
|
2361)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85815)
Posted 10 Apr 2018 by Richard Haselgrove Post: Yawn. Do I have to get up now? |
|
2362)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85808)
Posted 10 Apr 2018 by Richard Haselgrove Post: I could have had a lie-in.Henry and Hetty might keep you up.It's time to wake up the mods so they will be ready for tomorrow's outrage.Bad timing. I'm just off to bed! |
|
2363)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85805)
Posted 9 Apr 2018 by Richard Haselgrove Post: It's time to wake up the mods so they will be ready for tomorrow's outrage.Bad timing. I'm just off to bed! |
|
2364)
Message boards :
Web interfaces :
Frequency of BOINC status
(Message 85783)
Posted 7 Apr 2018 by Richard Haselgrove Post: It is true to say that SETI.Germany organises a special WOW! Challenge for SETI only once a year, to mark the anniversary (in August) of the recording of the famous WOW! signal. While the annual challenge is running, SETI.Germany manages to update progress hourly, even though the bulk export files remain on a 24-hour cycle, as Rob says. The challenge stats take several minutes to update each time, so my suspicion is that for the limited number of participants involved (*), they may have received special permission to collect the individual stats directly from the website for this purpose. Consult with the project administrator - Eric Korpela for SETI - and obtain their explicit permission before attempting anything like this. Projects have been know to block access by IP address if users abuse their hospitality by scraping web data. * edit - the WOW! challenges are by individual user pre-registration, so the numbers involved are kept under control. Web scraping would not be allowed, and probably wouldn't work, for the SETI population as a whole. |
|
2365)
Message boards :
The Lounge :
Needful Things
(Message 85780)
Posted 7 Apr 2018 by Richard Haselgrove Post: I've now got SSD boot drives on most of my machines, and they really do make a tremendous difference. Trouble is, they only get rebooted once a month for Patch Tuesday, so the only real gain is that I can move more quickly from machine to machine while patching. I have the development machines set up with dual monitors on dual KVMs: if I don't have the KVMs set correctly when they reboot, Windows gets the resolution detection all wrong and I'm stuck like that until the next reboot. |
|
2366)
Message boards :
The Lounge :
Needful Things
(Message 85777)
Posted 7 Apr 2018 by Richard Haselgrove Post: I have noticed it's a lot faster than before. Programs load and exit quicker, even the internet feels speedier.Interesting to read that, while the absolute headline speed has gone down from 3.30GHz to 3.20GHz. I always thought those gigahertz advertising wars were a bit of a fraud, and now we have the proof. Don't work harder, work smarter. |
|
2367)
Message boards :
The Lounge :
Needful Things
(Message 85769)
Posted 6 Apr 2018 by Richard Haselgrove Post: There are stories to be told of that '52, but they probably need a beer accompaniment. The fun one was the 24-hour field "most contacts" competition, with no mains power allowed. The picture doesn't make it clear, but it's a real brute, probably five feet across - minimum two adults or four schoolkids simply to move it. Getting it out to the school playing fields on the other side of town was an expedition in itself - and rigging the horizontal SW dipole from sectional guyed masts... |
|
2368)
Message boards :
Questions and problems :
Mac Pro (2013) (Trashcan), 10.13 (High Sierra), GPU crashes
(Message 85764)
Posted 6 Apr 2018 by Richard Haselgrove Post: Are there beta release notes anywhere? I've got 7.9.3 installed at this point, but curious if there's anything in there about GPU fixes, etc.Have a look through BOINC 7.9/7.10 Change Log on this message board. I don't remember anything recent, but we are pulling up patches from as far back as June last year - the v7.9.2 list is a long one - so worth looking. |
|
2369)
Message boards :
The Lounge :
Needful Things
(Message 85760)
Posted 6 Apr 2018 by Richard Haselgrove Post: My last post reminded me of a story from years ago, but I thought I'd better not tell it until you'd come out the other side. I was minding my own business on one side of the room, when I heard a 'pop' sound from the other side of the room. Turning round, I was in time to see a perfect 18" mushroom cloud rising from the middle of what would nowadays be an open PC case lying on the floor, and my schoolmate rocking back on his heels away from it. Back in the late 1960s, the case actually contained an amateur radio transmitter, 'amateur' in the sense that he was building it from components. Back in those days, radio transmitters - especially on school budgets - used thermionic valves and very high voltages. Our biggest set used anode voltages of 1,000V or more, powered by rotary converters from 12V batteries - it was army surplus kit like  Used by the British Army, this is a WS-52 set less accessories. (L-R) Receiver, common power supply and transmitter. The dark, mail-box shape atop the WS-52 was a post war modification which contained a new dynamotor because the original 1200v dynamotors were not reliable due inadequate HV insulation. It was cheaper to install a piggy back unit rather to retrofit a new one. At the top right is an antenna tuner. (Photo courtesy Wireless for the Warrior, http://www.jproc.ca/marconi/no52_set.html)We didn't have the dynamotor safety modification, so ours were the originals in the central narrow section, between the squarer receiver and transmitter sections. The one my friend was building would have been much lower power, but still had enough oomph to blow a very pretty mushroom. |
|
2370)
Message boards :
The Lounge :
Needful Things
(Message 85758)
Posted 6 Apr 2018 by Richard Haselgrove Post: If I am not heard from again.... ;-)We'll send you some sticks, a canister of smoke fluid, and a blanket. |
|
2371)
Message boards :
The Lounge :
Needful Things
(Message 85753)
Posted 6 Apr 2018 by Richard Haselgrove Post: Wouldn't cost much to replace the CMOS battery, just in case. |
|
2372)
Message boards :
Questions and problems :
Mac Pro (2013) (Trashcan), 10.13 (High Sierra), GPU crashes
(Message 85734)
Posted 5 Apr 2018 by Richard Haselgrove Post: There is a beta - please feel free to try it. Betas are always listed at https://boinc.berkeley.edu/download_all.php - a small 'All versions' link on the main download page. Current beta version is 7.9.3, to be replaced by 7.9.4 in the next couple of days. There's a significant fix to the appearance of one of the dialogs in 7.9.4, but it's cosmetic only - so far as we know, it works properly. Until you tell us different... |
|
2373)
Message boards :
BOINC client :
task switching limitation when running multiple tasks on same GPU?
(Message 85722)
Posted 4 Apr 2018 by Richard Haselgrove Post: Agreed. |
|
2374)
Message boards :
Projects :
Are projects responsible for "cheaters" ?
(Message 85721)
Posted 4 Apr 2018 by Richard Haselgrove Post: I had a look at the Collatz site and message boards yesterday, and could spot one very obvious cheat which I won't discuss here. BOINC HQ staff are aware of the problem and are on standby to help out if needed. BOINC provides the tools, but it's up to projects to implement and use them. Slicker thought it would be difficult to use the main one, in light of the nature of his project - it exists, after all, simply to prove or disprove a mathematical conjecture, and a single counter-example would be enough to disprove it. There are only so many ways you can compare "nope, this isn't a counter-example" with "nope, this isn't a counter-example". |
|
2375)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 85649)
Posted 2 Apr 2018 by Richard Haselgrove Post: I hope Uncle Lazarus pays a visit just over the border in West Yorkshire: http://www.bbc.co.uk/news/uk-england-leeds-43618517. All the heating in Silsden (the next small town up the valley from me) got cut off on Good Friday morning, and is only being put back on this evening. And I can attest from personal experience that it's been a horrible day to be without heating or hot food (fortunately I have both): it snowed all morning, and rained most of the afternoon. |
|
2376)
Message boards :
Questions and problems :
Older AMD APUs and Boinc under Ubuntu
(Message 85623)
Posted 2 Apr 2018 by Richard Haselgrove Post: SETI has applications for Open'C'L (computing on graphics chips) but not for Open'G'L (rendering graphics on graphics chips). |
|
2377)
Message boards :
News :
DENIS@home returns
(Message 85621)
Posted 2 Apr 2018 by Richard Haselgrove Post: It's just https://denis.usj.esHmmm. Found it in an fresh 14-day download today, and that says it's http://denis.usj.es/denisathome/, complete with keywords 9 1003 1012 In for a penny, in for a pound - I'll try attaching. Well, that gave me a clean attach, no error messages - but no tasks available, either. Edit - looks like the proper way to get the current XML version is to visit https://boinc.berkeley.edu/project_list.php, not any other of the urls I've tried in the past (although they all generated/displayed XML of various vintages) |
|
2378)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85607)
Posted 2 Apr 2018 by Richard Haselgrove Post: Yes, I wondered if the two might be connected. I got a peek at the SSP about 30 minutes ago (possibly linked to trying by http, rather than https). We're low on tasks (below 300K), and creation rate is very low, but we might hold up until they get to the lab. If today is a working day in the USA? |
|
2379)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85604)
Posted 2 Apr 2018 by Richard Haselgrove Post: Well, I got the home pages (after two hours), but then on to 'Establishing secure connection..' and 'Server took too long to respond'. Data transactions seem to be working OK, though. |
|
2380)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85585)
Posted 1 Apr 2018 by Richard Haselgrove Post: The strange thing today was that SETI was working fine when I got up, around midnight Berkeley time. Then, we couldn't get any work from around 3.30am to 6.30am: work started to flow, but I couldn't access the message boards. My browser said the delay was while it was establishing a secure connection, long before it tried to retrieve any data from the database. So, I'm speculating about infrastructure maintenance, rather than SETI servers? I haven't looked yet. |
|
2381)
Message boards :
BOINC client :
Thread Title Change Suggestion - done
(Message 85543)
Posted 29 Mar 2018 by Richard Haselgrove Post: Thanks - done. |
|
2382)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85523)
Posted 28 Mar 2018 by Richard Haselgrove Post: I expect a riot. :-)I don't. They're all too busy clocking their retry buttons for anything active like that! |
|
2383)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85437)
Posted 27 Mar 2018 by Richard Haselgrove Post: Stand aside from the doors, here they come. |
|
2384)
Message boards :
The Lounge :
These may not be with us long.
(Message 85413)
Posted 24 Mar 2018 by Richard Haselgrove Post: That's a lovely photo, Richard :) Thank you for posting it.Yes, I'm very pleased with it - the sense of movement and purpose, the impression I'm hiding in the bushes given by the depth of field on the leaves in the left foreground... This is a more honest account of the situation.  Somewhere, there's another one with the armed ranger who was standing not far away, watchfully, just in case. But I haven't got that one online. While I have the archive open:  Italy, I think. |
|
2385)
Message boards :
The Lounge :
These may not be with us long.
(Message 85407)
Posted 24 Mar 2018 by Richard Haselgrove Post: To mourn, but also to celebrate, Sudan's life.  That's a photograph I took in Kenya in 1986, and have framed on my wall. It's almost certainly not Sudan himself, but one can represent all of the species. |
|
2386)
Message boards :
GPUs :
GPU + Core usage
(Message 85399)
Posted 23 Mar 2018 by Richard Haselgrove Post: 3/22/2018 2:36:07 PM | | OS: Microsoft Windows 10: Core x64 Edition, (10.00.16299.00)You probably lost your NVidia driver during a Windows 10 update. Download and install a new one direct from nvidia.com |
|
2387)
Message boards :
Questions and problems :
Can't uninstall or update Boinc (Windows 7)
(Message 85393)
Posted 23 Mar 2018 by Richard Haselgrove Post: This message appears if you try to install the 64-bit version of BOINC on a machine running the 32-bit version of Windows 7. Try downloading boinc_7.8.3_windows_intelx86_vbox.exe instead. |
|
2388)
Message boards :
BOINC Manager :
Boinc opens the pages of the projects incorrectly
(Message 85392)
Posted 23 Mar 2018 by Richard Haselgrove Post: Repeating that test takes me to the url http://milkyway.cs.rpi.edu/milkyway/, and the full text appears in my browser. Could you please try clicking that link, and then separately clicking https://milkyway.cs.rpi.edu/milkyway/, and letting us know what happens? I'm just wanting to check whether this might be a http / https problem with Firefox. |
|
2389)
Message boards :
News :
DENIS@home returns
(Message 85388)
Posted 23 Mar 2018 by Richard Haselgrove Post: There is an option in Client configuration (cc_config.xml) which says:(unless you've politely declined to pester Google every time your internet connection goes down).What does it means? <dont_contact_ref_site>0|1</dont_contact_ref_site>That procedure was written into BOINC many years ago, when dial-up internet connections were unreliable. I now have a robust fibre-optic internet connection, and I don't need to rely on BOINC to tell me about the rare occasions when it goes down - I usually know about the problem before BOINC does. So, I choose not to let BOINC contact Google for this purpose. I did ask recently why these two seemingly unrelated features were combined under a single configuration option, and I was told that it was for the benefit of large corporate users of BOINC (or individual users of BOINC within a large corporate environment), where the internet traffic management of the organisation wish to lock down all internet access to 'authorised traffic/destinations only'. I'm not sure I'm convinced by that argument, but it's not worth going to the barricades over. Back to the original point - I was wrong. DENIS@Home won't re-appear in the 'attach to project' wizard quite yet - it's only been partially reinstated to just one of the BOINC lists. Users will need to look up the manual-attach url via https://boinc.berkeley.edu/projects.php - which turns out to be difficult, because the project itself tells you to use the 'attach' wizard, and doesn't display the url directly. I pulled this out of the home page: <!--<scheduler>https://denis.usj.es/denisathome_cgi/cgi</scheduler> --> <link rel="boinc_scheduler" href="https://denis.usj.es/denisathome_cgi/cgi">but I can't find a master url anywhere. You might have to ask at their forums: http://denis.usj.es/denisathome/forum_index.php |
|
2390)
Message boards :
News :
DENIS@home returns
(Message 85372)
Posted 22 Mar 2018 by Richard Haselgrove Post: Great news. Please, insert it in the list of available projects in boinc manager.It's on https://boinc.berkeley.edu/projects.php already: your computer should pick it up automatically within two weeks (unless you've politely declined to pester Google every time your internet connection goes down). |
|
2391)
Message boards :
The Lounge :
BOINC Politics thread
(Message 85357)
Posted 20 Mar 2018 by Richard Haselgrove Post: Mind you America is just as bad as here but with a big difference - There they massacre their kids in schools.This is the bit I don't get. What have we all got against kids? Over there, they shoot them in the name of freedom. Over here, they b*gger them in the name of sport. |
|
2392)
Message boards :
Web interfaces :
account manager re-attach bug: sync not picking up resource values
(Message 85325)
Posted 20 Mar 2018 by Richard Haselgrove Post: I think this is going to be a difficult one to resolve. I think we probably need to get someone to set up an Account Manager account (BAM! is probably better for this purpose than Gridcoin, because Willy has spent more time working with the BOINC developers to ensure that his BAM! works according to the specification). Make the account, record exactly how it's set up, request attach/synchronise from a BOINC client, capture and preserve the file the AM sends in reply with the settings, and see whether the Client gets settings that match what was set up in the AM. Then, do it all again under this re-attach / sync situation. Verify the AM account, attach, request sync, capture and preserve the new settings file, and see whether the client ends up with the wrong settings. Then, we can compare what happened at the various stages: did the AM send a faulty file, did the BOINC client mis-read the file, did something else interfere along the way? Then we can work out which part of the complex chain of events is responsible for any error. It's actually quite important to get to the bottom of this, because a new AM is being prepared for use: we need to be sure that the client is handling things properly so that we can test the new AM thoroughly. |
|
2393)
Message boards :
BOINC Manager :
Boinc Manager Suggestions
(Message 85289)
Posted 17 Mar 2018 by Richard Haselgrove Post: But I also don't think that the client can be made to switch between upload servers if the storage server is at fault.Interesting question. The client can certainly switch between *download* servers when the project specifies them in advance: Einstein does that, with <download_url>http://einstein2.aei.uni-hannover.de/download/hsgamma_FGRPB1G_1.20_windows_x86_64__FGRPopencl1K-nvidia.exe</download_url>
<download_url>http://einstein-dl.syr.edu/download/hsgamma_FGRPB1G_1.20_windows_x86_64__FGRPopencl1K-nvidia.exe</download_url>
<download_url>http://einstein-dl.phys.uwm.edu/download/hsgamma_FGRPB1G_1.20_windows_x86_64__FGRPopencl1K-nvidia.exe</download_url>- but they only do it with application files which can be mirrored around the world and kept unchanged for a long time and multiple downloads. They don't do it for single-task data files, and I don't know if the client would support it for uploads.CPDN does use different upload servers for different research types - Pacific North West study results go directly to Oregon, Australian study results go directly to Australia - but only to a single server in each region. It's not under user control. This sounds more like a server solution where the storage should warn when it's got just 15-20% storage left, that a human comes look. Saw Andy's answer on the CPDN list, where he said he wasn't aware of any space issues? That means their storage doesn't warn when it's reaching capacity.And did you see David's solution in c0c6cf7 - ah, those long-ago, far-away, innocent days when an upload server could run until it had 1 MB of free disk space? That brings timing into play as well - even if the server has 101 MB free at the start of your 100 MB upload, somebody else might have sneaked in and filled 35 MB of it while you were uploading. With regard to Andy's comment, Sarah came back the next day with "We have cleared off some space and intend to clear more in the coming days." - so it wasn't a false negative. In the meantime, a single user had uploaded 1.7 GB - the current 1 MB test is clearly inadequate in the modern era, and the projects which use huge network storage facilities need to implement some kind of early warning system. Perhaps we can invite Andy and Laurence to compare notes on Tuesday? |
|
2394)
Message boards :
BOINC Manager :
Boinc Manager Suggestions
(Message 85287)
Posted 17 Mar 2018 by Richard Haselgrove Post: Researching that old solution, I found http://boinc.berkeley.edu/trac/ticket/133: this problem shouldn't be occurring if it's the actual upload server which is out of disk space. That's why I phrased my upstream report in terms of transfers to a backing store. The ticket I was actually looking for is http://boinc.berkeley.edu/trac/ticket/139 |
|
2395)
Message boards :
BOINC Manager :
Boinc Manager Suggestions
(Message 85286)
Posted 17 Mar 2018 by Richard Haselgrove Post: I'm about sure that you can select tasks in the Transfers tab and press for them the Retry Now button. Only the selected tasks will then try the upload.Unfortunately, not quite. There are two sorts of transfer backoffs. One which shows an individual delay for an specific file until the next retry: the other is a project-wide backoff which applies to all files from that project. Selecting a single file for retrying will indeed reset the individual delay for that file, but it will also clear the project-wide delay, and other files whose individual delay since the last attempt has expired may be retried first. This means that you can't cherrypick the smallest file for the attempt. Also, the project-wide backoff isn't reapplied until three attempts have failed. Since the default setting is for two uploads to be tried concurrently, a single click on the 'retry' button will upload four files - up to 280 MB on Darmok's figures - under these conditions. This problem has plagued CPDN for many, many years. One of their project moderators - Thyme Lawn - developed code to apply 'suspend network activity' on a project-by-project basis, and took it as far as an experimental client/manager combination, which I tested. This was before the days of separate sub-projects uploading to different servers, so it might not be ideal now: but in any event, the concept was rejected by the core BOINC developers and abandoned. Oh, and the 'retry now' button is disabled until at least one file has been selected. |
|
2396)
Message boards :
The Lounge :
The "I don't want to be stuck in a time loop" thread
(Message 85279)
Posted 16 Mar 2018 by Richard Haselgrove Post: Würst? |
|
2397)
Message boards :
The Lounge :
BOINC Politics thread
(Message 85245)
Posted 15 Mar 2018 by Richard Haselgrove Post: Yea you're probably right bout the KKK but sure I saw a doco on the beeb maybe last year or the year before where a reporter " went inside " a " klan " meeting and seamed to me the KKK was fairly " into " the Bible , think it was on BBC 3 but can't be 100% sure where I saw this program or it's nameThere was a programme on BBC2 (repeated on BBC4) in association with the Open University - fronted by Neil Oliver. Scotland and the Klan: Scotland has exported many great things to the rest of the world, and people like Neil Oliver have often celebrated the disproportionate impact of its ideas and energy on places like America. The role of Scots in shaping the concept of the American Dream is a story often told, but could Scottish settlers have also had a hand in America's racist nightmare?The full programme is not currently available - only a couple of clips are available - but I remember Neil visiting the birthplace of the KKK and being shown the original commemorative plaque with the names of the six founders. Edit - found I've still got it on my digibox. According to the plaque, the KKK was founded on 24 December, 1865, in the law office of Judge Thomas M. Jones. |
|
2398)
Message boards :
BOINC Manager :
Boinc Manager Suggestions
(Message 85234)
Posted 15 Mar 2018 by Richard Haselgrove Post: The particular CPDN project failure which may have prompted this thread has been resolved for the time being, and uploads are being accepted again - although I don't know how quickly they're being accepted, or how big the queued backlog became before it was released. But it has happened to them before, and will likely happen to them again, so we still need to have some way of handling it automatically. |
|
2399)
Message boards :
The Lounge :
Old relics
(Message 85230)
Posted 15 Mar 2018 by Richard Haselgrove Post: I'll stop when it suits me not when it suits you.How does that chime with SPOT ON 100% !!! When in Rome you do as the Romans do. |
|
2400)
Message boards :
The Lounge :
BOINC Politics thread
(Message 85224)
Posted 15 Mar 2018 by Richard Haselgrove Post: And in my experience, demonstrably false.In the UK all Asian communities refuse to integrate into British life, and make no effort to learn the language. They are quite happy making whole cities and metropolitan areaa no go areas for white people. FGM, arranged marriages, Hala, Kosher, boiling dogs and cats, no rights for women, it all goes on. They are forced to apply for postal posts so their religious leaders can manipulate the vote. The Government won't stop it because they know their will be race riots on the streets of Britain,Have to say that is one of the most ante Asian post I think I have ever seen. Strong echoes of the vile "Britain First" party. I've lived and worked around one of the UK's most stereotypically 'Asian' cities - Bradford - for over 40 years. For 20 of those years, I worked as a freelance computer consultant, and did work inside several Asian businesses - solicitors, spice warehouse, bakery, mobile phone warehouse, disability charity. I don't speak any Asian languages: they all spoke my language. There were no communication difficulties at all. Although I was mainly contracted by 'middle class' owners and managers (maybe this is the perspective missing from the quoted post), I also had to work with spice rack loaders, bakers, and other working-class shop floor workers while debugging the programs I'd written for them. Again, no lack of English at that level either, although they - unlike me - were multilingual. Asian communities, like all communities, are mixed and varied. They contain the greedy and the generous, the good, the bad, and the ugly. When those individual quirks and differences are swept up in the generalisation "All Asians ...", there's only one word for it: racism. |
|
2401)
Message boards :
BOINC Manager :
Boinc Manager Suggestions
(Message 85222)
Posted 15 Mar 2018 by Richard Haselgrove Post: Welcome Darmok, and thanks for your suggestion. In fact, I don't think the situation is quite as bad as you suggest. We went through the 'multiple upload server' problem in some detail a few years ago, and in most cases, the current solution works well enough. Usually, when an upload servers fails, is unreachable, or is shut down for maintenance, your BOINC client gets that message immediately at the very beginning of the transaction: a few hundred bytes are exchanged in header negotiation, or a few dozen seconds are lost waiting for a timeout, but no major damage is done. If an upload fails for any reason, BOINC waits before retrying. At first, the waits are short, so that transient glitches are resolved quickly. If the retries fail consistently, the wait between retries is increased, and in addition, the whole project is inhibited from uploading until some event triggers a project-wide retry. To cope with the multiple-server situation, every new upload file generated by the project is tried once: maybe this one will be destined for a working server, and will go through at once. But if not, it joins the queue waiting to upload. The advantage of this system if that if a new upload intended for the failed server does get through (i.e. the project administrators have fixed the server), BOINC releases the project-wide inhibition, and all uploads intended for the broken server are retried automatically, with no action required at your end. There is, however, one situation where bandwidth is wasted as you describe - and this is one which is most commonly used by projects which have very large upload files to manage, thus compounding the problem. My understanding is that this happens when a project sets up a small 'gateway' upload server to handle the incoming files from users, but then transfers them to a much larger background data storage facility. If the background storage facility fills up or otherwise fails, the gateway server doesn't find out about it until the upload is complete and the gateway server attempts to transfer the entire file in one go. My understanding is that one project - Climate Prediction (CPDN) - is suffering this sort of problem at the moment. Maybe this is what prompted your suggestion? The project administrators are aware of the problem and are investigating. But it would be better if this case was handled by BOINC in the same way as I described above - with the gateway server rejecting the upload right at the beginning if the backing store is not ready to accept it. I think this would be better than adding a control to the manager for individual users: not everyone monitors their BOINC installation as closely as you obviously do, and almost everyone takes time off to sleep most days! More seriously, there's a risk that someone would turn off the uploads during a problem, and then forget to turn them back on again in a timely fashion after the problem is fixed. I do understand that the problem you've observed is frustrating, and for some users with metered bandwidth may even incur unexpected expense. I'll pass on your concern both to the individual project I've identified, and to the wider BOINC development team, to see what can be done. Edit - reported as Github issue #2411 |
|
2402)
Message boards :
Projects :
News on Project Outages
(Message 85207)
Posted 14 Mar 2018 by Richard Haselgrove Post: Maybe the baton has been passed. Einstein seems to be OK again now. |
|
2403)
Message boards :
Questions and problems :
BOINC "forgets" its stats
(Message 85195)
Posted 14 Mar 2018 by Richard Haselgrove Post: Where does BOINC keep its stats?Data folder, separate file 'statistics_url.xml' for each project url. Only one file is kept for each project, unlike client_state.xml which is protected by a file, _prev, _next rotation. A crash while the file is open for writing is likely to lead to corruption and data loss, so daily backups just after midnight are the only solution. |
|
2404)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85134)
Posted 13 Mar 2018 by Richard Haselgrove Post: Yes, just noticed that. I expected an hour earlier because of DST, but something like 2:20? And fixing an error message elsewhere on the website last week has (unnecessarily, IMHO) deprived us of the drying paint on the SSP during maintenance. |
|
2405)
Message boards :
The Lounge :
BOINC Politics thread
(Message 85121)
Posted 12 Mar 2018 by Richard Haselgrove Post: So unlike our own Nick Timothy. |
|
2406)
Message boards :
BOINC client :
Stats interval
(Message 85118)
Posted 11 Mar 2018 by Richard Haselgrove Post: You mean <save_stats_days>365</save_stats_days>in cc_config.xml If you're using a current copy of BOINC, it may have made a template file for you - that's in alphabetic order, and S for stats is below the out-of-format P section. It's fooled people before now. |
|
2407)
Message boards :
Questions and problems :
100% GPU load after BOINC is closed.
(Message 85100)
Posted 10 Mar 2018 by Richard Haselgrove Post: Windows Task Manager also show GPU usage but for me it's using non-compute engine and I can't find a way to change it.For me, I found evidence of activity like this, on the Performance tab:  Note that each of the four mini-graphs has a drop-down chooser above left: this shows activity for 3D and Copy, which I think was for an OpenCL task. It's not easy to match the application with the activity in Task Manager, though. |
|
2408)
Message boards :
The Lounge :
These are no longer with us, may they have peace.
(Message 85080)
Posted 9 Mar 2018 by Richard Haselgrove Post: Sir John Sulston, Nobel Prize winner and leader of British research into the human genome. He is being described by the BBC as a staunch socialist, who ensured that the results of his research were openly available to all humanity, and were not hidden for the private profit of pharmaceutical companies. |
|
2409)
Message boards :
Questions and problems :
100% GPU load after BOINC is closed.
(Message 85078)
Posted 9 Mar 2018 by Richard Haselgrove Post: OK, I think we can assume (for the time being) that it isn't the science applications which are causing the problem. The next thing to look at would be - when does the GPU usage start to ramp up? The only thing which BOINC itself does to/with a GPU is to query at startup what GPUs are fitted and what their specifications - memory, speed, driver version, supported languages - are. We did have a similar problem some years ago (but I think it was only present in macbooks fitted with dual GPUs - one low-power to prolong battery life on the road, the other high-powered for desktop power users). In that case, the 'capability test' triggered the high-power GPU to fire up and it wouldn't let go again. If you could test that, please? Suspend all projects (or finish all worked currently on the machine and set 'no new tasks' so nothing new is downloaded), then do whatever is needed (reboot, log off/on) to ensure the GPU shows as idle. Then, start the BOINC Client and the BOINC Manager (separately if you can), and watch to see if the GPU usage spikes at that stage - and if so, how long for. Remember to un-suspend the projects, or to fetch new work, when you're done. |
|
2410)
Message boards :
The Lounge :
Why am I doing this?
(Message 85071)
Posted 9 Mar 2018 by Richard Haselgrove Post: And David is still actively working on Nebula. He wrote this three days ago (in the repository, not on the web site): Eliminate separate program versions for SERENDIPAs you noticed (you have a 'special fundraiser' icon), SETI is preparing to add Parkes (southern hemisphere) as a data source. That's never been searched before. |
|
2411)
Message boards :
Questions and problems :
100% GPU load after BOINC is closed.
(Message 85070)
Posted 9 Mar 2018 by Richard Haselgrove Post: It would be worth looking at BOINC Manager just before you shut down, to see which project's science application(s) are running on your GPU when this happens. Then, after shutting down BOINC, use Task Manager (or Process Explorer) to see whether that particular science app is still in the task list. I know that at least one of the SETI@Home science apps for GPU does take an appreciable amount of time to notice that BOINC has requested a closedown (although my experience is with NVidia cards, not AMD). But once both BOINC and every related project application have closed and vanished from the list of running processes, there would be nothing left under our control to keep the GPU active. If a project science app does remain active for more than a minute (and that's being generous) after BOINC itself closes, you should take it up directly with the project concerned. |
|
2412)
Message boards :
The Lounge :
Future Boinc thoughts
(Message 85054)
Posted 8 Mar 2018 by Richard Haselgrove Post: I think it's unfair to categorise BOINC's outputs purely in terms of volunteers' personal scoring, but other research programmes in terms of scientific benefit. Only last week Einstein reported new-to-science discoveries from NASA's Fermi Gamma-ray Space Telescope mission: this is a massively funded international collaboration, and the data has been extensively researched already, but it still took the distributed computing approach to find these results. In the field of life sciences and medical research, I've personally contributed to 20 published scientific research papers since 2012 via just one BOINC project - GPUGrid (and there'll be more in the pipeline - scientific publishing is a slow business). |
|
2413)
Message boards :
Questions and problems :
can't run boinc in azure: /user/slots directory does not exist
(Message 85020)
Posted 6 Mar 2018 by Richard Haselgrove Post: In which case, you'll need to ensure that the secure sandbox account that BOINC is running under has write access to /home/wally Or, you need to find where the sandboxed account is using for its data folder (and storing the *real* event log messages): grant yourself read access to that folder. |
|
2414)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85017)
Posted 6 Mar 2018 by Richard Haselgrove Post: The usual starting time is a few minutes after 13:00 UTC each Tuesday. For the next couple of weeks, it is likely to be 12:00+ UTC because of Daylight Saving Time in the USA, before reverting to 13:00+ BST a couple of weeks later. Very rarely, the day is switched to Wednesday because of public holidays or other events. These times are not guaranteed, and variations are not usually signalled in advance - they depend on who is available and what needs to be done on the day. You are right that it takes longer than you remember, because the size of the things to be maintained - the databases - has grown much larger. We are routinely processing 120,000 tasks per hour (approaching 3 million a day) - we used to do maybe a third of that number. All thanks to returning volunteers like you, new recruits, more available work from Breakthrough Listen, and more powerful consumer hardware. Maintenance usually lasts at least 10 hours (start looking for signs of life around 23:00), but can take longer as the server hardware wears out and components (especially hard disk drives) need to be replaced. |
|
2415)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 85010)
Posted 6 Mar 2018 by Richard Haselgrove Post: The dreary outrage has begin once again.And unusually, this time they've disabled the upload and download servers as well. Maybe they have hardware to fix, maybe they have backlogs to clear. |
|
2416)
Message boards :
GPUs :
CUDA Toolkit
(Message 85008)
Posted 6 Mar 2018 by Richard Haselgrove Post: To run NVidia applications, you need two things: 1) A suitable hardware driver for your GPU. It's the user's responsibility to obtain and install this. 2) The runtime support files for the programming language used by the application programmer. It's usually the programmer's (or in our case, the BOINC project's) responsibility to supply these. CUDA is NVidia's proprietary programming language, and the toolkit contains vastly more than is needed simply to run compiled programs. Programmers need it, ordinary users don't. The toolkit does, however, contain everything necessary for a programmer to get started. So, there's a driver in there, and all the necessary library files. Downloading the toolkit is a way of getting hold of any missing components that the programmer has failed to pass on to you. I don't know the file names in the Linux context, but for Windows the ones you are most likely to need are: cudartxx_yy.dll cufftxx_yy.dll where xx is the bitness (32 or 64), and yy is the version number. You need the version which exactly matches what your programmer developed for. |
|
2417)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 84993)
Posted 2 Mar 2018 by Richard Haselgrove Post: Poking about, there might be a viable 64-bit file in /dl/BOINCx64_vbox/WEBx64/Web/DiskImages/DISK1 (right name and size) - I've only just discovered it and I need a break, so I haven't tested it. Have fun. Edit - yes, the file in DISK1 is viable and installs VBox v5.2.8 as it should |
|
2418)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 84991)
Posted 2 Mar 2018 by Richard Haselgrove Post: The 32-bit installer on a 32-bit version of Windows 7 seems to behave as expected from what we've seen over the last couple of days. David has refreshed the build, and it now includes VBox 5.2.8: that goes through all the VBox installation steps (including USB and networking), and shows up in the list of installed programs in Control Panel: I have BOINC configured not to use it (because this machine - at the full 4 GB - barely has enough RAM to do the things I ask of it), so it's not yet showing in the BOINC Event Log. David was going to refresh the 64-Bit version as well, but something went wrong: the 19 Feb build with broken VBox is still there. I'm going to start putting together a cherrypick PR so we can test everything in place as a v7.9.3 build. The 32-bit BOINC on 32-bit OS combo reports: 02/03/2018 10:26:55 | | Starting BOINC client version 7.9.2 for windows_intelx86 02/03/2018 10:26:59 | | OS: Microsoft Windows 7: Ultimate x86 Edition, Service Pack 1, (06.01.7601.00)David has put together a patch (awaiting approval), so that the first 'Starting client' line reports the build version rather than the environment version: I'll try to get that into v7.9.3 Edit - VBox detection is OK: 02/03/2018 10:56:44 | | VirtualBox version: 5.2.8 |
|
2419)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84986)
Posted 1 Mar 2018 by Richard Haselgrove Post: I remember that one. I wasn't in Bradford then: we lived in Manchester at the time, but I was staying with my grandparents when it started. They'd retired to a Cotswold village, and I can still see one of the access roads: a sunken road, flanked by high hedges, completely filled with a U-shaped showdrift with a single-track trampled line of footprints along the middle.Not as bad as 1963, couldn't get to the Poly, Bradford district was gripped by weather in March 1963Aren't you the "lucky" one....Actually, it can be quite fun once in a while. I have a low wall separating the yard from an open field: plenty of space for disposing of the fallout. If you catch it while it's fresh, soft and fluffy, over the wall it goes with the minimum of fuss. Even better if you can time it for the sunshine between the snow showers... |
|
2420)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 84981)
Posted 28 Feb 2018 by Richard Haselgrove Post: Just done exactly the same thing. With a deliberate 32-bit installation, I get 28/02/2018 19:04:12 | | Starting BOINC client version 7.9.2 for windows_x86_64 28/02/2018 19:04:12 | | OS: Microsoft Windows 10: Professional x64 Edition, (10.00.16299.00) 28/02/2018 19:04:12 | | Can't detect VirtualBox because this is a 32-bit version of BOINC; to fix, please install a 64-bit version.So I read it wrongly too. If it's confused all of us, I'd suggest it's a bad message. |
|
2421)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84977)
Posted 28 Feb 2018 by Richard Haselgrove Post: Aren't you the "lucky" one....Actually, it can be quite fun once in a while. I have a low wall separating the yard from an open field: plenty of space for disposing of the fallout. If you catch it while it's fresh, soft and fluffy, over the wall it goes with the minimum of fuss. Even better if you can time it for the sunshine between the snow showers... |
|
2422)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84972)
Posted 28 Feb 2018 by Richard Haselgrove Post: Send it over to blighty, we can then have a good moan about the wrong type of snow on the tracks.About 5 inches on my yard in the last 24 hours, judging by what had arrived since I shovelled it yesterday. And as soon as I finish today's clearance, it starts filling up again. |
|
2423)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 84969)
Posted 28 Feb 2018 by Richard Haselgrove Post: David says that the SETI server has been fixed to run the auto-attach test, and it's working for me with the plain-vanilla BOINC installer under Windows 10. The BOINC+VBox combined installer seems to have multiple problems, and gets an error message when attempting to attach: I've reported that back upstream, but clearly more testing is needed. I'm pondering that strange 32-bit message for VBox: my W10 log has 28-Feb-2018 11:32:25 [---] Starting BOINC client version 7.9.2 for windows_x86_64 28-Feb-2018 11:32:26 [---] OS: Microsoft Windows 10: Professional x64 Edition, (10.00.16299.00) 28-Feb-2018 11:32:26 [---] VirtualBox version: 5.1.26 as it should - that was an old install from previous testing, though it should have been updated to VBox 5.2.6 as part of my testing. Edit - found it. if (!strcmp(get_primary_platform(), "windows_x86_64")) {
msg_printf(NULL, MSG_USER_ALERT,
"Can't detect VirtualBox because this is a 32-bit version of BOINC; to fix, please install a 64-bit version."The primary platform "OS: Microsoft Windows 8.1: Core x64 Edition, (06.03.9600.00)" fails that test. I'll open an issue.
|
|
2424)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 84950)
Posted 27 Feb 2018 by Richard Haselgrove Post: At the moment, I'm getting Fatal error: Call to a member function update() on a non-object in /disks/carolyn/b/home/boincadm/projects/sah/html/inc/account.inc on line 28from SETI - probably because their server is still in maintenance mode. When I tried it yesterday, even the ?dev=1 modifier on the url gave me the current v7.8.3 BOINC installer - which won't perform the auto-attach trick. Best to wait on the SETI version of this test until the installation there has been checked - we'll post here when it's good to go. |
|
2425)
Message boards :
Questions and problems :
BOINC freezes Raspberry Pi
(Message 84901)
Posted 25 Feb 2018 by Richard Haselgrove Post: What's your thermal solution? Do you have a heatsink? Do you have a fan on (or directed at) the heatsink? What's the ambient temperature in your environment? |
|
2426)
Message boards :
Questions and problems :
Not requesting tasks: don't need (not highest priority project)
(Message 84855)
Posted 20 Feb 2018 by Richard Haselgrove Post: Er, boinctray.exe |
|
2427)
Message boards :
BOINC Manager :
Whenever I open the 'news' tab in boinc manager, my browser is opened instead at localhost
(Message 84823)
Posted 18 Feb 2018 by Richard Haselgrove Post: Which version of BOINC are you using, and which operating system? This sounds like a problem which was reported some months ago for Apple Mac, and fixed in the latest recommended v7.8.6 of BOINC. |
|
2428)
Message boards :
Questions and problems :
Not requesting tasks: don't need (not highest priority project)
(Message 84816)
Posted 17 Feb 2018 by Richard Haselgrove Post: Is there some reason Boinc would not detect that the computer was not in use for 30 minutes?Detecting 'computer in use' is usually described as interaction with the mouse or keyboard, but I think it will actually detect activity on any HID (Human Interface Device). There have been reports that if a high-sensitivity device like a gaming controller is connected to the computer that can generate false positive signals which get detected as 'in use'. |
|
2429)
Message boards :
The Lounge :
Science and Technology in the News
(Message 84785)
Posted 14 Feb 2018 by Richard Haselgrove Post: Meanwhile, Bitcoin energy use in Iceland set to overtake homes Iceland is facing an "exponential" rise in Bitcoin mining that is gobbling up power resources, a spokesman for Icelandic energy firm HS Orka has said.My opinion is that this craze has exceeded all reasonable limits. |
|
2430)
Message boards :
GPUs :
Question on installed GPU
(Message 84760)
Posted 13 Feb 2018 by Richard Haselgrove Post: I think that card should still be usable by SETI@Home, but don't expect blistering performance - it wasn't even state-of-the art (though not far behind) when GPU processing first became possible ten years ago. Gulp - that's not a phrase I expected to find myself typing! By all means attach to SETI, and see what happens. You'll have to be quick - SETI has regular maintenance on Tuesdays, starting in around 2 hours (13:00 UTC) - or patient: SETI should be accessible again around midnight UTC. Another thing to watch for: those early NVidia cards were quite power hungry - your electricity consumption will go up (and your room will become warmer). Your local climate will determine whether this is a good thing or a bad thing. |
|
2431)
Message boards :
GPUs :
Has anyone recompiled BOINC on Parallella to use the 18 cores available?
(Message 84756)
Posted 12 Feb 2018 by Richard Haselgrove Post: BOINC is a CPU-based 'command and control' program. Although OpenCL is available for CPUs, I think it's generally accepted that the added layers of complexity and runtime support would make it less efficient than the current native C++ code. Where OpenCL comes into its own is in programming GPUs, as you've correctly indicated by choosing this message board area for your question. GPU programming is the role of the science projects that run under BOINC, not BOINC itself. Parallellas were discussed when (and before) they were first launched, but IIRC the conclusion was that they could address too little memory per processing core for serious scientific processing to be attempted. Your best source for technical information and discussion is probably the thread PARALLELLA, RASPBERRY PI, FPGA & ALL THAT STUFF on the Einstein@Home message board - but be warned, there are over 900 posts spread over more than 5 years. |
|
2432)
Message boards :
BOINC Manager :
Building Boinc on windows 7/64bits 'Software not detected: Build tools...'
(Message 84751)
Posted 11 Feb 2018 by Richard Haselgrove Post: It seems like the code sub-directory name can be no longer than 13 characters. client_r_7-9-1 fails client__7-9-1 worksTested with command line only - it's quicker. The length of the top-level folder name seems not to matter. Updated documentation. |
|
2433)
Message boards :
Questions and problems :
Not requesting tasks: don't need (not highest priority project)
(Message 84746)
Posted 10 Feb 2018 by Richard Haselgrove Post: Doing a manual request on a single project may get 'not highest priority' because a different project has higher priority and is fetchable. If you allow all projects to fetch at will, then higher priority projects will be tried first. If they fail, then lower priority projects will be tried while the higher priority projects are inaccessible (server requested backoff, or suchlike). |
|
2434)
Message boards :
BOINC Manager :
Building Boinc on windows 7/64bits 'Software not detected: Build tools...'
(Message 84740)
Posted 10 Feb 2018 by Richard Haselgrove Post: Ah - cracked it. Path length limitation. Same folder renamed: D:\Sources_build\client_rel\win_build>buildenv type release platform x64
Initializing BOINC Build Environment for Windows
Software Platform Detected: Visual Studio 2013
D:\Sources_build\client_rel\win_build>build all
Building the World...
Building BOINC:
Success
Building BOINC Manager:
Success
Building BOINC Command:
Success
Building BOINC Screen Saver:
Success
Building BOINC Screen Saver Controller:
Success
Building BOINC Service Controller:
Success
Building BOINC System Tray:
Successand all the libraries are in the output folder.Using D:\Sources_build\boinc_7-9-1\win_build works too. |
|
2435)
Message boards :
BOINC Manager :
Building Boinc on windows 7/64bits 'Software not detected: Build tools...'
(Message 84739)
Posted 10 Feb 2018 by Richard Haselgrove Post: Tried the command-line variant too: D:\Sources_build\client_release_7-9-1\win_build>buildenv type release platform x64 Initializing BOINC Build Environment for Windows Software Platform Detected: Visual Studio 2013 Software NOT Detected: Build Tools... D:\Sources_build\client_release_7-9-1\win_build>build all 'build' is not recognized as an internal or external command, operable program or batch file. D:\Sources_build>fc D:\Sources_build\boinc_head\win_build\buildenv.cmd D:\Sources_build\client_release_7-9-1\win_build\buildenv.cmd Comparing files D:\SOURCES_BUILD\BOINC_HEAD\WIN_BUILD\buildenv.cmd and D:\SOURCES_BUILD\CLIENT_RELEASE_7-9-1\WIN_BUILD\BUILDENV.CMD FC: no differences encountered |
|
2436)
Message boards :
BOINC Manager :
Building Boinc on windows 7/64bits 'Software not detected: Build tools...'
(Message 84738)
Posted 10 Feb 2018 by Richard Haselgrove Post: Now I'm getting the same message when trying to compile the test v7.9.1 branch under VS2013 Comparing the output from the two builds, my previous 'head' build or v7.8.1 build (both of which still work) finishes 2> Generating Code... 2> boinc_cli_vs2013.vcxproj -> D:\Sources_build\boinc_7-8-1\win_build\.\Build\x64\Release\boinc.exe 2> Initializing BOINC Build Environment for Windows 2> Software Platform Detected: Visual Studio 2013 2> Executing: updatedepends.cmd 2> 2> sqlite3.dll: 3.8.1.0 3.8.1.0 2> libcurld.dll: 2> libcurl.dll: 7.42.1.0 7.42.1.0 2> ca-bundle.crt: - - 2> libeay32.dll: 1.0.2.1 1.0.2.1 2> ssleay32.dll: 1.0.2.1 1.0.2.1 2> zlib1d.dll: 2> zlib1.dll: 1.2.8.0 1.2.8.0 2> msvcp120d.dll: 2> msvcp120.dll: 12.0.21005.1 12.0.21005.1 2> msvcr120d.dll: 2> msvcr120.dll: 12.0.21005.1 12.0.21005.1whereas v7.9.1 finishes 2> Generating Code... 2> boinc_cli_vs2013.vcxproj -> D:\Sources_build\client_release_7-9-1\win_build\.\Build\x64\Release\boinc.exe 2> Initializing BOINC Build Environment for Windows 2> Software Platform Detected: Visual Studio 2013 2> Software NOT Detected: Build Tools... 2>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V120\Microsoft.CppCommon.targets(132,5): error MSB3073: The command ""D:\Sources_build\client_release_7-9-1\win_build\\buildenv.cmd" devenvdir "D:\Program Files (x86)\Common7\IDE\" type "Release" platform "x64" exec updatedepends.cmd 2>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V120\Microsoft.CppCommon.targets(132,5): error MSB3073: :VCEnd" exited with code 2.Stackoverflow suggests that this might be a copying error (confusion between folder and file name). The only other difference I can see between the builds is that v7.9.1 references a new file, project_list.cpp Looking through buildenv.cmd, the error message 'Software NOT Detected: Build Tools...' occurs if build.cmd doesn't exist in boinc_depends_win_vs2013 - but I'm using the same 'depends' folder for both build attempts, and it's there. Building boincmgr ends with the same error message. In both cases, the binary executable is created (I'll test them in the morning), but the normal set of DLLs are not copied to the release directory. Line 132 of Microsoft.CppCommon.targets is <Exec Command="%(PostBuildEvent.Command)$(_BuildSuffix)" Condition="'%(PostBuildEvent.Command)' != ''"/>I think I'm stuck there. |
|
2437)
Message boards :
The Lounge :
Needful Things
(Message 84730)
Posted 9 Feb 2018 by Richard Haselgrove Post: And didn't check the Wiki definition of the actual word used - which nowhere contains the e-word. |
|
2438)
Message boards :
The Lounge :
Needful Things
(Message 84728)
Posted 9 Feb 2018 by Richard Haselgrove Post: Settling down to watch Michael Portillo visiting the seismology department of UC Berkeley as part of his latest Great American Railroad Journey. |
|
2439)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 84721)
Posted 8 Feb 2018 by Richard Haselgrove Post: It would have been around 08:45 or 09:45 tomorrow morning, depending on when github counts 'n hours ago' from (I don't get self-notifications).Remind me when that 24 hours will be up?I'll probably forget so how about I remind you now :) I feel a cold coming on, so I'm going to dose myself with drowsy cough-medicine and take an early night: then I can branch and tag in peace tomorrow morning, without the americans peering over my shoulder. Thanks for clearing the merges. |
|
2440)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 84719)
Posted 8 Feb 2018 by Richard Haselgrove Post: We got the last functional pull request (#2260 - which you merged, thank you) into the code on Monday. I tipped the wink to Christian after our teleconference on Tuesday, and he put up a PR for the latest translations yesterday. I approved it this morning, but we're still awaiting merge. If it's still stuck after 24 hours, I'll break the rules and merge it myself - I seem to have that power now. That and Charlie's #2342. That's it. Branch (I wonder if I can do that now), tag (likewise), and build - which has to be David, unfortunately, because of the physical security of the code-signing machine. But if I tag, there's nothing to stop Charlie (Mac) and Laurence (Debian) starting work on their respective builds. Edit - yes, looks like I can both branch and tag. Remind me when that 24 hours will be up? |
|
2441)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 84717)
Posted 8 Feb 2018 by Richard Haselgrove Post: I think it's actually just been very quiet - there hasn't been any active testing recently, and preparation for the next round has mostly been carried out on Github or in private mail. The switchover to Google Groups was conducted by a technical support worker within the Space Science Laboratory at Berkeley, apparently without reference to BOINC core workers. Technical and administrative issues have been referred directly to the support worker concerned. I've sent out a test message and received a private reply from another of our (external) regular testers, so I think the list is usable for now. Which is a good thing, because we're about to start a new round of intensive testing - watch the usual spaces! |
|
2442)
Message boards :
Questions and problems :
Not requesting tasks: don't need (not highest priority project)
(Message 84672)
Posted 5 Feb 2018 by Richard Haselgrove Post: Every one of the 'not highest priority project' is from an 'update requested by user'. Try just letting the computer stand idle for 10 minutes - don't touch mouse or keyboard, don't run either of your exclusive apps - and see if it will ask CPDN (fails, offline), Malaria (dead project), and finally reach one of the live ones. |
|
2443)
Message boards :
Questions and problems :
BOINC and SSL intercepting IDS
(Message 84636)
Posted 2 Feb 2018 by Richard Haselgrove Post: Did you try the new version of the file? Is the CA you added a purely private one, or one we should add to the generic bundle for other people with the same problem? |
|
2444)
Message boards :
Questions and problems :
BOINC and SSL intercepting IDS
(Message 84630)
Posted 2 Feb 2018 by Richard Haselgrove Post: BOINC's SSL certificates are stored in a file called ca-bundle.crt, which you can find in the BOINC program directory. We are about to start using a new version of this file, which you can find in https://github.com/BOINC/boinc/tree/master/curl - it would be helpful if you could download this new file and see if it resolves your problem. If not, you should be able to add your own certificate using a plain-txt editor - despite the name, the file and the certificates within it are stored in simple ASCII format. |
|
2445)
Message boards :
BOINC Manager :
Building Boinc on windows 7/64bits 'Software not detected: Build tools...'
(Message 84628)
Posted 2 Feb 2018 by Richard Haselgrove Post: I use the VS2013 build system, as shown in Compile Client - Windows, and I'm also running under Windows 7/64: it's a couple of years since I set it all up, but I don't remember having any difficulty with the instructions there. You need both the BOINC source files and the VS2013 dependencies, expanded side-by-side in the same working directory. Then, double-clicking on the boinc_vs2013.sln file in win_build should launch the VS2013 development environment: select a sub-project in the right-hand pane (sounds like you want boincmgr), and then build from the main menu. I'm not familiar with the command-line approach, but I just tried following the instructions and it worked without error: D:\Sources_build\boinc_head\win_build>buildenv type release platform x64
Initializing BOINC Build Environment for Windows
Software Platform Detected: Visual Studio 2013
D:\Sources_build\boinc_head\win_build>build all
Building the World...
Building BOINC:
Success
Building BOINC Manager:
Success
Building BOINC Command:
Success
Building BOINC Screen Saver:
Success
Building BOINC Screen Saver Controller:
Success
Building BOINC Service Controller:
Success
Building BOINC System Tray:
Success
D:\Sources_build\boinc_head\win_build>You can see how I executed both 'buildenv' and 'build' from the win_build folder in my working tree.Please review all the steps you've taken so far, and post again if you can't get it to work. |
|
2446)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84594)
Posted 30 Jan 2018 by Richard Haselgrove Post: The 'other fact' being On Tuesday 2018 January 30th at 1430 UTC the Web site will be down for about 30 minutes. That is 1530 CET for Central Europe and 0930 EST for east coast North America.The site and work fetch is working fine for me in the meantime. |
|
2447)
Message boards :
Questions and problems :
Carolyn disk error message blocks my account now
(Message 84590)
Posted 29 Jan 2018 by Richard Haselgrove Post: The error message on the personal donation list page. |
|
2448)
Message boards :
Questions and problems :
Carolyn disk error message blocks my account now
(Message 84587)
Posted 29 Jan 2018 by Richard Haselgrove Post: Yes, I have that one too. |
|
2449)
Message boards :
BOINC client :
BOINC appears in home directory on Fedora
(Message 84582)
Posted 29 Jan 2018 by Richard Haselgrove Post: I've had a word with the Fedora package maintainer, and he directed me to the discussion in https://github.com/BOINC/boinc/issues/2262. I'm not a Linux person, but it sounds as if it should be configurable. The second user in that discussion seems to have resolved it by uninstalling BOINC and re-installing with v7.8.4 The issue is certainly on the radar - please let us know how you get on. |
|
2450)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84580)
Posted 29 Jan 2018 by Richard Haselgrove Post: Or it fell over of its own accord. It's back up now. |
|
2451)
Message boards :
BOINC client :
BOINC appears in home directory on Fedora
(Message 84568)
Posted 27 Jan 2018 by Richard Haselgrove Post: Are you aware of any specific event which might have triggered this change? Did you choose to update anything (either Fedora or BOINC), or are you aware of any automatic update around the same time? BOINC in general doesn't have any form of automatic update (although I can't speak specifically of the Fedora implementation). |
|
2452)
Message boards :
Questions and problems :
No way to run projects sequentially on host?
(Message 84567)
Posted 27 Jan 2018 by Richard Haselgrove Post: On the contrary, I feel that greater efficiency can sometimes be achieved by running different projects together. Some programs rely primarily on integer mathematics, others require floating point maths. Some programs are heavily optimised, making extensive use of SIMD (single instruction, multiple data) operations. Others, not so much. Some projects (exploring the effect of parameters on an algorthm) require very little data: others have to load large datasets. It's perhaps easiest to consider some hypothetical project which makes extremely heavy use of a single resource within the computing environment - it might be memory access, it might be the CPU's floating point unit. If multiple copies of that single project are running together, the stress will be multiplied. The memory bus may become saturated, the CPU may need thermal or power throttling. On the other hand, if that 'heavy' project is run alongside a 'lighter' project, say an integer maths algorithmic task, both may be able to run at full speed in their respective parts of the hardware. |
|
2453)
Message boards :
Questions and problems :
Carolyn disk error message blocks my account now
(Message 84542)
Posted 25 Jan 2018 by Richard Haselgrove Post: David was working on the special donation icon code last night - I got a notification at 00:17 UTC that new code had been committed to the SETI repository. That's round about the time these specific errors were first reported at SETI - I think Keith must just have caught David in the act. So yes please, Keith - retest now things have settled down and report. |
|
2454)
Message boards :
Questions and problems :
security hack? unauthorized gui_rpc attempt seconds after a new installation
(Message 84541)
Posted 25 Jan 2018 by Richard Haselgrove Post: You should also describe how this new computer is connected to the internet. Yes, IP addresses are used for GUI RPCs, just as they are used for RPCs to project servers. But an incoming call from France? That should be caught and blocked at least twice: once by the NAT translation in your router, and again by the firewall in your operating system. |
|
2455)
Message boards :
The Lounge :
BOINC Politics thread
(Message 84529)
Posted 24 Jan 2018 by Richard Haselgrove Post: I should think so to: The Presidents Club says it is to close after accusations of groping at a men-only charity dinner News coverage says that this dinner has been running annually for 33 years - i.e. since around 1985, the years of the Big Bang (financial deregulation), the Roaring Ferraris and the first upsprouting of Yuppies. I can't say I find that a coincidence. Their Charity Commission register entry shows some interesting overlaps. From their accounts: Annual event Income 2012 £141,754 2013 £264,389 2014 £694,250 2015 £1,103,855 2016 £1,590,139 All during the period of austerity after the banking collapse. |
|
2456)
Message boards :
Questions and problems :
Can't update BOINC client to 7.8.3. Keep getting Error 1325
(Message 84527)
Posted 24 Jan 2018 by Richard Haselgrove Post: Two questions: How far into the installation process do you get before it throws the error? What is the 'Data directory" location being used by your current v7.6.33? It's shown on about the fourth line of the Event Log after startup. C:\ProgramData (notice specifically drive C:) is the default location for BOINC - and many other programs - to use for working data. It's, by Microsoft default, a hidden folder, but it can be made visible via the Folder options control panel. If the problem happens very early in the installation procedure, the installer will be looking for the UN-install tool for the previous version. Problems at this point can often be solved by running a repair installation of your old v7.6.33 - the file is still available via http://boinc.berkeley.edu/download_all.php, if you've thrown your old copy away. After that, the upgrade should proceed normally. If the upgrade gets more than four or five steps in, you should have seen a screen telling you what the two separate installation locations will be, for programs and data. That screen has an 'Advanced' button in the lower-right corner: use that to choose locations which are valid on your computer. If the error message mentions a drive letter in addition to ProgramData, and if that drive letter is no longer valid on your machine, you'll have to get down and dirty with the Windows Registry Editor, regedit. Navigate to [HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup] and look at the key DATADIR. Mine reads "DATADIR"="D:\\BOINCdata\\" (in registry export format) - that's my choice. It would normally be "DATADIR"="C:\\ProgramData\\" Anything weird or not valid on your machine - correct it, or simply delete the key to restore the installation default. Check INSTALLDIR (the program directory) while you're there. |
|
2457)
Message boards :
Questions and problems :
Over-sized graphics?
(Message 84524)
Posted 24 Jan 2018 by Richard Haselgrove Post: I suggest you raise this on the Einstein message boards, if you haven't already. Their forum moderator Mike Hewson has himself played around with the graphics display program: I suspect the difference may be due to the different programming languages chosen by the programmers at the two independent projects. Mike may be able to advise better. |
|
2458)
Message boards :
The Lounge :
BOINC Politics thread
(Message 84522)
Posted 24 Jan 2018 by Richard Haselgrove Post: We've sometimes discussed men-only social and charitable clubs in this thread. The low-key local clubs no doubt do much good work without causing offense, but the top-most level of the pyramid has been exposed by London's Financial Times newspaper - not known for hype or fake news. Men Only: Inside the charity fundraiser where hostesses are put on show Great Ormond Street Hospital shuns scandal charity I have suggested to the BBC that the money raised should be donated instead to support rape crisis centres and women's refuges. |
|
2459)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84521)
Posted 24 Jan 2018 by Richard Haselgrove Post: I've just noticed that we passed a significant milestone five days ago. It's now less than 20 years until the end of Linux as we know it. The end of the Linux epoch That's also the date all our message board spammers come back to haunt us, which is why I noticed it! |
|
2460)
Message boards :
Questions and problems :
is there an exclude_cpu or ignore_cpu?
(Message 84481)
Posted 22 Jan 2018 by Richard Haselgrove Post: Any lost tasks would show on the project web site - they're only 'lost' from the client. They're also noted as resends in the event log. But yes - I do believe lost tasks bypass the "don't use..." selection. The wording has possibly changed over the years - the checkbox is labelled "Use CPU" at projects using the most recent version of BOINC. You should also see "Don't request tasks for CPU" in the project properties in BOINC Manager if that option is set on the project website.[That's using BOINC Manager - your screenshot seems to come from a different tool] And as an added wrinkle, when MilkyWay first added mt tasks (known as n-body) to their website, some years ago, they classed them as GPU tasks... |
|
2461)
Message boards :
Questions and problems :
BOINC 7.8.3 is not picking up the avx, avx2 flags in Ubuntu 16.04.3
(Message 84459)
Posted 21 Jan 2018 by Richard Haselgrove Post: Good thinking. My Windows machine gets fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 vmx smx tm2 pbe fsgsbase bmi1 smep bmi2into under 250 characters. |
|
2462)
Message boards :
Questions and problems :
BOINC 7.8.3 is not picking up the avx, avx2 flags in Ubuntu 16.04.3
(Message 84447)
Posted 20 Jan 2018 by Richard Haselgrove Post: I'd double-check in stdoutdae.txt that BoincTasks isn't truncating that feature string - although it's suggestive that the list ends with pni (aka SSE3). It's helpful that one of the administrators of LHC is becoming involved in maintaining the Linux distros - that might give you some traction. |
|
2463)
Message boards :
Projects :
News on Project Outages
(Message 84429)
Posted 18 Jan 2018 by Richard Haselgrove Post: Two new project outages: 1) Climate Prediction (CPDN): "We are upgrading the operating system on our Dev site server and we will take the dev site down from 14:00 UK time [UTC] today [Thursday 18 January]". This probably won't affect ordinary users, but may affect new application testers. 2) Einstein @ Home "We are going to shut down the project next Tuesday, Jan 23rd at around 10 AM CET for an upgrade of our database backend systems to make them ready for the years to come. We're going to upgrade hardware parts, operating systems as well the databases themselves, which is why we need to shut down the entire project, including the BOINC backend and this very website." |
|
2464)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84415)
Posted 16 Jan 2018 by Richard Haselgrove Post: Tap, tap.As I guessed, Eric is transferring the new server code recently tested at Beta (I spoke with him this evening). That requires an extra step of adding new fields into the BOINC database, which adds extra time. Wrap up warm. |
|
2465)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84412)
Posted 16 Jan 2018 by Richard Haselgrove Post: My it is awful quiet in the cafe... Wondering if everyone on both sides of the Atlantic are frozen in their tracks due to the weather?Very much so. 1°C here (positive!), but an inch of snow fell early evening, and nothing/nobody has moved since then. I actually drove down to the pub - not a problem if you take it gently on the hill - and found I was the only customer. No sign of any gritting in advance, although one gritter passed just as I turned in. Nothing on the way back either, but a safe arrival and a cosy end to the evening. |
|
2466)
Message boards :
GPUs :
Ability to stop CPU work while GPU is working and vice versa.
(Message 84404)
Posted 16 Jan 2018 by Richard Haselgrove Post: This is very similar to github #2214 - perm 'one at a time' from {CPU}, {nvidia GPU}, {intel gpu}. We've never found a simple, universal way of controlling temperatures of (especially) GPU devices from every manufacturer under every operating system, or to handle different usage preferences at different times of day or alongside different workloads. Sadly, I don't think it's likely to happen in the near future, either, unless a new volunteer comes along and has a lightbulb moment. |
|
2467)
Message boards :
The Lounge :
BOINC Politics thread
(Message 84398)
Posted 15 Jan 2018 by Richard Haselgrove Post: Read Bernie Sanders' global perspective feature on the same subject in Monday's Guardian - printed before the Carillion collapse became official. Let’s wrench power back from the billionaires |
|
2468)
Message boards :
The Lounge :
BOINC Politics thread
(Message 84375)
Posted 13 Jan 2018 by Richard Haselgrove Post: http://www.bbc.co.uk/news/uk-england-nottinghamshire-42671837: Part of Nottingham railway station built at the start of the last century is listed but that section escaped the worst of the damage, for which Ms Derry is thankful.Why can't we build fire-safe public buildings in the 21st century? |
|
2469)
Message boards :
Questions and problems :
No new tasks - machine
(Message 84374)
Posted 13 Jan 2018 by Richard Haselgrove Post: The only conventional shortcut we're missing is Ctrl-A for 'select all'. That would work (with some degree of usefulness) for all three lists - projects, tasks, and transfers. Remember that you can Ctrl-click to toggle individual lines for those times when you want 'all except...'. I might suggest that. |
|
2470)
Message boards :
Questions and problems :
No new tasks - machine
(Message 84372)
Posted 13 Jan 2018 by Richard Haselgrove Post: Multi-select all projects (mouse drag up from the bottom is easiest) and click 'No new tasks' once. Edit: Or for users who find the drag operation problematic - click first, shift-click last. |
|
2471)
Message boards :
News :
Client version 7.8.3 released
(Message 84371)
Posted 13 Jan 2018 by Richard Haselgrove Post: Many of us (myself included) are running version 7.8.3 with successful GPU outcomes. I'm sorry that v7.8.3 appears to coincide with GPU problems on your machine (and those of a very few other users), but unless you tell us exactly what the problem is, we can't address it. Work is likely to start soon on version 7.10. Now would be an excellent time to tell us what needs doing so that any corrections can be incorporated. In the meantime, v7.6.33 is available via the download all page, so that you can continue to contribute to scientific research. |
|
2472)
Message boards :
The Lounge :
BOINC Politics thread
(Message 84362)
Posted 12 Jan 2018 by Richard Haselgrove Post: I wonder who did the Nottingham station refurbishment? Even with an arson starter, it shouldn't have gone up like that. #TwinnedWithGrenfell ? |
|
2473)
Message boards :
The Lounge :
Needful Things
(Message 84344)
Posted 11 Jan 2018 by Richard Haselgrove Post: Needful thing #2: if you're going to build a brand-new house (date stone 2012) into a 240 year old canal bank, check that it's not leaking first. Or at least while you build. |
|
2474)
Message boards :
Server programs :
VGTU transfer problem
(Message 84342)
Posted 10 Jan 2018 by Richard Haselgrove Post: Tell the server administrator. I told SETI@Home about a similar problem: [http] Info: Empty reply from server [http] HTTP error: Server returned nothing (no headers, no data) which had been plaguing us all weekend, and they managed to fix it. |
|
2475)
Message boards :
The Lounge :
Needful Things
(Message 84341)
Posted 10 Jan 2018 by Richard Haselgrove Post: What's the etiquette when the pub quiz caller gets the question wrong? Answer - Bristol. Question - Which city's concert hall has been renamed from I don't do the quiz, because you need a team to catch the sport and pop music questions - so I just had a quiet word as I left in the pizza break. |
|
2476)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 84322)
Posted 9 Jan 2018 by Richard Haselgrove Post: My understanding was that drive manufacturers always quoted sizes in powers of 1,000: operating systems in powers of 1,024. They diverge quite quite quickly as the sizes grow. |
|
2477)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84317)
Posted 9 Jan 2018 by Richard Haselgrove Post: As I said last time around, when I spent a good afternoon logging with Wireshark what happened and digging through that data, those lines are only shown in the client's log, they're not sent to the server. It usually is a snippet of the latest news/notices, but if need be, resurrect #2101I think they're genuine received data, copied to the log as well as to their proper destination (notice, sched_reply, data file). I've copied a sample file to David's issue. |
|
2478)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84310)
Posted 9 Jan 2018 by Richard Haselgrove Post: IE says the Login and Sig Up links work now.Yes, I'm out and back in again. And the machine which M$ nuked is semi-alive again. I'd better wait for another automatic restore point before I try that update again. |
|
2479)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84307)
Posted 9 Jan 2018 by Richard Haselgrove Post: Not me - the only thing I've heard from David is a one-liner about fixing the configuration error very early Friday morning. Sent 20 minutes after a round-robin saying "Science United has not gone live." Off to find a sacrificial machine to log out of - probably not the machine which M$ just just sacrificed to the gods of "monthly security rollup". |
|
2480)
Message boards :
Server programs :
result XXXX is no longer usable
(Message 84296)
Posted 9 Jan 2018 by Richard Haselgrove Post: You'll have to give us a bit more detail about the scenario, then. |
|
2481)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84290)
Posted 8 Jan 2018 by Richard Haselgrove Post: Here with Chrome: traditional http works fine. https gives: This site can’t provide a secure connectionSounds like a re-run of the SU f-up. |
|
2482)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84284)
Posted 8 Jan 2018 by Richard Haselgrove Post: In my case, I was concentrating on data which was flowing - quite properly - in the opposite direction, from server to client. As far as I know, the whole data file was properly written to disk (I haven't checked), but some small-ish segments were copied into the <http_debug> log - as if the client hadn't noticed the join between headers and content, where the logging ought to have stopped. |
|
2483)
Message boards :
Questions and problems :
Boinc fails to start after updates
(Message 84280)
Posted 8 Jan 2018 by Richard Haselgrove Post: This is a vague and uncertain memory (I'm not a Linux user), but I think I've read that because of the way that the NVidia driver binds to the Linux Kernel, the driver has to be re-installed after every kernel update. Worth a try while you wait. |
|
2484)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84278)
Posted 8 Jan 2018 by Richard Haselgrove Post: OK, it happens under BOINC v7.6.33 as well, but rarely. Just done a bulk fetch from SETI (one request, got 48 tasks, and hence 48 download files). Of those, <http_debug> showed the contents (opening ~20 lines, maybe) of the scheduler reply and 3 of the 48 files. That sounds more like a bug than a deliberate policy. And tested again on a machine running BOINC v7.8.3. Got 62 files, just one of which (plus the scheduler reply) was copied into the log. I was wrong about the number of lines copied - about 35 from the scheduler, 64 lines of data. |
|
2485)
Message boards :
Server programs :
result XXXX is no longer usable
(Message 84277)
Posted 8 Jan 2018 by Richard Haselgrove Post: It's usually a very late reply, long after deadline, where the workunit has already received a quorum of successful results, validated, and been purged from the database. |
|
2486)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84275)
Posted 8 Jan 2018 by Richard Haselgrove Post: Yes, I've seen file contents 'leaking' into <http_debug> - clearly something's wrong. But it still seems to work, and I only turn on <http_debug> when I'm troubleshooting something more urgent. So I've never followed it up - largely because I'n reaching the point where I doubt that BOINC will ever fix any of the huge backlog of bugs. But I'll try going back to v7.6.33 sometime, to see if this is "New in v7.8.3", or alternatively some change in the way the servers are setting up the file transfers. |
|
2487)
Message boards :
Questions and problems :
Science United ???
(Message 84269)
Posted 7 Jan 2018 by Richard Haselgrove Post: Hopefully quicker than SynecdocheAnd how quickly could they fork it to allow apps that are not on the now hidden list?Well, just for the heck of it I went to CAS@home and looked for the list of applications. There was none.http://casathome.ihep.ac.cn/apps.php works just fine - they just don't advertise it on the home page. (That's the benefit of common standards and common software) |
|
2488)
Message boards :
The Lounge :
BOINC Politics thread
(Message 84267)
Posted 7 Jan 2018 by Richard Haselgrove Post: BBC editor quits over equal pay row She asked for the four international editors to be paid equally.That shows how wide the gap is. |
|
2489)
Message boards :
Questions and problems :
Science United ???
(Message 84263)
Posted 7 Jan 2018 by Richard Haselgrove Post: Well, just for the heck of it I went to CAS@home and looked for the list of applications. There was none.http://casathome.ihep.ac.cn/apps.php works just fine - they just don't advertise it on the home page. (That's the benefit of common standards and common software) |
|
2490)
Message boards :
Questions and problems :
Science United ???
(Message 84260)
Posted 7 Jan 2018 by Richard Haselgrove Post: Circa 2004 one project ran antrax briefly in collaboration with the dodWas that a separate project just grabbing the open-source software (as anybody can) and putting it out there, or was it one of the established projects doing a bit of contract work on the side? If it was an open-source grab and run, I don't think we have to worry so much. It would still be possible now, but people would have to find the project on the web and attach to it manually, just the same as they do now. The Alpha site for SU says: Eventually the selection of science projects, and the allocation of computing power among them, will be determined by an international committee of scientists. For now, all projects vetted by BOINC are included, and they have equal allocations.The projects list here is the same as the list of projects shown when you 'Add project' using BOINC Manager. It doesn't prevent a project going rogue after being added to that list, but it is one level of safeguard. I'm more worried about SU being developed 'under the direction' of just one man, with the intention of being moved "to a community-based model" once the money runs out. Having just sat (with Jord) for six months on a working group coping with the fallout from the last time that was attempted, wouldn't it be better to apply community direction right from the start, rather than leaving the community to pick up the pieces after the major decisions have already been made? |
|
2491)
Message boards :
Questions and problems :
Science United ???
(Message 84255)
Posted 7 Jan 2018 by Richard Haselgrove Post: Gary Charpentier has reposted a blog from the Planetary Society about the genesis of SETI@Home, and hence of BOINC. One paragraph caught my eye: In 1999, Berkeley released the result, SETI@home, and since then more than 8 million people have downloaded the program and donated spare computing power to help search for intelligent life. The open-source software, BOINC, on which SETI@home is based, is now used for other projects. This led to what Werthimer calls "the democratization of supercomputing," where users can choose individual research programs to assist.Is Science United in danger of de-democratising it again? |
|
2492)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 84229)
Posted 5 Jan 2018 by Richard Haselgrove Post: More to the point, the splitters on the Server Status Page are active again. |
|
2493)
Message boards :
Questions and problems :
New Install of BOINC Manager; cannot add project or account manager (Connection error)
(Message 84227)
Posted 5 Jan 2018 by Richard Haselgrove Post: We do development and testing here as well - all by volunteer effort. And yes, we are also human beings with our own frustrations. |
|
2494)
Message boards :
Questions and problems :
Yahoo Mail is dowe not working for hundreds of user
(Message 84226)
Posted 5 Jan 2018 by Richard Haselgrove Post: Yes, I'm one of them. In the UK, the major ISP British Telecom sub-contracts its mail services to Yahoo. That's not looking like a wise choice today. |
|
2495)
Message boards :
Questions and problems :
New Install of BOINC Manager; cannot add project or account manager (Connection error)
(Message 84220)
Posted 5 Jan 2018 by Richard Haselgrove Post: And for anybody else with neither patience nor diagnostic skills, the older versions 7.6.33/34 are listed on http://boinc.berkeley.edu/download_all.php, just one click away from the automatic download page. |
|
2496)
Message boards :
Questions and problems :
Science United ???
(Message 84218)
Posted 5 Jan 2018 by Richard Haselgrove Post: The URL to the insecure front page of the BOINC website is now again pointing at the BOINC website.Yes. It's been confirmed to me (private email) that it was a configuration error, now fixed. |
|
2497)
Message boards :
BOINC client :
Loss of Einstein WU Credit after updating Boinc Manager
(Message 84195)
Posted 4 Jan 2018 by Richard Haselgrove Post: That's the problem, I've created a new account with a different E-Mail address, DOH!!!Sorry, nope - the email address is the key identifier. The best thing to do is to finish and report current work under the new email address, then 'Remove' the project from BOINC Manager and 'Add' it back again, using the old email. |
|
2498)
Message boards :
BOINC client :
Loss of Einstein WU Credit after updating Boinc Manager
(Message 84194)
Posted 4 Jan 2018 by Richard Haselgrove Post: So, which project has lost these credits for 2,000,000+ Work Units, and where are you reading the credit information? The definitive credit record is shown on your account page at each individual BOINC project. They are exported back to the BOINC Manager on your computer for easy reference, and they are aggregated on a number of independent statistics sites, but the source and master record is always the project. |
|
2499)
Message boards :
Questions and problems :
Science United ???
(Message 84191)
Posted 4 Jan 2018 by Richard Haselgrove Post: Most of the 'home page' links here and in the Wikis - like the BOINC logo you see top-left as you read this - link to SU as well. This SUcks. |
|
2500)
Message boards :
Questions and problems :
Science United ???
(Message 84190)
Posted 4 Jan 2018 by Richard Haselgrove Post: Well, I've sent out an APB to project administrators to cover the situation until David wakes up in the Pacific morning. |
|
2501)
Message boards :
Questions and problems :
Science United ???
(Message 84188)
Posted 4 Jan 2018 by Richard Haselgrove Post: Existing projects, like for example SETI@Home, have a 'powered by BOINC' button on their front page. That links to the non-https versions of the BOINC url - hence, to an alpha test page for Science United. Existing projects like NumberFields@home which still use the traditional BOINC web interface have a 'Join our project' panel on the front page. Step 2 is 'download BOINC': again, this uses the non-https form of the link. This takes you to a login page for SU. I suppose this is one way to secure success for a new, private, non-community-managed, project: kill the one you're leaving behind. This, like so much recent BOINC development, hasn't been thought through at a full systems analysis level. |
|
2502)
Message boards :
Questions and problems :
New Install of BOINC Manager; cannot add project or account manager (Connection error)
(Message 84167)
Posted 2 Jan 2018 by Richard Haselgrove Post: It is certainly the case that the error message "The password you have entered is incorrect" is a last-chance fall-through for otherwise undiagnosed failures: see https://github.com/BOINC/boinc/blob/master/clientgui/BOINCBaseFrame.cpp#L517. (Please confirm that the exact wording of the message you saw is "The password you have provided is incorrect, please try again.": it makes searching much easier). Assuming that's the case, some other connection failure has occurred, and it's one I'd like to track down. BOINC consists of two local applications on your computer (the 'Client' and the 'Manager': they communicate by TCP/IP over port 31416, even on a single isolated computer). Communications failures can occur either because the Client is not running, or because the communications channel is blocked: it's hard to distinguish between the two. You've mentioned that you have purchased some very expensive hardware - how good are you on software? Could you compile and run a debug version of the Manager, if we need to explore alternative reasons for the display of this message? The password being discussed is the contents of the file "gui_rpc_auth.cfg", which should have been created automatically in your BOINC data directory. On a default single-computer installation, this file should have been created on installation, and read from the correct location when the Manager is first run: no security is needed. The authentication is provided to cover the case where the Manager is being used to control a remote machine over a local network. This isn't the first time it's prevented normal operation: we need to do some forensic research to discover why. |
|
2503)
Message boards :
Questions and problems :
Help: "git remote" doesn't work on Windows with git for windows commandline version
(Message 84138)
Posted 1 Jan 2018 by Richard Haselgrove Post: Speaking personally, I've always used the Windows GUI git tools, but command line principles should be the same. If you've cloned BOINC, you already have everything - master, all branches, all tags. From there, I export the code for tag 7.8.3 (or whatever), which gives me a single .zip file containing the code as at that version only. Then, I expand the zip into a new, separate, folder tree from where I can build, test modifications or anything else. Getting fixes back into the github source is another question entirely. Edit - should have said. v7.8.3 is a bit of a dog's breakfast, and is missing a lot of bugfixes which should have been included - you'll find a pull request from me (#2065) attempting to identify the fixes which were missed. Conversely, master (and hence 7.9.0) contains a lot of new development code which has never been tested: it doesn't surprise me it's broken. Best of luck. |
|
2504)
Message boards :
Questions and problems :
Quick question: where can I report problem for pre-release client? (7.9.0)
(Message 84136)
Posted 1 Jan 2018 by Richard Haselgrove Post: Depends what the problem was, and what exactly you mean by 'v7.9.0'. If it's a self-build from master, you can identify the cause of the problem, and it's reproducible on subsequent testing, raise an issue on Github. If you just want to compare notes with other testers, write to the boinc_dev email list, or simply post some details here. |
|
2505)
Message boards :
BOINC client :
Command-line-only client?
(Message 84127)
Posted 31 Dec 2017 by Richard Haselgrove Post: I can't see an equivalent of Debian's remote_hosts.cfg ...remote_hosts.cfg is an optional file that you would create yourself, if needed. It is part of the generic BOINC file-set, and not limited to any one distribution or even operating system. Details are in Controlling BOINC remotely: you also need to ensure that the data traffic (requests and replies) doesn't fall foul of any generic security traps like firewalls. SekeRob's <allow_remote_gui_rpc> is a perfectly acceptably alternative route, provided you can trust everybody with access to your local LAN. |
|
2506)
Message boards :
BOINC client :
Loss of Einstein WU Credit after updating Boinc Manager
(Message 84126)
Posted 31 Dec 2017 by Richard Haselgrove Post: The other possibility, if you didn't manage the folder locations carefully (though BOINC should remember that for you) is that you created a new account instead of re-using the old email address. If that happens, there will be two of you: the previous credits will be frozen, but the new credits will start from zero under the new account. |
|
2507)
Message boards :
Questions and problems :
Live Help by e-mail, voice or text not working?
(Message 84108)
Posted 30 Dec 2017 by Richard Haselgrove Post: Tried with Windows 7/Chrome, in three different Skype configurations: never installed, installed but not running, installed and active (signed in). Nada. Just the standard Chrome error screen This page isn’t workingfor the url http://boinc.berkeley.edu/help_vol.php?volid=xxx (and the same with https etc.) Edit - there is a file help_vol.php in https://github.com/BOINC/boinc-site/, but that's not to say it was properly deployed again when boinc-site was split from boinc. |
|
2508)
Message boards :
The Lounge :
BOINC Politics thread
(Message 84005)
Posted 23 Dec 2017 by Richard Haselgrove Post: I heard it was going to be €€€€€€€€€€€€€€€€€ !How easily some are appeased by irrelevant cosmetics... ;-)Some may think that, but I'm pretty sure that the printers that win the contract won't be seeing any cosmetics but plenty of ££££££££££££££££££££££££££ :-) |
|
2509)
Message boards :
GPUs :
Changing GPU and CPU allocation per task? (ex. 0.5 CPU + 0.5 GPU per task)
(Message 83916)
Posted 17 Dec 2017 by Richard Haselgrove Post: Yes, that's worth a try. Let us know how you get on. |
|
2510)
Message boards :
GPUs :
Changing GPU and CPU allocation per task? (ex. 0.5 CPU + 0.5 GPU per task)
(Message 83914)
Posted 17 Dec 2017 by Richard Haselgrove Post: Try setting those valuesUnfortunately, not true. The <gpu_usage> tag is not used in any way to control the behaviour of the application running on the GPU. So far as I know from talking with GPU developers, the GPU runtime support tools (installed as part of the GPU driver bundle) don't provide any API hooks to support throttling (though admittedly the developers I talk to are all motivated to provide maximum possible performance, and probably haven't even looked for any other way of programming). And beyond that, perhaps because there are no known tools, BOINC doesn't supply any way of passing such a 'slow down' instruction to GPUs. Your app_config fragment will operate as you describe to limit the number of tasks running on the CPU, but the single running task - once launched - will be free to grab every available resource and we're back to square one. Some project applications may interfere less than others with use of the GPU for its primary purpose of rendering images on screen: testing that to find an acceptable compromise will have to be done on the particular machine in question. I'm a user of NVidia cards: in general I find that applications programmed using the CUDA programming system are less intrusive than those coded using the rival OpenCL system. Solving this problem partly depends on how empl watches television. If he has a specific computer program which is used only for viewing TV, he can use <exclusive_gpu_app>important.exe</exclusive_gpu_app>to keep crunching and viewing separate: but if he uses a generic tool like a web browser, that may switch off BOINC's GPU computation too readily. |
|
2511)
Message boards :
The Lounge :
Needful Things
(Message 83896)
Posted 15 Dec 2017 by Richard Haselgrove Post: Sounds like it should be their test vehicle for autonomous driving. Edit - they're not very popular in the UK. Current models (less than 3 months old) are being offered at barely half price. |
|
2512)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83866)
Posted 13 Dec 2017 by Richard Haselgrove Post: I should hope not - 'ready to send' is way above high water mark.24 hours already, only another 5 to go :-)Not really, splitting is not happening as of yet And many of the other numbers are pretty screwy, too. |
|
2513)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83853)
Posted 13 Dec 2017 by Richard Haselgrove Post: Einstein has come back again, so that's a bit of a relief. No explanation yet where they went. |
|
2514)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83850)
Posted 13 Dec 2017 by Richard Haselgrove Post: Seti is still down.And Einstein has crashed under the strain. Going to be a fun day all round. |
|
2515)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83794)
Posted 12 Dec 2017 by Richard Haselgrove Post: Mine has a proper one-touch button which turns the blower on without delving through the touch-screen menu system and messing with your air-conditioning setting. That's probably fan speed 8, or perhaps 11: I don't know if has any electric heating, or just gets first dibs at whatever's flowing out towards the radiator, but it seems to work pretty well. |
|
2516)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83791)
Posted 12 Dec 2017 by Richard Haselgrove Post: Quite chilly up North, too, but we've learned how to deal with that.  First winter with the new car. It has a snazzy aluminium gear knob - which is bloody freezing to the touch after it's been standing out all night. So I wear heavy driving gloves, and they catch on the steering wheel radio controls and change the station without me noticing... Aargh. They've coated the windscreen with something odd, too, The frost turns into solid almost-but-not-quite-clear ice, which is much harder to scrape off. At least it's got a good demister on the inside. |
|
2517)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83786)
Posted 12 Dec 2017 by Richard Haselgrove Post: Early drop today, four minutes to one. And the feeder/scheduler had been playing up all morning - prepare for a long, cold, one while they investigate. |
|
2518)
Message boards :
Questions and problems :
New PC; getting Computation Error results
(Message 83771)
Posted 11 Dec 2017 by Richard Haselgrove Post: It might be worth asking if there is anything you can do to rectify the reported problems. In some cases, it's settings in the computer's BIOS which prevent the VM running, rather than the chip's capabilities themselves: that would be worth other readers here reading about, if you could come back and report what they say. |
|
2519)
Message boards :
Questions and problems :
New PC; getting Computation Error results
(Message 83769)
Posted 11 Dec 2017 by Richard Haselgrove Post: This sort of question is far better asked, and answered, at the project's own website. Taking your most recent failed task, http://www.cosmologyathome.org/result.php?resultid=61598493, you can read: 2017-12-11 07:52:57 (9448): Hardware acceleration failed with previous execution. Disabling VirtualBox hardware acceleration support.They are far better equipped to deal with that than we are. |
|
2520)
Message boards :
Questions and problems :
Silent operation issue.
(Message 83767)
Posted 11 Dec 2017 by Richard Haselgrove Post: There's also the problem that "clean slate" restoration wipes out any work in progress at the end of the session. If the hardware is of the order of ten years old, you'd need to pick projects which can * reliably supply work at the first request after switch-on * complete at least one task during an average session as well as being from an officially sanctioned institution. Otherwise, any energy used will be entirely wasted. |
|
2521)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83761)
Posted 11 Dec 2017 by Richard Haselgrove Post: 12/10/2017 10:18:39 PM | | Project communication failed: attempting access to reference siteAnd did you get through to the reference site? I usually get those messages when the router my ISP supplied for my house is going through its weekly restart and line speed test. |
|
2522)
Message boards :
The Lounge :
Needful Things
(Message 83750)
Posted 10 Dec 2017 by Richard Haselgrove Post: And now it's reached me. I'm not going anywhere today - did my shopping yesterday. :-) |
|
2523)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83749)
Posted 10 Dec 2017 by Richard Haselgrove Post: Jacob Klein has pointed out to me that each of the four mini performance graphs to the top right of my screenshot has a dropdown arrow beside the label. That opens a list of (I think) 12 different metrics to plot: compute_0 is the one that BOINCers are likely to want. Unfortunately, Task Manager: Does NOT remember that I last viewed the performance tab. DOES remember that I viewed the performance of the GPU. Does NOT remember that I want to view compute_0. That's just sloppy GUI design... |
|
2524)
Message boards :
The Lounge :
Needful Things
(Message 83744)
Posted 10 Dec 2017 by Richard Haselgrove Post: It's snowing!Hasn't reached here yet, but we're under a warning that it's heading towards us. |
|
2525)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83742)
Posted 10 Dec 2017 by Richard Haselgrove Post: And we all learned something along the way. This is the machine I took that screenshot from, showing the variation in task run time - that one was particularly quick. All tasks for computer 388426 |
|
2526)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83740)
Posted 9 Dec 2017 by Richard Haselgrove Post: And that is interesting.  Same machine running GPUGrid. That does use the GPU, quite heavily, but under the CUDA (v8.0) programming environment. And no load is registered. |
|
2527)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83739)
Posted 9 Dec 2017 by Richard Haselgrove Post: Well, blow me down - so it does. I still mainly use Windows 7, so I missed that. Here goes: That's running a SETI OpenCL task, which uses the GPU rather inefficiently - but at least it shows signs of life. This machine does switch to GPUGrid periodically, so I'll look to see if CUDA apps display similarly when I catch it happening. |
|
2528)
Message boards :
Questions and problems :
Unable to connect to the core client
(Message 83737)
Posted 9 Dec 2017 by Richard Haselgrove Post: I was going to ask: you say you tried starting the client manually. Did it keep running, or did it stop again? We had some problems with the client crashing in early testing under Windows 10, and eventually found some crash dump files in the (hidden) folder C:\users\[your account name]\Appdata\Roaming\BOINC You probably haven't wiped that, so it might be worth a look. |
|
2529)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83735)
Posted 9 Dec 2017 by Richard Haselgrove Post: Sure - and that certainly wouldn't show directly in Windows Task Manager. I was also pointing out, since Battu007 is new to BOINC and all its little quirks, that not everything shown on the BOINC Manager display can be taken at face value either - trying to fill in some background. At first glance, I'd say that the CPU showing at 42% for the GPUGrid task in [Windows?] Task Manager - 12:56 post - sounds plausible for GPUGrid on a dual-core Windows 10 machine - except that the opening log says it has 8 cores. We still have some way to dig here before we work out which tool is telling the truth. |
|
2530)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83730)
Posted 9 Dec 2017 by Richard Haselgrove Post: I've just started a new GPUGrid task on this machine, and Task Manager is showing a CPU load of around 7% (it flutters slightly). This is a quad-core machine, so that would equate to 28% of a core. A second monitoring tool, reporting on BOINC usage explicitly and averaging over 30 seconds, says that the same task is using about 24% of a single core. I think that's within the margin of error, and I trust them both. BOINC is lying to me, and echoing the 10% I put into an app_config.xml file years ago: without manual configuration, BOINC's statement of 95% is equally to be considered a lie. Neither 10% or 95% is obtained from actual measurement. |
|
2531)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83725)
Posted 9 Dec 2017 by Richard Haselgrove Post: Look at it in BOINC Manager, Advanced view. See what the 'Progress' column on the 'Tasks' tab is showing [Ah, you said 3.599% - that's good}. But BOINC has a sneaky way of indicating what I call "pseudo progress", to reassure users running tasks which don't report progress properly themselves. GPUGrid tasks run very uniformly from start to finish, and should start showing progress almost immediately - although the total predicted runtime is variable and not to be trusted: check that the indicated progress% increases steadily over time. I'm currently running them in a bit over 12 hours on desktop 970s, so your 12 hours on a mobile 960 may be over-optimistic - but you should be able to guesstimate when it gets a bit further. If GPUGrid progress increases uniformly over time, you would seem to be crunching OK and your GPU usage meter might be broken. |
|
2532)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83720)
Posted 9 Dec 2017 by Richard Haselgrove Post: The BOINC system - end-to-end, counting both the server and client together - makes a decision when work is fetched from the server, and assigns it to a particular class of device. Once you have been assigned a task, it will show in your BOINC Manager which device it will run on: switch to 'Advanced View' if you haven't got that displayed yet.You can even disable the Intel GPU in the config to test what happens.I tried that, but then the task only runs on CPU while both the GPU are sitting idle. The tasks themselves are generalist, and could be run on any device, but it's devilish tricky to change the assignment once made - that's not something to try in the first week. So the question becomes "why hasn't the server assigned any new work to my NVidia GPU?". There are certain project-specific limitations at SETI which prevent certain types of task being assigned to NVidia GPUs (they run inefficiently, so it's wasteful). So you may not get NVidia work at every attempt, but if you persevere, you should get some soon enough. I personally check <shed_op_debug> in the Event Log options list (Ctrl+Shift+F from the Manager) - that allows you to see exactly what is being requested and supplied, without overwhelming you with data. That might give a clue. |
|
2533)
Message boards :
GPUs :
Boinc using integrated Intel GPU instead of Nvidia GPU - Lenovo Y700
(Message 83715)
Posted 9 Dec 2017 by Richard Haselgrove Post: I'm not quite sure why that should be - it looks as if you've done everything right: you have in fact successfully completed one task on the NVidia chip, shown as 'cuda42' in http://setiathome.berkeley.edu/results.php?hostid=8407014&state=4. The 'use all gpus' wasn't strictly necessary - that only comes into play if you have two different GPUs using the same technology. That's only likely to happen in a desktop machine. But it'll do no harm - you may as well leave it there now. I'd just let it run for a while, and see how it settles down. BOINC has a learning curve while it finds out which applications run best, and how long they're going to take: and you have a learning curve while you find out how noisy the fans get when everything is running at full power. You may find you need a supply of extra cool air from a fan-assisted stand you can place the laptop on: at the very least, ensure that no air inlet vents are obstructed while you're crunching. Even lifting the machine half an inch off the desk on four props can help, and soft furnishings are a no-no. Watch out for that 'Suspending computation - user request' message at the very end of your log. It very much depends how you intend to use the laptop. You will make very different choices if it will mainly be static with mains power available, compared to your choices if you need to use it on the move and preserve battery life. If it's going to be static and mains powered - and if you can stand the noise - check that the 'Activity' menu says 'run always' for as many components as you dare to risk! There's also a question how Lenovo have integrated the two GPUs. It's possible that it will switch between them, according to demand, but the power circuits may not be adequate to run both at the same time. Again, take your time to watch how it behaves, and decide what combination of settings suits you best. It'll be fun, but warning - it can get addictive! Enjoy. |
|
2534)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83699)
Posted 8 Dec 2017 by Richard Haselgrove Post: Ouch. Richard, don't give it all away to the Americans! :)Well, at least the fact that they've reached a deal - or the announcement, which may or may not be the same thing - means the exchange rate has tipped in my favour overnight. If only I could coax a bank account number out of Eric - I don't want to hand it over to Visa International in handling fees and exchange rate rip-offs. |
|
2535)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83685)
Posted 7 Dec 2017 by Richard Haselgrove Post: Tory dinner at National Mining Museum cancelled I don't know what the dining facilities at the museum are, but I have visited it and taken the underground tour (and I have a genuine miner's safety lamp, purchased in the souvenir shop, hanging my living room). Note that the cancellation report says that former production miners who lost their jobs still work at the pit - they serve as underground guides, and their experience of working around mining machinery is crucial for the safety of visitors. I don't know whether the dinner would have included an underground tour - it seems likely that it would have - but that would have been the particularly insulting part, not the surface politics. |
|
2536)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83683)
Posted 7 Dec 2017 by Richard Haselgrove Post: Just as normal will do - I think the plan is that anything donated between now and 31st December 2017 will be counted for the hardware and the special icons. There'll be a parallel special appeal through the GPU Users Group, because this sort of kit is really their specialism. The plan is to make this a blitz appeal, over and done as quickly as possible. Let's see if we can hit the target by Christmas. |
|
2537)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83680)
Posted 7 Dec 2017 by Richard Haselgrove Post: Please visit the SETI Home Page for some exciting news :-) |
|
2538)
Message boards :
Questions and problems :
Can't connect to locahost
(Message 83674)
Posted 7 Dec 2017 by Richard Haselgrove Post: BOINC Manager actually attempts to connect, specifically, to the boinc client running on the local host. Can you independently verify that the client is currently running? Without that, the connection will inevitably fail. |
|
2539)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83642)
Posted 6 Dec 2017 by Richard Haselgrove Post: One from the North: 'Insensitive' Conservative dinner at coal museum Conservatives in one part of West Yorkshire say they won't cancel a party they've booked at the National Coal Mining museum at Wakefield, despite strong opposition.Maybe we could hand them each a pick and a candle and just leave them to tunnel their way out... |
|
2540)
Message boards :
BOINC client :
--get_old_tasks - not showing tasks from last 24 hours as stated?
(Message 83630)
Posted 6 Dec 2017 by Richard Haselgrove Post: Boinccmd manual updated. Reported to developers as #2268 |
|
2541)
Message boards :
BOINC client :
--get_old_tasks - not showing tasks from last 24 hours as stated?
(Message 83629)
Posted 6 Dec 2017 by Richard Haselgrove Post: I'd agree with that. I've tried with one of my faster machines: in the last 24 hours, it's completed 317 tasks, but --get_old_tasks only reports the final 16 of them (about [edit] 1 hour. Another problem with --get_old_tasks: for the few tasks it does find, it reports separately on "completed time" and "reported time", but the values are identical. I'm using <sched_op_debug> logging, so I get a timestamp for "got ack for task" every time. It doesn't match "reported time". Edit: found 9024f8cc18fe06b4c9d8bededda80355e4aaceb6: Client: change length of old-results log from 24 hours to 1 hourI'll update the documentation. |
|
2542)
Message boards :
BOINC client :
--get_old_tasks - not showing tasks from last 24 hours as stated?
(Message 83594)
Posted 5 Dec 2017 by Richard Haselgrove Post: --get_old_tasks operates by doing a GUI RPC against the currently-running client. I haven't traced it any further back than that, but I'd assume for the time being that it is retrieving results from the memory space maintained by the current BOINC client instance: in other words, only since the beginning of the current session. I'll come back and let you know if I find out any more. Edit: commit dd7c8335f3a0f743679c61610d5a3946cb454c30 supports that interpretation: Client: keep an in-memory list of reported tasks in the last 24 hours, and add a GUI RPC for fetching this list.Edit 2: Hmmm. Checking it out on the (old, slow) laptop in front of me, which has been running continuously since 17 November, I find 26 completed tasks reported in the message logs for the last 24 hours. But only two of them - both reported in the last four hours - appear in response to a --get_old_tasks. Back to the drawing board. |
|
2543)
Message boards :
Questions and problems :
Silent operation issue.
(Message 83585)
Posted 5 Dec 2017 by Richard Haselgrove Post: Yes, the possibility of attaching to a malicious project, or for malicious actors to insert code into a normal project, is probably the greatest risk you face. Your proposed plan of denying access to the Manager (and I'd suggest boinccmd too) would make it much harder for a library user to attach to a malicious project, and if you limit the range of pre-attached projects you set up to those with a proven academic sponsor and a track-record of reliable operation, you'll have exercised due diligence. If you also install BOINC 'as a service', the enhanced sandbox security (special limited-permission accounts) would make it almost impossible for anyone else to tamper with the installation without access to an administrative account on the machine. The point about opening firewall ports is a complete red herring. The only ports required are 80 and 443, the standard ports used by browsers to access the WWW. Any computer placed in the public access area of a library will have those open for the public to use anyway. You might find that the library browsers are configured to use a proxy server for access: BOINC can handle that, but you may have to supply proxy configuration details when you set them up. You may also need to get agreement for the authorised project servers to be whitelisted. The only non-standard port you might need to consider is the control port for remote monitoring. If you use the 'VPN termination on router' approach I suggested earlier, even that doesn't require a separate port: your monitoring machine will appear as a trusted insider on the network. You'll have to convince the authorities to allow you that, though. |
|
2544)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83583)
Posted 5 Dec 2017 by Richard Haselgrove Post: The provocative nature of much of the flame-posting has been considered, yes. |
|
2545)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83534)
Posted 4 Dec 2017 by Richard Haselgrove Post: FGM is primarily an African (mal)practice, not Indian. WHO statistics. |
|
2546)
Message boards :
Questions and problems :
Silent operation issue.
(Message 83496)
Posted 3 Dec 2017 by Richard Haselgrove Post: The machines are in public libraries, and I would not want the staff being pestered ...What's your relationship with those public libraries? If you have persuaded the staff to let you perform 'run as administrator' installations on the machines, I assume you have some sort of trust relationship with their employers. In that case, what I've done in the past (not for BOINC - more directly for the work needs of a commercial employer without onsite tech support) is to set up a secure VPN termination on the site's router which allows me to 'dial in' (in practice over broadband) from a machine on my home LAN and become a member of the remote site's LAN. When that's up and running, you could visit each remote machine using BOINC Manager and Remote GUI RPC. Except I still wouldn't do that. BOINC Manager's once-per-second fixed update interval is too fast for use over a WAN, so instead I'd use BoincTasks to consolidate all the remote machines onto one monitoring display, and set the update interval to something low enough not to saturate the VPN at either end. From what I know of public library computers, they won't be high enough specification to run GPU applications, and will be switched off when not in use. In which case, a service installation will be adequate and will silence the messages. |
|
2547)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 83482)
Posted 3 Dec 2017 by Richard Haselgrove Post: Any woman knows damn well if their man is playing away from home, whether they are a member of a men's club or not.My remarks were aimed at those sections of society where the men have claimed the positions and roles of power for themselves, and the women - especially in professions where physical appearance is significant - have to fight their way past groping hands and worse to reach any semblance of equality. Think Hollywood, business, politics, many sporting hierarchies - the list goes on. |
|
2548)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 83478)
Posted 3 Dec 2017 by Richard Haselgrove Post: Remember that I went through the 'all male' educational route from the age of 8 to 20 - my first encounter with the female of the species as an equal (at about a 10% ratio) was during my third year at university. Prior to that, the only females I had encountered (apart from my mother, from whom I was separated for maybe 75% of that time period) were in service roles - cooks and health providers. Those schools got me through the exams, but I don't think they gave me a good 'education for life', in any real sense of the word. I have a great fear that the 'elites' of various societies - not just in this country - may have gone through similarly skewed educations, and may inflict the same on their own sons, because it's the only system they know. And I also fear that the 'all male club' - service, social, or sporting - may largely provide a refuge for these men from a world they don't understand. If they channel their energies into service rather than sexual impropriety, that's certainly a positive outcome, but I don't think it excuses the entire system. |
|
2549)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 83474)
Posted 3 Dec 2017 by Richard Haselgrove Post: None of it is anything to do with being "a man" ...So why did it turn out that so many turned out to be male-only, or female opt-outs because they'd been excluded? Is that because men can't be trusted to behave themselves when there are females in the same club? See recent events... |
|
2550)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83460)
Posted 2 Dec 2017 by Richard Haselgrove Post: You missed Sophos from the UK. Even as a non-customer, I got good service from them when I reported a 'fresh-in-the-wild' virus to them, some 15 years ago. |
|
2551)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83455)
Posted 2 Dec 2017 by Richard Haselgrove Post: It's more likely a job that lets you hold state secrets on your employer's computer. No, I wouldn't want one of those either. |
|
2552)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83453)
Posted 2 Dec 2017 by Richard Haselgrove Post: Executive summary: don't use it if you're cleared to hold state secrets on your computer. |
|
2553)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83439)
Posted 1 Dec 2017 by Richard Haselgrove Post: Yes, I've had to use System Restore on mine a few times over the years. One thing I'm learning: the (commercial) AV I'm using occasionally pops up an alert saying that the machine needs to be rebooted so that an update can be completed. I've usually delayed it until 'Patch Tuesday' so I can get both reboots out of the way together (being 10 years old and still on a derivative of the original factory OS, it takes a long time to boot through all the junk). But that's turned out to be a false economy: a few times after a 'double update', I've had to restore back a few steps and then re-install both the M$ updates, and the AV, separately. So now I'd recommend doing any required reboots singly, before starting on the next update task. |
|
2554)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83417)
Posted 1 Dec 2017 by Richard Haselgrove Post: I had one yesterday that had been crunching for SETI@Home for even longer than I have. But still didn't realise that the 'problem' he was complaining about was the consequence of his own inappropriate actions. The usual thing: "I'll fix that server by nuking my own computer" |
|
2555)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83413)
Posted 1 Dec 2017 by Richard Haselgrove Post: The generic Berkeley SETI Research Centre is on twitter, and Eric K is on twitter (or so I hear from third parties - I'm not, so I can't confirm personally). But you can't fit much of an event log into |
|
2556)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83410)
Posted 1 Dec 2017 by Richard Haselgrove Post: Also "I haven't learned to read yet" day. |
|
2557)
Message boards :
Questions and problems :
nuisance log messages, "unrecognized report_on_rpc/"
(Message 83405)
Posted 30 Nov 2017 by Richard Haselgrove Post: It shows up in the checkin notes for 2011: David 24 Apr 2011
- client: simplify the semantics of sticky files:
- All sticky files are reported on each scheduler RPC
- If a scheduler reply says to delete a file, clear its sticky flag
In particular:
- remove the "send file list" tag in scheduler RPC replies
- remove FILE_INFO::marked_for_delete
- remove FILE_INFO::report_on_rpc
- remove the request_file_list programEinstein are using a fork from very old server code, because they didn't accept the changes to CreditNew made in 2010. I suggest you start a conversation with Bernd Machenschalk at Einstein to report your findings, and ask whether it would be safe to remove that tag from their code, remembering that people may still be using pre-2011 clients.
|
|
2558)
Message boards :
Questions and problems :
boinc cause cpu throttling
(Message 83400)
Posted 30 Nov 2017 by Richard Haselgrove Post: I can't think of any mechanism by which running BOINC (or more likely, the project science applications which use the CPU intensively and continuously) would cause throttling other than for thermal reasons - I don't think you can disable the ultimate safety overheat protection. Are you monitoring the CPU core temperatures at the same time as you are measuring frequency? |
|
2559)
Message boards :
BOINC client :
Message: Task postponed: Suspicious pulse results
(Message 83379)
Posted 29 Nov 2017 by Richard Haselgrove Post: #1978Well, I tried. #2065 |
|
2560)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83377)
Posted 29 Nov 2017 by Richard Haselgrove Post: I just hope that nobody tells Trump that the people of North Korea are Muslim. (they're not)This is getting a long way beyond a joke.His daddy was a member of KKK. He was born with a silver spoon. He was brought up as a racist. He can't change and he doesn't want to. |
|
2561)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83374)
Posted 29 Nov 2017 by Richard Haselgrove Post: This is getting a long way beyond a joke. Trump account retweets anti-Muslim videos |
|
2562)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 83372)
Posted 29 Nov 2017 by Richard Haselgrove Post: As I have already posted on the main SETI board: Apparently the Master Science database (that's the long-term storage of completed results, not the day-to-day BOINC server that we interact with every day) crashed after 241 days of continuous running. It's restarted, and appears to be running OK, but I guess they're keeping it lightly loaded overnight so they can run further tests in daylight. |
|
2563)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83368)
Posted 29 Nov 2017 by Richard Haselgrove Post: Exactly. The tracks of the East Coast mainline are used by, from memory and probably not an exhaustive list, trains from:East Coast mainline would be the first service to try the new model from 2020.At the moment Network Rail, which is state-owned, looks after the track and other infrastructure while train services are operated by private companies. Mr Grayling said Network Rail was a "big blob at the centre" and said it "makes sense" to run trains and the track together.Only if it is fully nationalised, just tinkering will make things worse. Remember Railtrack anyone? ScotRail CrossCountry Northern TransPennine Express Grand Central East Midlands Trains Great Northern - and that's just the passenger operations. If [edit: Virgin] East Coast control the points and signals, will they block their rivals so their own trains go through first? That's what happened before the Railway Grouping Act of 1921. |
|
2564)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83363)
Posted 29 Nov 2017 by Richard Haselgrove Post: The setup is the same as for the railways. Network Rail run the system, the operating companies lease the rolling stock and provide the services. The infrastructure is government backed, the end product is a commercially based provision.Chris Grayling must have reading your posts. Apparently there's a proposal to put the trains and the track back together again - although without spending any money. |
|
2565)
Message boards :
BOINC client :
Message: Task postponed: Suspicious pulse results
(Message 83349)
Posted 28 Nov 2017 by Richard Haselgrove Post: Lovely jubbly. Commit SHA or PR#, please, so I can see if it's been deployed yet? |
|
2566)
Message boards :
GPUs :
Nvidia 2 way SLI Gpu(s) - No Load
(Message 83348)
Posted 28 Nov 2017 by Richard Haselgrove Post: 1) SLI cards should compute automatically as two separate devices, no need to change configuration. 2) No need for <use_all_gpus> either, because SLI should only be used with two identical cards - BOINC will use both automatically. 3) Only downside of SLI is that the card speeds will be locked together, so both will run at the speed of the slower (usually the hotter one). Checking project preferences is the first thing to try. |
|
2567)
Message boards :
BOINC client :
Message: Task postponed: Suspicious pulse results
(Message 83329)
Posted 28 Nov 2017 by Richard Haselgrove Post: You're probably right about the project name being listed (perhaps Bill can confirm, but I'm pretty sure that 'Suspicious pulse results' are only reported by SETI). But what we're missing is precisely which 'Task' has been 'postponed': knowing the name would enable us to look up the stderr_txt on the project website (if the project's server code is new enough to have that search-box facility). |
|
2568)
Message boards :
BOINC client :
Message: Task postponed: Suspicious pulse results
(Message 83321)
Posted 28 Nov 2017 by Richard Haselgrove Post: That will likely be a GPU app from SETI@Home - we had a spate of those recently. Remind me to open an issue for the task, app, or project name to be included in the Event Log for cases like this. |
|
2569)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83318)
Posted 28 Nov 2017 by Richard Haselgrove Post: It goes against the grain for me to say it, but the commercial electricity suppliers seem to be much more 'enlightened' (pun intended) than local government blue-badge providers when it comes to priority care. Here is your local version: http://www.ukpowernetworks.co.uk/internet/en/power-cuts/priority-services-during-a-power-cut/ Scroll to the bottom for online registration. |
|
2570)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83316)
Posted 28 Nov 2017 by Richard Haselgrove Post: It most certainly isn't routed through different distribution channels in London.I do hope gas and electricity are kept separate during distribution... |
|
2571)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83313)
Posted 28 Nov 2017 by Richard Haselgrove Post: The 'at risk' register might be quite widely drawn. I seem to have got recorded on my local one, even though the only possible grounds for doing so would have been age (60+ when I found out) and living alone. They even sent me a 'periodic review' questionnaire to check I still qualified - I told them I hadn't got any younger. You should be able to register online, through your local equivalent of https://www.northernpowergrid.com/care. Another keyword worth searching for is 'priority', though that seems to have disappeared from the url since I renewed my registration in April this year. Edit - I note from the form I filled in that as well as a pretty professional list of medical dependencies (including 'refrigerated medicine'), it explicitly lists "Household in fuel poverty" as a reason why additional support is required. People who self-disconnect because of a prepayment meter might be able to initiate some interesting conversations with their landlord on that basis ;-) |
|
2572)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83307)
Posted 28 Nov 2017 by Richard Haselgrove Post: Plus of course mobile phones.Do your local cellfone base stations have UPSs? |
|
2573)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83306)
Posted 28 Nov 2017 by Richard Haselgrove Post: Gas and electric all comes from the same National distribution networks, but you choose which supplier you buy it from and how much you pay for it.But is routed through different distribution channels locally, perhaps using gasholders (Keighley until recently) or substations (Bingley) respectively. My last power cut was when the Bingley substation was (locally) submerged in riverwater. I was amused to discover the same outage advice as you've subsequently posted: Do not call your electricity supplier. Electricity suppliers sell consumers energy and are not responsible for the power lines that connect to home or business to our network.In other words, your electricity supplier has no role in supplying electricity. Could we get them under the Trades Descriptions Act? |
|
2574)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83298)
Posted 28 Nov 2017 by Richard Haselgrove Post: Postal deliveries, other technical services, and local government administration all have different boundaries. I get my post from Keighley, telephone from Bingley (and I think electricity too), and governance from Bradford.and no power outage here, in South London, strange.Croydon was in Surrey with a CR postcode but was amalgamated into Greater London in 1965. Official postal address is still Surrey. I think the power outage was fairly local to Central London, |
|
2575)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83296)
Posted 28 Nov 2017 by Richard Haselgrove Post: It would help if the Government did some governing, instead of infighting. |
|
2576)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83294)
Posted 28 Nov 2017 by Richard Haselgrove Post: Well, the BBC is only part-time these days. Posted at 18:00 27 NovMaybe they got stuck in the tube on the way home and haven't recharged their phones yet? Edit - bookmark http://ukpower.ukpowernetworks.co.uk/fault-map. And buy a bigger UPS! |
|
2577)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83292)
Posted 28 Nov 2017 by Richard Haselgrove Post: Apparently it was epic with a National Grid failure.Well, it was a good day to bury bad news. Nobody except Annie noticed. And I thought she was north of the river? |
|
2578)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 83288)
Posted 27 Nov 2017 by Richard Haselgrove Post: I've just finished being power-failured. Five hours of oblivion wreaking havoc in a cupboard looking for candles more substantial than cake ones.That's pretty epic, even by African standards. Do we need to start sending food parcels to London? |
|
2579)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83283)
Posted 27 Nov 2017 by Richard Haselgrove Post: Ah yes 12" LP's I remember them.I remember that one particularly well. I don't know how I first came across Roy and Leon, but I heard they were putting on a gig at Bingley Arts Centre - probably soon after that one came out, so 1975 or 1976 - and I knew it would be worth going. So I had to cycle back to Bradford afterwards with two LPs under my arm, and probably several pints inside me... |
|
2580)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83274)
Posted 27 Nov 2017 by Richard Haselgrove Post: With the exception of Britannia, 'digging coal', and possibly 'property in Spain', I reckon the lyrics have lasted 40 years pretty well. As for Harry not getting any younger, they used up William to take our mind off the early stages of austerity in 2011: obviously they kept 'the spare' until needed. I just hope this means there won't be another crisis until young George is old enough to do his duty. Do all American 12" vinyl pressings come with a 12-page booklet, with background, lyrics, commentary, and erudite footnotes explaining British slang terms? My British one just had one column of notes on the back cover of the sleeve - but then, I suppose we were all fully aware of the political situation without written aids. The lyrics aren't even in the song book of the same name. |
|
2581)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83267)
Posted 27 Nov 2017 by Richard Haselgrove Post: It's official - Britain has finally reached rock bottom (due to Brexit, recession, and anything else you care to mention). So, as predicted 40 years ago, it's time for a Royal Proclamation..... Coats Off for Britain (lyrics) Coats off for Britain! says Her Majesty, From the bottom of her very gracious heart, Trying to boost the nation With a Royal Proclamation, To encourage everyone to play his part. While politicians palter And the bills remain unpaid, Her step will never falter And her smile will never fade. Dutifully doing What she's born and bred to do, Working hard from morn till late, Receiving dreary Heads of State, Arranging Royal Weddings And a Royal Birth or two, For me and you, To pull us through. And on the Royal Yacht Britannia they'll be singing then---Congratulations to Megan and Harry - though a mixed-race American divorcee has resonance... |
|
2582)
Message boards :
Questions and problems :
broken ?
(Message 83189)
Posted 24 Nov 2017 by Richard Haselgrove Post: There can be two, quite separate and distinct, reasons for the continual 'connecting to...' message: 1) A TCP/IP communications problem - perhaps a firewall, security, or corporate group policy issue. 2) There's nothing to connect to, because the client hasn't started or has crashed immediately after startup. We need to advise people how to distinguish between the two cases, and react appropriately depending on what they discover. |
|
2583)
Message boards :
Questions and problems :
Resource share questions/possible issue
(Message 83133)
Posted 22 Nov 2017 by Richard Haselgrove Post: Resource share is based on a long-term work share, not a short term one - I think the time is weeks/months rather than days.To amplify that statement: projects never 'send' work, projects respond to 'requests for work'. The BOINC client on your machine requests enough work to keep it busy for a certain number of seconds (based on your cache settings and usage history), and it's the job of the server to supply that amount of work. Unfortunately, the server's estimate of how many tasks are needed to keep you busy for the given number of seconds can be, ahem, over-enthusiastic. This is particularly a problem for newly attached computers, and for computers which aren't available for BOINC to use for very many hours each day. BOINC - the project server, in this case - will eventually learn how much work to send in response to your requests, but by then you may have a minor case of indigestion. The best advice when attaching to a project for the first time is to keep your cache settings modest until the learning period is over. |
|
2584)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83114)
Posted 21 Nov 2017 by Richard Haselgrove Post: Brexit? http://www.dw.com/en/britain-loses-seat-on-international-court-of-justice-for-first-time-ever/a-41460450No, that's a United Nations elected position, not (directly) anything to do with the EU or Brexit. Apparently, Lebanon lobbied very hard to get the Asian place, leaving India to muscle in on the European seat - or something like that. No British judge on world court for first time in its 71-year history But I wouldn't be surprised if our general incompetence over recent years had influenced electors thoughts towards 'time for someone new', quite apart from the examples in that Guardian article. |
|
2585)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83090)
Posted 21 Nov 2017 by Richard Haselgrove Post: Neither the Tories nor Labour are fit to rule as a majority government.Absolutely. Basically, because both the Tories (as you would expect) and Labour (shamefully, under Tony Blair) decided to govern by money alone. And good money people don't get elected to Government: they don't stand as MPs, because they're too busy making money. As a result, the privatised contract-wallahs can run rings round the government - any government - because they're on home territory and know how to rig the pitch. And to prove the point, NHS overcharged by millions for key drug. The amount the NHS paid per pack rose from around £4.46 in 2007 to £258.19 by July 2017, an increase of almost 6,000%. |
|
2586)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83084)
Posted 21 Nov 2017 by Richard Haselgrove Post: And so it came to pass. Privatisation is a catastrophe, warns godfather of forensics (2012) Drug driving cases dropped over forensics (2017) |
|
2587)
Message boards :
Questions and problems :
WU doesn't run
(Message 83080)
Posted 21 Nov 2017 by Richard Haselgrove Post: Did you look at the Event Log? |
|
2588)
Message boards :
Questions and problems :
Installing VirtualBox with Boinc
(Message 83018)
Posted 17 Nov 2017 by Richard Haselgrove Post: To clarify: there's no way of changing the VirtualBox installation drive during a combined BOINC+VB installation operation. But there's nothing to stop you doing two separate installations: download VB from Oracle, and use the non-VB BOINC installer. The net effect will be the same - BOINC will still recognise VirtualBox even if installed separately. |
|
2589)
Message boards :
Questions and problems :
Cannot connect to client
(Message 83008)
Posted 16 Nov 2017 by Richard Haselgrove Post: Charlie has had a quick look, and confirmed that there's nothing new in v7.8.4 that should be getting in the way. He has suggested trying to re-install v7.8.3 with a freshly-downloaded copy of the installer (in case of corruption of the original), though I must confess I'm sceptical about that idea - in my experience downloads either fail completely or get through 100% correct: very rarely do they get corrupted in transit these days.My suggestion is that you don't need the upgrade, and you should revert to v 7.8.3: in the meantime, we'll let Charlie know that the memory size update may need to be revisited.Thank you Richard, but the problem was already happening with v 7.8.3. I upgraded only to see if it might fix the problem. So reverting won't fix :-( I would suggest that you have a look at the adjacent thread Any known problems with OSX 10.13 APFS? - there are some useful posts in there about additional tests you could perform and information you could provide us with. Ignore the APFS in the title - the consensus seems to be that it's a red herring, and the actual problem is similar to yours. |
|
2590)
Message boards :
The Lounge :
BOINC Politics thread
(Message 83002)
Posted 16 Nov 2017 by Richard Haselgrove Post: The wife sounds fairly strong, but not in a good way - I wouldn't trust her to use her strength for the good of the rest of the country. |
|
2591)
Message boards :
The Lounge :
Needful Things
(Message 82971)
Posted 14 Nov 2017 by Richard Haselgrove Post: Interesting that when the moderation tools were created, 'obscene' was given 'top of the bill' status. Now it's spam. How society changes... |
|
2592)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 82962)
Posted 14 Nov 2017 by Richard Haselgrove Post: Just don't pay in advance is what I think now...Absolutely right. I stopped the cheque last night. |
|
2593)
Message boards :
Questions and problems :
Cannot connect to client
(Message 82929)
Posted 13 Nov 2017 by Richard Haselgrove Post: I'm afraid I'm not a Mac specialist, and we're on the bleeding edge here... So far as I know, the only reason for releasing BOINC v7.8.4 for Macs was because some ATI/AMD graphics cards were reporting negative amounts of video memory. Your HD 5750 reported "1024MB available" under BOINC v7.8.3, which sounds perfectly reasonable to me. My suggestion is that you don't need the upgrade, and you should revert to v 7.8.3: in the meantime, we'll let Charlie know that the memory size update may need to be revisited. |
|
2594)
Message boards :
Server programs :
my boinc-server project doesn't send to client a application
(Message 82861)
Posted 9 Nov 2017 by Richard Haselgrove Post: I think "Couldn't create host record in database" is the most significant. That will be a message sent from the server to be displayed to the user for debugging. If the server daemons can't write to the database, no progress can be made. There may be additional information in the server logs. |
|
2595)
Message boards :
The Lounge :
Needful Things
(Message 82849)
Posted 9 Nov 2017 by Richard Haselgrove Post: Which is very firmly either 'on' or 'off', and takes a discernable amount of time to switch between the two states? If the car happens to be moving when the message finally makes it all the way from the switch to the mechanism, your passengers sue you for whiplash. It's also very difficult to get the disengagement timing exactly right for, say, a hill start out of a junction into a gap in a line of moving traffic. I've got one on my new car, and I much preferred being able to ease the handbrake gradually with a directly-cabled lever, for smoother starts in particular.Handbrake? Oh, you mean the electronically applied and disengaged by a switch all stop fail safe. That's right, a switch not a pedal or lever.and holds the car on the footbrake.I hate those guys in front of me. I only have a foot pressing down on the clutch, nothing else. I do, however, agree that high-level brake lights tend to destroy night vision for the following driver in close traffic, and I try to disengage the foot brake as quickly as possible when the traffic has come to a complete stop. |
|
2596)
Message boards :
Web interfaces :
No Info on Cross-Project Certificates
(Message 82826)
Posted 9 Nov 2017 by Richard Haselgrove Post: You're trying to set up the user account details on a project server web-site, right? Looking at a working project, the link should be calling cert_all.php on your own server - the actual call is at https://github.com/BOINC/boinc/blob/master/html/inc/user.inc#L191 The ultimate data will be retrieved from netsoft-online, which assembles all the official cross-project data from known projects for BOINC to use centrally. But that call is made a lot further down the web include spaghetti. |
|
2597)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82807)
Posted 8 Nov 2017 by Richard Haselgrove Post: Sheep trained to recognise Barack Obama Doesn't say whether they can recognise Donald Trump, or tell the difference between them. We're going to have to be careful with that 'sheeple' joke at the next election - they might make a better job of it than we have. |
|
2598)
Message boards :
Questions and problems :
How can I recover my account authenticator from my Macintoch OS machine and reconnect the Windows 7 box to the project?
(Message 82774)
Posted 8 Nov 2017 by Richard Haselgrove Post: If your mac is attached to SETI, you will have the file. BOINC always has a data directory: the problem is how to find it. If all else fails, restart BOINC and look in the Event Log: the location will be listed around 4 lines down. Similarly, the Windows machine - once BOINC is loaded and running - will also have a data directory: the same procedure will tell you where it is. Just copy the file from the machine which has it, to the machine which hasn't got it yet. Any which way will do - local network, USB stick, even email it to yourself. Apart from the possible problem I mentioned earlier with line-ending formats, it should just pick up the settings and start crunching. It's as easy as that. |
|
2599)
Message boards :
Questions and problems :
How can I recover my account authenticator from my Macintoch OS machine and reconnect the Windows 7 box to the project?
(Message 82767)
Posted 7 Nov 2017 by Richard Haselgrove Post: The authenticator you require is contained in the file account_setiathome.berkeley.edu.xmlin your BOINC data directory - the root of the folder structure, not the separate SETI@Home project folder. For machines running the same OS, I'd simply say "copy that file to the other machine, and you'll be attached straight away, automatically". But crossing from *nix to Windows, you may hit formatting problems because of different end-of-line conventions. On a Windows machine, try opening the file in Notepad and looking for normal formatting. If there aren't any newlines, open it in Wordpad instead and save it from there - that auto-converts the line format. |
|
2600)
Message boards :
GPUs :
Dual GPUs but only one with load
(Message 82753)
Posted 7 Nov 2017 by Richard Haselgrove Post: Moo! Wrapper Windows app 1.04 (ati14) is quite old, dating from 6 Sep 2014. I was reminded of commit befb90f, from around the same time. I presume those two have worked together for the last three years: which does raise the possibility that we broke something in 7.8? |
|
2601)
Message boards :
GPUs :
Dual GPUs but only one with load
(Message 82737)
Posted 7 Nov 2017 by Richard Haselgrove Post: Could you check the command line BOINC used to start the GPU tasks? This Super User Q&A shows more than many enough ways to do that.BOINC v7.8.2 will normally pass "which device to use" instructions via init_data.xml, rather than the command line - the command line option is present and used for legacy compatibility purposes only. I don't know whether the Moo Wrapper server has been kept up-to-date to use the newer mechanism. @ Wayne, Although I don't think it's implicated in this case, BOINC v7.8.2 was buggy - I'd recommend that you update to v7.8.3 when convenient and check again. |
|
2602)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 82735)
Posted 7 Nov 2017 by Richard Haselgrove Post: Getting ready.....Ditto. Fetching to the task limit, but the server isn't co-operating. Edit - pulled 64 Arecibo VLARs out of the feeder. Hope that helps. |
|
2603)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82715)
Posted 6 Nov 2017 by Richard Haselgrove Post: She's cute. :)You might think so, I couldn't possibly comment ;-) But on previous form, and again this time, she's stupid - almost terminally so. Which is a big turn-off for me. Priti Patel apologises over undisclosed Israeli meetings She said she "regretted the lack of precision in the wording" of her previous statement about the trip.- otherwise known as being "economical with the truth". |
|
2604)
Message boards :
The Lounge :
You couldn't make it up (but some did)
(Message 82712)
Posted 6 Nov 2017 by Richard Haselgrove Post: Long, complicated, passwords are probably helpful against an extended dictionary attack (i.e. automated guesswork). But they provide no protection at all against a stolen password file. Which do you think was more likely to be used to obtain access to a Yahoo account for use as a spam sender? |
|
2605)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82701)
Posted 6 Nov 2017 by Richard Haselgrove Post: Chris, I think you have to check your references more carefully.More than 150,000 workers whose firms are signed up to the voluntary living wage rate are set to get a pay rise. It is separate from the government's compulsory National Minimum Wage (NMW) and the National Living Wage (NLW).How can any Government pay by law, wages that are below the National Living Wage? 1) That article is about commercial company wage rates, not rates paid by the Government. 2) The Government has defined, in law, that the National Living Wage is different from, and lower than, the - er - National Living Wage which had previously been defined outside Government. Yes, they can do that. But the workers concerned are going to get the externally-set rate, which is £1.25 per hour higher than the Government's cut-price version. If it only takes 30p to reach the new level, they're being paid more than the legal minimum already. The Government may be being shamed, but their laws aren't being flouted. The level of the voluntary living wage is calculated annually by the Resolution Foundation, a not-for-profit research and policy organisation. |
|
2606)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 82696)
Posted 6 Nov 2017 by Richard Haselgrove Post: OK, one alert on each board, then. We'll call it a draw :-) |
|
2607)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 82693)
Posted 6 Nov 2017 by Richard Haselgrove Post: Richard was approx 30 minutes behind me :-)I thought I watched it live and posted at the end, you caught up on iPlayer. Ahead :P Just been listening to daytime TV, which had a marvellous jump-cut from an analysis of BrightHouse's business methods to a live speech by Theresa May at the CBI - again extolling the free market as the best possible political system. That's "free" as in free to rip off the poor, free to declare profits overseas, free to relocate manufacturing in China, free to clad poor people's housing in fire accelerants. If she'd made that speech in Parliament, it would have been worth at least another 100,000 votes for Jeremy Corbyn, I reckon. It made no sense to anyone outside her own gilded ivory tower. |
|
2608)
Message boards :
Web interfaces :
SPAM Link in Cross-project statistics!
(Message 82687)
Posted 6 Nov 2017 by Richard Haselgrove Post: Team Starfire gave up running their stats page some years ago, and allowed the domain registration to lapse - you got redirected to the new owner's choice of entertainment. The Starfire link was removed from BOINC's web page template, again some years ago: SZTAKI can't have been keeping their server up to date. |
|
2609)
Message boards :
The Lounge :
You couldn't make it up (but some did)
(Message 82670)
Posted 6 Nov 2017 by Richard Haselgrove Post: Richard, unless they are in your imap outbox your account may have simply been spoofed. But take all appropriate scrubbing actions because something got your contacts list.They're not in my outbox. Because I use webmail, the contact list is stored on the server, and is accessible by anyone with access to the hacked password. A warning for anyone who uses a cloud email service. |
|
2610)
Message boards :
The Lounge :
You couldn't make it up (but some did)
(Message 82665)
Posted 5 Nov 2017 by Richard Haselgrove Post: Well, it's finally happened. After 16 years, my email account has finally been hacked - I must have been saying something interesting, to someone in Kazakhstan. Ignore any emails with links in the last two hours, if you're boring enough to be on my contacts list. |
|
2611)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82662)
Posted 5 Nov 2017 by Richard Haselgrove Post: Well, I'm more interested in what "The Great and the Good" (as they used to call themselves) have been up to, but a bit of light entertainment from the BBC will give us time to catch our breaths. BBC News seem to have cancelled all leave for the weekend - nice to see someone putting a bomb under the Tory party on bonfire night - but the key feature is that this is just the British highlights of an international co-production. It will be interesting to see what other countries pick out for their showreels. Will also be fun to see how Trump handles Wilbur Ross and the Russians from Japan. |
|
2612)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82659)
Posted 5 Nov 2017 by Richard Haselgrove Post: Now available on iPlayer |
|
2613)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82656)
Posted 5 Nov 2017 by Richard Haselgrove Post: Ooh, this is going to be fun. The Queen, the Conservative Party, and President Trump, all in the first 30 minutes. Another hour tomorrow. Paradise Papers: Tax haven secrets of ultra-rich exposed |
|
2614)
Message boards :
The Lounge :
Needful Things
(Message 82629)
Posted 4 Nov 2017 by Richard Haselgrove Post: the engine switches of at traffic lights and then restarts when the gas pedal is pressed.My new car has that too - well, the cheap version, where the engine only switches off if you put the gearbox in neutral and take your foot completely off the clutch pedal. That's a bit clunky, so I tend only to use it at canal swing bridges, railway level crossings, and extremely complicated multi-phase junctions. It's very pleasant to sit there in peace with no strain on the clutch leg. If your battery is flat, or your starter motor has failed, you wouldn't have set out on the journey, would you? If it's the car's final journey to the scrapyard, you can turn off the 'eco' feature. |
|
2615)
Message boards :
The Lounge :
Needful Things
(Message 82626)
Posted 4 Nov 2017 by Richard Haselgrove Post: It's required on every car sold in the EU since January 2017.I bought a new car this year (ordered as a factory build in February, supplied in May), so I suppose I've got it, according to that rule (we are still in the EU for the time being, after all). But I don't think I do. What I do have is short-range parking sensors, which bleep and show on the dashboard which corner of the car is in danger of hitting something. 'Green' is no more than a metre away, 'Red' perhaps 10cm. It's very useful for snuggling up close to stone walls in my courtyard without damaging the paintwork, but it's never distracted me while driving, even in heavy traffic. |
|
2616)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 82604)
Posted 4 Nov 2017 by Richard Haselgrove Post: Yes, I heard Eric say that too - and it was Eric Korpela, the image was on the SETI@Home forum, and that dealing with the lawyers wasted time he didn't have. He didn't say who had posted the image, though. |
|
2617)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 82594)
Posted 4 Nov 2017 by Richard Haselgrove Post: A very good question. I'd guess, most often, some other fool was persuaded to give it to them.Fools & their money are soon partedSo how the devil did they get together in the first place? |
|
2618)
Message boards :
The Lounge :
Needful Things
(Message 82517)
Posted 2 Nov 2017 by Richard Haselgrove Post: No probs Richard, Mac stuff is way beyond me :-)And beyond TBar, by the looks of things. |
|
2619)
Message boards :
The Lounge :
Needful Things
(Message 82514)
Posted 2 Nov 2017 by Richard Haselgrove Post: Apparently all mushrooms are fungi, but not all fungi are mushrooms.Yes, that's the fundamental basis of the modern Taxonomic classification for all life on earth - not just mushrooms, and not just in Yorkshire. See the sidebar alongside the section headed 'Kingdoms'. (Sorry, you just caught the backwash from an argument with TBar at SETI Main) |
|
2620)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82508)
Posted 2 Nov 2017 by Richard Haselgrove Post: I saw the resignation, and the statement, slightly differently, and it left me puzzled. The single incident we know about in Michael Fallon's past is relatively minor on the scale of sleaze, and has been openly 'forgiven and forgotten' by the female journalist concerned. We've all seen ministers clutching at straws and clinging to office after far worse. Also, the statement was very dignified and appropriate - almost old-fashioned. The whole combination left me feeling that there is more to come. Perhaps?
We'll have to wait and see. The next 18 months are going to be even more interesting, and I guess the chances of another general election in that time have gone up, rather than down. |
|
2621)
Message boards :
The Lounge :
Needful Things
(Message 82507)
Posted 2 Nov 2017 by Richard Haselgrove Post: Well, all I know is that having been to West Yorkshire, that I never saw any Mushrooms of any sort, magic or not.Well, fungi do grow in West Yorkshire. This was taken during my uncle's visit last year.  Apparently they're known as 'Earthstars', so they fit this board's various interests rather well. |
|
2622)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82501)
Posted 1 Nov 2017 by Richard Haselgrove Post: Well, one of them has gone. Reshuffle time. |
|
2623)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82488)
Posted 1 Nov 2017 by Richard Haselgrove Post: From what I remember Labour did their best to bankrupt the country last time they were in government they may not of started out to but they did , just a inconvenient fact people seam to forget .Shouldn't that one have been posted in the 'Myths and conspiracy theories' thread? From what I remember, there was a network of Building Societies, mainly based in the north and west of England, outside the cosy financial cliques of London. They mostly handled small-scale, simple, stuff - personal savings and mortgages. Some of them had existed for 100 years or more. During the 1990s (actually starting with Abbey National in 1989), a group of 'carpetbaggers' joined a number of societies and tried to persuade the other members/owners of the societies to 'de-mutualise'. This process was made possible by the Building Societies Act of 1986 - remember who was in power then? The de-mutualised building societies became banks owned by shareholders - their former members - who mostly sold their shares to other banks for a quick profit. The new banks tried to join the 'big boys' in the London and international money-go-round, and were quickly spotted as the naive suckers they were, and were taken for a ride - sometimes taking their new partner-owners with them. Remember Nat West, Bank of Scotland - and of course Northern Rock, which started the collapse of the whole house of cards in the UK? The same thing was happening in the United States, only bigger and brasher. We got caught up in that, too. The whole nest of crooked gamblers in the banking industry imploded while Gordon Brown and Alistair Darling were in charge, and - at the time - they were widely praised for the way they handled a crisis not of their making. They kept the banks open - no ordinary customers lost money. (The new shareholders in Northern Rock - many of them former employees - did lose money and are still owed it, but that's a small part of the story) The myth that Gordon Brown and Alistair Darling had caused the crisis was put about by the Tory party and the right-wing press at the subsequent 2010 election, and sadly the Labour party didn't insist on fact-checking. I agree that the Labour party made some big financial mistakes during 1997-2010. One of them was allowing betting shops to expand and put FOBTs in every run-down high street (gambling became the opiate of the masses, alongside cheap booze). Another was building schools and hospitals on the never-never under PFI, where again Labour was taken for a ride by the financiers and lawyers who drew up the contracts for continuing repayments and inflated running costs. And the PFI contractors hired some pretty dodgy builders too, as Scottish schoolchildren have found out. But at least Labour did build public facilities like schools and hospitals, unlike this lot. |
|
2624)
Message boards :
News :
Client version 7.8.3 released
(Message 82468)
Posted 31 Oct 2017 by Richard Haselgrove Post: I've completed my full scan with updated McAfee, and no malware whatsoever was found or reported. Clean bill of health, including BOINC v7.8.3 and attached projects. |
|
2625)
Message boards :
News :
Client version 7.8.3 released
(Message 82447)
Posted 31 Oct 2017 by Richard Haselgrove Post: I have a machine with a fully-licensed McAfee subscription, also running v7.8.3: I'm doing a forced update now. I'll let you know how I get on when it finishes. This is taking a while: the update ran for 35 minutes with 0% progress, before I killed it. After a reboot, it ran better: I'm now 36% of the way through a full scan. My BOINC programs are installed in C:\BOINC, and we're well past that point - I watched the scanner go through that folder, and it found nothing. I'm using version 20.4, build 20.4.168, engine version 3152.0, updated/created today. I'll post a full report when its finished, but it's beginning to look as if: a) you might have a corrupted McAfee installation, or b) the BOINC files might have got infected on the way to your machine. My copy was downloaded direct from Berkeley on 04 October 2017. |
|
2626)
Message boards :
News :
Client version 7.8.3 released
(Message 82445)
Posted 31 Oct 2017 by Richard Haselgrove Post: Using an alternative service to display the image that 2tone was trying to show us. 
|
|
2627)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 82440)
Posted 31 Oct 2017 by Richard Haselgrove Post: I hope you're sure that your client isn't a vampire. |
|
2628)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82407)
Posted 30 Oct 2017 by Richard Haselgrove Post: ...even he cannot uniaterally order a nuclear strike single handedly on N Korea.Sadly, I don't think that's true. Two quotes from this Boston Globe article dated 08 August this year (I'm quoting because there is limited public access before the paywall kicks in): Trump has “unchecked authority to order the use of conventional or nuclear weapons against North Korea,” Bruce G. Blair, a nuclear security expert at Princeton University’s Program on Science and Global Security, said in an e-mail. “I believe that both options are being prepared right now.”The position seems to be that Trump can issue the order by himself, with no checks and balances whatsoever. But he can't fly the aeroplane, or press the launch button, or whatever, by himself. So, the question becomes - will anybody lower down the chain of consequences refuse to obey the order? Will anybody risk what is almost certainly a career-ending, court martial, act? Will the sort of servicemen (or women) selected for positions in that chain of command have the mindset to over-rule their Commander-in-Chief? We know Stanislav Petrov did something like that from the other side, although there was no direct order. How do you think the US service personnel will react? |
|
2629)
Message boards :
The Lounge :
Needful Things
(Message 82389)
Posted 30 Oct 2017 by Richard Haselgrove Post: The device it's connected to by any chance?Partly that, and also the ring-main wiring embedded in your walls. If that starts smouldering because of extended over-current (possibly aided and abetted by an elderly high-resistance joint - i.e. a hotspot), things get very nasty. |
|
2630)
Message boards :
The Lounge :
Needful Things
(Message 82329)
Posted 29 Oct 2017 by Richard Haselgrove Post: And every time you bought a cinema ticket for one of her films, or bought one of her books (I forget - deliberately - exactly what she was famous for), you subsidised her concorde lifestyle. |
|
2631)
Message boards :
The Lounge :
Needful Things
(Message 82325)
Posted 29 Oct 2017 by Richard Haselgrove Post: It was becoming uneconomical to run at the time anyway.It was operationally uneconomical from day one, and only survived as long as it did because the development costs had been subsidised by the two governments, the ongoing political breast-beating, and the sky-high fares paid by the vainglorious few (or their customers). The Sunday Telegraph |
|
2632)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 82323)
Posted 29 Oct 2017 by Richard Haselgrove Post: SETI is about the scientific search for extra-terrestrial intelligence (that's what it says on the tin), not about leader boards.And it's nothing to do with SETI:It is when it slews the leader boards! |
|
2633)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 82321)
Posted 29 Oct 2017 by Richard Haselgrove Post: Chris dear, please calm down. Yes, it's the gridcoin lot, and yes, it's a lot of them sharing one account, not just one guy. And it's nothing to do with SETI: they work across the spectrum of BOINC projects. It's been discussed on this board already. If it's anybody's fault, it's the Western capitalist financial system which has brainwashed everybody into seeing every interpersonal transaction in purely financial terms. "Freedom" (so passionately advocated by the West) includes the freedom for individuals to earn money by crunching if somebody offers them that facility. The best of luck in swimming against the tide if you want to pursue that one. |
|
2634)
Message boards :
The Lounge :
Needful Things
(Message 82317)
Posted 29 Oct 2017 by Richard Haselgrove Post: The local regional airport is about 10 miles east of my house, and because of the same prevailing winds, most of the regular fare of holiday charter jets heads directly towards me. Most take a left turn and head south towards the sun, some continue straight on (to Dublin, perhaps?) and pass near enough directly overhead. In its later years, when it was clear that Concorde wouldn't be expanding its route list of scheduled passenger services, the spare planes were used for those short excursion flights and regularly visited LBA. They were the only planes I could ever hear from my downstairs workroom, rush along the corridor and up the stairs, and still be in time to see fly overhead and off towards the Irish Sea. They were beautiful, but boy were they loud. |
|
2635)
Message boards :
Questions and problems :
Transfers are not shown in the Transfer tab
(Message 82277)
Posted 26 Oct 2017 by Richard Haselgrove Post: That was a lucky guess on my part! No, the historic lists only appear via the message log: that's viewable for the current session (since last re-start) only. But data for previous sessions is accessible via the files 'stdoutdae.txt' and 'stdoutdae.old' in your BOINC data directory. If you're interested in history, you can also look at the various 'job_log_<project name>.txt' files, also in your data directory: they keep a record of the tasks you have completed. |
|
2636)
Message boards :
Questions and problems :
update to 7.8.3 service problem
(Message 82276)
Posted 26 Oct 2017 by Richard Haselgrove Post: Yes, there are many ways to fix a broken computer. Among them are: * Go and have a nice cup of tea, and try again later. * Switch it off and switch it on again. * Have a long sulk - preferably overnight - and try again in the morning. * Spend hours explaining your problem to somebody else, even if they aren't listening. I'm glad at least one of them worked for you ;-) |
|
2637)
Message boards :
Questions and problems :
BOINC crashes my linux machine
(Message 82274)
Posted 26 Oct 2017 by Richard Haselgrove Post: Post any error messages or logs? |
|
2638)
Message boards :
Questions and problems :
update to 7.8.3 service problem
(Message 82267)
Posted 26 Oct 2017 by Richard Haselgrove Post: I'm guessing this is on Windows: which version of Windows? And if Windows 10, which build or sub-version? Have you tried right-clicking on the installer file in your download directory, and selecting 'Run as Administrator' from the context menu? Edit - if this is a Windows Server, specifically a Domain Controller (so Small Business Server and up), you may find that installing 'as as service' is impossible. We used not to allow the installer to run at all: now that restriction has been lifted, but Microsoft still doesn't allow our installer to create system services on a domain controller. So BOINC can only be installed in User mode on DCs - hardly ideal, but better then nothing. |
|
2639)
Message boards :
Questions and problems :
Transfers are not shown in the Transfer tab
(Message 82265)
Posted 26 Oct 2017 by Richard Haselgrove Post: Thu Oct 26 08:39:32 2017 | Rosetta@home | Started upload of rb_10_22_78445_121028__t000__3_C1_SAVE_ALL_OUT_IGNORE_THE_REST_524101_105_0_r948990948_0I think that's the longest one, and it only took 17 seconds. Transfers only show in BOINC Manager while they are actively in progress - you would only see that one happening if you happened to be looking at the transfers tab during that 17 second interval. Most of the others took 4 seconds or even 2 seconds - that's a 'flash and it's gone'. Are you saying that you have never, ever, seen a transfer visible on the transfers tab, or were you expecting (perhaps) to see a historic list of the transfers which have happened previously? |
|
2640)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82259)
Posted 26 Oct 2017 by Richard Haselgrove Post: Southend hospital set to pilot Airbnb-style scheme. I've just sent the BBC this message: Does anyone in Essex have a spare room any more? I thought it was Government policy to phase them out under the so-called 'bedroom tax' - certainly not to pay their owners a subsidy. |
|
2641)
Message boards :
News :
Client version 7.8.3 released
(Message 82257)
Posted 26 Oct 2017 by Richard Haselgrove Post: Both of those logs relate to project science applications, rather than BOINC itself - WCGRID_SCC1_VINA_7.08_WINDOWS_X86_64 from World Community Grid and MINIROSETTA_3.78_WINDOWS_INTELX86.EXE from Bakerlab. As far as I know, there's been no change in the way BOINC launches application binaries in the new version. Do you have any confirmation logs which point directly to BOINC being the culprit, rather than the project files? For what it's worth, McAfee Adaptive Threat Protection (ATP) "analyzes content from your enterprise and decides what to do based on file reputation, rules, and reputation thresholds." - in other words, it's not saying these 'are' viruses, it's saying 'these files might be bahaving a bit like viruses, and we don't know enough about them to be certain'. These are known as 'false positives', and usually clear up after a few days after enough people complain to the virus company in question and persuade them to check the programs more carefully. You can help the process along by passing your report direct to McAfee. |
|
2642)
Message boards :
Questions and problems :
Notices tab sends me to localhost
(Message 82237)
Posted 25 Oct 2017 by Richard Haselgrove Post: Are you running the latest recommended version of BOINC for Mac OS X - version 7.8.3? I think they fixed that (it was an OS change) - and if you use the screensaver module, I think they fixed that too. |
|
2643)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82178)
Posted 23 Oct 2017 by Richard Haselgrove Post: And this country's in trouble. QED. |
|
2644)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82176)
Posted 23 Oct 2017 by Richard Haselgrove Post: If a major investor pulls out the company is in dire straits.You could equally say that if a major company pulls out of a country (or pulls its tax out of a country), then that country is in dire straits. |
|
2645)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82170)
Posted 23 Oct 2017 by Richard Haselgrove Post: BBC Newspaper review wrote: Elsewhere on Monday, business leaders take to the Financial Times to "sharply criticise the state of capitalism".Unfortunately, the article is behind a paywall (there's capitalism for you), so I can't link it. Oh, OK then. 
|
|
2646)
Message boards :
The Lounge :
BOINC Politics thread
(Message 82152)
Posted 22 Oct 2017 by Richard Haselgrove Post: Back to real-world politics: the appointment has been cancelled. The question is - who the hell thought it might have been even remotely appropriate in the first place, and why? Did money change hands? (Everything seems to involve money these days) |
|
2647)
Message boards :
Questions and problems :
Can't update BOINC
(Message 82126)
Posted 20 Oct 2017 by Richard Haselgrove Post: What operating system? |
|
2648)
Message boards :
Questions and problems :
Install Error Issues - Please help
(Message 82115)
Posted 19 Oct 2017 by Richard Haselgrove Post: Very difficult to advise you as so many things have changed, so many attempts have been made, so many details are vague. But my approach would be: Registry - did you clean both HKEY_CURRENT_USER and HKEY_LOCAL_MACHINE? BOINC has entries in both - references to installation drives are in HKEY_LOCAL_MACHINE. Failure to uninstall is usually due to a missing boinc.msi for the OLD version. Most easily solved - after sorting out the registry, see line above - by re-installing the OLD version to another location. If the old drive is completely dead, you'll get a new clean installation: most of my drive failures have still allowed data to be recovered via an external caddy and recovery tools, so placing the BOINC data directory on the temporary drive enables recovery of history, current work etc. Then, once you have control of the system, place the data directory in its final location on the replacement drive. Run the BOINC installer for the version you're planning to run, using the 'advanced' screen to point it to the data directory in its new location. And off you go. |
|
2649)
Message boards :
Questions and problems :
Lost multi-GPU BOINC capability with Windows 10 Fall Creators Update
(Message 82111)
Posted 19 Oct 2017 by Richard Haselgrove Post: There is one alternative possibility: that you have coincidentally suffered a hardware failure of the primary slot. I'd suggest a brief test with the currently-working 750 Ti card (solo) in the primary slot. |
|
2650)
Message boards :
Questions and problems :
Boinc.msi install error
(Message 82108)
Posted 19 Oct 2017 by Richard Haselgrove Post: It is more likely that Windows is trying to find and use the boinc.msi file associated with v7.6.33, which it will use to un-install the previous version. The simplest work-round, if that file has been removed from the Windows installer cache by a clean-up operation, is to re-install the old version and then try the upgrade again. If you don't have your original downloaded installer file, you can still download a fresh copy from http://boinc.berkeley.edu/download_all.php. Make sure you choose the correct (64-bit or 32-bit) version for both installations. |
|
2651)
Message boards :
Questions and problems :
Lost multi-GPU BOINC capability with Windows 10 Fall Creators Update
(Message 82099)
Posted 19 Oct 2017 by Richard Haselgrove Post: Your given log says something different:Note that that part of the log refers to app version, not to any current tasks. Once you have run a given application type - even just once as a test - the <app_version> record for that application, and the corresponding downloaded files, remain in client_state.xml and the project folder respectively - for ever. (unless you do a 'reset project') I thought there was an issue on Github (it might even have been originated by the Einstein project) requesting a mechanism for removing stale app_version records from client hosts at server request, but I can't find it just now. Edit - I've just looked at my oldest working machine, and client_state.xml lists 22 different <app_version> records for Einstein, and the project directory lists 30 application binary files totalling 445 MB - that includes several graphics app files, with the oldest binaries dating back to 2009. I'm not even running Einstein on that machine at the moment! |
|
2652)
Message boards :
API :
BOINC Wrapper: Output file absent (Computation Error)
(Message 82094)
Posted 19 Oct 2017 by Richard Haselgrove Post: 'Output file [is] absent' is usually a symptom, rather than the cause of a problem: the application has crashed before it has a chance to do any science, so the scientific output never gets created. The wrapper has exited with status 0x8b, or 139. From the error number list, that might be #define ERR_GETS -139
// gets() or fgets()but I'm afraid that's as far as my knowledge goes.
|
|
2653)
Message boards :
BOINC Manager :
Improvement for a next release of Boinc
(Message 82084)
Posted 18 Oct 2017 by Richard Haselgrove Post: hello, for what I want to do, it's not possible with the xml file that's why I ask if it's possible to make this "improvment" :)Yes, you're correct: it isn't possible to do what you wish via the xml file, and it isn't possible with any current version of BOINC - which doesn't mean it would never be possible - it just needs some better programming. For what it's worth, this enhancement is the most commonly-requested of user enhancements: it first surfaced in 2006, in what is now called Issue #41 (that was before GPUs were added to the computing capability of BOINC, which is why it only mentions CPUs - but if we could get it implemented, I'm sure the demand to include GPUs as well would be irresistible.) Other people have also tried - Issue #202, Issue #303, Issue #935 etc., but so far all requests have fallen on deaf ears. |
|
2654)
Message boards :
The Lounge :
Needful Things
(Message 82083)
Posted 18 Oct 2017 by Richard Haselgrove Post: The good news was that London and the SE missed all the bad weather. The even better news is that The State pension is going up next year.Yes. Only Tory-voting pensioners are protected from having to pay the bankers back for their misdemeanours. |
|
2655)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 82064)
Posted 17 Oct 2017 by Richard Haselgrove Post: Pictures (in the early days of Kodachrome) were rarer and more precious. We came back from that American trip with what turned into a single box of 65 slides, which we transferred to high-quality glass mounts - I can also remember 'helping' my mother do that on the dining room table.... Salvaging that box was a high priority when I had to break up my mother's house for sale, after she died two years ago. It wasn't far from the surface of the archaeology :-) |
|
2656)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 82059)
Posted 17 Oct 2017 by Richard Haselgrove Post: Likely Vetter Mountain Lookout, but sorry to report the Station Fire got it in 2009.Yes, you were right. Found the box of 60 y.o. slides...  
|
|
2657)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 82046)
Posted 17 Oct 2017 by Richard Haselgrove Post: Too close http://abc7.com/brush-fire-burns-near-mount-wilson-observatory/2542698/One of the few places I have a certain memory of, from my family visit to CA in 1956/57 (age 4½). Apart from a tour inside the observatory itself - my father was on sabbatical at CalTech, so we may have had special access - I remember being taken to visit a fire-watcher's tower in the hills. Something I'd never seen, or had need of, before: our British hills tend to attract enough rain not to require constant watching. |
|
2658)
Message boards :
News :
Client version 7.8.3 released
(Message 82020)
Posted 16 Oct 2017 by Richard Haselgrove Post: Please note that this version is not yet recommended for users of the Spanish translation. Traducir google wrote: Tenga en cuenta que esta versión aún no se recomienda para los usuarios de la traducción al español. |
|
2659)
Message boards :
News :
Client version 7.8.2 released
(Message 82009)
Posted 15 Oct 2017 by Richard Haselgrove Post: I would, however, suggest that you wait until at least v7.8.3 is officially 'recommended', which should be "real soon now". v7.8.2 has several significant issues. v7.8.3 has ... fewer. Unless you wish to run the Spanish language version - in that case, wait until at least v7.8.4, or watch for announcements. |
|
2660)
Message boards :
Projects :
News on Project Outages
(Message 81882)
Posted 9 Oct 2017 by Richard Haselgrove Post: Einstein is downOliver Bock wrote: The core switch of our host university's network failed yesterday evening, forcing us off-line. Services seem to be restored now.And they are. |
|
2661)
Message boards :
The Lounge :
BOINC Politics thread
(Message 81871)
Posted 8 Oct 2017 by Richard Haselgrove Post: In the meantime by their own admission they are dyed in the wool left wing Labour with an on off love affair with the Lib Dems...I think the Guardian itself would hope that there was a closer link between evidence and conclusion. Where did that 'wing' come from? |
|
2662)
Message boards :
The Lounge :
BOINC Politics thread
(Message 81863)
Posted 7 Oct 2017 by Richard Haselgrove Post: GuardianIn addition, follow the link to the Scott Trust Limited, and consider the difference from publishing empires controlled by Lords Beaverbrook and Dacre, Rupert Murdoch, Richard Desmond etc. |
|
2663)
Message boards :
The Lounge :
BOINC Politics thread
(Message 81860)
Posted 7 Oct 2017 by Richard Haselgrove Post: Seen by whom, one wonders? This reader sees it as a paper that doesn't follow an establishment (or proprietorial) editorial line, and one which publishes free-thinking articles from a variety of points of view. |
|
2664)
Message boards :
News :
BOINC Workshop 2017
(Message 81849)
Posted 6 Oct 2017 by Richard Haselgrove Post: Where can I find the Google Drive of the talks? Would be awesome to watch them. Thanks!Start from Message 80830 The google drive link is in that post, and you can follow the talks in sequence from my notes. (Edit - they're audio only, so I'm afraid you'll have to listen, not watch) |
|
2665)
Message boards :
BOINC Manager :
Trivial issue I noticed.
(Message 81839)
Posted 6 Oct 2017 by Richard Haselgrove Post: <report_results_immediately>0|1</report_results_immediately>After the intrusive and badly formatted <proxy_info> block (9 lines plus terminator), the <options> list should continue with the 'R' section of the alphabet - <rec_half_life_days> is next on my version. The file ends normally with proper XML termination <vbox_window>0</vbox_window>
</options>
</cc_config>
|
|
2666)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81835)
Posted 6 Oct 2017 by Richard Haselgrove Post: Charlie Fenton wrote a long email to the boinc_projects mailing list yesterday, explaining why changes are needed, how they're supposed to work, and what needs to be done. In particular, he covered your point: he has designed the new graphics API calls to be fully backwards compatible. A single graphics app built with the new tools will work with both old-style and new-style Mac screensaver controllers. But as always, the proof will be in the testing... The email is archived at Update your project graphics for Mac OS 10.13, but that link is only accessible to current list subscribers. |
|
2667)
Message boards :
BOINC Manager :
Trivial issue I noticed.
(Message 81834)
Posted 6 Oct 2017 by Richard Haselgrove Post: The feature exists and is fully operational: it is documented on the client configuration page of the user manual: <report_results_immediately>0|1</report_results_immediately>Reading that back, I see the (outdated) reference to a one-day delay: I'll change that. I think the 60-second delay has also been removed, but I'll check further before changing it. The only features which do not exist are the per-project settings, and a GUI control for setting it. Given the general advice to avoid the setting, I expect that the policy of setting the feature to 'off' by default will remain in place. You need to think of BOINC as an integrated project, combining both the client that you run, and the servers that manage the work supply and return. Sometimes, decisions are made which favour the server side of the balance, and assist project administrators who have to purchase, maintain and manage those servers. This seems to be a case in point: opening a project database, locating task and workunit records, and updating them is a computationally expensive task which is most efficiently performed in batch mode (Rom Walton once wrote a detailed blog post explaining the costs, but I don't have a link handy). |
|
2668)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81830)
Posted 6 Oct 2017 by Richard Haselgrove Post: Version 7.8.3 is now available for testing. This should allow BOINC to run with the failed CPDN tasks in situ, and also allow the Mac screensaver to show project graphics again once projects have updated their apps. If anyone still has a copy of BOINC which won't start, please test and report back. Download the new version from http://boinc.berkeley.edu/download_all.php |
|
2669)
Message boards :
Questions and problems :
I see credit listed for completed tasks but Recent Average Credit stays at zero
(Message 81829)
Posted 5 Oct 2017 by Richard Haselgrove Post: See the many, many threads at CPDN concerning their bespoke credit-granting script and how it's having problems keeping up with the increasing number of historical data records in their database. |
|
2670)
Message boards :
BOINC client :
Possible 7.8.2 error? - see inside
(Message 81827)
Posted 5 Oct 2017 by Richard Haselgrove Post: No, I don't think that's a likely explanation. You know the Milkyway application better than I do: you might like to take a look at one of the valid 17 KB files yourself - it did seem rather repetitive to me, but I didn't have the patience to count to five... I did line up the two files at the bottom, and the structure seemed identical back up to what seemed like a normal start (valid case), intrusive 'called boinc finish (0)' (invalid case). That's what made me suspect a complete task loop, rather than any form of sub-tasking. Edit - when you get a chance to test and report using v7.8.3, could you please include the segment of the client Event Log which covers the processing of the test task. You'll be familiar with the BOINC scenario "Task %s exited with zero status but no 'finished' file",
"If this happens repeatedly you may need to reset the project."- I would wonder whether this might have appeared around the time of the failures.
|
|
2671)
Message boards :
BOINC client :
Possible 7.8.2 error? - see inside
(Message 81823)
Posted 5 Oct 2017 by Richard Haselgrove Post: OK, I've looked at an invalid task, and I've found a valid task run with the same application with roughly the same runtime - and yes, that proved to have been run on the GT 430 under BOINC v7.6.33. So, it's plausibly a valid comparison. I extracted and saved stderr_txt from each task. The first thing to note is that the file for the invalid task is 64 KB (and shows signs that this is the final 64 KB of a longer file - it's truncated): the valid file is 17 KB. Secondly, the invalid file contains the phrase 'called boinc_finish(0)' five times: the valid file contains it only once, at the very end. So, I conclude that the application is looping. This could be because of the known, reported, bug in v7.8.2, where a slot directory is not fully cleansed after use. This bug has been fixed in v7.8.3, so I don't intend to look any further until test results are available for v7.8.3 - please let us know when you have some. To save information for future investigation, we're talking about MilkyWay@Home v1.46 (opencl_nvidia_101) on host 616064 |
|
2672)
Message boards :
Questions and problems :
output file absent and computation error
(Message 81812)
Posted 5 Oct 2017 by Richard Haselgrove Post: First, forget about 'output files absent'. If the science application crashes, it produces no scientific output. That is inevitable, and not diagnostic. Yes, stderr_txt is wiped from the client computer after the crash, but it can be found in two places: a) Before the result is reported to the server, it can be found embedded in client_state.xml b) After the result has been reported and acknowledged, it can be seen in the task record on the server |
|
2673)
Message boards :
Questions and problems :
Cannot find slots directory under windows7
(Message 81797)
Posted 4 Oct 2017 by Richard Haselgrove Post: Juha's reply immediately prior to your post still applies. If anything was written to stderr before the application crashed, it will be sent to the server as part of the task report, and can be viewed from the task list. It's cleaned off the client machine and can't be viewed there. Although BOINC applications are designed to run without a display window (with any output directed to a file), I sometimes find that trying to run an application directly on the target machine manually - either in a command window, or simply by double-clicking it - provides clues to the problem, like the name of a missing DLL. |
|
2674)
Message boards :
Questions and problems :
Cannot find slots directory under windows7
(Message 81773)
Posted 4 Oct 2017 by Richard Haselgrove Post: Are you able to see the BOINC data directory? The location is listed among the startup messages in the BOINC Event Log - near the top, IIRC. The default location for the data directory, in C:\ProgramData, is a hidden directory, but you can see the contents by pasting the location into the address bar in Windows Explorer. The slot directories are in a 'slots' sub-directory of the data directory. If there is no slot directory there, then I would presume that the application failed to launch and there probably won't be a stderr_txt either. |
|
2675)
Message boards :
BOINC Manager :
Trivial issue I noticed.
(Message 81772)
Posted 4 Oct 2017 by Richard Haselgrove Post: There is another, relatively new, feature in BOINC. If you set 'No new tasks' for a project, that automatically sets 'report results immediately' for that project. This was done to ensure that completed work was reported in a timely fashion if, for example, the user was preparing to go on vacation, or to retire a computer. Any work in the queue that was run and completed before the holiday/retirement would be reported before the machine was shut down. |
|
2676)
Message boards :
BOINC Manager :
Trivial issue I noticed.
(Message 81765)
Posted 4 Oct 2017 by Richard Haselgrove Post: GPUGrid does that too. You'll probably see the tag <report_immediately/>in the <result> section for their workunits - in other words, it's an option set on their server and sent out with the tasks. |
|
2677)
Message boards :
The Lounge :
BOINC Politics thread
(Message 81752)
Posted 3 Oct 2017 by Richard Haselgrove Post: A current case in point: Tech giants sorry for false news about Las Vegas gunman Google and Facebook have apologised after their algorithms led to the promotion of inaccurate information about the Las Vegas shooting.Another area where tech algorithms are active - some would say over-active - is alleged copyright violation. Copyright sometimes to have a higher priority than terrorism or sexual exploitation when the role of 'publisher' is questioned. |
|
2678)
Message boards :
Questions and problems :
BUG REPORT: 100%CPU usage ofboinc.exe when --no_gui_rpc switch specified
(Message 81700)
Posted 2 Oct 2017 by Richard Haselgrove Post: David found what ailed this and is putting a fix into a PR.Good. But that's one I would leave until v7.10.x, and let the real programmers have a good look at it. |
|
2679)
Message boards :
BOINC Manager :
BOINC Manager - Windows version 7.8.2 (64) -Column Sort
(Message 81694)
Posted 2 Oct 2017 by Richard Haselgrove Post: That's odd. I'm running a v7.9.0 built from master, with all current patches included - and column sort works as designed on both tabs. Now I have to delve into the registry again to reset my preferred 'native' sort order on those tabs :-D I say it's odd, because so far as I know nobody else has previously reported this as a problem, and there are no recent patches - i.e. since mid-June, when the v7.8.x release code was separated from master - designed to address any difficulties in this area. There is growing pressure to get another update ready soon, so we should test for that. Since this is a Manager-only query, I'll move the post into the dedicated message board area, and see if we can encourage others to test. |
|
2680)
Message boards :
Questions and problems :
Projects: "... needs more disk space. You currently have 0.00 MB available" [Linux]
(Message 81691)
Posted 2 Oct 2017 by Richard Haselgrove Post: A confusing cause of this error message can be that the boinc data directory is not write & execufe enabled for the user. Guess how I found that one :-(That sounds to me like a general stylistic error in BOINC code. We ought to add to the 'BOINC coding style' manual that when iterating through a cascade of possible error conditions, the final one should always be "Aw shucks - something unexpectedly bad happened", rather than a specific statement like 'zero disk space available'. A similar condition happens when a Windows Manager fails to connect to a client. It always falls through to 'incorrect password' when something unlisted happens. And having added it to the coding style, we have to find and eradicate where the bad practice occurs. At the next full code review. Rob, could you verify by experiment that the problem is still as you describe - setting the directory read only causes the 'no disk space' error to show? (i doubt no-execute applies in this case, but might still be worth checking). If verified, I'll raise it as an 'Issue'. |
|
2681)
Message boards :
Questions and problems :
BUG REPORT: 100%CPU usage ofboinc.exe when --no_gui_rpc switch specified
(Message 81673)
Posted 1 Oct 2017 by Richard Haselgrove Post: 0d2a0ad9a3124776c45458f83fddfe75b0924117 davidpanderson committed on Jul 23, 2005 + - Core client: add "-no_gui_rpc" cmdline option. + If present, don't create listening socket for GUI RPC. + (Useful if bind() doesn't work).See how that compares with what David remembers. |
|
2682)
Message boards :
The Lounge :
BOINC Politics thread
(Message 81663)
Posted 1 Oct 2017 by Richard Haselgrove Post: In a democracy, leaders cannot tell untruths with impunityI'm sure ours do. |
|
2683)
Message boards :
BOINC Manager :
Trivial issue I noticed.
(Message 81660)
Posted 1 Oct 2017 by Richard Haselgrove Post: Actually, which system does matter. On the larger projects like CPDN and Einstein, the upload server (where the files go) and the database server (where the housekeeping is done - keeping track of tasks, processing time, and credit) may be on different machines - sometimes even on different continents. Because the transactions are so different, there's no way of 'combining' them as you suggest. |
|
2684)
Message boards :
BOINC Manager :
Trivial issue I noticed.
(Message 81646)
Posted 30 Sep 2017 by Richard Haselgrove Post: Many years ago, the delay between 'ready to report' and actual reporting was a maximum of 24 hours. In recent versions of BOINC, that's been brought down to 1 hour maximum. There are many other triggers for a possible earlier report (like the need for new work), so you shouldn't delay the science unduly by not report reporting manually. |
|
2685)
Message boards :
BOINC Manager :
Trivial issue I noticed.
(Message 81633)
Posted 29 Sep 2017 by Richard Haselgrove Post: Sorry, sometimes my attempts at English humour don't translate too well. I should know that by now. |
|
2686)
Message boards :
BOINC Manager :
Trivial issue I noticed.
(Message 81631)
Posted 29 Sep 2017 by Richard Haselgrove Post: Assuming you're using Windows: In Windows, highlighting is applied to the control that has 'focus'. When you select a project, the project list control has focus, and you see highlighting. When you click the 'update' button, the button takes the focus and the list loses focus - so the highlighting is removed (by Windows), although the project remains selected in the list. When you switch from the projects tab to any other tab, and back again, BOINC sets the focus to the list control - and Windows highlights the selected row. That's how Windows works these days. Complain to Bill Gates, not us. |
|
2687)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81621)
Posted 28 Sep 2017 by Richard Haselgrove Post: The Mac developer is anxious to make a new version of BOINC for Mac because Apple has changed the way the screensaver works in an operating system revision. We have a fix for the crash being discussed in this thread ready to go, so I've suggested that we combine the two and make a new version and make one which can work with the crashed task, as well as solve the screensaver problem. I can't promise a timetable, but keep an eye on this thread. |
|
2688)
Message boards :
Questions and problems :
Boinc client crashes on Mageia Linux 5 (64 bits)
(Message 81597)
Posted 26 Sep 2017 by Richard Haselgrove Post: Again, you have one of the faulty CPDN batch 658 tasks: wah2_pnw25_c6bi_190312_49_658_011243953_0 Refer to http://boinc.berkeley.edu/dev/forum_thread.php?id=11853 - don't worry about the reference to Mac OS X in the title, these tasks fail on both Linux and OS X equally. The same removal advice will apply. |
|
2689)
Message boards :
Questions and problems :
Cannot connect to client
(Message 81586)
Posted 26 Sep 2017 by Richard Haselgrove Post: Crashing and restarting at the point where the last item written to the event log is 'Running CPU benchmarks' is exactly the symptom we discovered in the CPDN batch 658 reports last week - see the thread I linked previously. |
|
2690)
Message boards :
Questions and problems :
Cannot connect to client
(Message 81584)
Posted 26 Sep 2017 by Richard Haselgrove Post: Your symptoms are exactly the same as several other users running climateprediction.net on Linux or OS X. See explanations and workrounds in the thread BOINC client crashes immediately at startup (macOS 10.12.6) (This thread used to refer to version 7.8.2 in the title, but we've verified that the problem isn't version-specific) |
|
2691)
Message boards :
Questions and problems :
How to properly attach project via console?
(Message 81577)
Posted 25 Sep 2017 by Richard Haselgrove Post: If this is related to your current thread - "BOINC portable" for Windows hosts - at SETI, I presume that you are trying to run on a Windows machine without running the Windows BOINC installer. If that is the case, none of the registry values normally found in "HKLM\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup\" will have been set by the installer. You will be missing INSTALLDIR and DATADIR, among others. Best to use command line parameters for everything you need, as Juha suggests. |
|
2692)
Message boards :
Questions and problems :
Can't install BOINC 7.8.2 on Win10 to replace previous version because a drive letter reassignment between versions occurred
(Message 81556)
Posted 23 Sep 2017 by Richard Haselgrove Post: This has been a known problem with earlier versions of BOINC and Windows, and I expect the same procedure will work. You will need, I'm afraid, to edit the Windows registry, using the tool regedit Find the following values (this is my custom setup) [HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup] "INSTALLDIR"="D:\\BOINC\\" "DATADIR"="D:\\BOINCdata\\"Change the two path names (the first is for BOINC's programs, the second for data) to match your new drive letters and preferred folder structure. |
|
2693)
Message boards :
The Lounge :
BOINC Politics thread
(Message 81541)
Posted 22 Sep 2017 by Richard Haselgrove Post: Oh, groan. Everybody has been jumping on either side of this story, depending on their political slant. The choice isn't between Uber and black cabs: London has had what they call minicabs, and the rest of the country calls private hire, for about 50 years. All that will happen on the 30th (extended by however long the appeal takes) is that minicab offices will get busy, minicab owners will start hiring drivers, and Uber drivers will start applying for the jobs. There'll be an opening for programmers to integrate smartphone booking apps into the minicab offices' existing booking systems, that's all. Life will go on - unless it's come to a halt for other reasons, of course. |
|
2694)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81536)
Posted 22 Sep 2017 by Richard Haselgrove Post: And finally: Right now i have suspended the CPDN project how will we know it is safe to resume work ?CPDN users don't seem to be discussing it much on their forums, but the staff certainly are - see Jord's message 81516. Unfortunately, I got thrown off that feed, but while Jord is still receiving it, I'm sure he'll keep you posted. |
|
2695)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81534)
Posted 22 Sep 2017 by Richard Haselgrove Post: Well, every workunit is different, but yours does match the same pattern: <name>wah2_pnw25_c8b7_190312_49_658_011246534_0</name> Those are the bits to pick out. |
|
2696)
Message boards :
The Lounge :
Needful Things
(Message 81521)
Posted 21 Sep 2017 by Richard Haselgrove Post: Since I've had to go through all this recently, and I've still got the box of papers handy... Both my father's January 1953 copy certificate, and my stepfather's August 1926 original certificate - issued under Acts of 1836 to 1874 - only show the 'given' names (two, in each case). Leaving the surname to be deduced from the current names of their respective fathers and mothers (the same in both cases, as it happens): former surnames of the women are also listed. So, it all seems very standardised, as you might expect. Oh, I should have answered. No. |
|
2697)
Message boards :
The Lounge :
BOINC Politics thread
(Message 81518)
Posted 21 Sep 2017 by Richard Haselgrove Post: Alternatively (for car drivers only): watch your wing mirror throughout the 'park, switch off, unbelt, open door' process. Develop "situational awareness", or in English, know what's going on around you. Not just when parking. |
|
2698)
Message boards :
The Lounge :
Needful Things
(Message 81502)
Posted 20 Sep 2017 by Richard Haselgrove Post: On my mother's certificate, there are two first names, and no surname - "me me them", as you suggest. |
|
2699)
Message boards :
The Lounge :
Needful Things
(Message 81500)
Posted 20 Sep 2017 by Richard Haselgrove Post: I can't answer that question directly - but I have in front of me a 1976 certified copy of my mother's 1930 birth certificate. Her first name is spelled 'Jennifer', when she'd been known as 'Jenifer' for the previous 45 years. That certainly worried her - she had them go back and check it. And I had to make sure that all variants were listed on her death certificate, just in case somebody refused to pay up on an insurance policy or something. So she has three names on her death certificate: "Jenifer ... otherwise Jennifer ... formerly known as Haselgrove". The only suggestion I have for your problem is that you can always get a certified, pukka, copy of a birth certificate, which is accepted wherever a birth certificate is needed. Unless you were born in a country with a less well established, or less accessible, public records service than the UK... The copy UK certificate issued under the 1953 Act (which still seems to be the way of doing things here) says on the back that an additional name (such as one given in Christian baptism) can be added to the register, and hence to the certificate, if the certificate of baptism or naming is dated less than 12 months after the birth was registered. |
|
2700)
Message boards :
Projects :
Enigma@home CPU tasks and Android.
(Message 81493)
Posted 20 Sep 2017 by Richard Haselgrove Post: Dunno. Why don't you ask them on their message board? |
|
2701)
Message boards :
News :
Client version 7.8.2 released
(Message 81485)
Posted 19 Sep 2017 by Richard Haselgrove Post: Are you connected to the CPDN project? If so, please read recent posts in the 7.8.2 client crashes immediately at startup (macOS 10.12.6) thread. |
|
2702)
Message boards :
Questions and problems :
Install Problem - Windows 10
(Message 81472)
Posted 19 Sep 2017 by Richard Haselgrove Post: Just tried it on my Windows 10 test machine (Dell Optiplex business desktop, with x64 Pro version, updated to current consumer patch level, no insider build). "It worked for me" - created accounts, required reboot, ran CPU tasks but couldn't find GPU. All as expected. I removed the previous 'user' installation before testing - this will have been the first attempt at a service installation on this hardware, either under Windows 7 (I have dual boot) or Windows 10. I can try again tomorrow if anyone has anything in particular they want me to test. I always start the BOINC installer by double-clicking the file in the downloads folder - not from the browser/download manager - and I let UAC take care of the Administrator request. |
|
2703)
Message boards :
Questions and problems :
7.8.2 Failure (doesn't run with PrimeGrid)
(Message 81438)
Posted 19 Sep 2017 by Richard Haselgrove Post: We have a v7.6 release branch, and building from it will give you v7.6.33 or, with a single update, v7.6.34 for Mac. Build from that - even use the existing binaries - and update the "with VBox" installers to the new VBox. No change at all to BOINC code or builds while we catch our breath. |
|
2704)
Message boards :
Questions and problems :
7.8.2 Failure (doesn't run with PrimeGrid)
(Message 81436)
Posted 19 Sep 2017 by Richard Haselgrove Post: Could be. Why is 7.8.2 still recommended?I think that was a question I asked myself the night it was released - I got the 'dirty slot' error within 45 minutes, as you know (although we didn't fully understand the details at that stage). With one set of developers saying "I want to build a new one now", and another set of developers saying "we can do one, but not until December", I think I'm going to repeat my call for a v7.6.35 with VBox v5.1.28 while we argue about it. |
|
2705)
Message boards :
Questions and problems :
7.8.2 Failure (doesn't run with PrimeGrid)
(Message 81434)
Posted 19 Sep 2017 by Richard Haselgrove Post: 'Disk usage limit' could be the 're-using dirty slot' problem we know about already. |
|
2706)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81404)
Posted 18 Sep 2017 by Richard Haselgrove Post: Good idea. My other correspondent has sent me some log files. That client is also failing immediatly after starting the 'run benchmark' process, but with one extra line: 17-Sep-2017 14:17:10 [---] Running CPU benchmarks 17-Sep-2017 14:17:10 [---] Received signal 15 17-Sep-2017 14:17:09 [---] cc_config.xml not found - using defaults 17-Sep-2017 14:17:09 [---] Starting BOINC client version 7.6.34 for x86_64-apple-darwinWith extra 'Signal 15', if that helps anyone. |
|
2707)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81399)
Posted 18 Sep 2017 by Richard Haselgrove Post: Thanks George. There seem to be two problems there: 1) Why do Linux and Mac CPDN tasks fail after one month? Dunno, but you might talk to the CPDN programmers about vsyscall - mentioned in this thread, I think, else search recent threads. Certainly for Linux, that function is being removed from Linux: if it's being called during the month-end file shuffle, that might cause the failure on recently updated Linux kernels. 2) Why does BOINC crash when the CPDN app crashes? We seem to have established via this thread (and the final test was exactly a repeat of your own procedure) that BOINC can't run when it has a huge stderr_txt and a huge pile of failed uploads for an unreported result. That seems to be the result of old and neglected code, perhaps dating back several versions. There is a fix in the pipeline, but no date for a test build yet: it was omitted from v7.8.2, despite being available then. |
|
2708)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81392)
Posted 18 Sep 2017 by Richard Haselgrove Post: That's more-or-less what we expected. Ask him to check batch 658 - that's the PNW that I've got. Data, not application. Edit - note on the CPDN front page that batch 658 was submitted on 15 September. That matches - and my first one crashed, though that was a 651. |
|
2709)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81390)
Posted 18 Sep 2017 by Richard Haselgrove Post: And thank you for yours. We know what we need now, and the fix is already available - we just need a new build so we can test it properly. |
|
2710)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81386)
Posted 18 Sep 2017 by Richard Haselgrove Post: I've re-activated my CPDN account, and with any luck I should get a WAH2 PNW in about 10 minutes (end of 1 hour backoff). If I disappear off the face of the earth, you'll know where I've gone... Otherwise, I might be able to see how it works under v7.8.2 for Windows. Edit - well, that worked better than I expected: 18/09/2017 18:05:52 | climateprediction.net | [sched_op] CPU work request: 10726.27 seconds; 4.00 devices 18/09/2017 18:05:54 | climateprediction.net | Scheduler request completed: got 4 new tasks 18/09/2017 18:05:54 | climateprediction.net | [sched_op] estimated total CPU task duration: 2309063 secondsOne and three spares. Why does the client request work for 4 devices, when all cores are busy with three CPU tasks and one OpenCL support job? Edit2 - actually, handy. One CAM model with 12 upload files, two AFR with 14 files, and one PNW with the 51 files. I'll start with an easy one. |
|
2711)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81384)
Posted 18 Sep 2017 by Richard Haselgrove Post: Richard, a shot into the blue, but the site this work was done at was having internet connection issues this weekend. Might this be related, i.e. only a temporarily failed upload triggers the client suicide?I think it's unlikely - BOINC is designed to cope with that sort of thing. My ISP resets my line once a week, and that doesn't cause any problems while it's down. But thanks for adding it to the report - we're probably going to need every scrap of information we can get. Secondly, is there anything but a clean reset that I can do to get my installation running again?The reset will be the quickest and easiest, but if you feel up to performing a little experiment first, it would be very helpful. Could you open the file 'client_state.xml', please, with a plain text editor - not a fancy XML editor. Then search for the line <name>wah2_pnw25_c4ci_190312_49_658_011241397_0</name>It will be in a block of XML that looks like ...
</file_ref>
<file_ref>
<file_name>ozone_hist_N96_1899_1910v2.gz</file_name>
<open_name>ozone_hist_N96_1899_1910v2.gz</open_name>
</file_ref>
</workunit>
<result>
<name>wah2_pnw25_c4ci_190312_49_658_011241397_0</name>
<final_cpu_time>15880.410000</final_cpu_time>
<final_elapsed_time>16233.776360</final_elapsed_time>
<exit_status>0</exit_status>
<state>3</state>
...If you could go back to the beginning of the <workunit> section, and down to the end of the </result> section, and remove absolutely everything - the whole<workunit> ... </workunit> <result> ... </result>segment, including those lines. Then save the file, and try BOINC one more time. If it starts normally, we have our smoking gun. Remember to stop work fetch from CPDN as soon as you get control again! If that doesn't work, just reset things - I doubt I'm going to come up with any more ideas tonight. |
|
2712)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81380)
Posted 18 Sep 2017 by Richard Haselgrove Post: One other small anomaly in pbro's bundle of files is a zero-length file called 'all_projects_list_temp.xml' (there's a full-size 'all_projects_list.xml' as well). We had problems with a zero-length RPC (different file) recently - could this indicate another problem causing or caused by the client crashing at startup? File is dated 18/09/2017 10:21, which is in the middle of the sequence of crashes with v7.8.2 - in fact the file coincided with "Version change (7.6.34 -> 7.8.2)", so it's part of the testing just before I got the files. |
|
2713)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81377)
Posted 18 Sep 2017 by Richard Haselgrove Post: We might have something there. I'm looking through pbro's client_state.xml, and a second user - MR, you know who you are - has also sent me a set of files including client_state.xml. Both users have a failed CPDN WAH2 PNW task in their client_state, showing 50 files (51 actually - zips 1 to 49, restart, out) plus a crash dump - about 14 KB in total for the <result> section. MR also mentions suffering the "attached to Einstein@home twice" problem due to the changed master url - that's another complication we could do without. Edit - I've posted in Jim1348's thread at CPDN, asking if users have failures/successes with the 51-file WAH2 PNW workunits. Most Linux CPDN failure reports tend to come from bad 32-bit library installation - I haven't found any discussion specific to these tasks yet. |
|
2714)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81373)
Posted 18 Sep 2017 by Richard Haselgrove Post: Got it, thanks, and opened OK. I'll start looking, but I may be some time... Well, the first obvious clue is from stdoutdae.txt: the client has been restarting continuously. It seems normal, until this happens: 18-Sep-2017 15:16:39 [---] (to change preferences, visit a project web site or select Preferences in the Manager) 18-Sep-2017 15:16:39 [---] Using account manager BOINCstatsBAM! 18-Sep-2017 15:16:39 Initialization completed 18-Sep-2017 15:16:39 [---] Running CPU benchmarks 18-Sep-2017 15:16:39 [---] Starting BOINC client version 7.8.2 for x86_64-apple-darwinabout 120 times in the snippet you sent me. So, init completed, start running benchmarks, crash. The next line, on my v7.8.2 for Windows (and many previous versions) is 09-Sep-2017 21:35:39 [---] Suspending computation - CPU benchmarks in progressso the crash is very quick. BTW, BAM! is not implicated - about half the crashes happened before BAM! was attached. |
|
2715)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81369)
Posted 18 Sep 2017 by Richard Haselgrove Post: Different handling of permissions in the v7.8.2 and v7.6.33 installers is interesting and useful to know - thanks. But it doesn't explain why v7.8.2 couldn't be re-started following the original crash. |
|
2716)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81367)
Posted 18 Sep 2017 by Richard Haselgrove Post: I seem to have run afoul a similar problem: BOINC 7.8.2 on Mac OS X 10.11.6 does no longer start. The manager complains about failing to start the client three times and then gives up, prompting for another three attempts with the prompt: "The BOINC client has exited unexpectedly 3 times within the last 3 minutes. Would you like to restart it again? (No/Yes)"Well, I didn't expect to get another example while typing the previous reply! If your installation is still borked and not starting, maybe we can try and look to see what the problem is - I think there must be some file corruption somewhere. Could you please archive up (zip, 7z, rar - whatever your favourite tool is) stderrdae.txt, stdoutdae.txt, and client_state.xml - perhaps plus any other xml files with timestamps around the crash time. Then, either put them in a dropbox we can access, or email them direct to me at (looking to my name on the left) initial dot surname at btinternet dot com. Thanks. Edit - your task was 20729988 - that looks like Weather At Home 2 for the Pacific North West region. One of the applications built earlier this year. |
|
2717)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81366)
Posted 18 Sep 2017 by Richard Haselgrove Post: From what you're saying, it does seem plausible that the CPDN are the originating cause of the problem. They have a varied selection of Linux applications - some deployed earlier this year, and some dating back as long as 2011. The first step might be to establish which app is running at the time? That's for you and CPDN to sort out between you - you might like to alert them to the vsyscall issue in newer kernels, and verify that the call isn't used even in their oldest apps. But my concern on this board is - why should an application crash on (we presume) one project nuke an entire BOINC installation so thoroughly that even a re-installed older version can't run? That we still have to get to the bottom of. |
|
2718)
Message boards :
The Lounge :
The problem with that is...
(Message 81351)
Posted 17 Sep 2017 by Richard Haselgrove Post: Even harder when the venison is still running wild through the forest!What's wrong with getting your protein from a bit of red meat or chicken?It's a bit difficult when the meat is still strolling around the farmyard |
|
2719)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81346)
Posted 17 Sep 2017 by Richard Haselgrove Post: ... stacktrace mentioned vsyscall. Its disabled by default in later Linux kernels like 4.10 that comes with Ubuntu 17.04. I've linked to a thread over at Einstein that talks about it. From the information so far I suspect its a separate issue to the OSX one.A quick search on github suggests that vsyscall isn't used in the boinc client code, but is used in science applications from various projects. There is also Christian Beer wrote: You need to set the kernel option vsyscall=emulate in order to run applications compiled with older glibc. |
|
2720)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81328)
Posted 16 Sep 2017 by Richard Haselgrove Post: Jim has provided the following stack trace from, if we've got it right, BOINC v7.8.2 running under Ubuntu 17.04 [vsyscall] |
|
2721)
Message boards :
Questions and problems :
BOINC client crashes immediately at startup (macOS 10.12.6)
(Message 81325)
Posted 16 Sep 2017 by Richard Haselgrove Post: I'm seeing similar remarks in a Linux thread at GPUGrid (start from message 47880). I've quoted your stack trace (for which, many thanks) in an attempt to persuade them to locate and post something similar. The reports are similar: * Manager loses connection * Restart not possible * Restart still not possible after reversion to previous (known working) version. That makes me wonder whether, perhaps, a necessary data file gets so badly corrupted that reading it at startup causes the client to crash: the most obvious candidate would be client_state.xml (or _prev, _next), but it could be anything written and subsequently read by the client. If anyone finds themselves in this predicament, please check your boinc data files, or drop them into a dropbox somewhere that we can examine them from. |
|
2722)
Message boards :
Questions and problems :
BOINC Data on USB HDD
(Message 81323)
Posted 16 Sep 2017 by Richard Haselgrove Post: One other thing it could be is the bug that's in 7.8.2 which causes the client to not clean up the slot directories and it reusing them while there are still other files in there. There is no client out yet with a fix for that and I don't know how many versions you should go back to be before that bug was introduced. (Richard?)That bug was for Windows only, so not relevant here. For the record, it was introduced on Jan 31 this year (2017, for readers in the future) with commit bbd34f70d315336310e82b8d651730f1913a2778 - so any v7.6 release will be clear of it. |
|
2723)
Message boards :
Questions and problems :
BOINC Data on USB HDD
(Message 81311)
Posted 16 Sep 2017 by Richard Haselgrove Post: Sam 16 sep 12:10:01 2017 | LHC@home | Starting task x6lLDm2SNDrnDDn7oo6G73TpABFKDmABFKDmiOLKDmABFKDmZVGOTm_0There might be something in std_err. Task 157411792 hasn't been reported yet. It might be worth keeping an eye on it - the previous task 156526891 has quite a lot of detail. |
|
2724)
Message boards :
BOINC client :
RPC WU credit
(Message 81289)
Posted 15 Sep 2017 by Richard Haselgrove Post: No. The credit for an individual task is calculated and stored on the server, and not transmitted back to the client - only the totals and RAC for each project are transmitted back. BOINC does maintain a log for every task run on the client, which includes both the CPU time and the elapsed time. The logs are kept indefinitely - I have some going back 10 years. The files are kept in the root of your BOINC data directory, and take the form "job_log_[project url]" Project sites purge their databases by time, rather than by number of tasks. SETI purges after 24 hours, Einstein purges after 7 days, and so on. |
|
2725)
Message boards :
GPUs :
0MB , 0MB available Radeon Pro 560
(Message 81276)
Posted 15 Sep 2017 by Richard Haselgrove Post: Mind you, I don't think World Community Grid has any GPU apps - either ATI/AMD or Intel: you client has probably gone into indefinite backoff. A single click on the project 'update' button will probably trigger a request, but you won't get any actual work until you attach to a GPU-supporting project. |
|
2726)
Message boards :
Questions and problems :
setting up Boinc Server for my personal use.
(Message 81274)
Posted 15 Sep 2017 by Richard Haselgrove Post: I think it would be best if I asked somebody like Christian Beer to pass by this thread - he prepared the server VM you're trying to use, and hearing your difficulties first-hand might inspire improvements (or links!) to the documentation. In the meantime, you might explain what type of application you're intending to deploy from the server, and process using the other machines on your LAN - then any help offered can be more precisely targeted. |
|
2727)
Message boards :
Questions and problems :
setting up Boinc Server for my personal use.
(Message 81267)
Posted 15 Sep 2017 by Richard Haselgrove Post: Yes and yes. |
|
2728)
Message boards :
News :
Client version 7.8.2 released
(Message 81240)
Posted 14 Sep 2017 by Richard Haselgrove Post: Only the release manager knows, and he won't tell.I suspect the word is "can't", rather than "won't": he's been the one making the changes, and I have a strong suspicion that he hasn't been keeping notes. So, he can't tell you what he's done, and he won't put in the hours of research needed to work it out. We can probably split it up into sections and work it out between us, but it's going to be a while. |
|
2729)
Message boards :
GPUs :
Limit GPU works
(Message 81218)
Posted 14 Sep 2017 by Richard Haselgrove Post: You can use the '<max_concurrent>' flag in an app_config.xml file, described in the application configuration section of the user manual. |
|
2730)
Message boards :
News :
Project management changes proposed
(Message 81217)
Posted 14 Sep 2017 by Richard Haselgrove Post: 'Help desk expert' is a tag applied manually by the board administrators, and is their personal judgement of the value of that individual's posting to the community at large. Not every expert can be an expert in everything. I know a fair bit about Windows, but I'm a complete novice in Linux, OS X, and Android. But I can't remove my 'expert' tag if I feel the need to post in a thread discussing one of those operating systems. Most 'experts' have been tagged on the basis of their knowledge of the BOINC software and its interaction with operating systems and the projects that use it. We don't have any 'experts' in documentation except Jord (Ageless). Please excuse us when we try - like any other user here - to help in areas outside our direct expertise. |
|
2731)
Message boards :
The Lounge :
Needful Things
(Message 81168)
Posted 13 Sep 2017 by Richard Haselgrove Post: ... am impressed with the way it shows how the wind swirls around in the UK before thumping into mainland Europe.And how it channels down through the Aire Gap. Fortunately, the people who founded my village a thousand years ago knew to build in a fold of the hills, sheltered from the north-west. |
|
2732)
Message boards :
The Lounge :
The problem with that is...
(Message 81167)
Posted 13 Sep 2017 by Richard Haselgrove Post: We proper townies know that the first rule of the countryside is that you keep gates SHUT ...Proper townies obviously can't read. "Leave all gates as you found them" has been written in the The Country[side] Code since at least 1981. |
|
2733)
Message boards :
The Lounge :
Needful Things
(Message 81162)
Posted 13 Sep 2017 by Richard Haselgrove Post: Hasn't everyone in that 1970 video put on weight? I'm sure they used to be much skinnier. |
|
2734)
Message boards :
Questions and problems :
Boinc client wants to read b11:0
(Message 81156)
Posted 13 Sep 2017 by Richard Haselgrove Post: If it looks like a disk drive, BOINC might be evaluating available disk space. We had a problem in Windows recently where, instead of simply reading file sizes, BOINC opened the file for reading and triggered a full virus scan. |
|
2735)
Message boards :
The Lounge :
Needful Things
(Message 81067)
Posted 11 Sep 2017 by Richard Haselgrove Post: Hmm. wasn't aware of that so I'll look into on next trip to opticians which is October. I've used the same optician for 26 years & knowing the type of work I was doing he suggested it.Well, it's possible that the Specsavers advice only applies to the precise type of photochromic glass that they use. You can test your existing ones easily enough: look at them indoors, then take them outside into bright sunlight* and see how much they darken. Then go and sit in a car, and see if they lighten again? Your polaroid pair, how good are they when you're driving?They're fine - the varifocal bit allows me to read both the road and the instruments with minimal strain, and the anti-glare matches what I was used to for sunny days: not a total blackout (which you wouldn't want for driving anyway), but good enough for round here - to the extent I sometimes walk back into the house and forget that I'm wearing them, then wonder why it's so dark in here. * (might be a flaw in the plan at this point, given the sunlight levels in the UK at the moment...) |
|
2736)
Message boards :
The Lounge :
Needful Things
(Message 81052)
Posted 11 Sep 2017 by Richard Haselgrove Post: Likewise. I've managed for most of my life without optical assistance. From age ~50, I started to need reading glasses: by last year, I needed distance assistance too, and switched to varifocals. Specsavers advised me against photochromic for driving, because the darkening effect is triggered (they said) by ultraviolet, and car windows can filter out the UV. Instead, I got a plain pair and a polaroid pair with the same prescription - that's good for me, because I've used non-prescription polaroid glasses for years when driving, as an anti-glare measure. |
|
2737)
Message boards :
The Lounge :
The problem with that is...
(Message 81035)
Posted 10 Sep 2017 by Richard Haselgrove Post: There is a semi-samizdat (and borderline illegal: i.e. illegal over the border in Scotland) trade in green-top unpasteurised farm-bottled milk. There is some hope that producers (the farmers) prepared to jump through the legal hurdles for a discerning purchasing public also have the welfare of the real producers (the cows) in mind as well. |
|
2738)
Message boards :
Questions and problems :
BOINC can't access internet
(Message 81007)
Posted 9 Sep 2017 by Richard Haselgrove Post: You want the raw file: https://raw.githubusercontent.com/BOINC/boinc/master/curl/ca-bundle.crtApologies. I did test the right-click thing before posting, and got the actual file. Browsers differ. |
|
2739)
Message boards :
The Lounge :
Needful Things
(Message 81001)
Posted 9 Sep 2017 by Richard Haselgrove Post: That's why I prefer to get it out of the way 30 minutes from my front door, when I can still remember what I've packed and where. |
|
2740)
Message boards :
The Lounge :
Needful Things
(Message 80999)
Posted 9 Sep 2017 by Richard Haselgrove Post: They have bateaux mouches in Las Vegas? I must try it sometime. |
|
2741)
Message boards :
The Lounge :
Needful Things
(Message 80997)
Posted 9 Sep 2017 by Richard Haselgrove Post: While I'm here - Jord asked me for proof that I'd really been in Paris. Best I could do... 
|
|
2742)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 80994)
Posted 9 Sep 2017 by Richard Haselgrove Post: I went and found a toilet and gave that hand a thorough wash. There was no sign that the skin had been broken at any point, and there's been no sign of anything malign growing since Tuesday, which is when I took that picture - using my other hand to un-pocket the camera and aim it blind. |
|
2743)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 80992)
Posted 9 Sep 2017 by Richard Haselgrove Post: Sometimes the tables are turned. I was 'befriended' (or seen as lunch) at Montmatre... 
|
|
2744)
Message boards :
The Lounge :
Needful Things
(Message 80990)
Posted 9 Sep 2017 by Richard Haselgrove Post: Although quite why one would want to drink German lager, flavoured with Mexican tequila in France I'm not quite sure .....This is it (sorry Jord)  à la CdG - but it tastes much better in a chic Parisian diner, watching chic Parisiennes striding past the plate-glass windows, with chic Parisiennes at the other tables, and served au pression by chic Parisienne waitresses... :-))) |
|
2745)
Message boards :
News :
Client version 7.8.2 released
(Message 80986)
Posted 9 Sep 2017 by Richard Haselgrove Post: It was just the 'Downolad' in the advertisement that amused me... (it was more prominent in the original formatting) |
|
2746)
Message boards :
The Lounge :
Needful Things
(Message 80984)
Posted 9 Sep 2017 by Richard Haselgrove Post: CdG is better now - it has a fixed-rail automatic shuttle (think DLR) between the railway stations and the main terminals (not including T3, natch). No more unguided tours of the car-parks by navette. |
|
2747)
Message boards :
The Lounge :
Needful Things
(Message 80978)
Posted 9 Sep 2017 by Richard Haselgrove Post: Well, Jet2 sent me a 'welcome back' email, with an invitation to fill in a satisfaction survey with - unusually, but welcome - plenty of opportunity to express myself in my own words. They should know my views on remote airport ground handling provision now. |
|
2748)
Message boards :
Questions and problems :
When does the client communicate with the project server
(Message 80976)
Posted 9 Sep 2017 by Richard Haselgrove Post: Reporting takes place whenever the client feels the need to contact the project server, subject to limitations whenever a delay between contacts is enforced. John McLeod used to maintain a list of about a dozen possible triggers: the commonest one is "need more work". The "5 minute cycle" that Rob mentions is specific to the SETI@Home project: other projects have used backoffs between 7 seconds and an hour. Reports of completed tasks are batched whenever possible, for efficiency in inserting the results into the project database. But nowadays, no task will be held, completed, within your local client for longer than an hour if communication is possible: older sources will cite that ultimate delay as 24 hours, but it's changed. |
|
2749)
Message boards :
The Lounge :
Needful Things
(Message 80975)
Posted 9 Sep 2017 by Richard Haselgrove Post: Jet2 & KLM both go from Leeds/Bradford direct to Paris. Flybe goes direct from Doncaster/Sheffield.Thanks for trying, Chris, but I know my own local airport better than that. The LBA Flight Guide is clear that only Jet2 flies direct: other sources make it clear that all KLM flights require a change of planes in Amsterdam. I've done that shuffle when travelling to Rome and Berlin: I can assure you that the walk from one side of Schiphol airport to the other takes a damn sight longer than 2 minutes (though at least they transport your heavy luggage for you). Had I been Richard I would have taken the train from Leeds to Kings X, then a 2 minute walk to St Pancras Intl and the Eurostar to Paris Gare Du Nord. Leeds airport is 9 miles from Leeds city centre, and Charles De Gaulle airport is 14 miles from Paris city centre. requiring transport at both ends. However it's Richards personal choice how to travel which suits him best.The trip from Leeds to London by train is fine, even adding in the (fast and comfortable) local commuter train to Leeds. Eurostar from London to Paris is also fine, but the interchange in London isn't 2 minutes: it's a short walk, followed by an airport-style strip search and a compulsory check-in delay. I just don't want that hassle after I've already settled in to the journey. I use a local minicab firm to take me to the airport, door-to-door: they drop me within seconds of the Jet2 bag-drop, which is fast and efficient (though quite busy this trip). At the other end, CDG has one TGV station and two RER stations within the airport complex itself. I've travelled as far as Avignon in the deep south direct from the airport without a change of train: you couldn't get a direct train from Heathrow to, say, Perth (a similar distance). The RER into the middle of Paris may not be particularly fast, but it's frequent and efficient, and links you into the entire Metro network. For this trip, I was able to book into a hotel just round the corner from an RER stop on the direct line. Even if I'd travelled by Eurostar, I'd still have been using that same RER B train for the final leg from the Gare du Nord. |
|
2750)
Message boards :
Questions and problems :
BOINC can't access internet
(Message 80963)
Posted 8 Sep 2017 by Richard Haselgrove Post: For other operating systems, BOINC doesn't rely on the certificates installed globally by the operating system: instead it uses certificates supplied by BOINC in a file called 'ca-bundle.crt'. You could ask your QNAP developer if he included that file: it's normally placed in the same folder as the boinc client binary. If he didn't include the file, it can be downloaded from https://github.com/BOINC/boinc/tree/master/curl The QNAP NAS link that Jord posted seems to be mostly about allowing authorised users to access your QNAP device securely from outside the building: that's going in the wrong direction. You're inside the building, and want to reach out to BOINC projects outside it. The instructions for how to create your own certificates may still be useful in another context, but I don't think they'll help with the initial problem. |
|
2751)
Message boards :
The Lounge :
Needful Things
(Message 80957)
Posted 8 Sep 2017 by Richard Haselgrove Post: Trouble is, if I flew BA, I'd have to change planes in Heathrow - and that's even worse. Similarly, that's why I don't use Eurostar - it's so bloody London-centric. It's about time Yorkshire started digging its own escape tunnel...Bloody low-cost airlinesAh well you see, now you know why I fly BA everywhere. |
|
2752)
Message boards :
News :
Client version 7.8.2 released
(Message 80943)
Posted 8 Sep 2017 by Richard Haselgrove Post: During installation on Windows, i have message from my AVG Antivirus that "C:\Windows\Boinc.scr" is infected and antivirus block it.French airport WiFi advertisement: Free AVG AntivirusWould you trust this AV? |
|
2753)
Message boards :
The Lounge :
Needful Things
(Message 80942)
Posted 8 Sep 2017 by Richard Haselgrove Post: Just how many bottles can you get in your hand baggage????Getting Desperados through security is probably even trickier than most other fluids - I'll just have to finish this tinnie before the bag drop opens. Bloody low-cost airlines - there are precisely two retail outlets land-side of this entire terminal, but at least Relay (the French equivalent of W H Smith) sells sleepy-juice. I can confirm that Desperados is brewed in France. (and at least the airport has free power and free WiFi for laptops) |
|
2754)
Message boards :
Questions and problems :
BOINC can't access internet
(Message 80924)
Posted 8 Sep 2017 by Richard Haselgrove Post: You need to sort out which bit of internet access is failing. Set the <http_debug> message log flag, and look to see which server it is trying to reach at the time: is DNS resolution working, is the app and port allowed through any firewall the QNAP might be running - that sort of thing. Development versions on their own shouldn't make a difference, though unless you know exactly when the source code was pulled, you can't track down exactly what its status might be. |
|
2755)
Message boards :
The Lounge :
Needful Things
(Message 80916)
Posted 7 Sep 2017 by Richard Haselgrove Post: Since I'm living above an American diner called 'Indiana' - Desperados, with a touch of tequila. Dunno where it was brewed, but it went down a treat. As did the next one. |
|
2756)
Message boards :
The Lounge :
Needful Things
(Message 80910)
Posted 7 Sep 2017 by Richard Haselgrove Post: ... it seems that Richard is playing an absolute blinder out there ...I'll leave that to others to decide, but you'll have heard me commenting and questioning occasionally on the recordings. Now, I'm f---ing shattered: beer calls. Over and out. |
|
2757)
Message boards :
News :
BOINC Workshop 2017
(Message 80909)
Posted 7 Sep 2017 by Richard Haselgrove Post: Well, the open discussion session has closed without a plenary, the room has cleared, and we've all gone our separate ways: I for one am exhausted. I can only attend part of the Hackfest tomorrow, and I have to pack for travelling home by lunchtime - so no more blogging from me, I fear. Someone else will have to tell you what gets done there. |
|
2758)
Message boards :
News :
BOINC Workshop 2017
(Message 80900)
Posted 7 Sep 2017 by Richard Haselgrove Post: Sadly workshop link has gone down ...Yes, I think it was deliberately taken down because of the difficulty of covering multiple small group discussions with a single static camera |
|
2759)
Message boards :
The Lounge :
You couldn't make it up (but some did)
(Message 80894)
Posted 7 Sep 2017 by Richard Haselgrove Post: I am the beneficiary of an agreement struck between a previous owner of my house and the pre-1974 local authority: for the princely sum of 50 pence per year, the current successor to that local authority guarantees my right to light through a window installed in to 1960s |
|
2760)
Message boards :
News :
BOINC Workshop 2017
(Message 80893)
Posted 7 Sep 2017 by Richard Haselgrove Post: Thank you for your interest. For reference, all the talks I've recorded and described here are available on Google Drive. |
|
2761)
Message boards :
News :
BOINC Workshop 2017
(Message 80892)
Posted 7 Sep 2017 by Richard Haselgrove Post: And finally, Laurence Field on HTCondor. That's the last of the advertised speakers from the workshop programme. We're now breaking into small group discussions, which I won't even try to record: if we come together into a final plenary, I might try to grab that. |
|
2762)
Message boards :
News :
BOINC Workshop 2017
(Message 80889)
Posted 7 Sep 2017 by Richard Haselgrove Post: And back from lunch. This one is more for scientists than BOINC administrators - Marius Millea on how to create a science app using boinc2docker. |
|
2763)
Message boards :
News :
BOINC Workshop 2017
(Message 80886)
Posted 7 Sep 2017 by Richard Haselgrove Post: Short and sweet - more in the life of a server operator, by Keith Uplinger of WCG. Now we're off for lunch. |
|
2764)
Message boards :
News :
BOINC Workshop 2017
(Message 80884)
Posted 7 Sep 2017 by Richard Haselgrove Post: And now the report on the BOINC Pentathlon 2017. |
|
2765)
Message boards :
News :
BOINC Workshop 2017
(Message 80883)
Posted 7 Sep 2017 by Richard Haselgrove Post: We're back. Laurence Field's talk on alternative GUIs is up. |
|
2766)
Message boards :
News :
BOINC Workshop 2017
(Message 80881)
Posted 7 Sep 2017 by Richard Haselgrove Post: Marius Millea's talk on setting up a BOINC server using Docker is posted. Another technical one, that will be difficult to follow without the live interactive demo - but of interest to anyone thinking about setting up their own project, or just wanting to see what BOINC admins have available to work with. Coffee break time. |
|
2767)
Message boards :
News :
BOINC Workshop 2017
(Message 80880)
Posted 7 Sep 2017 by Richard Haselgrove Post: Gridcoin talk posted. |
|
2768)
Message boards :
News :
BOINC Workshop 2017
(Message 80878)
Posted 7 Sep 2017 by Richard Haselgrove Post: OK, just setting up for the next day's session - lining up the video camera. To work. Video should be on, talk just starting. |
|
2769)
Message boards :
BOINC client :
Can max RAC be simulated from previous results
(Message 80876)
Posted 7 Sep 2017 by Richard Haselgrove Post: Yes, I found that one via another thread. Too bad the Excel file link doesn't work. Anyone has that file or knows where I can find it?I used to have it and got it tweaked so that it worked. I don't think it's on an active machine, though. I'll look through my box of cast-off hard disks when I get back from Paris at the weekend. |
|
2770)
Message boards :
News :
BOINC Workshop 2017
(Message 80868)
Posted 6 Sep 2017 by Richard Haselgrove Post: .....or frogs legs and snails being France?I ordered Coq au Vin, but they'd run out. |
|
2771)
Message boards :
News :
BOINC Workshop 2017
(Message 80858)
Posted 6 Sep 2017 by Richard Haselgrove Post: And David Anderson's proposal for TDB concludes the entertainment for today. Quick packup for dinner - see you tomorrow. |
|
2772)
Message boards :
News :
BOINC Workshop 2017
(Message 80853)
Posted 6 Sep 2017 by Richard Haselgrove Post: That was quick - Matt Blumberg GridRepublic up. One more. |
|
2773)
Message boards :
News :
BOINC Workshop 2017
(Message 80852)
Posted 6 Sep 2017 by Richard Haselgrove Post: And Nils Hoimyr CERN - we're running late. Possibly two more tonight. |
|
2774)
Message boards :
News :
BOINC Workshop 2017
(Message 80851)
Posted 6 Sep 2017 by Richard Haselgrove Post: Andy Bowery CPDN is uploaded |
|
2775)
Message boards :
News :
BOINC Workshop 2017
(Message 80850)
Posted 6 Sep 2017 by Richard Haselgrove Post: This afternoon's session is short talks about projects. First, Steven Clark of nanohub. Many speakers use slides - by convention these are uploaded to the workshop page later. |
|
2776)
Message boards :
News :
BOINC Workshop 2017
(Message 80846)
Posted 6 Sep 2017 by Richard Haselgrove Post: I'll keep making audio recordings as well, so we have a full set. |
|
2777)
Message boards :
News :
BOINC Workshop 2017
(Message 80845)
Posted 6 Sep 2017 by Richard Haselgrove Post: Marius says "make sure you have clicked http link and downloaded client software". |
|
2778)
Message boards :
News :
BOINC Workshop 2017
(Message 80842)
Posted 6 Sep 2017 by Richard Haselgrove Post: Tecchies appeared just as we were heading upstairs for coffee - and nicked my network cable. Quick bit of furniture re-arranging later, I have another. Could you test video again now and report back, please? They say it's working (though you'll see an empty room and a blank screen - we're on break). |
|
2779)
Message boards :
News :
BOINC Workshop 2017
(Message 80839)
Posted 6 Sep 2017 by Richard Haselgrove Post: Any news?Marius was sitting behind me and has seen your question. I'll pester him over coffee. |
|
2780)
Message boards :
News :
BOINC Workshop 2017
(Message 80838)
Posted 6 Sep 2017 by Richard Haselgrove Post: Christian Beer's talk is complete and uploaded. This one is long (75 minutes) and technical - may not be for casual listeners. For the interactive section, start at https://github.com/BOINC/boinc/issues |
|
2781)
Message boards :
News :
BOINC Workshop 2017
(Message 80835)
Posted 6 Sep 2017 by Richard Haselgrove Post: will there be a live video feed later?We're not all back from lunch yet, and Marius is at the front trying to get help - but no signs of life yet :-( I'll keep you posted. He knows you're asking. |
|
2782)
Message boards :
News :
BOINC Workshop 2017
(Message 80834)
Posted 6 Sep 2017 by Richard Haselgrove Post: Would help if, I assume Richard on his laptop, could type more gently.I was trying ;-) I've collected an external keyboard from my hotel during the break, and I've put that on a different table - should be better for the next talk. Edit - the beginning of David's talk is poor quality because I was sitting towards the back. I moved to the front part way through, and the others should be clearer from the start. Can't do anything about the cough, thought. Christian is just starting. |
|
2783)
Message boards :
News :
BOINC Workshop 2017
(Message 80830)
Posted 6 Sep 2017 by Richard Haselgrove Post: I'm going to try uploading raw audio recordings of the talks shortly after they complete. Please report if the quality is good enough to make this worth while. Files will appear on Google Drive - David's opening talk is there already. Kevin Reed, speaking now, will be next. Link updated to show all files - Kevin's talk is there now. Breaking for lunch. |
|
2784)
Message boards :
News :
BOINC Workshop 2017
(Message 80829)
Posted 6 Sep 2017 by Richard Haselgrove Post: Just get us audio, forget about the video. Audio is enough and not as taxing as trying to stream 720p or 1080p video as well!Just jury-rigged a wired connection, should be better now. Will try for video after lunch. Edit - it turns out the port needed for the conferencing software I was using is blocked by the official networks in the building, both wired and WiFi. So we're out of ideas until lunch... Sorry about that. |
|
2785)
Message boards :
News :
BOINC Workshop 2017
(Message 80827)
Posted 6 Sep 2017 by Richard Haselgrove Post: Apologies to all for the breakdown in the workshop feed. We're going to try and jury-rig something during the first coffee break, and get the main feed working after lunch. |
|
2786)
Message boards :
The Lounge :
BOINC Politics thread
(Message 80807)
Posted 5 Sep 2017 by Richard Haselgrove Post: Greenwich - at a guess, 1300's?Nope. According to Wikipedia, it (the greenwich observatory, not any village of the same name) was founded on 22 June, 1675, by Charles II. The Greenwich Meridian wasn't established until 1851, largely at the request of the railways, who wanted a common time standard so their timetables would work. Fat lot of good that did. |
|
2787)
Message boards :
The Lounge :
BOINC Politics thread
(Message 80806)
Posted 5 Sep 2017 by Richard Haselgrove Post: Postscript to the observatory story. The site today is in 'Boulevard Arago', a name unfamiliar to me. But looking up the meridian, In the early 19th century, the Paris meridian was recalculated with greater precision by the astronomer François Arago, whose name now appears on the plaques or medallions tracing the route of the meridian through Paris.So that explains the road name. And perhaps why there's an empty plinth further along the road, with a sign explaining that Arago's statue (along with many others) was destroyed by the Vichy government during the second world war. Dangerous to be an astronomer, even a long-dead one, in times of trouble. |
|
2788)
Message boards :
The Lounge :
BOINC Politics thread
(Message 80804)
Posted 5 Sep 2017 by Richard Haselgrove Post: While I'm bored and in the mood - and waiting for the traffic noise to die down so I can get some sleep - here's another one. Tomorrow, I'm attending a meeting in a seminar room at the French Royal Observatory - well, strike the 'Royal' some 230 years ago, obviously, but you get my drift. Looking at my street plan, I see a long, linear, feature passing northwards from the observatory through Jardin du Luxembourg and the front door of the Louvre. This afternoon I was walking downhill from the viewpoint at the Sacre Coeur (Montmartre - the observatory is visible and marked), when I passed a 'history of Paris' wall-plaque. On this spot (it said in rough translation) the prime meridian for all French maps was set in 1670, followed by an obelisk-like stone marker in 1675. That seems to be on the same alignment, too. Can anyone remember when Greenwich was founded, without looking it up? |
|
2789)
Message boards :
The Lounge :
BOINC Politics thread
(Message 80802)
Posted 5 Sep 2017 by Richard Haselgrove Post: Oh dear, I seem to be turning into Chris S. I've just written this on a BBC iPlayer comment page: I am in Paris on business, using my UK laptop which knows my details and will be attached to my UK home TV licence when that becomes a requirement. My hotel TV doesn't offer BBC Worid. I can't even listen to BBC Radio, because it's broadcasting sports and that's broken too. In the old days, the BBC used to speak to the world - not any longer, it seems.Shall I sign his name before I post it? PS - It's Wales away to Moldova they won't let me listen to. I think I'll just go to bed early. |
|
2790)
Message boards :
BOINC client :
Can max RAC be simulated from previous results
(Message 80745)
Posted 4 Sep 2017 by Richard Haselgrove Post: The best description of RAC I know was in the late lamented Unofficial BOINC Wiki. Here's a stab at the Wayback Machine rendering of it: https://web.archive.org/web/20080311233928/http://www.boinc-wiki.info:80/Recent_Average_Credit |
|
2791)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80739)
Posted 4 Sep 2017 by Richard Haselgrove Post: If it was signed by the authors it might be worth summat on Ebay?No, the project team was based at MIT in the states (as was the programming language). There were contributors from Germany, India, Iran, Norway, and Turkey, but not one single Brit. They gave me the London edition of the paperback, priced one pound: I don't need to get that back. |
|
2792)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80730)
Posted 3 Sep 2017 by Richard Haselgrove Post: Don't forget Boscombe Down. I had a certain level of security clearance for some stuff that I supported at one point. If I told you I would have to shoot you :-))That's OK, I'll pass, thanks. My work at AERE wasn't secret. I was doing some modelling work on global oil pollution for an outpost of the Department of the Environment that was based behind the wire. On my first day there, they gave me a copy of the Club of Rome's "Limits to Growth" report: that was my primer for the 'DYNAMO' programming language we were going to use. I see the copy still on my shelf was published and copyright March 1972: this must have been around June-July 1972, so that pretty much counts as a first (popular paperback) edition. |
|
2793)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80723)
Posted 3 Sep 2017 by Richard Haselgrove Post: Ah...... still damn clever, and why Wiltshire?? How many Uni's there?I had a summer job one student year at the Atomic Energy Research Establishment, Harwell*. If that isn't in Wiltshire, it ought to be - I couldn't name the alternative, because of its association with the Other Place. * That's the peaceful one, not to be confused with AWRE Aldermaston. |
|
2794)
Message boards :
News :
Project management changes proposed
(Message 80722)
Posted 3 Sep 2017 by Richard Haselgrove Post: Changes to governance document requires a vote of the PMC as outlined in the governance document.The existing governance document was correctly linked by the questioner, and remains in force unless and until a revision is approved. As with all governing instruments, the bar for changing the status quo was set deliberately high (in our case, requiring a 75% affirmative vote). See how well the Founding Fathers protected the Constitution of the United States: you wouldn't want that document re-writing every four years as political fashions change. |
|
2795)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80720)
Posted 3 Sep 2017 by Richard Haselgrove Post: ... by students ...Clever people, students - I was one once. And with lots of spare time on our hands during the long summer holidays. |
|
2796)
Message boards :
News :
Project management changes proposed
(Message 80707)
Posted 3 Sep 2017 by Richard Haselgrove Post: I think it was the size (transmission bandwidth) of the sched_request files which was the concern for some projects. But this is very far from the project management debate: if people want to go on discussing task limits, I'll create a thread for you in another area and you can take the discussion there. |
|
2797)
Message boards :
The Lounge :
Needful Things
(Message 80689)
Posted 2 Sep 2017 by Richard Haselgrove Post: Obviously everybody fled from the message board when they saw I'd been made a mod. That's why there haven't been any posts. Glares around. Anybody want to contradict? |
|
2798)
Message boards :
The Lounge :
Needful Things
(Message 80679)
Posted 2 Sep 2017 by Richard Haselgrove Post: Thanks Rob - I'll raise one in your direction too. So far, no commiserations are needed - it's been eerily quiet in here, after frantic last-minute preparations yesterday (you don't see all the paddling that goes on below the waterline). Maybe it's the calm before the storm: ask me again when I get back from Paris next weekend. |
|
2799)
Message boards :
Questions and problems :
is pentium 4 computer useless for boinc
(Message 80678)
Posted 2 Sep 2017 by Richard Haselgrove Post: And to prove that last point, I have one of the very early Pentium 4 computers running on my account: SETI host 4060784. It contributes a negligible amount to my overall production, but I can afford the electricity, and I keep it running for sentimental reasons: it was my mother's old computer, before she passed away a couple of years ago. |
|
2800)
Message boards :
The Lounge :
Needful Things
(Message 80668)
Posted 1 Sep 2017 by Richard Haselgrove Post: Yup, that'll do. At first I got an ad about Alpacas in Peru, to a soothing background of plinky-plonk music. For a moment I thought that was what you were suggesting! |
|
2801)
Message boards :
The Lounge :
Needful Things
(Message 80666)
Posted 1 Sep 2017 by Richard Haselgrove Post: Here is the News!That needs a BONG - new committee formed - BONG - hold the front page - BONG - read all about it... |
|
2802)
Message boards :
The Lounge :
Needful Things
(Message 80663)
Posted 1 Sep 2017 by Richard Haselgrove Post: Ooooo goody, fresh blood, nothing like some fresh blood.....Ooooo, and buttons too. Which ones will I press first? |
|
2803)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80652)
Posted 1 Sep 2017 by Richard Haselgrove Post: Explain that one!There's not much difference in form and function (but a big difference in materials) between an astronaut's suit and a diving suit. According to Wikipedia: The first diving suit designs appeared in the early 18th century. Two English inventors developed the first pressure-proof diving suits in the 1710s.- so the concept was known before the end of the first construction period for the cathedral |
|
2804)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80623)
Posted 31 Aug 2017 by Richard Haselgrove Post: No - he's just on TV! |
|
2805)
Message boards :
Questions and problems :
Since Windows 10 Insider Build 16226, computer is always "busy"
(Message 80619)
Posted 31 Aug 2017 by Richard Haselgrove Post: Short for, cannot reproduce with last official release!It might be helpful to try that test again with the v7.8.2 'release candidate' build announced Tuesday - just to be on the safe side. |
|
2806)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 80609)
Posted 30 Aug 2017 by Richard Haselgrove Post: Precisely. I'll start writing an email :-( Edit - another wall of text duly dispatched. |
|
2807)
Message boards :
BOINC client :
Problems with OpenCL in Ubuntu using open source driver mesa
(Message 80607)
Posted 30 Aug 2017 by Richard Haselgrove Post: From that list, probably the best Linux GPU support community is at Einstein. Try BRP6-OPENCL-ATI WITH LINUX MESA OPENCL |
|
2808)
Message boards :
BOINC client :
Problems with OpenCL in Ubuntu using open source driver mesa
(Message 80603)
Posted 30 Aug 2017 by Richard Haselgrove Post: And your tasks are coming from - which project? |
|
2809)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 80601)
Posted 30 Aug 2017 by Richard Haselgrove Post: It was David's decision only to implement one of the reported problems with v7.8.1 |
|
2810)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 80595)
Posted 30 Aug 2017 by Richard Haselgrove Post: It gets better:  The word has been in the code since Jul 27, 2015: it was one of the first community contributions out of the traps after the community-based governance model was announced on 3 Jul 2015. A volunteer implemented trac ticket 79, and we didn't know how to deal with it. |
|
2811)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 80583)
Posted 30 Aug 2017 by Richard Haselgrove Post: The version I downloaded last night:  
|
|
2812)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 80561)
Posted 29 Aug 2017 by Richard Haselgrove Post: But other - reported and fixed - problems are still present. - the word 'daemon' is presented to users in the options dialog - 'OK' is still used instead of 'Save' in the Computing preferences dialog - GPU detection fails when 'boinc' is started from the command line, instead of 'boinc.exe' |
|
2813)
Message boards :
Questions and problems :
What is the currently recommended VirtualBox version?
(Message 80500)
Posted 28 Aug 2017 by Richard Haselgrove Post: VirtualBox is an independent product, supplied by Oracle. BOINC makes it available as a combined download for convenience, but you don't have to use the BOINC-supplied version. Currently, BOINC is offering VB 5.0.18 with the (somewhat elderly) recommended version of BOINC, and VB 5.1.26 with the next alpha test version of BOINC. I wouldn't recommend installing a test version of BOINC just to get a newer version of VB - this one isn't finished yet. Instead, get VB 5.1.26 from https://www.virtualbox.org/wiki/Downloads |
|
2814)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80451)
Posted 24 Aug 2017 by Richard Haselgrove Post: It appears that some people are preoccupied or obsessed with a certain idea, which is a known medical condition called OCD.Yes, I'd noticed. |
|
2815)
Message boards :
The Lounge :
BOINC Politics thread
(Message 80375)
Posted 20 Aug 2017 by Richard Haselgrove Post: Carved on tablets of stone?There's much about ancient civilisations to amaze & also ponder on, unlike today's civilisation. What will future historians be able to understand about the "Digital Age"?Likely the first thing they find is an archive of Rump's tweets. |
|
2816)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80358)
Posted 20 Aug 2017 by Richard Haselgrove Post: Terrestrial Pole offset: It is estimated that the pole offset from the Giza geographical co-ordinates changes at about 0.31 arc minutes per century; thus over 4500 years the offset is in the order of 13.1 arc seconds, a value far too small to be considered.One arc degree is about 70 miles, so one arc second is about 30 metres: 13.1 arc seconds is over 400 metres, which is significant when you're measuring to half an inch. |
|
2817)
Message boards :
Questions and problems :
Boinic does not obey project preferences
(Message 80341)
Posted 19 Aug 2017 by Richard Haselgrove Post: You'll probably need to set cpu_sched_debug logging to see what's really happening, but the output is pretty opaque. Look for any reference to EDF. Else, suspend a goodly number of CPU tasks, and see if the GPU picks up again. If it does, your cache is too large and you're in deadline trouble. *Reduce* cache size, and let the suspended tasks off the leash gradually while you burn through the cache. Remember to un-suspend *all* tasks when you need more for the GPU. |
|
2818)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 80337)
Posted 19 Aug 2017 by Richard Haselgrove Post: Coordinates of the pyramid - expressed as degrees north of the equator, using 360 degrees to a full circle. OK, the Babylonians had that around 2,400 BC - we can use those units. But that number of decimal places implies an accuracy of measurement of about half an inch at both ends. Good going. Speed of light - expressed in metres per second. Seconds are based on the length of the day, divided by 24*60*60. The Babylonians didn't start using base-60 for time until about 300 - 100 BC, so that's a bit anachronistic. (both references so far from https://math.stackexchange.com/questions/340467/why-is-a-full-circle-360-degrees) Metres - weren't defined until after the French Revolution, so they were unlikely to have been predicted as a unit of measurement at the time the pyramid was built. But they were defined in terms of the distance from the equator to the pole, so somebody else might have had the same idea earlier. (reference: https://en.wikipedia.org/wiki/History_of_the_metre) But the same number of decimal places implies the same accuracy of measurement: you're right, the Egyptians must have had some help. |
|
2819)
Message boards :
Questions and problems :
File transfer problems
(Message 80300)
Posted 18 Aug 2017 by Richard Haselgrove Post: If you enable the http_debug logging flag, you'll get additional output which is likely to be useful. In cases like this, http_debug is more useful than http_xfer_debug. Depending on the platform and version of BOINC you're running, it may be possible to enable logging flags from a graphical interface: press Ctrl+Shift+F in the Manager. |
|
2820)
Message boards :
Projects :
Remove project FiND@Home
(Message 80030)
Posted 6 Aug 2017 by Richard Haselgrove Post: I agree the transfer to Grid Republic did not go smoothly, but there has been no official announcement of closure yet. We should at least try to make contact with them before we lose the project for good. |
|
2821)
Message boards :
Projects :
For the betterment of BOINC
(Message 79997)
Posted 4 Aug 2017 by Richard Haselgrove Post: The decision of the working group is not to make internal working documents (including minutes) available at this stage. We are a long way from considering individual pull requests at this stage - we are a working group, not the PMC. Our role is to provide advice on how decisions are made in the future, not to make those decisions ourselves or review previous decisions. And believe me, that's tough enough already. |
|
2822)
Message boards :
Projects :
News on Project Outages
(Message 79987)
Posted 3 Aug 2017 by Richard Haselgrove Post: SETI has in large part recovered from the effects of the nearby fire, and should be functioning normally again. |
|
2823)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 79981)
Posted 3 Aug 2017 by Richard Haselgrove Post: 300,000Kg of CO2I don't mind how much CO2 tRump dumps into USA airspace - it's his country, and he can do what he likes: we shouldn't interfere in the internal affairs of another country. But woe betide him if any of that CO2 should invade another country's airspace, like mine. |
|
2824)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79964)
Posted 3 Aug 2017 by Richard Haselgrove Post: Woops splat - now not getting to the website here, and the DNS has definitely gone for a hike in the woods :-(Working OK here, though it looks like I might lose it in the next 10 minutes :-( setiathome.berkeley.edu
----------------------------------------
Record Name . . . . . : setiathome.berkeley.edu
Record Type . . . . . : 1
Time To Live . . . . : 506
Data Length . . . . . : 4
Section . . . . . . . : Answer
A (Host) Record . . . : 208.68.240.110
|
|
2825)
Message boards :
Projects :
News on Project Outages
(Message 79962)
Posted 3 Aug 2017 by Richard Haselgrove Post: SETI@Home Front page news: Mail problems and fire near SSL.From what I'm seeing here, I think the DNS problems have already caused client access failures. |
|
2826)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79961)
Posted 3 Aug 2017 by Richard Haselgrove Post: Looks like DNS has gone again. Haven't they got a UPS for that switch yet? Edit - and like last time, that's really slowing down posting here. Edit 2 - see front page news at SETI. |
|
2827)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79948)
Posted 2 Aug 2017 by Richard Haselgrove Post: Which planet are they trying to protect? Any chance of it being this one? |
|
2828)
Message boards :
Projects :
News on Project Outages
(Message 79885)
Posted 31 Jul 2017 by Richard Haselgrove Post: You'd be surprised! 31/07/2017 22:10:18 | Einstein@Home | Reporting 18 completed tasks 31/07/2017 22:10:20 | Einstein@Home | Scheduler request completed: got 33 new tasksThanks for the quick response. |
|
2829)
Message boards :
Projects :
News on Project Outages
(Message 79882)
Posted 31 Jul 2017 by Richard Haselgrove Post: Do you have any idea about the scheduler, rather than the message boards? I've been reporting tasks intermittently all evening (more "couldn't connect to server" than anything else). But now I've just had 31/07/2017 21:50:48 | Einstein@Home | update requested by user 31/07/2017 21:50:53 | Einstein@Home | [sched_op] Fetching master file 31/07/2017 21:50:53 | Einstein@Home | Fetching scheduler list 31/07/2017 21:50:56 | Einstein@Home | [sched_op] Got master file; parsing 31/07/2017 21:50:56 | Einstein@Home | [sched_op] Found 0 scheduler URLs in master file 31/07/2017 21:50:56 | Einstein@Home | [error] No scheduler URLs found in master file 31/07/2017 21:50:56 | Einstein@Home | [sched_op] Deferring communication for 1 days 00:00:00 31/07/2017 21:50:56 | Einstein@Home | [sched_op] Reason: 4 consecutive failures fetching scheduler listThat's your temporary home page, I think. That machine is going to be dry overnight. |
|
2830)
Message boards :
Questions and problems :
Task selection: Earliest Deadline First versus First In First Out
(Message 79873)
Posted 31 Jul 2017 by Richard Haselgrove Post: It doesn't need a schedule, because it's purely historical - it works by keeping an eye on how often and how long BOINC has been running in the past. Which has one side effect - BOINC is lousy at predicting when you're going on holiday and switching the machines off. |
|
2831)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79859)
Posted 31 Jul 2017 by Richard Haselgrove Post: Einstein has gone off air?"The server is temporarily unable to service your request due to maintenance downtime or capacity problems. Please try again later." It's usually capacity, or very brief maintenance. 'Later' may mean five minutes, or may (as now) be half an hour. |
|
2832)
Message boards :
Projects :
B612, aka Orbit@Home
(Message 79847)
Posted 30 Jul 2017 by Richard Haselgrove Post: https://b612foundation.org/And can be made clickable! (as you found) |
|
2833)
Message boards :
BOINC client :
Suggestion - allow for updating the deadlines
(Message 79835)
Posted 29 Jul 2017 by Richard Haselgrove Post: I've just replied to a query from a project administrator, who is seeing a spate of ERROR_UNSTARTED_LATE tasks. A previous volunteer running the same project had reported a spate of EXIT_TIME_LIMIT_EXCEEDED errors. Both of these are client-generated errors, but I suspect arise from a common cause: poor runtime estimation by the project server. In the specific case of this particular project, I suspect (but haven't had confirmation yet) that a run of very quick-finishing tasks has driven the APR values stored in the host_app_version table to impossible heights: I've seen evidence of a CPU supposedly running at over 200 GFlops. I've given the usual advice: Ensure rsc_fpops_est is accurate Set rsc_fpops_bound to a high multiplier Mark infeasible runtimes as outliers - all of which we've known about for years (runtime outliers were documented in 2014, although I remember the original discussion as being earlier than that). As things stand at the moment, it's very hard for a volunteer to recover from EXIT_TIME_LIMIT_EXCEEDED and ERROR_UNSTARTED_LATE errors - it can be easiest to force-generate a new HOSTID (itself not easy), and thus start with a fresh HAV row. The ERROR_UNSTARTED_LATE (client) errors could be addressed by the OP's suggestion, and it should be an issue on GitHub. I couldn't find an existing one this morning - I'll search again, more carefully, and start one if needed and nobody beats me to it. |
|
2834)
Message boards :
Projects :
For the betterment of BOINC
(Message 79817)
Posted 27 Jul 2017 by Richard Haselgrove Post: I have to wonder what is the returnFor the group, I've pointed out http://boinc.berkeley.edu/wiki/Publications_by_BOINC_projects which is understated and outdated. But would something like that, perhaps with a paragraph from the abstract of each paper, count as a return? |
|
2835)
Message boards :
Projects :
For the betterment of BOINC
(Message 79803)
Posted 26 Jul 2017 by Richard Haselgrove Post: Sorry, not without authority from the group itself. I'll ask.Can you provide a link please?Also making sure that the outcome of the discussions are correctly recorded, and actions properly placed....Our secretary (minute taker) is remarkably efficient, making written minutes available online within an hour of the end of the meeting. Even I can remember if they're accurate for about that long... |
|
2836)
Message boards :
Projects :
For the betterment of BOINC
(Message 79799)
Posted 26 Jul 2017 by Richard Haselgrove Post: Also making sure that the outcome of the discussions are correctly recorded, and actions properly placed....Our secretary (minute taker) is remarkably efficient, making written minutes available online within an hour of the end of the meeting. Even I can remember if they're accurate for about that long... |
|
2837)
Message boards :
Projects :
For the betterment of BOINC
(Message 79793)
Posted 26 Jul 2017 by Richard Haselgrove Post: Are these teleconferences open to the public or recorded and published online?Not open to the public, I'm afraid. It's a good question about recordings - I was there, and I don't honestly know (it hasn't been mentioned). I'll ask next time - recordings might be useful to some future historian. |
|
2838)
Message boards :
Projects :
For the betterment of BOINC
(Message 79792)
Posted 26 Jul 2017 by Richard Haselgrove Post: A litlle note: update the copyright in Boinc Manager, now is "(C) 2003-2015" with 7.6.33 version.v7.6.33 was released over a year ago - OK, in 2016 (5 Jun 2016), so your point is valid. But more than the copyright notice needs updating... |
|
2839)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79784)
Posted 26 Jul 2017 by Richard Haselgrove Post: Unlawful, unconstitutional, unanimous. Slam dunk :-) |
|
2840)
Message boards :
BOINC Manager :
What's new in 7.8.0??
(Message 79766)
Posted 25 Jul 2017 by Richard Haselgrove Post: I've just heard that there was an operational oopsie behind the scenes at LHC - a mistake cancelled a number of active tasks. It happens. But I think BOINC is definitely off the hook for this one. |
|
2841)
Message boards :
Questions and problems :
What are the reasons that some people avoid BOINC project?
(Message 79753)
Posted 24 Jul 2017 by Richard Haselgrove Post: I think memory is the more likely cause. BOINC-spawned project applications run at lowered priority, and modern operating systems are pretty good at multitasking. |
|
2842)
Message boards :
BOINC Manager :
What's new in 7.8.0??
(Message 79744)
Posted 23 Jul 2017 by Richard Haselgrove Post: I agree with the moderator who posted the reply: this is more likely to be a project server (or operations) problem, not something caused by the BOINC Client version. I do know that the Sixtrack administrator has been struggling with workflow and validation problems recently, and may have had to cancel a batch of tasks. |
|
2843)
Message boards :
Questions and problems :
GPU resumes when told to snooze
(Message 79728)
Posted 22 Jul 2017 by Richard Haselgrove Post: Did you use the snooze option in the notification area icon? If you used then that is what it's supposed to do, pause CPU/GPU computing for 30 minutes or 1 hour, I can't remember how long the pause is.One hour. |
|
2844)
Message boards :
Projects :
For the betterment of BOINC
(Message 79722)
Posted 22 Jul 2017 by Richard Haselgrove Post: Gary Charpentier asked in an unrelated thread I didn't see if there was a time for a final report.No, there isn't. I suspect it largely depends on how well and how fast we get along, how well our ideas are accepted by the existing practitioners, and how well you the community keep us provided with ideas and suggestions. But we do have one pretty clear initial aiming point, to be able to present at least an initial report. That's at the BOINC Workshop being held in Paris from Wednesday 6 to Friday 8 September 2017 (see separate thread) - that's less than 7 weeks away. So get your thinking caps on, please. I had a short break booked for this week, before I was invited to join this working group, which is why I've been off the boards for a few days. I'm back now, though quite tired from what turned out to be a busy trip. Let me sleep a bit tonight and tomorrow, and hopefully I'll get up to full speed to help Jord during Monday and Tuesday. |
|
2845)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79673)
Posted 18 Jul 2017 by Richard Haselgrove Post: *plonk down to wait for the news*Fortunately audio only, so they'll never know. |
|
2846)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79669)
Posted 18 Jul 2017 by Richard Haselgrove Post: Good, at least Eric will then be free for that teleconference we'll have tonight. Hint hint, news on this will be coming later tonight or tomorrow. :)If the leash is untied a little? |
|
2847)
Message boards :
Questions and problems :
cc_config.xml not being read by BOINC?
(Message 79654)
Posted 17 Jul 2017 by Richard Haselgrove Post: Even after a night's sleep, I'm afraid I'm out of ideas on this one. Oh, just one last try: If you go to Options, Event Log Options in BOINC Manager and make a minor change (which you can undo later), BOINC will write out a complete pro-forma cc_config.xml file. If you edit that version of the file, BOINC should be able to read it back in. You might also be helped to work out what the difference is between your file and BOINC's file. |
|
2848)
Message boards :
Questions and problems :
cc_config.xml not being read by BOINC?
(Message 79648)
Posted 16 Jul 2017 by Richard Haselgrove Post: And where did you put it? From your log, it should be in C:\ProgramData\BOINC |
|
2849)
Message boards :
Questions and problems :
cc_config.xml not being read by BOINC?
(Message 79646)
Posted 16 Jul 2017 by Richard Haselgrove Post: How are you creating/editing the cc_config.xml file? It needs to be strictly text only. The simplest advice is: Use Notepad Save using ANSI encoding Save as type 'All files' Otherwise, Windows tends to save it as cc_config.xml.txt with the final four characters hidden. |
|
2850)
Message boards :
BOINC client :
The old scheduling problem strikes again
(Message 79617)
Posted 15 Jul 2017 by Richard Haselgrove Post: Am i right in saying, the GPU task actually falls back to last checkpoint and restarts from that point?They should do, provided the developer has implemented checkpointing correctly. |
|
2851)
Message boards :
BOINC client :
The old scheduling problem strikes again
(Message 79609)
Posted 15 Jul 2017 by Richard Haselgrove Post: And why is it removing the task from memory in violation of the leave suspended tasks in memory setting, is it to clear the gpu's memory?Yes, GPU apps are always removed from memory when suspended. GPUs don't have the facility to swap stale memory images out to a paging file on disk, so a suspended task would always continue to occupy real, physical, memory, which might be in short supply. |
|
2852)
Message boards :
Questions and problems :
Suggestion: BOINC only active when screen saver is running
(Message 79604)
Posted 14 Jul 2017 by Richard Haselgrove Post: I think it's a fair point that there should be be a setting, or combination of settings, which allows a 'screensaver only' mode of operation. So - is the problem that this particular user didn't find it, or does it in fact not exist? I remember discussions about how difficult it was to detect that a non-BOINC application was actively running full screen, as would be the case when watching movies. Could that be the problem still? |
|
2853)
Message boards :
BOINC Manager :
Compiling BOINC Manager 7.7 in Ubuntu 14.04.5
(Message 79590)
Posted 13 Jul 2017 by Richard Haselgrove Post: That's another dialog which could use an 'apply' button. |
|
2854)
Message boards :
Questions and problems :
I'm using Ubuntu 16.04, from the terminal i can start BOINC just fine however I can't start the manager at all. How can I fix this?
(Message 79587)
Posted 13 Jul 2017 by Richard Haselgrove Post: Some things I should have added, and will add for the benefit of future readers: One advantage of using the Manager GUI tool is that it issues an automatic 'Read local prefs file' after an edit, so the new setting takes effect immediately. If you create or edit the file at the command line, you have to issue a separate boinccmd --read_global_prefs_override(or wait until the next client restart) before they take effect. One disadvantage of using the Manager GUI tool is that it creates a fully-populated override file, taking control of all web-editable values. My hypothesis in suggesting a single-entry override file was that web settings would remain in control for all other values. But that hypothesis is untested. [The command-line approach seemed appropriate in this case because the OP seemed familiar with that mode of working] |
|
2855)
Message boards :
Questions and problems :
I'm using Ubuntu 16.04, from the terminal i can start BOINC just fine however I can't start the manager at all. How can I fix this?
(Message 79582)
Posted 13 Jul 2017 by Richard Haselgrove Post: Yes, if you use the _override version of the file (the version that the manager, er, manages), that's persistent across project contacts.I have BOINC set to use local preferences, shouldn't this remove that issue?Or - even more simply - you could make the same change via the global preferences page of any project your computer is attached to.Better do it there, else at the next contact with the project website, the value that the project sends out overwrites the manually edited value in the global_preferences.xml file. |
|
2856)
Message boards :
Questions and problems :
I'm using Ubuntu 16.04, from the terminal i can start BOINC just fine however I can't start the manager at all. How can I fix this?
(Message 79575)
Posted 13 Jul 2017 by Richard Haselgrove Post: You might be asking the wrong question here. The Manager won't give you thread (or CPU core ID) level control, even if you get it running. The best you can do is limit the total number of cores that BOINC is allowed to use, and let your other program pick the ones it wants first: BOINC will use whatever's left over, under OS control. The manager would let you create a global_prefs_override.xml file using a GUI tool, but that's all it would do. You could create the file yourself, containing something like <global_preferences> <max_ncpus_pct>62.5</max_ncpus_pct> </global_preferences>(to use five out of eight cores for BOINC) Or - even more simply - you could make the same change via the global preferences page of any project your computer is attached to. |
|
2857)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79453)
Posted 4 Jul 2017 by Richard Haselgrove Post: Just so long as they've automated the restart too... ;-)I think there is an automatic component to a lot of things associated with the Tuesday outages.It's down.And on the 4th of July, too. That (slightly) surprised me. |
|
2858)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79449)
Posted 4 Jul 2017 by Richard Haselgrove Post: It's down.And on the 4th of July, too. That (slightly) surprised me. |
|
2859)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 79402)
Posted 29 Jun 2017 by Richard Haselgrove Post: Any female astronauts in training or are they all promised in marriage?Well, they've just comprehensively beaten the West Indian women in World Cup cricket, having beaten the English last week. I don't know if that'll help them get into space, but it'll give them something to brag about to any prospective husbands. Cricket bat skills should translate to a rolling pin? |
|
2860)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79347)
Posted 27 Jun 2017 by Richard Haselgrove Post: Sorry, you don't know who I was talking to, and you don't know what we were talking about. Suffice to say that my report of yesterday's DNS outage reached the right place, and I know that they know what happened and why. Let's leave it there. |
|
2861)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79345)
Posted 27 Jun 2017 by Richard Haselgrove Post: Lost comms - forgot to plug a switch into a UPS.Yup, that was pretty much my reply too. The reply was "Our assumptions about their setup often don't match reality." |
|
2862)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79339)
Posted 27 Jun 2017 by Richard Haselgrove Post: Ah, I may have a picture of it! When I went on the lab tour in 2011, I was shown ancillary kit in a couple of other rooms apart from the closet. The kit there was the back-room support gear and may well have included the DNS thingy. Er? power outage? Do we not have UPS's anywhere?No idea - the conversation didn't need to go into that level of detail. I think they've been given a wake-up call and there'll probably be internal discussions between the teams involved. |
|
2863)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79336)
Posted 26 Jun 2017 by Richard Haselgrove Post: I have it on good authority that the DNS server belongs to the Space Science Laboratory - and they certainly know SETI exists.Yep it is a DNS problem. If you can be bothered to do anything about it , Richard has shown you how to do it. Me? i'm spending the day in the garden with the DT crossword and a beer (or 2 or 3).And for once you can bitch at the UNI. ADNS1.berkeley.edu has no clue Seti exists. Did Eric forget to pay the bill before he went on vacation? ;-) Apparently, it kept running through a power outage, but lost comms with the rest of the network. |
|
2864)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79330)
Posted 26 Jun 2017 by Richard Haselgrove Post: And neither of the previous two posts are visible in their respective tabs yet, although they show in new tabs. |
|
2865)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79329)
Posted 26 Jun 2017 by Richard Haselgrove Post: I ask, because it's taking ages for my browser to re-display the thread after I've posted to it (normally immediate). |
|
2866)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79328)
Posted 26 Jun 2017 by Richard Haselgrove Post: Someone remind me to ask Jord about this curiosity when he gets back from holiday, please... As you can see, my last post is stamped 14:00:00 UTC. But it was 6-7 minutes before the 'Last Post' column was updated on the index pages. Do we have a master/replica database structure here, and might the current DNS problems delay transaction processing? |
|
2867)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79327)
Posted 26 Jun 2017 by Richard Haselgrove Post: It'll be interesting to watchYep it is a DNS problem. If you can be bothered to do anything about it , Richard has shown you how to do it. Me? i'm spending the day in the garden with the DT crossword and a beer (or 2 or 3).And for once you can bitch at the UNI. ADNS1.berkeley.edu has no clue Seti exists. Did Eric forget to pay the bill before he went on vacation? ;-) Domain expires: 31-Jul-2017 |
|
2868)
Message boards :
Projects :
SETI
(Message 79320)
Posted 26 Jun 2017 by Richard Haselgrove Post: SETI@Home's DNS servers are down. These IP addresses provide basic connectivity: 208.68.240.118 setiboincdata.ssl.berkeley.edu # upload server Oct 2016 208.68.240.119 boinc2.ssl.berkeley.edu # download server Oct 2016 208.68.240.126 setiboinc.ssl.berkeley.edu # scheduler Oct 2016 |
|
2869)
Message boards :
Projects :
News on Project Outages
(Message 79319)
Posted 26 Jun 2017 by Richard Haselgrove Post: SETI@Home's DNS servers are down. These IP addresses provide basic connectivity: 208.68.240.118 setiboincdata.ssl.berkeley.edu # upload server Oct 2016 208.68.240.119 boinc2.ssl.berkeley.edu # download server Oct 2016 208.68.240.126 setiboinc.ssl.berkeley.edu # scheduler Oct 2016 |
|
2870)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79317)
Posted 26 Jun 2017 by Richard Haselgrove Post: $ ping setiathome.ssl.berkeley.edu ping: cannot resolve setiathome.ssl.berkeley.edu: Unknown hostLike I said, DNS is down. These are the notes I made last time. 208.68.240.118 setiboincdata.ssl.berkeley.edu # upload server Oct 2016 208.68.240.119 boinc2.ssl.berkeley.edu # download server Oct 2016 208.68.240.126 setiboinc.ssl.berkeley.edu # scheduler Oct 2016 |
|
2871)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 79308)
Posted 25 Jun 2017 by Richard Haselgrove Post: Getting DNS errors on Main project upload server Main project scheduler Beta project website |
|
2872)
Message boards :
Questions and problems :
Stop BOINC.
(Message 79271)
Posted 24 Jun 2017 by Richard Haselgrove Post: Change it. That's a known problem: it should only come back if you mess with the BIOS settings of the host computer - which BOINC can't do. Mind you, you do need to have enabled the right class of virtualisation first - I hope your adviser gave you that too. |
|
2873)
Message boards :
Questions and problems :
Stop BOINC.
(Message 79263)
Posted 24 Jun 2017 by Richard Haselgrove Post: I would question that advice. What value, from what, to what? |
|
2874)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79253)
Posted 23 Jun 2017 by Richard Haselgrove Post: Followed by an open window (on one of the hottest nights of the year), which let the fire out of the flat and into the cladding.In this case it was apparently an exploding fridge. HTF that happens is beyond me.Yes, to make plastic start with petrol.I was approaching it from the point of view of the materials manufacturer/supplier. Reynobond is patented - Composite panel with a foamed plastic core, US 6455148 B1 - which gives it a certain amount of monopoly protection.Probably because "fire" and "plastic" mix very very well. |
|
2875)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79250)
Posted 23 Jun 2017 by Richard Haselgrove Post: There are two clues if you read back through the thread.It's sad, really.Is "it" whom I suspect? |
|
2876)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79246)
Posted 23 Jun 2017 by Richard Haselgrove Post: The PM count since mid-day - so in the last six hours - has now reached seven. The latest exchange was statement - contradiction with evidence - 'all I know is...'. It's sad, really. |
|
2877)
Message boards :
Questions and problems :
Stop BOINC.
(Message 79244)
Posted 23 Jun 2017 by Richard Haselgrove Post: If you have the Manager open (and you can open it again safely if you closed it): File menu "Shut down connected client..." and (potentially) answer the questions from there. For those other readers who don't use the Manager: boinccmd --quit |
|
2878)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79241)
Posted 23 Jun 2017 by Richard Haselgrove Post: That was the Regional fire control centre in Taunton that was due to be sublet last year? Yes, those were never a very good idea - Yorkshire fire crews sent to wrong address by Cornwall control centre. And the PFI financing model was even worse: drawn up by Tony Blair and Gordon Brown when they were trying to ape the Tories, weren't they? Trouble is, the Tories had about 200 years head start in 'how to line your friends' pockets'. Edit - I think it's a bit similar to the way that the de-mutualised British Building Societies were taken for a sucker punch by the American 'bad mortgage' derivatives market. And we all know what happened then... |
|
2879)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79239)
Posted 23 Jun 2017 by Richard Haselgrove Post: Now Chris has queried the phrase 'Fires connected to fridge freezers and other electrical appliances are relatively common' used in a BBC report. I dug out (From FIRE0601 on https://www.gov.uk/government/statistical-data-sets/fire-statistics-data-tables - most recent year only.) Total Accidental 28428 Faulty fuel supplies 2384 Faulty appliances and leads 4328 Misuse of equipment or appliances 10197 Chip/fat pan fires 1755 Playing with fire 212 Careless handling of fire or hot substances 2864 Placing articles too close to heat 3456 Other accidental 3226 Unspecified 6 Most common cause after misuse - sounds reasonable to me. That's precisely why it's important that fires are contained within the flat and don't spread. |
|
2880)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79237)
Posted 23 Jun 2017 by Richard Haselgrove Post: He hasn't left, just switched to PMs - two so far since my last post. We've been talking about the power of local councils to employ Direct Labour Organisations. Since they're private, I can't tell you about the Scar House Reservoir. Oops :-) |
|
2881)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79234)
Posted 23 Jun 2017 by Richard Haselgrove Post: Generally the full time Council officials are simply the equivalent of, as you say, local civil servants, there is generally no specific expertise there other than in administrative matters.I think that may be a fair statement of current political fashion, but it deeply undervalues the role traditionally played by local government in planning and building our towns and cities. There are some telling statements relevant to the current controversy in Which councils employ the most in-house architects?: The decision by Croydon bucks the trend of recent decades, which has seen councils ridding themselves of their architects departments.I can't find Kensington & Chelsea in the list of 19 London boroughs which still employ at least one architect - maybe yours is missing too? |
|
2882)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79232)
Posted 23 Jun 2017 by Richard Haselgrove Post: WHICH grade of Reynobond was specified at the Grenfell Tower (and others), and WHICH grade was fitted?An important distinction - watch for news from Camden. Anybody else getting a whiff of John Poulson and T. Dan Smith here? |
|
2883)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79228)
Posted 23 Jun 2017 by Richard Haselgrove Post: Reynobond make no claim that their [PE] product meets any fire safety regulations or is non combustible.Which is why they make separate, more expensive, fire resistant products called Reynobond FR and Reynobond A2. Councillors are ordinary people like you and me often with no specific technical expertiseWhich is why they hire council officers (the local equivalent of civil servants) to write recommendations for them. Chancellor Philip Hammond has said he thought the Grenfell cladding was banned in the UK.Under aggressive quick-fire questioning on a Sunday morning television show. I see nothing in his career in governemt (but see *) that would qualify him to make such a statement - except maybe it's banned on Defence properties, which Nick 666 told us here (**) are built to different standards? The TV shows are designed to trick unwary politicians into saying things they shouldn't have and thus get some lively headlines for the lunchtime news: I think this is another example. * Maybe we should send the building inspectors to check out Castlemead Homes? ** Nick's post was at the other place |
|
2884)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79206)
Posted 22 Jun 2017 by Richard Haselgrove Post: I was approaching it from the point of view of the materials manufacturer/supplier. Reynobond is patented - Composite panel with a foamed plastic core, US 6455148 B1 - which gives it a certain amount of monopoly protection. The patent makes much of architectural uses, and cites the Reynobond trade mark. It mentions that the outer metal sheets are "made from a corrosion-resistant aluminum alloy" (american spelling), but the word 'fire' (as in testing, retardant, etc.) appears nowhere. |
|
2885)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79201)
Posted 22 Jun 2017 by Richard Haselgrove Post: The next question I would ask is whether there is any connection between the owners of building materials companies, and the politicians responsible for drafting (or not drafting) building regulations - party donations, shared membership of rolled-up-trouser-leg societies, that sort of thing. |
|
2886)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79197)
Posted 22 Jun 2017 by Richard Haselgrove Post: I've been listening to the statement made by Theresa May in the House of Commons, and the questions she was asked and (barely) answered - enquiry this, investigation that, testing the other. The biggest question seems to be whether flammable cladding was permitted under building regulations. If if was forbidden, then an awful lot of builders and building inspectors will be passing through the courts in the coming months and years. If it was (and still is) permitted, then the Houses of Parliament should be spending their time passing a one-clause bill 'for the avoidance of doubt' saying that no flammable panel should be attached to a building higher than xx metres after, say, 12 noon today. |
|
2887)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79194)
Posted 22 Jun 2017 by Richard Haselgrove Post: And the part I've emphasised has led to the situation we're discussing today.The Greater London Council (GLC) was the top-tier local government administrative body for Greater London from 1965 to 1986. It replaced the earlier London County Council (LCC) which had covered a much smaller area. The GLC was dissolved in 1986 by the Local Government Act 1985 and its powers were devolved to the London boroughs and other entities. A new administrative body, known as the Greater London Authority (GLA), was established in 2000.I would think that many locally Tory controlled London Councils would have a word to say about a new GLC! And who was the leader for many years? Red Ken! If London wishes to vote for Red Ken, then London should be able to vote for Red Ken. That's democracy. |
|
2888)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79190)
Posted 22 Jun 2017 by Richard Haselgrove Post: I thought this was a democracy ...Has anybody else thought of re-establishing the GLC, so that there is democratic control over matters applicable to all London boroughs? |
|
2889)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79188)
Posted 22 Jun 2017 by Richard Haselgrove Post: Since it mentions Paget-Brown by name, I think it's time for a re-post of the letter from a resident of R B K & C. My local tax rebate is blood money |
|
2890)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79157)
Posted 21 Jun 2017 by Richard Haselgrove Post: Bill-by-bill at a glance. But I wouldn't bother glancing at it. Al-Jazeera spotted that Donald Trump was missing, too. And also noticed that control of Executive pay had vanished. "new powers to license a wide range of new commercial spaceflight..." Let's hope that's their exit strategy. |
|
2891)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79155)
Posted 21 Jun 2017 by Richard Haselgrove Post: OK, back to Al-Jazeera for some real news. |
|
2892)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79154)
Posted 21 Jun 2017 by Richard Haselgrove Post: Was that it? What a mealy-mouthed set of platitudes - about the only thing she didn't mention was motherhood and apple pie. And how much time was wasted by how many people - like the guy who carried in the crown she didn't wear. About the only interesting thing was that there's going to be a state visit - but by the King & Queen of Spain, not by Donald Trump. |
|
2893)
Message boards :
Questions and problems :
Problem: Boinc Notices trying to connect to localhost
(Message 79120)
Posted 19 Jun 2017 by Richard Haselgrove Post: Other threads on this same message board suggest that BOINC v7.6.34 for Mac was released to address this very problem. |
|
2894)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79110)
Posted 19 Jun 2017 by Richard Haselgrove Post: What follows is a thought experiment only - I'm not sure where it'll lead. Let's assume that most human settlements started with a pretty homogeneous community. Perhaps the odd Roman, Viking, minstrel or trader wandered through, but basically uniform. So, what happens when the first 'outsider' likes the look of the place, and decides to settle? Round here, the phrase is often "they're driving house prices down" - which might be a good thing, no? In London, maybe the phrase is "they're driving house prices up", I don't know. How are prices set - free market, willing buyer, willing seller. A key driver is what the buyer is willing and able to pay - remember gazumping? It's a human thing, not something intrinsic to the house. So, why do immigrants tend to cluster together in defined areas of towns and cities? Is it because of the comfort from familiar languages, familiar foodstuffs in the market? (almost certainly yes). Or is it because that the previous wave of migrants decide that they're not comfortable with their new neighbours, and move away? (in my experience, also yes). |
|
2895)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 79100)
Posted 19 Jun 2017 by Richard Haselgrove Post: Einstein is back up - operationally and website. |
|
2896)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79099)
Posted 19 Jun 2017 by Richard Haselgrove Post: Should there be extra security, or extra welcome mats? |
|
2897)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 79096)
Posted 19 Jun 2017 by Richard Haselgrove Post: There was an emergency home page. Posted in 'project outages' here. |
|
2898)
Message boards :
Projects :
News on Project Outages
(Message 79094)
Posted 19 Jun 2017 by Richard Haselgrove Post: Einstein@Home We are experiencing some technical difficulties. The Web site is down. There is an unknown connection problem with AEI/Univ. of Hannover. We hope to have things working soon! |
|
2899)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79089)
Posted 18 Jun 2017 by Richard Haselgrove Post: We should be able to sell them some second-hand aesthetic cladding in a week or two. |
|
2900)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79087)
Posted 18 Jun 2017 by Richard Haselgrove Post: 6 things that could topple Donald Trump's border wall is worth a read as well. |
|
2901)
Message boards :
Projects :
recently denied permission, lhcathome
(Message 79084)
Posted 18 Jun 2017 by Richard Haselgrove Post: It would have been a problem at CERN, but it's fixed now. |
|
2902)
Message boards :
News :
NSF funds new model for BOINC
(Message 79048)
Posted 15 Jun 2017 by Richard Haselgrove Post: Their role, like all projects, is to get science done - in their case, complicated and time-consuming science. It isn't their role to provide work for volunteers who offer resources which are unsuited to their needs - other projects are available for that. We disagree, so I accept that we should close this side-discussion. |
|
2903)
Message boards :
Promotion :
An Idea about promoting BOINC in Local School System
(Message 79045)
Posted 15 Jun 2017 by Richard Haselgrove Post: Project granted credit comparison should get you started. But note:You need to be aware different projects assign values to credit differently, so comparing different projects is similar to comparing currencies. If a value in Row A and Column B is greater than 1, that indicates that project A is awarding more credit per cpu second than project B.- so no allowance is made for either project supporting GPUs. |
|
2904)
Message boards :
News :
NSF funds new model for BOINC
(Message 79043)
Posted 15 Jun 2017 by Richard Haselgrove Post: boboviz wrote: In which way they "hinder" the project?Well, as you know, Jacob Klein posted in reply to the post from Stefan that I quoted - asking many of the same questions that Rob Smith has just hinted at. Neither has posted since, so we are - as so often - working somewhat in the dark. BOINC actually consists of three parts:
There are issues with all three which could fit the pattern of complaints from both users and administrators at GPUGrid. Some examples: |
|
2905)
Message boards :
News :
NSF funds new model for BOINC
(Message 79039)
Posted 15 Jun 2017 by Richard Haselgrove Post: Maybe, but imagine how much easier their life would be if BOINC was helping, rather than hindering?It's interesting that the original article I linked makes no mention of BOINC: it credits GPUGrid and NVidia, but not the glue that binds them together. I think that's inevitable - infrastructure should be boring, invisible, and reliable. The second quote should remind us of the ongoing need for maintenance of the existing BOINC framework, which has been sadly neglected for the last two years. GPUGrid has hugely intensive computational needs - their tasks are regularly taking longer than 24 hours on my GTX 970 cards, and that's longer than they would wish. Hardware, drivers, and operating systems are all evolving, and BOINC has to keep up or become irrelevant. I don't see how GPUGrid could be supported under TBD.Wrong example. |
|
2906)
Message boards :
Questions and problems :
Why no GPU tasks?
(Message 79034)
Posted 15 Jun 2017 by Richard Haselgrove Post: Do I understand correctly that this lineThat's what it's telling us, but we're not 100% sure why. From your second message: scheduler RPC backoff (237.83 sec)You need to wait approx 4 minutes and try again. The server will refuse your request for work if you try too soon. But 2017-06-15 03:22:53 | | [work_fetch] --- state for NVIDIA GPU ---is different, but you're not out of the woods yet. |
|
2907)
Message boards :
Questions and problems :
Why no GPU tasks?
(Message 79031)
Posted 15 Jun 2017 by Richard Haselgrove Post: But why, then, is the active work fetch declaring 'account manager prefs'? |
|
2908)
Message boards :
Questions and problems :
Why no GPU tasks?
(Message 79028)
Posted 15 Jun 2017 by Richard Haselgrove Post: The GRC account manager has No ATI and No Intel checked, because I don't have those. No NVIDIA is *not* checked, I just looked.Whatever the state of the checkbox, your BOINC client thinks the state is "don't use": 2017-06-14 20:06:40 | | [work_fetch] --- state for NVIDIA GPU --- 2017-06-14 20:06:40 | | [work_fetch] shortfall 432000.00 nidle 1.00 saturated 0.00 busy 0.00 2017-06-14 20:06:40 | Einstein@Home | [work_fetch] share 0.000 account manager prefs 2017-06-14 20:06:40 | LHC@home | [work_fetch] share 0.000 no applications 2017-06-14 20:06:40 | Milkyway@Home | [work_fetch] share 0.000 account manager prefs 2017-06-14 20:06:40 | SETI@home | [work_fetch] share 0.000 account manager prefs That's the one that answers your opening question. I'm not familiar with the working of account managers, so someone else will have to guide you from here. I just did a work fetch, which failed because Seti does not have work units at the moment.Not quite. The real reason is likely to be the last one in the list: 2017-06-14 20:06:43 | SETI@home | This computer has reached a limit on tasks in progress SETI has a hard limit of 100 CPU tasks in progress. That sounds OK - you can re-allow work fetch for other projects and turn off the logging flag while you explore the account manager. |
|
2909)
Message boards :
Questions and problems :
Why no GPU tasks?
(Message 79023)
Posted 14 Jun 2017 by Richard Haselgrove Post: I'm afraid we're going to need at least one cycle of <work_fetch_debug> to work that out. Press Ctrl-Shift-F to bring up the picker, then check work_fetch_debug and 'OK': after the first cycle has appeared in the event log, reverse the process. It puts a lot of gobbledegook into the event log every minute, but if you post one cycle here, we might be able to pick some bones out of it. |
|
2910)
Message boards :
The Lounge :
BOINC Politics thread
(Message 79005)
Posted 13 Jun 2017 by Richard Haselgrove Post: Maybe that Cornwall thingy you found is just the start of it being rolled out nationwide, Richard. You know... your country needs you sort of thing, to extract your ill people...I assumed it was just the start of - how do you spell it - Corwexit? |
|
2911)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78995)
Posted 13 Jun 2017 by Richard Haselgrove Post: And back. Looks like it's going to be OK. |
|
2912)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78991)
Posted 13 Jun 2017 by Richard Haselgrove Post: OK, prep done, on my way. Bradford Royal Infirmary, ward 29, room 3 if you need to send a search party. |
|
2913)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78990)
Posted 13 Jun 2017 by Richard Haselgrove Post: GPU only, two cards in machine. I run a short cache six days a week, but stock up on Tuesday mornings.Edit:Hoarder. Is that for CPU and GPU together? |
|
2914)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78987)
Posted 13 Jun 2017 by Richard Haselgrove Post: Patient is elderly and severely disabled - no question of walking. She has: own electric wheelchair, own (directly employed) care worker on site with her. Will be a chair and taxi job to get her home - no problem, she's done that before. But we'll probably need to borrow a hospital hoist to transfer her from bed to chair. |
|
2915)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78985)
Posted 13 Jun 2017 by Richard Haselgrove Post: This looks helpful. Why is Cornwall the only hospital trust to put this online? |
|
2916)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78980)
Posted 13 Jun 2017 by Richard Haselgrove Post: Well, I may get to see a little bit of that today - just had an urgent call to spring somebody from hospital against - it sounds like - the administration's advice. Bout of pneumonia, now sorted out - but "if I have to spend another night in here, somebody's going to get hurt, and it won't be me".And more seriously, if she can end austerity that easily, it wasn't needed in the first place. A change of direction after the bankers ran away with the money in 2008, sure, but how hard and how long was pure Tory ideology.Well it's crushed me under its wheels unfortunately. It's forced my daughter to live abroad, and saddled my son with almost £40 000 of debt, whilst containing multiple ATOS (who have done very very well out of it) insults like my other half being able to get something out of his top pocket meaning he is fit for work. So it's not been a lot of fun I must say, and I don't expect that to change much to be honest, but we'll see :) So, searching for a pro-forma self discharge letter for her to sign - anybody got any advice/experience? |
|
2917)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78974)
Posted 13 Jun 2017 by Richard Haselgrove Post: And more seriously, if she can end austerity that easily, it wasn't needed in the first place. A change of direction after the bankers ran away with the money in 2008, sure, but how hard and how long was pure Tory ideology. |
|
2918)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78971)
Posted 13 Jun 2017 by Richard Haselgrove Post: Uploads and reporting are still ticking away, not getting any replacements. Outrage cometh.Just the work mix. No tasks sent Tasks for CPU are available, but your preferences are set to not accept them Tasks for AMD/ATI GPU are available, but your preferences are set to not accept them Tasks for Intel GPU are available, but your preferences are set to not accept them Edit: 13/06/2017 09:01:29 | SETI@home | Scheduler request completed: got 172 new tasks |
|
2919)
Message boards :
GPUs :
OpenCL 2.0 "File not found" FFS!
(Message 78970)
Posted 13 Jun 2017 by Richard Haselgrove Post: 'Output file absent' more commonly means the application has crashed without producing any output, not that anything has deleted it. The 'file not found' in the thread title refers to the AMD support page on the web - I get the same here. |
|
2920)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78969)
Posted 13 Jun 2017 by Richard Haselgrove Post: Austerity is over! Theresa May's found a magic money tree!I like that link on the right more - for us foreigners: Confused by UK politics? You're not alone. Although I would think even the residential Brits are a bit confused by now. :)Strong and stable, strong and stable. Oh, it's now the opposite. And if you believe that, you'll believe anything. |
|
2921)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78956)
Posted 12 Jun 2017 by Richard Haselgrove Post: Yes, that's a good link. But it gets better... The DUP (10 MPs) are strongly against gay marriage. The Scottish Conservative party (13 MPs - up from 1 MP last time) is led by Ruth Davidson, who is an 'out' lesbian engaged to be married to her female partner. So if Theresa May hooks up with the DUP, she risks alienating the Scots (nothing new there) and going down by three votes... |
|
2922)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78954)
Posted 12 Jun 2017 by Richard Haselgrove Post: US appeals court blocks 'travel ban' A US appeals court has upheld a decision blocking President Trump's revised "travel ban" on people from six mainly Muslim nations. |
|
2923)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78949)
Posted 12 Jun 2017 by Richard Haselgrove Post: My (former) MP made his one and only TV appearance about half way through the short version. I won't spoil the punchline for you ;-) |
|
2924)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78947)
Posted 12 Jun 2017 by Richard Haselgrove Post: I'll dip a tentative toe in the water... It's off. It's back on. It may be back off again. Her Majesty will not be amused - especially if she has to miss Ascot ("A horse race, m'lud"). The trailers for tonight's Panorama special are looking interesting. |
|
2925)
Message boards :
Questions and problems :
Why no GPU tasks?
(Message 78943)
Posted 12 Jun 2017 by Richard Haselgrove Post: Well, a K2000 should do the trick, but... Einstein is offline (power surge - see elsewhere on this board) LHC is a CPU / VM project only MilkyWay is predominantly an ATI/AMD project, although they do have NVidia apps. SETI is up and running, but I've had several responses (to an NVidia request) like 12/06/2017 13:59:38 | SETI@home | Tasks for CPU are available, but your preferences are set to not accept them 12/06/2017 13:59:38 | SETI@home | Tasks for AMD/ATI GPU are available, but your preferences are set to not accept them 12/06/2017 13:59:38 | SETI@home | Tasks for Intel GPU are available, but your preferences are set to not accept themIt's possible that BOINC has gone into backoff because of previous failures. I'd suggest you (temporarily) set 'No new tasks' for all projects except SETI, and click 'update' for SETI in an attempt to force a work request. Once you have work - it may take several goes, each five minutes apart - reset everything to normal. If that doesn't kick it into life, post again. I agree your initial log message (blank reason for NVidia fetch) is unusual: you may need to enable some debug log flags. |
|
2926)
Message boards :
News :
NSF funds new model for BOINC
(Message 78941)
Posted 12 Jun 2017 by Richard Haselgrove Post: Two straws in the wind from GPUGrid today: New publication: Complete protein–protein association kinetics! Scientists reveal first details at the atomic interactions between proteins(Google Translate version of original article in Spanish) GPUGRID active users are falling dramatically! Fixing bugs with BOINC is relatively pointless from our perspective (and time-intensive). We are considering rather other options like moving out of it, but don't ask when or how as it's more an idea than a scheduled plan.It's interesting that the original article I linked makes no mention of BOINC: it credits GPUGrid and NVidia, but not the glue that binds them together. I think that's inevitable - infrastructure should be boring, invisible, and reliable. The second quote should remind us of the ongoing need for maintenance of the existing BOINC framework, which has been sadly neglected for the last two years. GPUGrid has hugely intensive computational needs - their tasks are regularly taking longer than 24 hours on my GTX 970 cards, and that's longer than they would wish. Hardware, drivers, and operating systems are all evolving, and BOINC has to keep up or become irrelevant. I don't see how GPUGrid could be supported under TBD. |
|
2927)
Message boards :
Questions and problems :
Why no GPU tasks?
(Message 78940)
Posted 12 Jun 2017 by Richard Haselgrove Post: What sort of NVidia GPU do you have? - model number and driver version, please. Also, please post the rest of the startup lines from the Event Log, covering the GPU detection phase. |
|
2928)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78922)
Posted 11 Jun 2017 by Richard Haselgrove Post: We have another general election, and keep on having elections until we get it right. |
|
2929)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78920)
Posted 11 Jun 2017 by Richard Haselgrove Post: calm down and get behind the Prime MinisterAnd form an orderly queue. et tu, Brute? |
|
2930)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78906)
Posted 11 Jun 2017 by Richard Haselgrove Post: I wouldn't mind everyone else's views on this. And do so in the open, so that everyone can read it all.Well, I've let that marinade for a morning, and I think it's well enough stewed to post what were my first thoughts on reading it. We live in interesting political times, and it would be a great shame to have to close down all discussion of them. And discussion, to be interesting, has to be between people of differing prior experience, differing prior assumptions, differing prior views. We might even change each others' minds! But it also has to be both polite and thoughtful. The ad hominem attacks have got to stop. [I was going to add that we ought to extend that rule to comments on the politicians too, but after the belly laugh Mark's comments on Ben Gummer have just given me, I haven't the heart. Whatever happened to his, I presume, sister Cordelia, who was force-fed a hamburger on camera at the height of the BSE crisis?] So, I propose that Jord leaves the thread open for debate, and goes away on holiday without a care in the world. BUT: leaves instructions to the other mods that at the first sign of name-calling or other trouble, they lock it - no warning, no PM to anybody, no post in thread, just lock it and walk away. No appeal either. And with that, back to politics. What says anyone to the idea that we adopt the French voting system? They seem to have dug themselves out of a hole with the minimum of fuss. |
|
2931)
Message boards :
News :
NSF funds new model for BOINC
(Message 78882)
Posted 10 Jun 2017 by Richard Haselgrove Post: The quoted text simply becomes part of your reply post. You can delete some or all of it, as you wish - either at the time, or up to an hour later with the 'edit' button. |
|
2932)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78880)
Posted 10 Jun 2017 by Richard Haselgrove Post: The question now is: will May survive to lead the Tories for the next 5 years?Or even the next five days. The BBC are reporting with some authority (though no named sources) that... Norman Smith, Assistant political editor wrote: Posted at 11:56 10/06/17Edit: both advisers have resigned. Theresa May is weakened, and the Tory back-benches have tasted blood. |
|
2933)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78860)
Posted 9 Jun 2017 by Richard Haselgrove Post: The question now is: will May survive to lead the Tories for the next 5 years?Not if the Evening Standard has anything to do with it. |
|
2934)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78859)
Posted 9 Jun 2017 by Richard Haselgrove Post: it's the party political machinery which pulls all the strings - mostly in private.With one exception - each party has at least one conference per year, with the main ones being in the early autumn (September/October). And again for the main parties, many of the set-piece speeches are televised. I suspect the Conservative party conference this year will have more viewers than usual... |
|
2935)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78856)
Posted 9 Jun 2017 by Richard Haselgrove Post: OK, if no-one else wants to, I'll try. It's all based on constituencies - 650 of them at the moment: geographical areas. People living in that area get one vote, for one candidate of their choice, and the person with the most votes is the Member of Parliament for that area. Nothing proportional, no votes to transfer. In theory, practically anybody can put themselves forward as a candidate (and some very strange people do), but in practice the vast majority are chosen and supported by political parties: very occasionally, independent candidates get elected because of some special issue they've taken up in a local area, but it's very rare. There were none last night. Again theoretically, all the power resides in the hands of those 650 directly-elected MPs. But in practice, it's the party political machinery which pulls all the strings - mostly in private. Since every member has a party affiliation which got them nominated and elected in the first place, they tend to vote in herds: so the party with the most elected members usually votes their own members into positions of executive power like Prime Minister and other departmental ministers. Enough to get you started? |
|
2936)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78852)
Posted 9 Jun 2017 by Richard Haselgrove Post: Can there be a coalition with the DUP when you consider the Brexit border issues?And the DUP's inability to power-share with Sinn Fein, and inability to address the renewable energy scandal, to name but two. |
|
2937)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78851)
Posted 9 Jun 2017 by Richard Haselgrove Post: Mine has also returned from blue to red - two rather lacklustre tories since the late Bob Cryer held it. But the majority is only 249, so we're in for massive leafleting and doorstepping when the inevitable rematch is called. This House of Commons has no chance of lasting five years. |
|
2938)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78848)
Posted 9 Jun 2017 by Richard Haselgrove Post: It's them, or a coalition with the DUP. I know which I'd prefer. |
|
2939)
Message boards :
Questions and problems :
e-mail from this project down?
(Message 78845)
Posted 9 Jun 2017 by Richard Haselgrove Post: I got my delayed SETI mail overnight. I'm surprised to hear it attributed to DNS, but whatever they tweaked, it's worked. |
|
2940)
Message boards :
Questions and problems :
e-mail from this project down?
(Message 78839)
Posted 8 Jun 2017 by Richard Haselgrove Post: Checking them out, all the emails I got came from this site. Neither SETI site have liberated theirs yet. |
|
2941)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78838)
Posted 8 Jun 2017 by Richard Haselgrove Post: That's the question of the night. BBC European editor is saying the EU wanted a single strong leader - any leader - above all else. And that seems the least likely outcome at this early stage. |
|
2942)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78836)
Posted 8 Jun 2017 by Richard Haselgrove Post: But the exit poll in 2015 was a lot closer than the campaign opinion polls. We wait and see - the first two real results didn't exactly confirm the exit poll. |
|
2943)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78833)
Posted 8 Jun 2017 by Richard Haselgrove Post: Wow, did I hear that exit poll right? This is going to be an interesting (and long!) night. |
|
2944)
Message boards :
Questions and problems :
e-mail from this project down?
(Message 78832)
Posted 8 Jun 2017 by Richard Haselgrove Post: And here too. Somebody's found the stopper, and pulled their finger out. |
|
2945)
Message boards :
Questions and problems :
Where is my GPU?!
(Message 78823)
Posted 8 Jun 2017 by Richard Haselgrove Post: The later post indicates that the AMD device is now detected as OpenCL as well. But it isn't clear to me whether the real question is one of detection, or one of whether the AMD card is being used. MilkyWay is possibly not the best project to use for this. My understanding is that MW requires double-precision maths support from the GPU, and the HD 6350 is an entry-level device without DP support (source https://en.wikipedia.org/wiki/Radeon_HD_6000_Series). That might account for the computation errors. Edit - look in the table of specifications under 'Processing Power' - the double precision column shows N/A until you reach the HD 6930. |
|
2946)
Message boards :
The Lounge :
BOINC Politics thread
(Message 78807)
Posted 8 Jun 2017 by Richard Haselgrove Post: Looks like they're having some fun in Chatham for the next two weeks :-)Well, there's a precedent... Fascinating link, Sirius - thanks for sharing it. Yes, I'm one of the ones who wasn't taught about that episode in school, and hadn't picked up on in my developing interest in non-school history in later life - even though I've been following the anniversary stories of the Great Fire the year before and (to a lesser extent) the Plague the year before that. But just 21 years after the Battle of Medway came The Glorious Revolution, when the English King was deposed and replaced by, yes, a Dutchman (and his wife). Many parallels between the second half of the 17th century and the first half of the 21st - religious sectarianism (protestant/catholic and sunni/shia), to name but one. Let's hope it doesn't take 21 years to pass a replacement Bill of Rights, after Theresa May's threat on Tuesday night to tear up the existing one. Correction from another link - a Dutch woman (and her husband). |
|
2947)
Message boards :
The Lounge :
Discussion thread for These are no longer with us, may they have peace
(Message 78719)
Posted 6 Jun 2017 by Richard Haselgrove Post: Adnan Khashoggi Donald Trump bought his old yacht! Margaret Thatcher’s role in securing controversial £42bn arms deal with Saudi Arabia revealed What we’ve learned from fifty years of Saudi arms deals |
|
2948)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 78717)
Posted 6 Jun 2017 by Richard Haselgrove Post: Jeff is looking into the SSP warnings, and is asking if anyone saw similar events prior to my report here '[As of 3 Jun 2017, 8:40:05 UTC]'? I've pointed out that was only 24 hours after a web-server crash and reboot by data centre staff - that'll be what made the messages seem important to me. They've put in some extra diagnostics and will monitor. At which point, another email flashes up: "I think we fixed it." |
|
2949)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78704)
Posted 6 Jun 2017 by Richard Haselgrove Post: I have received a PM from Chris which will remain private, but is so wrong on so many levels that I won't encourage him by replying. And I'll take another timeout from this thread, too. See you all on the other side of the UK General Election. |
|
2950)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78693)
Posted 6 Jun 2017 by Richard Haselgrove Post: I see Jord. So your little mates Sirius and Gary are allowed to have the last word are they, with no right of reply?Chris, I'm sorry to say: you're beginning to sound like the old 'Saturday Night Mark'. To the rest of us, Jord seems to be doing exactly what a volunteer mod should do: let the conversation meander along without interference until it starts to turn unpleasant, and then step in quickly to calm things down. That's how I tried to play it when I was a mod at SETI, and how I would play it now if I was in a hot seat again. Mark wasn't banished on Saturday nights because he was called Mark: he was banished, or given six hours to 'cool off' (euphemism), because he was making the place uncomfortable for everyone else. It's the behaviour which is addressed by mods - mod intervention isn't directed ad hominem either. Chris, please stop trying to earn yourself a reputation like Mark's. Edit - cross-post with #78691. Feel free to remove if I was too late. |
|
2951)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78683)
Posted 6 Jun 2017 by Richard Haselgrove Post: Thank you. |
|
2952)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 78674)
Posted 6 Jun 2017 by Richard Haselgrove Post: It would seem a good thing to do. Thank you.Done. It appears that during these episodes, the remaining information is unreliable: upload and download servers, for example, are shown as 'disabled' when they are in fact running normally. |
|
2953)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 78672)
Posted 6 Jun 2017 by Richard Haselgrove Post: Biggest set of warnings yet [As of 6 Jun 2017, 11:00:03 UTC]: 133 warnings on the SSP. I think I'd better report them in. |
|
2954)
Message boards :
Questions and problems :
e-mail from this project down?
(Message 78655)
Posted 5 Jun 2017 by Richard Haselgrove Post: If you haven't done so yet, you should also ask your ISP if they block *.berkeley.eduWell, I got the email from David to <boinc_projects> about hiding the zip code request. BT put it through X-Junkmail-Premium-Raw: score=9/50,refid=2.7.2:2017.6.5.204815:17:9.975,ip=128.32.13.237,rules=__CSHC_RCVD, __TO_MALFORMED_2, __TO_NAME, __TO_NAME_DIFF_FROM_ACC, __HAS_FROM, FROM_EDU_TLD, __HAS_MSGID, __SANE_MSGID, DATE_TZ_NA, __USER_AGENT, __MOZILLA_USER_AGENT, __MIME_VERSION, __SUBJ_ALPHA_END, __HAS_LIST_ID, __HAS_LIST_UNSUBSCRIBE, __HAS_LIST_HELP, __HAS_LIST_SUBSCRIBE, __CTE, __CT, __CT_TEXT_PLAIN, FROM_SAME_AS_TO_DOMAIN, __TO_IN_SUBJECT, __ANY_URI, __HTTPS_URI, __URI_WITH_PATH, __URI_NO_WWW, __PHISH_SPEAR_DETAILS, __CP_URI_IN_BODY, __PHISH_SPEAR_ACCOUNT_1, __URI_IN_BODY, __URI_NOT_IMG, __NO_HTML_TAG_RAW, BODYTEXTP_SIZE_3000_LESS, BODY_SIZE_800_899, __MIME_TEXT_P1, __MIME_TEXT_ONLY, __URI_NS, HTML_00_01, HTML_00_10, __HAS_LIST_HEADER, BODY_SIZE_5000_LESS, __FROM_DOMAIN_IN_RCPT, __TO_REAL_NAMES, __SINGLE_URI_TEXT, SINGLE_URI_IN_BODY, __PHISH_SPEAR_STRUCTURE_1, BODY_SIZE_1000_LESS, BODY_SIZE_2000_LESS, TO_IN_SUBJECT, __MIME_TEXT_P, BODY_SIZE_7000_LESS, URI_WITH_PATH_ONLYand it passed. I'm not going to investigate any further tonight. |
|
2955)
Message boards :
Documentation :
Drill Down by ZIP?
(Message 78653)
Posted 5 Jun 2017 by Richard Haselgrove Post: The usefulness probably varies from country to country. If the ZIPs are allocated in sequence across a country, you could generate a nice visualisation of where your computing power comes from. But UK post codes are alphabetic, and are derived from the name of the nearest large (or sometimes not-so-large) town. Mine starts with a 'B', but I'm in the middle of the country, not tucked away in a corner. Getting a geographic location from a UK post code is possible, but it requires proprietary software and a licensed copy of the Postal Address File - neither easy nor cheap. I doubt any UK BOINC project would bother. |
|
2956)
Message boards :
Questions and problems :
e-mail from this project down?
(Message 78650)
Posted 5 Jun 2017 by Richard Haselgrove Post: I got PMs (one for hide, one for restore), but no e-mails. Nothing trapped in the spam filter. |
|
2957)
Message boards :
Questions and problems :
e-mail from this project down?
(Message 78646)
Posted 5 Jun 2017 by Richard Haselgrove Post: Add another project to the 'no emails' list - SETI Beta. Just got a PM, but no e-mail - preferences are set for 'immediately', as usual. |
|
2958)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78629)
Posted 5 Jun 2017 by Richard Haselgrove Post: It would be foolish to think that MI5/6 and GCHQ were not aware of the views expressed on Boinc projects threads. When does mischievous become subversive? In the same way the American NSA are well aware of those who post in Seti politics.Eh? So what? We were talking about 'those in power': I very much doubt that any members of the Cabinet or Shadow Cabinet post on any SETI message boards... |
|
2959)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78624)
Posted 5 Jun 2017 by Richard Haselgrove Post: Rules of the game has changedTook me a long time to find the source, and the context of what was actually said - nearly twelve years ago - makes it sound very different. Neglecting that the tense has changed twice in two successive misquotes (BBC from BBC, and you from BBC), the passage reads: [the Prime Minister] said British hospitality had been abused and people should know the "rules of the game are changing".That's a direct copy'n'paste. The 'he' at the end gives the game of quotations away: it was Tony Blair. I've played a game of quotations, but the real point is that those in power woke up to terrorism not being a game at least a decade ago - if they ever thought it was: I doubt the Conservative party members blown out of their beds by the Brighton conference bombing ever held that view (ask Norman Tebbit). To suggest otherwise is mischievous. |
|
2960)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78620)
Posted 5 Jun 2017 by Richard Haselgrove Post: The owner of the Daily Mail couldn't possibly support Tottenham, could he? (I genuinely don't know, and I don't care either) |
|
2961)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78617)
Posted 5 Jun 2017 by Richard Haselgrove Post: "I did my bit... but the authorities didn't do their bit"."At the moment, I'm taking those sorts of reports as 'he said ... he said' hearsay. We don't know exactly what phone call was made - what number was dialled, how clearly the concern was expressed. And I doubt many callers to the dedicated tip-off lines keep a careful log of their calls. The police have said - I think referring to the Manchester bomber - that they have checked their logs and can't find any record of the calls having been made. So we're stuck, and it's too early for a similar records search to have been conducted for the possible London tip-offs. But the assertion is worrying enough that I hope there will be a proper trawl through the paperwork and a public statement made of what is found. And possibly operational changes may need to be made: I think it's possible that calls have been made and followed up, but that the follow-up wasn't visible to the caller and wasn't reported back to them. We know from complaints on this very board how discouraging it can be if bug reports aren't acknowledged and acted on: that applies in the real world too! Meanwhile, are other people as alarmed as I am by this little snippet buried amongst all the London coverage on the BBC? Saudi Arabia, Egypt, Bahrain, the United Arab Emirates and Yemen have cut diplomatic ties with Qatar Another Gulf War in the making? Wannabe superpowers take sides? Another tidal wave of refugees? That's the last thing we need right now - or ever. |
|
2962)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78599)
Posted 4 Jun 2017 by Richard Haselgrove Post: I was wondering about those 50 shots - from eight separate police weapons - too. But under the circumstances of last night: three perpetrators, all three carrying large bladed weapons, and all three visibly wearing what appeared to be suicide belts, I hope it'll be adjudged appropriate force - even though the suicide belts were found to be inert after the event. It seems (unless something as yet unknown is found during the enquiry) to be a clear case of 'suicide by cop', and they got what they must have expected. It also seems that by the time the shots were fired, bystanders were already sheltering inside buildings or running away: the attackers were - judging by where the bodies were filmed - out on the street, and reasonably well isolated from the general public. A single third-party wounding, however regrettable, is a better outcome than might have been the case. But it's a vivid demonstration of the firepower authorised for the police, let alone the army, under current circumstances. I still wonder if that's wise. Imagine at the concert two weeks ago: indoors (covered foyer), single attacker, concealed bomb. At what point would armed police (if present) have decided that his demographic, his apparel, his demeanour, justified summary execution? Yet if armed police had chosen to drop him without danger to the parents and children sharing the public space, would we have judged that action appropriate, too? In the end, it might have saved more lives than were lost last night. It's very, very difficult. As I said last time, I don't think we can beat the terrorists unless we 'win smart': if we simply uprate the hardware without re-writing the manual, I don't think the outcome would be a win. |
|
2963)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78588)
Posted 4 Jun 2017 by Richard Haselgrove Post: Here's another 'hindsight is wonderful' thought. According to both Theresa May and the chief of police, there were three active killers in this latest incident, and all three had been shot dead by police within eight minutes. I haven't heard even a hint that terrorist activity continued after those eight minutes. But I am hearing that residents, including tourists staying in local hotels, were evacuated in a hurry - some in nightclothes or underwear, and without their mobile phones - and are only now being allowed to return to their accommodation. And thus regain the ability to tell their families that they are safe. In the UK, we have a habit of going into lock-down mode after any incident, while 'procedures are followed'. Roads remain closed for hours after accidents, for example. But a Belgian tourist just interviewed on TV said 'that's what the terrorists want - to disrupt normal life'. Aren't police tactics doing that for them? Obviously, opening the area back up too soon would have been disastrous if, say, a time-bomb had been left in the crashed van. But wouldn't it be better to move to a policy of rapid risk assessment and return to normal life where possible, rather than the current 'total control'? |
|
2964)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78579)
Posted 3 Jun 2017 by Richard Haselgrove Post: Many more exec errors on the server status page (both columns) at the 23:30:03 UTC refresh. Data preserved, but I won't try to make sense of it until tomorrow. |
|
2965)
Message boards :
Documentation :
Drill Down by ZIP?
(Message 78572)
Posted 3 Jun 2017 by Richard Haselgrove Post: It's an optional field under 'Account information' - in the UK called Postal code (well, near enough). I've filled mine in. |
|
2966)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 78566)
Posted 3 Jun 2017 by Richard Haselgrove Post: If 'all your files' includes your BOINC data directory, then make sure the new BOINC install points to the same place, and it'll pick up just where it left off - including work in progress. No attaching needed. |
|
2967)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 78557)
Posted 3 Jun 2017 by Richard Haselgrove Post: Well, if the have the manpower to investigate and cure the problems, rather than simply apply brute force, there's some smoking evidence here. But I'm not holding my breath. |
|
2968)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 78554)
Posted 3 Jun 2017 by Richard Haselgrove Post: Just preserving a set of error messages I saw at SETI this morning, in case anyone wants to do any debugging. [As of 3 Jun 2017, 8:40:05 UTC] Server status Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/mysql_connect_test oscar' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 326 Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/mysql_connect_test carolyn' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 334 Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/paddym_connect_test' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 342 Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/marvin_connect_test' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 350 Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/apache_running_on_host synergy' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 369 Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/nginx_running_on_host georgem' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 386 Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/nginx_running_on_host vader' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 392 Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/apache_running_on_host bruno' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 406 Warning: exec(): Unable to fork [/usr/bin/ssh synergy /bin/ps ww 10388 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 Warning: exec(): Unable to fork [/bin/ps ww 1709 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 Warning: exec(): Unable to fork [/bin/ps ww 1889 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 Warning: exec(): Unable to fork [/bin/ps ww 2072 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 Warning: exec(): Unable to fork [/usr/bin/ssh georgem /bin/ps ww 21836 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 |
|
2969)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78553)
Posted 3 Jun 2017 by Richard Haselgrove Post: I used to get free poppadoms at another takeaway while I waited, but they sold out to different owners and the quality went right down - I stopped going there. |
|
2970)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78550)
Posted 3 Jun 2017 by Richard Haselgrove Post: I find that wearing a full beard gets me a comradely smile of recognition from both the local Sikh bus-driver and the staff of the Bengal Tiffin takeaway. |
|
2971)
Message boards :
Questions and problems :
WCG keeps inserting itself as my account manager
(Message 78545)
Posted 3 Jun 2017 by Richard Haselgrove Post: Does it really say exactly that? I don't use an account manager, so I can't reproduce exactly, but I've looked in two places: My Event Log says 30-May-2017 13:05:30 [Einstein@Home] General prefs: from Einstein@Home (last modified ---)(there should be a date-stamp in there, but because of a long-standing bug in BOINC, it gets eaten during the transfer process) And the local preferences dialog says Using local preferences. Click "Use web prefs" to use web-based preferences from http://einstein.phys.ewn.edu/In both cases, the word is 'from' - that's where my chosen preferences must have been changed most recently. But it doesn't say 'managed by' - and I don't see any of the management tools like 'synchronise'. If you feel that the dialog header should be modified when Account Manager tools are activated (i.e., when you have successfully attached your BOINC client to an Account Manager), I suggest you open an issue at https://github.com/BOINC/boinc/issues. |
|
2972)
Message boards :
Questions and problems :
WCG keeps inserting itself as my account manager
(Message 78530)
Posted 2 Jun 2017 by Richard Haselgrove Post: WCG won't be acting as an account manager, but just as custodian of your most recent set of preference changes. Anything you (or perhaps BAM) set using any project website is passed from site to site as your computer contacts each project in turn: the timestamp and location of the most recent change is passed as part of this process, so that the collective group of projects you're attached to knows what is a change, and what is simply a re-confirmation of what it knows already. |
|
2973)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78526)
Posted 2 Jun 2017 by Richard Haselgrove Post: Couldn't open file '....cpt': No such file or directory is OK if it happens right at the beginning of a task: it's Einstein's way of working out if you're starting a new task from scratch, or picking up in the middle of something you were doing yesterday. Either is possible, and both work fine. |
|
2974)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78508)
Posted 2 Jun 2017 by Richard Haselgrove Post: Small straw in the UK election wind: I went down to collect my Guardian newspaper as usual, at lunchtime. I saw that the racks of newpapers for general sale were practically empty, and feared that the wholesaler had failed to deliver (again). But no - deliveries had arrived as usual, and they had simply sold out - much higher demand than usual. According to the shop assistant behind the counter, people had been buying a range of different papers, and one she asked had said that it was in order to get a range of different viewpoints on the election. So, perhaps, people are being more engaged and thoughtful than usual about where to place their vote? That could affect the outcome in any direction, of course, but maybe the simplistic 'tory walkover' we all assumed at the beginning was wide of the mark? |
|
2975)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78498)
Posted 2 Jun 2017 by Richard Haselgrove Post: ... is it just the web server that's down, or everything?Just the web server - work distribution (uploading, downloading, reporting, fetching) is working normally, as is validation and credit updating. Nothing to worry about. |
|
2976)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78492)
Posted 2 Jun 2017 by Richard Haselgrove Post: The question being asked here is whether Trump can withdraw before the next US presidential election. |
|
2977)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78489)
Posted 2 Jun 2017 by Richard Haselgrove Post: Interesting piece in the Washington Post on companies and states that oppose Trump.Theresa May has refused to sign a letter of protest drafted by the other three European G7 members. Under the Tories, the UK more and more comes to resemble the 'Airstrip One' of George Orwell's 1984. |
|
2978)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78486)
Posted 2 Jun 2017 by Richard Haselgrove Post: Here's an interesting twist in the UK general election: Conservative candidate charged over election expenses That's expenses mis-stated for the last election, rather than this one, but better late than never. Should have been dealt with ages ago, but full marks to the Crown Prosecution Service for not kicking it into the long grass during purdah. |
|
2979)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78453)
Posted 1 Jun 2017 by Richard Haselgrove Post: If the President of Syria simply states that there are no chemical weapons, does that make the MOAB (sorry, different target, 59 Tomahawk cruise missiles) a war crime?Would increased climate-damaging emissions count as the use of chemical weapons - a crime against humanity?Well not if the person in charge says there are no such things as "climate-damaging emissions". It would save a lot of criminal court time if every defendant could simply say "there's no such thing as the jemmy you say you saw in my hand - case dismissed". |
|
2980)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78451)
Posted 1 Jun 2017 by Richard Haselgrove Post: Well, 36 minutes late, but he's said it: The United States will withdraw from the Paris climate accord. Would increased climate-damaging emissions count as the use of chemical weapons - a crime against humanity? If so (and following his own precedent), are we now authorised to drop a MOAB against, say, Andrews Air Force Base? |
|
2981)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78446)
Posted 1 Jun 2017 by Richard Haselgrove Post: I only see two votes - one each in the House and the Senate - that needn't take long :)Could somebody please impeach him as quickly as possibleImpeachment isn't a quick process. Read this Wiki part for what's needed. Our 'Brexit Bill' - the European Union (Notification of Withdrawal) Act 2017 - had to go through twelve steps: it was tabled on 26.01.2017, and received Royal Assent on 16.03.2017 I make that 52 days - the USA should be able to manage their simpler process in 10 days? |
|
2982)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78423)
Posted 31 May 2017 by Richard Haselgrove Post: Trump 'poised to quit Paris climate deal' Could somebody please impeach him as quickly as possible, and replace him with a President with at least a passing knowledge of scientific method and interpretation of evidence? |
|
2983)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78399)
Posted 30 May 2017 by Richard Haselgrove Post: Yes, indeed - François Piednoël and the skulltrails. Google found him on twitter - boasting this morning about 7.5 GHz. Same old François. |
|
2984)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78385)
Posted 30 May 2017 by Richard Haselgrove Post: At the back of my mind there is the memory of the three day outages...I think that may have been one of the experiments aimed at getting the Ntpcker analysis to run - allowing those tools to have clear access to the science database without fighting for resources with the daily inflow of new results. But we'll have to wait until either the News or Technical News archives are searchable before that ghostly memory can be confirmed or refuted. |
|
2985)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78269)
Posted 26 May 2017 by Richard Haselgrove Post: Have walked, will now go for a long drive. Make that two days at least. It's not moderators, it's not administrators, it's you. Look in the mirror. |
|
2986)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78259)
Posted 26 May 2017 by Richard Haselgrove Post: @Richard - Am I not allowed to respond to personal attacks?'Allowed' = free speech. Of course you are. 'Go on and on and on about it' = personal judgement. I'm out of here, for the rest of the day at least. |
|
2987)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78255)
Posted 26 May 2017 by Richard Haselgrove Post: Oh stop with the melodramatics please.And so say all of us. Please drop all the ad hominem sniping - all of you - and talk about grown-up politics instead. I've had a long post about the usefulness of the military / armed police response brewing all morning (on the whole I think it's a bad idea, and would focus the energies of public authorities in the wrong direction): that applies both on the micro scale (armed guards at the entrance to every public space in the western world) and the macro scale (laser guided bombs, cruise missiles, armed drone aircraft in the skies over the Middle East). I'd give my reasoning, but I don't think anyone's listening: so I'll save my breath and go out for a walk in the sunshine instead. |
|
2988)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78248)
Posted 26 May 2017 by Richard Haselgrove Post: Every time I look at a small-town or regional newspaper website - whether for my own local area in the north of England, or following a news link that somebody has posted on a message board - it seems to have the same panel of "promoted stories" towards the bottom of the home page, sometimes with "recommended" or even "recommended for you" added. It seems to be a global syndication service for clickbait. Sometimes (like this one) it's 'too good to be true' financial offers (= advertising), sometimes it's weird photo-stories with no obvious purpose (= fake news). I avoid both types unless I'm very, very, very, bored. |
|
2989)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78225)
Posted 25 May 2017 by Richard Haselgrove Post: Can you?No, I can't - and that's why I tend to be slow and careful in my posts to threads like this. |
|
2990)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78223)
Posted 25 May 2017 by Richard Haselgrove Post: I wasn't suggesting that Cressida Dick was gung-ho, but re-read Seven [police] mistakes that cost De Menezes his life - especially the contradictory reports of what happened in the final moments. I think it's common to both accounts that Jean Charles was under restraint ('pinned back in his seat') when he was shot from close range. |
|
2991)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78219)
Posted 25 May 2017 by Richard Haselgrove Post: I've just come back from the corner shop after getting the daily paper, and a couple of grocery items. There were 5 people in the shop including me. Some chap in the queue was spouting on about "We'll never beat 'em wot's the point of trying?". He was rounded on by two others and told in no uncertain terms what they thought of his comments. He slunk out red faced after paying for his goods.Just so long as 'beating' them involves evidence, arrest and trial to a professional standard: we could do without collateral damage from either gung-ho policing (Charles de Menezes?) or amateur vigilantism. Message boards, bloggers - much the same community, I fear. Katie Hopkins: I rest my case. |
|
2992)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78216)
Posted 25 May 2017 by Richard Haselgrove Post: You were talking about "people ... attracted to bulletin boards": that's the posting population at large, and I think your assertion is false. A mod team of seven is unrepresentative and irrelevant.I can't say I'd noticed that in, say, the Politics thread at SETI that spawned this spin-off.Are you being serious Richard? The majority of the Seti mod team are BHL's and their leader leans so far left, he hangs out of the side window when driving his car!!! |
|
2993)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78213)
Posted 25 May 2017 by Richard Haselgrove Post: Taking a helicopter view of the current situation I need to remind myself that the sort of people that are attracted to bulletin boards in the first place are far more likely to be left wing than anything else. I class myself as centre right, so it is not surprising that my views clash with the majority. As Mark says "Am I bovvered?".I can't say I'd noticed that in, say, the Politics thread at SETI that spawned this spin-off. But I would suggest that scientific / research bulletin boards in general, and SETI in particular, might be expected to attract people who take a satellite rather than a helicopter viewpoint: those who remember the impact of the original 'earthrise' photograph from moon orbit, and are more likely to take a global perspective where national and political boundaries are invisible. |
|
2994)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78186)
Posted 24 May 2017 by Richard Haselgrove Post: Oh, great. This is all we need. Democracy has retreated into a bunker, hidden away from the Demos. Palace of Westminster closed to non-passholders, all events cancelled after UK security threat raised to critical(source: Parliament's own website) |
|
2995)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 78184)
Posted 24 May 2017 by Richard Haselgrove Post: The question is, what do you mean by "win"? Recall the old fable about the bet between God and the Devil (I can't remember which religion's god/devil, and it doesn't matter). Oh, it was the Wind and the Sun: One fable describes a contest between The Wind and The Sun. They had spotted a traveller clad in a heavy overcoat, and they both claimed to be able to get him to remove his coat in a shorter time than the other took.It's looking as if this particular scumbag came from a Libyan family, but was born in Manchester while the family had exiled themselves from Gaddafi's regime. We (the West, including the UK) took the Wind's technique for removing Gaddafi's coat: look how well that turned out in the medium term - we don't know the long-term outcome yet, but it's not looking pretty. I think Bernie is right to say that we won't "win" against terrorism if we continue the tactics of the last 20 years: putting up yet more walls against 'others' (people not like 'us'), and lobbing the occasional bomb over them. Using the parts of the world where we don't happen to live as demonstration ranges for our armaments industry isn't the way to get the world's resistance movements to take their coats off. (Remember, we - the UK government - were actively preparing to use the same terrorist tactics against 'invaders' or 'occupiers' if the second world war had ended differently) I don't have any neat solution packages to offer, but they have to involve thoughtfulness, understanding, dialogue, patience... |
|
2996)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 78155)
Posted 23 May 2017 by Richard Haselgrove Post: I haven't tried any replies recently, but I've not noticed any significant slowdowns with accessing either the website or the data servers. Some glitches, sure, but they normally clear within a minute or two: nothing like the extended issues that people keep reporting here. I suspect that network routing issues are at least as significant as internal problems when it comes to accessing Berkeley. And it's probably going to get worse - did anyone notice FCC votes to overturn net neutrality rules a couple of days ago? it'll take a long time for that to have any visible effect, but it's coming. |
|
2997)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 78134)
Posted 22 May 2017 by Richard Haselgrove Post: Wouldn't the craft E.T arrived in be seen as a UFO by those living at the time?Or, more likely, magic. As in the famous Arthur C. Clarke quote, "Any sufficiently advanced technology is indistinguishable from magic." |
|
2998)
Message boards :
The Lounge :
You couldn't make it up (but some did)
(Message 78108)
Posted 21 May 2017 by Richard Haselgrove Post: To which, add two other factors. 1) Different varieties of potato crop at different times of the year - real food is seasonal. And some varieties - notably those sold as "new" potatoes - make lousy chips. 2) Real British chips were fried in beef dripping, but many fast-food joints have switched to vegetable oils. Opinions vary about the quality of the resulting product... |
|
2999)
Message boards :
Questions and problems :
e-mail from this project down?
(Message 78107)
Posted 21 May 2017 by Richard Haselgrove Post: It seems to be a common problem, slowly creeping across a number of BOINC-powered websites. I've mentioned it at SETI, there's an active thread at GPUGrid, and I think I've seen others too. It would be helpful if a sysadmin with Linux server (not just BOINC code) experience could check out the reports and server logs, to see if recent changes in either BOINC or Linux OS code have broken the message-passing linkage between BOINC, SendMail, and any other related components. |
|
3000)
Message boards :
The Lounge :
Popular Urban Myths & Conspiracy Theories Part 2
(Message 77740)
Posted 4 May 2017 by Richard Haselgrove Post: Conspiracy theory of my own: the Royal Family deliberately chose to announce this news today because every politician has been banished from the airwaves, and thus can't muscle in on their story. [There are local elections in much of the UK today, and by law radio and TV are forbidden from broadcasting 'campaign' material while polling stations are open - as they are in most democratic countries around the world. But since no politician can be allowed to see a microphone without immediately grasping it to campaign for, or against, something-or-other, that's suspended General Election broadcasting as well. Phew. We get a day off to enjoy the lovely Yorkshire sunshine.] |
|
3001)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77724)
Posted 2 May 2017 by Richard Haselgrove Post: I have an HP 950C that keeps doing that, it's likely the rubber rollers on the paper transport mechanism have worn smooth. I just push the paper stack a bit harder into it and it usually picks a sheet up in the end :-))I remember that we used to need to keep a matchbox near the office Gestetner - no matches, just the box. And not safety matches, either - proper sandpaper, to freshen up the rubber pad on the paper-pusher. There's evidently some technology they've still not got to work properly. |
|
3002)
Message boards :
Questions and problems :
How exactly exclusive_app works?
(Message 77707)
Posted 1 May 2017 by Richard Haselgrove Post: I'm on linux ...Sorry, my mind must still have been in the Mac thread next door. |
|
3003)
Message boards :
Questions and problems :
How exactly exclusive_app works?
(Message 77703)
Posted 30 Apr 2017 by Richard Haselgrove Post: Then maybe you could consider something like py2app - "Create standalone Mac OS X applications with Python", as it says on the tin. You would only need to compile the script(s) which interfere with BOINC - then the (unique) name of the resulting application would be your entry in exclusive_app. You could continue to run your other, non-intrusive, scripts in interpretive mode with python3 as now. |
|
3004)
Message boards :
Questions and problems :
How exactly exclusive_app works?
(Message 77700)
Posted 30 Apr 2017 by Richard Haselgrove Post: Have you tried simply python3 on its own, without the name of the script the interpreter is currently executing? |
|
3005)
Message boards :
Questions and problems :
100% CPU utilisation on Mac Book Pro
(Message 77698)
Posted 30 Apr 2017 by Richard Haselgrove Post: Far as I know, there's no WCG custom version of BOINC for Linux or OS-X, only for Windows. They simply host a copy on their download page of the latest 'recommended', that is, the code vetted version, becauuuuse their uptime is tad over Berkeley's.It wasn't the BOINC client behaviour that I was worried about, but the signal handler in the BOINC API code that's compiled into the science application: it's the API which gives the science application the ability to react to the various commands (suspend, resume, exit and so on) that the client sends it. Of course, as Jord points out, the commonest reason for not reacting to a signal is when no signal is sent because of a "run always" user preference... |
|
3006)
Message boards :
Questions and problems :
100% CPU utilisation on Mac Book Pro
(Message 77693)
Posted 30 Apr 2017 by Richard Haselgrove Post: The science applications at the projects will use the processor 100%, but are set to be at the lowest priority, meaning that anything else with a higher priority will take their CPU cycles fully away from the science applications.BOINC itself will continue to run continuously even at that 1% setting: but BOINC uses very little CPU time to perform its management and supervisory function. It will send instructions to the WCG application to run for 1 second, and then suspend for 99 seconds: if your CPU monitor continues to see 100% utilisation by the WCG apps during the 99 second 'idle' time, then it's possible that the Mac version of the WCG app is failing to respond to these 'command and control' messages. |
|
3007)
Message boards :
Questions and problems :
Win 10 bad pool caller on boinc suspend/exit
(Message 77685)
Posted 29 Apr 2017 by Richard Haselgrove Post: I've already called 'possible faulty driver' on 381.89 at GPUGrid (message 47141), but no confirmation or even corroboration yet. But in a previous message (47121) I identified that the user had updated drivers between 'last good task' (an ambiguous driver in the 376 range), and 'first task failure' (apparently under 381.89) |
|
3008)
Message boards :
Questions and problems :
Quit BOINC and tasks from command line
(Message 77674)
Posted 29 Apr 2017 by Richard Haselgrove Post: Only one 'e' in detach. If either is the default place for your BOINC as well, it becomes "C:\Program Files\BOINC\boinc.exe" --detach_console or "C:\Program Files (x86)\BOINC\boinc.exe" --detach_console, you add double quote marks around the path, because of the space(s) in the path. Leave the command to close the console outside the quotes.Why don't your colour <code> tags work for me? |
|
3009)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77447)
Posted 18 Apr 2017 by Richard Haselgrove Post: Now for some good news, Seti went down @ 8:17 PDT. Enjoy today's outrage.18/04/2017 15:51:36 | SETI@home | Project requested delay of 303 seconds 18/04/2017 15:56:39 | SETI@home | Sending scheduler request: To fetch work. 18/04/2017 15:56:42 | SETI@home | Project is temporarily shut down for maintenance Times are BST, so UTC+1 - and the real work of the project was down well before 08:00 PDT. Maybe they stopped for a coffee and a natter on their private boards before starting on the next bit? |
|
3010)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 77426)
Posted 18 Apr 2017 by Richard Haselgrove Post: And there I was, just thinking that the last faint shadow of David Cameron was being airbrushed out of history. |
|
3011)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 77419)
Posted 18 Apr 2017 by Richard Haselgrove Post: Theresa May to seek snap election for 8 JuneI'm perfectly happy to vote, as often as needed (and I don't even get a vote in the local council elections this year: voting is a rare privilege under this country's version of 'democracy') But I'll be leaving the country during the election campaign that precedes it... ;) |
|
3012)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 77374)
Posted 14 Apr 2017 by Richard Haselgrove Post: A war between North Korea and the USA isn't a world war. The original Korean war wasn't a world war either, so why would this be different?The assassination of an Archduke in Sarajevo wasn't a world war either, but it became one, because everyone seemed to be itching for a fight, and there was no commonly-accepted arbiter of "what's right for the world" at the time. I think those conditions are close to being replicated, now that the people in power no longer have memories of how horrible previous world wars were, with the relatively primitive weapons of the time. |
|
3013)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77336)
Posted 12 Apr 2017 by Richard Haselgrove Post: In the BOINC code it still is server_status.php: https://github.com/BOINC/boinc/blob/master/html/user/server_status.phpYes, and SETI does have a lot more servers to show the status of than your average BOINC project. (Main project, that is, not Beta - that doesn't need the extra bits.) But the users asked for the change, and David provided it. We were all there at the time: Message 1837602. |
|
3014)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77320)
Posted 11 Apr 2017 by Richard Haselgrove Post: Not rocket science - it's the one you get from the navigation bar. I think DNS is back - so everybody else is clogging up my nice fast link ;-) |
|
3015)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77310)
Posted 11 Apr 2017 by Richard Haselgrove Post: Just in case we need them later (if the servers come back up before the DNS), here are - I think - the current server addresses. 208.68.240.118 setiboincdata.ssl.berkeley.edu # upload server Oct 2016 # 208.68.240.126 setiboinc.ssl.berkeley.edu # scheduler Oct 2016 # 208.68.240.127 boinc2.ssl.berkeley.edu # download server Oct 2016Upload is the only one we can use during maintenance - and it's accepting files just fine. Leave the other two commented out until/unless you need them. |
|
3016)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77306)
Posted 11 Apr 2017 by Richard Haselgrove Post: Richard - did you notice the date stamp on the second entry on the list of outages:Yes, but they were all scheduled maintenance/outages - planned in advance. And yes, they do notify users in advance, for the problems they know about at least. But they don't seem to schedule work to start at 3 AM PDT, which is when our DNS went down (well, between 03:05 and 03:10 PDT, anyway). At the time I posted, the website was working normally, with everything shown as running and up-to-date on the SSP - but now it's all showing as routine maintenance. I presume they can let that run as normal using fixed IP addresses internally, while they fight with the DNS machine for the rest of us. |
|
3017)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77298)
Posted 11 Apr 2017 by Richard Haselgrove Post: No problem accessing the web site, but the aforesaid 'shedload of DNS errors' accessing the data servers - specifically upload and scheduler, but probably download as well. There's a lot of router updating work scheduled by IST this month (IST Service Status), but our particular 208.68.240.* address range doesn't seem to be listed. With no results coming back in, they should notice quickly enough... |
|
3018)
Message boards :
BOINC Manager :
NEED HELP
(Message 77180)
Posted 5 Apr 2017 by Richard Haselgrove Post: There is already an active discussion of this problem in the Questions & Problems area, under Notices tab goes to web page. It might be better to read the diagnosis so far in that thread, and contribute any further symptoms there too. |
|
3019)
Message boards :
BOINC Manager :
how to disable automatic use of VirtualBox?
(Message 77178)
Posted 5 Apr 2017 by Richard Haselgrove Post: BOINC file cc_config.xml, documented in Client configuration options. Use the setting <dont_use_vbox>0|1</dont_use_vbox>(Note that this is a feature of the BOINC client, rather than the Manager per se) |
|
3020)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 77168)
Posted 5 Apr 2017 by Richard Haselgrove Post: 1713 , That's about tea time here...You should just have time to nip up the road to Lincolnshire and stage yourself a re-enactment of the Boston Tea Party. Another example of our (in-)glorious colonial past. (Read the Tea_Act to see how duty-free trade via the Nederlands was treated as smuggling) |
|
3021)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 77164)
Posted 5 Apr 2017 by Richard Haselgrove Post: So, Chris wants us to cherry-pick just one outcome from the Treaties of Utrecht from 1713, does he? Doesn't fancy taking his holidays in Newfoundland or on the shores of Hudson Bay?  I love the description of 18th. century politics in the Responses section of the article I linked: Oxford and Bolingbroke persuaded the Queen to create twelve new "Tory peers" to ensure ratification of the treaty in the House of Lords.Then as now, British domestic politics cast a long shadow over the rest of the world. I suspect that the 'Free Trade' Tories were most interested in Gibraltar as a repair and re-provisioning port on the Triangular trade route - the British gained a monopoly over the slave trade by the same treaty. I presume you want to revive that, too? @ Chris, if you want to reply to that question, please do so here under your own account, for all to read. Boycotting the thread, then asking others to post on your behalf, seems to me to carry a touch of... What was that word you used about me in PM, again? |
|
3022)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77050)
Posted 30 Mar 2017 by Richard Haselgrove Post: It's back.Only the website. Still no science. |
|
3023)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77031)
Posted 30 Mar 2017 by Richard Haselgrove Post: Evidence? The science database is the source for Nebula data, which is under active development. 'Disabled' implies work in progress, not a failure. |
|
3024)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 77029)
Posted 30 Mar 2017 by Richard Haselgrove Post: Seti is down again .......This time, it is Informix... SETI@home science database paddym Disabled SETI@home science database: The informix database that contains final science products returned by SETI@home clients. |
|
3025)
Message boards :
Questions and problems :
Run Boinc without the screen graphic displayed.
(Message 77020)
Posted 29 Mar 2017 by Richard Haselgrove Post: Launch BOINC Manager? Again, although that runs, continuously but minimised, in the most common BOINC configuration, it isn't necessary, and it can be started and stopped independently of the 'Client' which manages the actual science. Or you can use a third-party application like BoincTasks, which can monitor multiple computers running BOINC from a central point. |
|
3026)
Message boards :
Questions and problems :
pogs: Notice from BOINC
(Message 77006)
Posted 29 Mar 2017 by Richard Haselgrove Post: Setting the 'Notice reminder interval' to "never" (Options | Other options...) will make the notices less obtrusive. But that's a very blunt instrument - you can't filter notices individually, and you might miss something more important. |
|
3027)
Message boards :
Questions and problems :
Run Boinc without the screen graphic displayed.
(Message 76987)
Posted 28 Mar 2017 by Richard Haselgrove Post: That's fine. Use your operating system tools to choose a screen saver to your liking. BOINC's screensaver is an entirely optional component, and it isn't essential to either install or run it for the scientific work to be done. In fact, the science will probably be done more quickly if you choose not to use the BOINC screensaver. |
|
3028)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76966)
Posted 28 Mar 2017 by Richard Haselgrove Post: Seti's outage (and the BOINC server for that matter) was related to something running internal, overwhelming the system: Nebula. Since Einstein is on the receiving end of that, that may be the coincidence.I think David (himself - not anybody else) should add this information to his own Nebula blog. I don't mind the oopsie - David is attempting stuff that hasn't been done before, and sh*t happens: lessons will be learned. In fact I'm glad to be told that our speculations about other possible causes have proved to be wide of the mark - this one is remediable. But if volunteer-based distributed computing is to survive, we need to have regard to the wise words of the BOINC Wiki: Recruiting and retaining volunteers |
|
3029)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 76948)
Posted 27 Mar 2017 by Richard Haselgrove Post: Can anyone confirm whether there was a protest march in London against the war in Yemen over the weekend? Only "news outlet" I've found so far that made mention of it was RT.There was a march planned for Saturday, but I can't confirm what the subject matter was - it's got very much swamped by the car attack, and may even have been called off because of road closures. No bombs this weekend, that I know of. I did like this image/video of a silent vigil on Westminster bridge on Sunday. Women link hands on Westminster Bridge to remember victims |
|
3030)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76941)
Posted 27 Mar 2017 by Richard Haselgrove Post: ... but there isn't an Einstein is down thread, lolThere's a News on Project Outages thread, where Cary has already posted. |
|
3031)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 76928)
Posted 27 Mar 2017 by Richard Haselgrove Post: I only see 1109 commits... My Tortoise is working (up to a point) - you have mail. |
|
3032)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76894)
Posted 26 Mar 2017 by Richard Haselgrove Post: I still have a bad feeling about that Informix database being stressed to its limits...Why do we keep speculating about Informix? It's a commercial, industrial-strength database system (now owned and maintained by IBM) that SETI uses for just one purpose: to hold the project's 'Crown Jewels' - the results (so far) of our searches for ET. It has nothing to do with our daily processing - our tasks, our gollum points, our messages (public and private). Those all live in a different database system entirely, and which - like the Informix database - seems to be running entirely as normal this weekend. What has gone down is the web server, running (IIRC) Apache. That must be irritating for the staff, as it is for us. But I really don't think that mis-identifying the problem helps us towards a cure. |
|
3033)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76821)
Posted 25 Mar 2017 by Richard Haselgrove Post: The only problem with leaving the boards down ...... is that users can't provide technical support to other users, or help them learn about the science behind the project. |
|
3034)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76819)
Posted 25 Mar 2017 by Richard Haselgrove Post: Eric has been away from Berkeley for the last week, on vacation with Angela. His only contact is via a mobile phone - I doubt he'd want to try and restart the system using that, even if it was technically possible.Mark, did you ever heard anything about what Eric found when he went to look at the system? If it was said I missed it somewhere ...No, sadly, I have not. SETI will be in the capable hands of Jeff this weekend, if anyone. |
|
3035)
Message boards :
GPUs :
How to exclude multiple GPUs for a given Project, using GPU_Exclude ?
(Message 76724)
Posted 23 Mar 2017 by Richard Haselgrove Post: Um, or maybe not. Multiple blocks, or multiple lines within each block? Try Jord's answer first, he wrote the FAQs... ;-) |
|
3036)
Message boards :
GPUs :
How to exclude multiple GPUs for a given Project, using GPU_Exclude ?
(Message 76723)
Posted 23 Mar 2017 by Richard Haselgrove Post: I think the instructions in the Client Manual are pretty clear. Under <exclude_gpu>, there is: You may include multiple <exclude_gpu> elements.- in other words, repeat everything between (and including) <exclude_gpu> ... </exclude_gpu> as many times as needed, changing an item or items within each block. |
|
3037)
Message boards :
The Lounge :
Popular Conspiracy theories
(Message 76647)
Posted 22 Mar 2017 by Richard Haselgrove Post: On this side of the pond all illegal activities engaged in by two or more are conspiracies.Much the same here. And the maximum sentence for a conspirator is usually much greater than the maximum sentence for an individual performing the same act on their own - a fact sometimes used by the prosecuting authorities in cases where they wish to 'send a message'. |
|
3038)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76639)
Posted 22 Mar 2017 by Richard Haselgrove Post: The server page seems to show work is there ... Everything is behind at 20 hours.So the status page is showing that there was work available 20 hours ago - no more, no less. Most people exchange old work for new work, when available. With 630,681 tasks available, and 103,489 tasks being returned each hour, simple arithmetic that all the available work would have been claimed within a little over six hours. |
|
3039)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76600)
Posted 22 Mar 2017 by Richard Haselgrove Post: Web pages may be down, but the server is accepting results. |
|
3040)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76559)
Posted 20 Mar 2017 by Richard Haselgrove Post: To my surprise, this year I heard some media commentary (for the first time I can remember) on something called the 'Equilux', when days and nights are equal. I'd always subscribed to the common (but, it turns out, false) belief that that happened on the Equinox. Equinox, Equilux, and Twilight Times seems to make a genuine, and credible, attempt to explain how, and why, the differences arise. |
|
3041)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 76546)
Posted 20 Mar 2017 by Richard Haselgrove Post: Finally!And according to the BBC, the Commission is far closer to being ready to commence negotiations than the UK is - they're expecting a formal reply, setting out the initial basis for discussions, within 48 hours. At least we'll be able to see what the words on the paper actually say - and find out whether the two positions are as far apart as I rather fear they might be. |
|
3042)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 76508)
Posted 17 Mar 2017 by Richard Haselgrove Post: BBC analysis is now available on their website, with more detail. George Osborne to become editor of London Evening Standard Edit - the nearest parallel I can think of is Silvio Berlusconi - but without the bunga bunga parties (so far). |
|
3043)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 76503)
Posted 17 Mar 2017 by Richard Haselgrove Post: The BBC is reporting that George Osborne has been appointed Editor of the London Evening Standard - subject to the mildest of regulatory approvals. George Osborne was Britain's 'Chancellor' (Finance Minister) from 2010 to 2016 - responsible for the policy of 'austerity' after the financial crash, and a prominent supporter of continued EU membership. He is still a serving Member of Parliament, and apparently intends to remain so. The London Evening Standard is a high-profile, even dominant, newspaper covering Britain's capital city. It's owned by the dissident Russian oligarch (and former KGB agent) Alexander Lebedev. It takes a robustly partisan editorial stance on issues such as the election of the powerful Mayor of London. To me, this feels wrong on so many levels. What do other countries think about serving politicians taking editorial control of major newspapers? |
|
3044)
Message boards :
GPUs :
Can not configure GPU's
(Message 76485)
Posted 16 Mar 2017 by Richard Haselgrove Post: No square brackets [ ] They are used in documentation to indicate that the item is optional, but can't be used in the actual file. |
|
3045)
Message boards :
GPUs :
Fedora 25 GPUs sometimes there but reported missing
(Message 76449)
Posted 14 Mar 2017 by Richard Haselgrove Post: But look at the second log, after the BOINC restart - BOINC detects the GPU, and even starts running an Einstein task. It's purely a timing thing - the drivers haven't initialised before the first attempt, so BOINC gets no response when it queries them. But later, everything is ready, and works as planned. Edit - but it might be worth a try to install the newer CUDA drivers and a later version of BOINC to iron out some of those remaining warning messages. |
|
3046)
Message boards :
The Lounge :
Popular Conspiracy theories
(Message 76429)
Posted 14 Mar 2017 by Richard Haselgrove Post: This one should start a new variant of the Roswell story. Crashed internet balloon mistaken for UFO |
|
3047)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76415)
Posted 14 Mar 2017 by Richard Haselgrove Post: United States Time Zone Map (with current time in each zone) |
|
3048)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76411)
Posted 14 Mar 2017 by Richard Haselgrove Post: Well it was 6 am here when I check, which means it was 4 am in Berkeley when it went down. So they really are starting very early these days.Are you sure? It's 05:50 in Berkeley now, and maintenance only started 40 minutes ago. |
|
3049)
Message boards :
GPUs :
Fedora 25 GPUs sometimes there but reported missing
(Message 76409)
Posted 14 Mar 2017 by Richard Haselgrove Post: The operating system is starting BOINC too early in the initialisation process, before all components are fully available for use. There have been posts over the years with guidance on changing the boot sequence, but I'll leave it to the Linux specialists to pick out one appropriate to Fedora 25. |
|
3050)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76406)
Posted 14 Mar 2017 by Richard Haselgrove Post: There's one fewer time zone between America and Europe this week (and next week), because of different policies about the starting dates of what we call "summer time". I do wonder, however, how much 'daylight' there is to save on the east coast of the United States today. ;) |
|
3051)
Message boards :
Questions and problems :
boinccmd --get_tasks gives read failure
(Message 76362)
Posted 11 Mar 2017 by Richard Haselgrove Post: Consider enabling debugging switches to cc_config.xml of yours. <task_debug/> and <slot_debug/> might be a good start. Also showing logs might be worth it.I'd suggest gui_rpc_debug too. Although he's not asking about the actual GUI (BOINC Manager), boinccmd uses the same rpc calls to query the client. |
|
3052)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 76318)
Posted 9 Mar 2017 by Richard Haselgrove Post: Although that is certainly true, I don't think it answers Chris's point. Both SETI and Einstein have similar server topologies: there are a pair of 'transactional' (BOINC) servers - primary and read-only replica - and a separate scientific repository for completed and validated scientific work. Only the transactional servers are visible in the sense that Chris and other volunteers from the general public are asking about - why do (or don't) they need regular maintenance? Matt Lebovsky answered that in SETI's technical news threads, many years ago - too long ago for me to be bothered to look up the exact reference. But in essence, he wrote that the actual backup of the BOINC server took only a small part of the maintenance interval: the majority of the time was spent compacting and re-indexing the transactional tables so that the database could function efficiently for the next operational cycle. I suspect that the major difference between the two projects is the transaction rate, and hence the degree of database fragmentation, together with the sheer size of some of SETI's more slowly changing tables. Einstein is sending out and receiving ~1.3 million workunits per week: SETI is handling the same number per day (typically over 110,000 results per hour, these days). Re-indexing SETI's tables will have to include the vastly larger user, host and message board archives: Einstein has taken steps to reduce their data overheads (e.g. combining multiple data segments into fewer, larger, workunits, and de-coupling the message boards from the BOINC database framework). But the biggest difference is in resourcing. Einstein (by my count) credit 14 contributors across their Wisconsin and Hanover sites, where SETI only credits six - and at least half of those are in supervisory, rather than operational, roles. If SETI had a similar staff-to-workload ratio as Einstein, I'm sure they could work out some optimisations similar to Einstein's, but for the time being they merely have the capacity to stop the wheels falling off the BOINC wagon - and although we can no longer ask Rom Walton for help with database optimisation, I suspect BOINC's data structure was never designed in the expectation of handling numbers on this scale. Remember how SETI had to make some emergency alterations a few years ago, when first task and then workunit IDs outgrew the 31-bit address space allowed? |
|
3053)
Message boards :
BOINC client :
Assigning a CPU only project to less than 1 whole core
(Message 76122)
Posted 24 Feb 2017 by Richard Haselgrove Post: project_max_concurrent = maximum amount of jobs per project.Note: A project may have more than one application. |
|
3054)
Message boards :
Questions and problems :
connecting to boinc remotely for rosetta@home
(Message 76041)
Posted 20 Feb 2017 by Richard Haselgrove Post: Most of what you asked is in Controlling BOINC remotely The only item missing is the default port number, which is 31416 |
|
3055)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 75934)
Posted 14 Feb 2017 by Richard Haselgrove Post: And we're back up, but still stuck at two splitters instead of seven (or fourteen). I've emailed the lab: could somebody else keep their toes to the toaster while I sleep, please? |
|
3056)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 75905)
Posted 14 Feb 2017 by Richard Haselgrove Post: 5:17 14/02/2017 13:09:51 | SETI@home | Project is temporarily shut down for maintenanceis my earliest record here. on the wet side of the USA.Makes a change for the left side to be the wet side. Maybe they want to be in and out of the lab early, so they can all go out for a swim? |
|
3057)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 75898)
Posted 14 Feb 2017 by Richard Haselgrove Post: SSP doesn't show Centurion as offline - just twiddling his thumbs because of lack of data.And Guppi splitters remain out of service ...Yep you are right Jim, Centurion is still off line :-( I had heard that it needed a new OS, but not sure what is up with it now. I think they're just checking with BL that they've got their +/- signs right this time, after the two short test batches. |
|
3058)
Message boards :
GPUs :
Different projects for 2 gpus
(Message 75812)
Posted 9 Feb 2017 by Richard Haselgrove Post: Certainly can. You can either let BOINC allocate tasks by itself, or you can use the <exclude_gpu> configuration option to restrict its freedom of action - I use that to restrict the more demanding projects to use the faster GPU only. |
|
3059)
Message boards :
The Lounge :
BOINC's Politics thread
(Message 75773)
Posted 7 Feb 2017 by Richard Haselgrove Post: For a good few years, it remained a small building society until greed took over.Yes, I got my business account with National Girobank (biggest branch network in the country - every village post office, including the little shop at the bottom of my street) purchased from under me, first by Alliance and Leicester, and then again by Santander. They're closed now - both my account, and the village shop. But the mass take-over of mutual building societies wasn't initiated by the banks. First, there was an active movement by so-called 'carpetbaggers', who moved from society to society, putting up 'demutualisation' resolutions at annual meetings and urging members to 'release their locked-in capital': most members were allotted shares and promptly sold them for a once-off windfall of around £1,000. It was only after these shares became tradeable on the open market (and were naturally scooped up by bigger competitors) that the rash of mergers started. There's an extremely close correlation between the list of former mutual building societies and the list of banks which failed in 2007/08/09 - starting with Northern Rock. I wonder why we don't hear anything from the carpetbaggers any longer? I was lucky enough to already have a personal account with the Nationwide, which has remained resolutely mutual throughout the crises, and continues to provide a current account service with full banking facilities. |
|
3060)
Message boards :
Questions and problems :
Curious Message
(Message 75695)
Posted 3 Feb 2017 by Richard Haselgrove Post: The message doesn't come from BOINC - it comes from the SETI@Home project. Breakthrough Listen |
|
3061)
Message boards :
Questions and problems :
** resolved ** GPU not being used / CPU High Priority Mode / Einstein
(Message 75646)
Posted 2 Feb 2017 by Richard Haselgrove Post: Do the people experiencing this problem use app_config.xml files, specifically files which include <max_concurrent> elements? I have identified a bug which has similar effects, and reported it as issue #1677 - after twice trying to get David to investigate the issue through private emails. I demonstrated the effect of the line 130 bug by adding additional debug logging to a private build and sending him the resulting output, but received no reply. |
|
3062)
Message boards :
Questions and problems :
strange occurance maybe some insight would be appreciated?
(Message 75412)
Posted 20 Jan 2017 by Richard Haselgrove Post: If the project's multi-core applications are written using the supported multi-threading tools, the number of cores they use can be controlled by using the --nthreads parameter on the command line, as shown in the extended format for app_config.xml in the application configuration documentation.You can also create app_config.xml for this app and specify there that app requires 0.5 CPU and 1.0 GPU. This is probably better than reducing CPU count to 75%, at least is used to be for older BOINC Client version.If using only app_config.xml multi-threaded tasks would still use 4 cores. <app_version>
<app_name>Application_Name</app_name>
[<plan_class>mt</plan_class>]
[<avg_ncpus>x</avg_ncpus>]
[<ngpus>x</ngpus>]
[<cmdline>--nthreads 7</cmdline>]
</app_version>
|
|
3063)
Message boards :
News :
New recommended client for Windows and Mac
(Message 75368)
Posted 18 Jan 2017 by Richard Haselgrove Post: Does it look like this one? Is the wording in the text area exactly the same? 
|
|
3064)
Message boards :
Questions and problems :
Running tasks on an APU
(Message 75309)
Posted 14 Jan 2017 by Richard Haselgrove Post: I would imagine that PrimeGrid is performing almost exclusively integer mathematics, allowing the CPU's floating point unit to draw close to zero power. You would very sophisticated monitoring software (or a thermal imaging camera) to find out which logical part(s) of your CPU package are actually running at that full 100%. |
|
3065)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 75172)
Posted 10 Jan 2017 by Richard Haselgrove Post: And GPUGrid's brand new server dropped offline (for the second time) on Sunday morning. There seems to be a lot of it about - I hope David is keeping an open mind for possible common causes and hence common solutions for the entire BOINC community. |
|
3066)
Message boards :
Questions and problems :
BOINC installing to non existent drive
(Message 74777)
Posted 14 Dec 2016 by Richard Haselgrove Post: Another pair of unresolved issues from the olden days are https://github.com/BOINC/boinc/issues/824 http://boinc.berkeley.edu/trac/ticket/847 I found that the Windows installer (but not BOINC itself) distinguishes between local and network drives. A normal installation will default to local disks: an administrative installation (setup /a) will default to network drives. I din't have any reason to check whether "normal installation to network drive" was forbidden, but it might be an explanation for the installer's refusal to use the NAS on Q: |
|
3067)
Message boards :
Questions and problems :
Breakthrough Listen
(Message 74761)
Posted 13 Dec 2016 by Richard Haselgrove Post: What does the subject Title phrase mean?It's one of a number of components of the Breakthrough Initiatives organisation. As their website says, The Breakthrough Initiatives were founded in 2015 by Yuri and Julia Milner to explore the Universe, seek scientific evidence of life beyond Earth, and encourage public debate from a planetary perspective. |
|
3068)
Message boards :
Questions and problems :
BOINC installing to non existent drive
(Message 74731)
Posted 10 Dec 2016 by Richard Haselgrove Post: This is one of the very many bugs which have been reported to BOINC developers over the years, but never attended to. By chance, I was looking up the references for a renewed attempt at getting developer attention a couple of days before you started this thread. Try reading through the workrounds in https://github.com/BOINC/boinc/issues/792 http://boinc.berkeley.edu/trac/ticket/815 |
|
3069)
Message boards :
Questions and problems :
remote connection failed unless I started using REMOTE_HOSTS.CFG
(Message 74671)
Posted 7 Dec 2016 by Richard Haselgrove Post: I cannot account for why this system requires that hosts file. I had something strange happen while testing the first install. I got an error -155 when attempting to execute bonccmd.exe "Authorization Denied" or something like that. That was in the program file\boinc directory so it should have worked. However, after reinstalling boinc a second time, that error went away. I originally executed the boinc install program using "run as administrator" but since I am the administrator that should have made no difference whether I selected it or not (one would think).It sounds as if you executed the boinccmd program from the BOINC program directory - as you should. But the authorisation key (password) is stored in the gui_rpc_auth.cfg file, which lives in the BOINC data directory. Since the two directories are different, there needs to be a way of finding one from the other: that's the DATADIR key in the Windows registry. It sounds as if your first installation failed to set that key properly, but the second did its job. |
|
3070)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74659)
Posted 7 Dec 2016 by Richard Haselgrove Post: Thinking some more, I wonder whether it may be an optimisation. PHP is an interpreted (and fairly simple-minded) programming language like BASIC, which is designed to run on servers and interface with databases - so it's often used on shopping sites. It might simply be too slow to look up all those tape progress bars and daemon status traffic-lights every time a user loads or refreshes the page: running the code once every 10 minutes, and redirecting the outputs to static HTML and XML files, may simply reduce the server load and result in shorter loading times. |
|
3071)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74651)
Posted 7 Dec 2016 by Richard Haselgrove Post: That's absolutely fine as a general statement of theory, but unfortunately there's a fly - a big fly - in SETI's specific ointment. The SETI-specific sah-status page starts life as sah_status.php. Yup, a PHP I wrote to David the day after the 'new look' changes started, and asked him to fix the XML output version while he was working on the file, and got a nice little note back: "I incorporated your change; works now. Thanks. -- David". And it did work, and has gone on working ever since. But: (and this is the strange thing) - the typo is still present, unfixed, in that SVN file. So, there must be another, operational, copy somewhere else, which I haven't found yet. That page must still run the php code, otherwise also available in XML wouldn't work. There must be a good reason why they run PHP to generate an HTML page, and only show us the HTML result: perhaps it simplifies keeping it going during maintenance. But it shouldn't be beyond the wit of a programmer to expose the raw page as PHP, so the navigation remains consistent across the site. |
|
3072)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74615)
Posted 6 Dec 2016 by Richard Haselgrove Post: Chris S wrote: (in the wrong thread)I think it was nearer 5am.. |
|
3073)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74613)
Posted 6 Dec 2016 by Richard Haselgrove Post: ... Which leaves Windows development, and here we're lacking knowledge, know-how, good documentation. If you feel it's up your street, contact David about it. He'll fill you in on all the exact requirements.Well said, Jord. I didn't see this until I had completed and sent my email to the boinc_dev mailing list: I think we're on the same page, but I was very deliberate in saying that these issues - including funding any additional personnel that may be necessary - should be considered and actioned by the Project Management Committee as a whole, not by David as an individual. And if anybody counters that they don't have that remit, then they should be re-constituted so that they do. |
|
3074)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74588)
Posted 5 Dec 2016 by Richard Haselgrove Post: I'd invite David to compare two pages. BOINC Forum advanced search SETI Forum advanced search Here, it's pretty clear what to do: fill in the boxes and click the button. Seti's page is less successful. 1) The left-hand column background has greater contrast with the page background. Final row looks odd. 2) The search button face colour is the same as the left column background colour. 3) There's another button with a much more inviting 'click me' face colour... Since all search parameters are cleared when you step back from the BOINC home page (been there, clicked that), you have to start filling in your 'advanced search' all over again. |
|
3075)
Message boards :
Questions and problems :
Notice Received. Need assistance to comply.
(Message 74585)
Posted 5 Dec 2016 by Richard Haselgrove Post: Actually, you don't "need" to do anything. This particular project has moved - well, it has more work still to do, so better wording might be 'is in the process of moving' - to servers provided for them by a third party. But they are doing it in a way which ensures that both the old and the new addresses can be used in parallel - certainly for the time being, and quite possibly permanently. Treat the current messages as 'informational', at most 'advisory', but certainly not 'obligatory'. If the wording changes, come back here, or go direct to the project website, and ask again. |
|
3076)
Message boards :
Questions and problems :
BOINC not uploading or downloading SETI@Home tasks
(Message 74545)
Posted 1 Dec 2016 by Richard Haselgrove Post: And so am I. I don't understand why it should have been necessary: normally I'd put that down to a bad peering gateway between two backbone carriers, somewhere between your local ISP and Berkeley. But that should affect all three computers equally.... I wouldn't worry about it. HTTP 1.0 should be plenty good enough for SETI. |
|
3077)
Message boards :
Questions and problems :
BOINC not uploading or downloading SETI@Home tasks
(Message 74539)
Posted 1 Dec 2016 by Richard Haselgrove Post: Thanks Jord.You'll find it's there already, in alphabetical order in the lower half. Just change the 0 to a 1 between the tags. (don't add a duplicate new row - the results can be undefined) |
|
3078)
Message boards :
Questions and problems :
BOINC not uploading or downloading SETI@Home tasks
(Message 74534)
Posted 1 Dec 2016 by Richard Haselgrove Post: Just for reference, I was employed as a computer/network technician for 15 years prior to my retirement in September.Ah - I'm self-taught, working as a consultant to small not-for-profit organisations - also retired now. But as a one-man-hand, I was my own final line of defence, and had to solve all the problems I threw at myself. I started just as networks were becoming relevant in small offices, so my first networking experience was the sort that requires a stepladder, an electric drill, and an 18" masonry bit... I've found the emails relating to the reset commands. It was given to me by Symantec technical support in 2006, in the format "netsh winsock reset catalog" (reading MSDN, I now think the word 'catalog' was redundant). But it worked then, well enough for me to re-install my client's copy of NAV 2006, which was the objective of the call-out. Is #3 a new machine, supplied with Windows 10, or an upgrade? Either way, I'd suggest a visual check of the hosts file - I've seen some malware put bad entries in there (though none specific to SETI), and some AV products put in blocking redirects to localhost. Windows 10 has the same file in the same place as always - C:\Windows\System32\drivers\etc\hosts - and an easier way of making hidden folders visible. The only other thing I can think of at the moment is to run ping and tracert tests on boinc2.ssl.berkeley.edu - but it all seems OK if you can establish the initial connection before the 'Empty reply from server' at the next step. I'll keep scratching my head, but there's nothing else there at the moment. |
|
3079)
Message boards :
Questions and problems :
BOINC not uploading or downloading SETI@Home tasks
(Message 74527)
Posted 1 Dec 2016 by Richard Haselgrove Post: OK, so let's say it's narrowed down to computer #3 and it's connection to the SOHO. First thing I'd try is plugging the cable into a different SOHO port - ports have been known to go bad. Unlikely, but it's probably the simplest possible test. (you didn't say whether you'd tried other websites from #3) And could you try an alternate ethernet cable for #3? After that, it's software. Have you ever played around with alternative DNS settings? Sometimes, in the bad old days, we discussed overriding DNS for a failed SETI download server by setting a manual IP entry in the local hosts file. Ring any bells? If you have, remove the manual IP - SETI's network is much more robust these days. If that suggestion doesn't apply, I have come across a Windows network reset command which got me out of a similar problem when an AV program went bad on me, several years ago - I'll need to go and look it up again, to make sure I get it right. Edit: here are a couple of versions. They both sound 'right' (viable alternatives) from memory - and I'll source them - but we should be careful. 1) Assuming that you're using standard Windows networking - no special add-ons provided by your employer, for example. 2) Run these commands from an 'elevated' - "Run as Administrator" command prompt. 3) Reboot the computer afterwards, to allow the standard settings to be re-created. netsh int ip reset (source: MSDN) netsh winsock reset (source: The Windows Club) I've checked, and both forms are still valid internal commands in Windows 10 Pro. |
|
3080)
Message boards :
Questions and problems :
BOINC not uploading or downloading SETI@Home tasks
(Message 74519)
Posted 30 Nov 2016 by Richard Haselgrove Post: ... can't request work: some download is stalled.That's been pretty hard-wired into BOINC since - well, pretty much for ever. There's no point in being allocated more work if you can't download the work you've already been allocated - you won't be able to download it. So, concentrate on the download problem by itself. Curing that will cure the other. Not a good time of night for me to read through all that log - I'll have a look in the morning, if nobody else has chimed in by then. |
|
3081)
Message boards :
Questions and problems :
BOINC fails to load after reboot
(Message 74493)
Posted 29 Nov 2016 by Richard Haselgrove Post: Well, after a timeout for a good meal, and another hour or more wrestling session, I think I have a working, updated, Windows 10 computer with a 1050Ti running NV's 375.95 and all showing/manageable downstairs. But - |
|
3082)
Message boards :
Questions and problems :
BOINC not uploading or downloading SETI@Home tasks
(Message 74484)
Posted 29 Nov 2016 by Richard Haselgrove Post: There are two SETI@Home download servers, and their IP addresses alternate every five minutes, using round-robin to balance the load. boinc2.ssl.berkeley.edu
----------------------------------------
Record Name . . . . . : boinc2.ssl.berkeley.edu
Record Type . . . . . : 1
Time To Live . . . . : 291
Data Length . . . . . : 4
Section . . . . . . . : Answer
A (Host) Record . . . : 208.68.240.119
Record Name . . . . . : boinc2.ssl.berkeley.edu
Record Type . . . . . : 1
Time To Live . . . . : 291
Data Length . . . . . : 4
Section . . . . . . . : Answer
A (Host) Record . . . : 208.68.240.127It may be worth trying again later, to see if you get .119 instead - but the one you did try should be valid.Do all your three machines use the same (if any) anti-virus program - they are often the cause of network problems, though more often with program (executable) files, rather than data files like these. |
|
3083)
Message boards :
Questions and problems :
BOINC fails to load after reboot
(Message 74482)
Posted 29 Nov 2016 by Richard Haselgrove Post: After two more hours, it's reached 12%. Good thing I didn't have anything urgent to do on it. |
|
3084)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74478)
Posted 29 Nov 2016 by Richard Haselgrove Post: BTW, that was an 'on the record' statement - the whole Bootstrap CSS adventure is being broadcast to all BOINC developers and projects on the mailing lists - including that exchange, and several requests for help. |
|
3085)
Message boards :
Questions and problems :
BOINC not uploading or downloading SETI@Home tasks
(Message 74477)
Posted 29 Nov 2016 by Richard Haselgrove Post: SETI's maintenance outage has been cancelled for this week, so all servers should be responding normally now. |
|
3086)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74474)
Posted 29 Nov 2016 by Richard Haselgrove Post: Message from Eric: We're cancelling this week's outage because of that.'That' being my report that "With SETI now down for maintenance, both the main page and the server status page show the barest minimum information - much less than before the changes." |
|
3087)
Message boards :
Questions and problems :
BOINC not uploading or downloading SETI@Home tasks
(Message 74469)
Posted 29 Nov 2016 by Richard Haselgrove Post: Try setting the 'http_debug' Event Log debug flag instead of (or as well as) 'network_status_debug' - that should provide more details on that "transient HTTP error". Note that SETI is currently down for maintenance, and will be for several more hours. That shouldn't affect the downloading of the files you're trying now, but it will prevent you being allocated any new work for the time being. But given all the changes at SETI recently, it's always possible that today's outage may include more servers than usual, but there's no direct evidence of that yet. (I've got nothing waiting to download, but uploads are going through normally) |
|
3088)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74466)
Posted 29 Nov 2016 by Richard Haselgrove Post: I guess they don't want us knowing about when things are coming back anymore, or be able to look at all the static web pages with information, like their work e-mail addys.I believe in cock-ups more often than conspiracies. I doubt anybody *wanted* that - they (or more likely 'he', I'm not getting the impression this is a collective effort) just didn't think or plan ahead. Ah, "I didn't think". The saddest three words in the English language. |
|
3089)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74463)
Posted 29 Nov 2016 by Richard Haselgrove Post: Well, interesting. With the now new outage even the SSP cannot be reached. (BTW, for those looking, it's sah_status.html, not server_status.php; while for other BOINC projects the latter is the server status page, it never was for Seti).The main 'front page' - which usually looked at least like a live project during weekly maintenance, even if missing all dynamic information, now looks like an empty scaffold - the same as the SSP. |
|
3090)
Message boards :
Questions and problems :
BOINC fails to load after reboot
(Message 74461)
Posted 29 Nov 2016 by Richard Haselgrove Post: Well, the "kicking" (remove all the devices that may have had dodgy drivers, like the GTX 1050Ti GPU and the wireless network card) didn't work - I still got  But it's in a room with a decent Cat5 feed, and a second computer, which makes looking up troubleshooting tips easier. Edit - and on Cat5, it's downloading the online version at the proper 40 Mbits, instead of the weasly 0.6 Mbits which is all it would do yesterday. Edit2 - the 'preparing to update' stage reached 100%, and prompted for reboot. We might have lift-off. (The hardware, minus extras, is certified compatible with 1607 by the manufacturer) 'Working on updates 2%. This will take a while. Your computer will restart several times. The coffee machine is downstairs.' And there's no space between the preview button and the edit button. |
|
3091)
Message boards :
Questions and problems :
BOINC fails to load after reboot
(Message 74460)
Posted 29 Nov 2016 by Richard Haselgrove Post: Currently I'm on 10.0.10586, but it wants to install 1607 the next time I restart.So does mine - and it failed several times yesterday, saying the current version (which is up-to-date, except from that one installation) isn't compatible with the patch. I've picked up a few hints from M$ and OEM websites, and I'm taking it upstairs to give it a good kicking next. |
|
3092)
Message boards :
Questions and problems :
BOINC fails to load after reboot
(Message 74457)
Posted 29 Nov 2016 by Richard Haselgrove Post: That looks plausible - I'll try it next time the testbed is in Windows 10 mode. But if my experience is commonplace - can't recall BOINC Manager to the screen after 'close to tray' with top-right 'X' - we may need to address it in development. |
|
3093)
Message boards :
Questions and problems :
BOINC fails to load after reboot
(Message 74452)
Posted 28 Nov 2016 by Richard Haselgrove Post: Further news - I managed to regain local control by opening a second manager from the command line with boincmgr /m That'll be enough while I wait for the Windows 10 version 1607 update to download - it's been running for about a couple of hours so far, at about 1.5% of my rated internet speed. |
|
3094)
Message boards :
Questions and problems :
BOINC fails to load after reboot
(Message 74451)
Posted 28 Nov 2016 by Richard Haselgrove Post: My turn to ask a question - for help - on this. I'm experimenting for the first time with Windows 10 - Professional, 64 bit, if that matters. Both the BOINC client and the BOINC Manager are running. I can see the tasks remotely using my preferred and long-established BoincView environment, and I can attach a full 'New BOINC Manager window' from a remote computer to the running client. But despite following my own advice, I can't invoke the local BOINC Manager window for the copy running locally on the Windows 10 machine. I have two icons visible to me: one in the 'Most used' program list in the Start menu, and one I've pinned myself to the Quick Launch area of the task bar. Clicking either of these has no effect, presumably because of the standard 'only run one copy at once' setting for BOINC Manager. The problem is, that there's no icon visible in the notification area that I could right-click or double-click on to bring the running copy to the foreground. Is this common with Windows 10? BOINC was installed via the development v7.6.33 Windows installer, and I have updated the binaries manually to home-builds from Master (last week - after the 'Apply' button was added to the Event Log options dialog). |
|
3095)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74441)
Posted 27 Nov 2016 by Richard Haselgrove Post: I agree. These posts are getting more and more erratic, and less and less relevant. Chris, please take a break and a chill pill.DA doesn't realise that designer stubble went out in the 70's.I am totally gobsmacked, what the hell does whether someone grows a beard or not has stubble or not have anything to do with this subject. |
|
3096)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74432)
Posted 27 Nov 2016 by Richard Haselgrove Post: Serendipitously - and unconsciously - I find that my first sentence uses the words "Bootstrap website designer". Maybe the gap between that and "website developer" is wider than appears at first glance, |
|
3097)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74427)
Posted 27 Nov 2016 by Richard Haselgrove Post: Without - honestly - meaning to disparage David Anderson in any way, I think it's clear from the moment-by-moment changes at SETI that he wasn't, before this sequence of attempted changes, an experienced Bootstrap website designer. Provided he sticks with it, he will be one by the end of the exercise - one more string to his CV, and probably several more grey hairs as well. I hope that, having started, he does persevere to the end, because I actually quite like the direction the site is going in, and I think it shows promise - but the public teething troubles are painful. Two learning points for developers anywhere. 1) When is it best to do the job yourself (learning on the job - even though you lack all the knowledge you might need when you start), and when is it best to hire in help who are already ahead of you in that particular sphere? (taken as a given that not every developer who puts themselves forward as an experienced practitioner is really as experienced as they claim to be - I suspect Einstein may have found themselves in that position) 2) When you develop something that's never been tried before (and I've been in that position myself), you can get away with almost anything: nobody has any prior expectations of you, and if it works at all, everyone is amazed and congratulates you. When the time comes for version 2, 3, 4, the expectations are already in place, and the world is much less forgiving. How much of the upgrade can you do in a private closed development, how much with a small-scale test environment, and how much in the full public view of your existing audience? How do you choose the right transition points between those stages? |
|
3098)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74417)
Posted 26 Nov 2016 by Richard Haselgrove Post: Found another one - when sending private messages, the preview label merges into the text without distinction, unlike when previewing a post here. 
|
|
3099)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74406)
Posted 26 Nov 2016 by Richard Haselgrove Post: By the way, those that read among the different forums, the quoted text isn't bold. It's just that it's 17.5 pixels big (it was 18.75px) that makes it bolder. However, if you add a [size /] tags to it resizing it back to e.g. 10, you'll see the text isn't bold. Has anybody expressed any justification for making quoted text 17.5 px? I'd prefer it to be the same size as default body text like this, if that's OK with everybody else. (and without that extra blank line afterwards, please) |
|
3100)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74397)
Posted 26 Nov 2016 by Richard Haselgrove Post: I agree about distraction and diversion of scarce development resources, especially at a time when the recommended client versions are dated 17 Dec 2015 (Windows and Mac OS X), 28 Feb 2014 (Linux - excluding distro-specific repo versions), and 3 Jul 2016 (Android - widely reported as broken, with the older development version suggested instead). The big benefit of BOINC's internal forum software is that it is hooked in to the back-end user, host, and task data - showing Gollum points for those that like such things, and allowing helpdesks to function. Onto specifics...It was a commonplace while BOINC was being developed, by the same people who had been somewhat overwhelmed by the public response to the original, stand-alone, SETI@Home Classic project. I've only been here since 2006 - I stuck with SETI Classic until closedown day, and didn't ever get involved with the forums there - but it was already taken for granted before I joined BOINC. May i suggestYou're right, I should get out more, and go and collect my black-ink-on-white-dead-trees newspaper. Wearing my prescription, varifocal, polaroid sunglasses, of course. edit - fixed list format. |
|
3101)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74396)
Posted 26 Nov 2016 by Richard Haselgrove Post: I was just starting to work as a freelance computer consultant for small not-for-profit organisations when Directive 90/270/EEC - display screen equipment came into force. Key features include Annex (1)(b) Display screenand Annex (3)(e) the principles of software ergonomics must be applied, in particular to human data processing.At the time, almost all display screens used CRT technology, and I was appreciated for curing a number of headaches by adjusting screen refresh (flicker) rates before I started work. The tools were there (or were, once Windows became commonplace), but workplace users didn't know that they were available or how to set them to best advantage - and the default values set by manufacturers were horrible. I found that 60Hz (on a CRT) was unbearable for even a few seconds, 72Hz was annoying, and I wasn't truly comfortable at anything below 85Hz. I don't know whether there's any US equivalent to the EC regulations (particularly the one about software ergonomics), but UC Berkeley's public status probably makes them applicable if they exist. |
|
3102)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74393)
Posted 26 Nov 2016 by Richard Haselgrove Post: Further to the eye-strain discussion: HAL9000 has pointed out that SETI task information pages switch contrast from white-on-black to black-on-white half way down - at the start of stderr_txt. As he says, "which does feel like someone flipping a light on in your face." |
|
3103)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74390)
Posted 26 Nov 2016 by Richard Haselgrove Post: Hoping to make this as a serious point... The idea behind BOINC was to encourage people to become involved in multiple branches of science - to attach to multiple projects, so that work would continue to flow if one project or other didn't have anything that needed doing from time to time. And many of the people who have the time to be involved in multiple project seem to be at or approaching retirement age. Being (actively) involved in multiple projects means visiting multiple websites. I find that my 64-year old eyes can read this bright white site quite comfortably, and they can read SETI's dark slate site quite comfortably. But they don't like switching between the two. Any optometrists amongst us care to comment whether that counts as exercise for the eyes (good), or strain (bad)? |
|
3104)
Message boards :
Projects :
New project appeared in my boinc manager itself !!!
(Message 74377)
Posted 25 Nov 2016 by Richard Haselgrove Post: You missed possibly the most important message: 25/11/2016 10:39:39 | Fight Neglected Diseases | [error] New code signing key doesn't validate They are moving the project back-end services to a new location under the auspices of GridRepublic, and there are clearly teething troubles: I suggest you keep an eye on recent posts in the project message-board thread at http://findah.ucd.ie/forum_thread.php?id=267 Edit - looks like they've sorted out the code signing key problem, and work is downloading, crunching, and being returned without problems. But I think they still have some less critical problems to sort out. |
|
3105)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74355)
Posted 24 Nov 2016 by Richard Haselgrove Post: I could tell you....BOINC Adam, judging by the path listed in the error message. |
|
3106)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74352)
Posted 24 Nov 2016 by Richard Haselgrove Post: SETI Beta is down! 🙀And back. Who's making syntax errors on Thanksgiving? |
|
3107)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74350)
Posted 24 Nov 2016 by Richard Haselgrove Post: Looks like work is continuing on the SETI Beta website makeover, with mixed results. Since I can't post there for reasons which will become apparent, I thought I'd record the error message here: Parse error: syntax error, unexpected ';' in /disks/carolyn/b/home/boincadm/projects/beta/html/project/project.inc on line 105 Edit - panic over, it's back. |
|
3108)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74333)
Posted 23 Nov 2016 by Richard Haselgrove Post: One positive comment: I do find that the 'green-on-dark-grey' does make links embedded in message board answers stand out much more clearly - like my Beta message 60165 - than the light-blue-on-white used here. I sometimes wonder whether people notice my smaller links at all. SETI Beta does have a particular problem with green-on-blue for user names because, by definition, everybody who posts there is a 'special person' - we're all volunteer testers. But there should be consistency whether links are underlined or not, and that sort order dropdown list is gross. |
|
3109)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74322)
Posted 23 Nov 2016 by Richard Haselgrove Post: I trust referring to a car engine is not disrespectful to Professor Porco?From her wikipedia biography, she's a NASA advisor on, among things, "the Solar System Road Map Development Team". I'm sure she'd love to drive your six cylinders on Mars... :-) |
|
3110)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74319)
Posted 23 Nov 2016 by Richard Haselgrove Post: Actually - and possibly this has only happened in the 10 minutes between you writing that and me starting this reply - she's back online and replicating as we speak. Replica seconds behind master 64,780 Please pardon my pedantry, but could we please speak of her using the spelling of her name used on the page we're both referring to? I speak as someone who inherited - with no choice in the matter - a surname variously written Ha {s,ss,z} {el, le} gr {o,a} ve. I'm most intrigued by the people who only know my name by reading it - i.e. I've never met them or spoken to them on the telephone - and have no idea how I pronounce it. But they clearly hear a sound in their head, because when they come to write my name back out - most commonly, on these or SETI's message boards - they use an alternative phonetic spelling from the one they first read. It's no big deal, but just feels mildly disrespectful - and I think we should show some respect to Professor Porco. |
|
3111)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74313)
Posted 23 Nov 2016 by Richard Haselgrove Post: Richard, is that a bit like only being fed chicken soup ...well, wouldn't you? ... while getting over a serious illness?More like while you're coming round after a big operation under anaesthetic, and before they take the stitches out. |
|
3112)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 74309)
Posted 23 Nov 2016 by Richard Haselgrove Post: Caroline is running but still offline again. That makes a lot of sense. My reading of the situation (but without any inside knowledge) is that they may have replaced Carolyn's [*] failing RAID card during the unscheduled outage on Monday, and gently re-introduced her into the server cluster for light duties only yesterday. After a few days burn-in to check that the new card (or whatever they did) is working properly and reliably, only then will it be time to expose her to the full hurly-burly of database replication. * The server was named after Carolyn Porco, planetary scientist. |
|
3113)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74150)
Posted 17 Nov 2016 by Richard Haselgrove Post: Max-width: 600px I know we're in a lounge here, but this is supposed to be a development and debugging board, and sometimes one needs to refer to a screenshot. Even though I know how to crop and scale before posting, nowadays my BOINC Manager windows are full HD (1920 x 1080) - 600 px (the minimum width) for the event log, and the remaining 1320 px for the main Manager window beside the log. So a full Manager screenshot (still without scaleable fonts) will be shrunk to less than half size. Read that if you can. |
|
3114)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74144)
Posted 17 Nov 2016 by Richard Haselgrove Post: Thanks goodness we have the scrollbars for over-size code blocks now, but why are they visible in that post, when the code isn't big enough to need them? Fixed size text (well, code) boxes in a re-sizeable window? That's just undone about about 20 years of GUI refinement. |
|
3115)
Message boards :
The Lounge :
The Einstein & BOINC forums make-over discussion
(Message 74139)
Posted 17 Nov 2016 by Richard Haselgrove Post: Thanks goodness we have the scrollbars for over-size code blocks now, but why are they visible in that post, when the code isn't big enough to need them? |
|
3116)
Message boards :
Questions and problems :
BOINC fails to load after reboot
(Message 74076)
Posted 14 Nov 2016 by Richard Haselgrove Post: Are you sure that it isn't running? That's not meant to be a trick question: it's just that BOINC is designed to start automatically (by default) every time the user logs on, but it can be completely invisible to the user. BOINC consists of two main components: boinc.exe (usually referred to as 'the client') boincmgr.exe (referred to as 'BOINC Manager') - check using Windows Task Manager that both of these are running. The client is completely invisible - it has no user interface at all. All interaction with the user is done via BOINC Manager. Assuming BOINC Manager is running, the way to bring it back to the state you saw it in immediately after installing it is to find the BOINC icon in the notification area (bottom right of screen - used to be called 'system tray' in older versions) and double-click it. Again, by default the icon is hidden when running under Windows 7 - click on the 'show hidden icons' triangle first to find it. |
|
3117)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 73759)
Posted 4 Nov 2016 by Richard Haselgrove Post: Edit: No, that doesn't work. Well, it would probably have worked if you had already submitted the public key so it had something to validate against. But full marks for trying. For the record, when I set up the git/tortoise repo clone on that machine in Feb 2016, I didn't load a putty key, and I still don't need one to update my local copy of the BOINC source code from github. I didn't need one to clone the dependencies from 'http://boinc.berkeley.edu/git/boinc_depends_win_vs2013.git' either: the authentication request is new with gitolite. |
|
3118)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 73756)
Posted 3 Nov 2016 by Richard Haselgrove Post: Yes, but... The context for that is Write Access to the remote Git repository - in other words, for code contributors only. Direct write access to the main repository is currently available via the SSH protocol with public-key authentication. If you want direct write access, contact David Anderson and/or Rom Walton. They'll ask you for your public key. Here's how you can create a public/private key pair and use it after you've been granted access to the repository: I was asking for read access only. It's perfectly fair, even under Open Source, to restrict graffiti scribblers and other vandals by requiring pre-authorisation. But that isn't what we're talking about here. |
|
3119)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 73741)
Posted 3 Nov 2016 by Richard Haselgrove Post: OK, fair enough - these things happen. Is it worth asking Christian, or putting out a general request on boinc_dev - or should we just wait until normal service is resumed? Hopefully my upload will allow mambo36 to get started while we wait. |
|
3120)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73739)
Posted 3 Nov 2016 by Richard Haselgrove Post: Deja Vu, all over again. 03/11/2016 14:25:01 | SETI@home | Computation for task blc3_2bit_guppi_57451_26010_HIP69732_OFF_0022.20906.831.17.26.25.vlar_1 finished 03/11/2016 14:25:03 | SETI@home | Started upload of blc3_2bit_guppi_57451_26010_HIP69732_OFF_0022.20906.831.17.26.25.vlar_1_0 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Info: Connected to setiboincdata.ssl.berkeley.edu (208.68.240.118) port 80 (#11264) 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Sent header to server: POST /sah_cgi/file_upload_handler HTTP/1.1 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Info: We are completely uploaded and fine 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: HTTP/1.1 403 Forbidden 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: Date: Thu, 03 Nov 2016 14:24:47 GMT 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: Server: Apache/2.2.15 (Scientific Linux) 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: Content-Length: 335 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: Connection: close 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: Content-Type: text/html; charset=iso-8859-1 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: <!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN"> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: <html><head> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: <title>403 Forbidden</title> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: </head><body> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: <h1>Forbidden</h1> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: <p>You don't have permission to access /sah_cgi/file_upload_handler 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: on this server.</p> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: <hr> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: <address>Apache/2.2.15 (Scientific Linux) Server at setiboincdata.ssl.berkeley.edu Port 80</address> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Received header from server: </body></html> 03/11/2016 14:25:04 | SETI@home | [http] [ID#12679] Info: Closing connection 11264 03/11/2016 14:25:05 | SETI@home | Backing off 00:02:54 on upload of blc3_2bit_guppi_57451_26010_HIP69732_OFF_0022.20906.831.17.26.25.vlar_1_0 Out of curiosity, why does "We are completely uploaded and fine" - a relatively new addition - appear at that point in the log, where it patently isn't true? |
|
3121)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 73736)
Posted 3 Nov 2016 by Richard Haselgrove Post: OK, I've emailed Charlie and David to ask about the password request, but I haven't heard back - it's a bad time of day for communicating with west coast USA. In the meantime, I've uploaded my saved copy of the VS2013 prerequisites to OneDrive: https://1drv.ms/f/s!AvWGSkp3v4ONh2YvHYK50QPORVfv That's the 'export' file from a git clone: it should expand to the correct subfolder structure as shown in Windows build dependencies |
|
3122)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 73734)
Posted 3 Nov 2016 by Richard Haselgrove Post: Yes, I saw that Charlie had made that change in the Wiki. I've managed to change the origin url in my clone, but now it wants:  and doesn't let me go any further. Rom was still making commits to the BOINC github repository as recently as last week: my understanding was that he was still employed 'in the family' by WCG, and was participating in the BOINC project as a volunteer like the rest of us, under WCG's wing. |
|
3123)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 73720)
Posted 3 Nov 2016 by Richard Haselgrove Post: I've found an email discussion about moving the dependency files - start from http://lists.ssl.berkeley.edu/pipermail/boinc_dev/2015-November/022113.html and follow the thread. But it rather peters out into generalised "wouldn't it be good if...?", and I haven't yet found any conclusive evidence that anybody decided to act on it, or not. Ironic, since the final few notes (continuing into December 2015) are all about keeping the documentation up to date. Jord, I think it would be helpful if you could prod David, Rom, Charlie and Christian when you get a moment - they seem to be the main protagonists. (edit - but see http://boinc.berkeley.edu/gitweb/. No sign of a move - may just be a server malfunction) |
|
3124)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 73719)
Posted 2 Nov 2016 by Richard Haselgrove Post: I did exactly what you're trying to do earlier in the year (mid-February), and got a successful git clone from the repository address. I've just tried to pull updates (if any) into my local copy, and got this response: git.exe pull --progress --no-rebase -v "origin" fatal: repository 'http://boinc.berkeley.edu/git/boinc_depends_win_vs2013.git/' not found git did not exit cleanly (exit code 1) (1420 ms @ 02/11/2016 23:02:11) So, I think we need to ask the development team if anything has deliberately changed, and if so, to document it. In the meantime, I do have a complete working copy, but at over 400MB (!), it's going to be a problem distributing it. D:\Sources_build>dir *.zip
Volume in drive D is New Volume
Volume Serial Number is 86EF-84C1
Directory of D:\Sources_build
15/02/2016 16:04 46,630,368 boinc_7-6-22.zip
15/02/2016 16:05 46,640,815 boinc_7-6-23.zip
15/02/2016 16:06 45,825,505 boinc_7-6-24.zip
15/02/2016 16:07 45,825,474 boinc_7-6-25.zip
03/03/2016 09:48 45,820,523 boinc_7-6-29.zip
05/10/2016 12:43 45,859,649 boinc_7-6-33.zip
14/02/2016 22:08 421,822,353 boinc_depends_win_vs2013.zip
31/08/2016 10:31 55,212,990 boinc_head.zip
15/02/2016 16:09 55,015,564 boinc_head_2016-02-15.zip
15/02/2016 19:48 5,841 Fixed solution files.zip
10 File(s) 808,659,082 bytes
0 Dir(s) 964,244,217,856 bytes freeI've just checked, and the source code pull from github (which I set up with the same tools on the same day) is working properly, so I don't think the problem is at my end. |
|
3125)
Message boards :
Questions and problems :
How to accommodate periodic internet connections
(Message 73710)
Posted 2 Nov 2016 by Richard Haselgrove Post: Deadlines are set by each project individually. The majority of projects allow at least a week, and several much longer. Look around for a project which matches your interests, and ask what their deadlines are. |
|
3126)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73709)
Posted 2 Nov 2016 by Richard Haselgrove Post: uploading is painfully slow. For uploads, you can click the 'Retry Now' button as often as you like - though beware RSI in your mouse finger, and boredom in other parts of your consiousness. There's no server limit. The only limit which the server imposes is when requesting new work - that 5:03 backoff is enforced by the server too. (You can ask - but you'll get nothing - and be told to wait another 5:03 for your pains.) |
|
3127)
Message boards :
Projects :
recently denied acces to SETI
(Message 73694)
Posted 2 Nov 2016 by Richard Haselgrove Post: Any ideas? I've posted the few ideas I do have in the lounge area. |
|
3128)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73690)
Posted 2 Nov 2016 by Richard Haselgrove Post: And likewise "You don't have permission to access /forum_index.php on this server." - whatever page you try to access is reported back in the error message. It started about 03:45 Berkeley time: this site is still up, and there's no sign of any planned outage on the IST page. So I'm guessing it's a breakage, not planned maintenance. I believe that SETI has more servers than we're familiar with on the SSP, and one of the back-office ones has responsibility for managing the access levels for different classes of user to the rest of the farm. Hopefully, it's just this 'access control' server which has gone down, which reduces the work that will be needed to get everything back up again (perhaps just a simple reboot). I can't think of any other scenario which would deny access to the web server, the upload handler, and the scheduler all at the same time in the early hours of the morning. Unless ET is getting p***ed off at having her privacy invaded? |
|
3129)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73674)
Posted 1 Nov 2016 by Richard Haselgrove Post: Project down for maintenance Einstein pre-announced a hardware upgrade for their web-server (only) - the actual science project is running as normal under the hood. |
|
3130)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73669)
Posted 1 Nov 2016 by Richard Haselgrove Post: SETI data servers were taken offline before 05:30 Berkeley time, and the forum pages before 06:00 - that's really early, even by Jeff's standards. Did he have a bad halloween, perhaps? |
|
3131)
Message boards :
Questions and problems :
Suggestion: Graphical Device Usage
(Message 73583)
Posted 26 Oct 2016 by Richard Haselgrove Post: ... in such case <exclude_gpu> should be applied to all projects. That's also available already, with <ignore_ati_dev>N</ignore_ati_dev> For details, refer to Client configuration. |
|
3132)
Message boards :
GPUs :
Windows, GPU and Remote Desktop
(Message 73471)
Posted 20 Oct 2016 by Richard Haselgrove Post: WDDM drivers (Windows Vista and onwards) run in the logged-on user's security space, and that is destroyed when a different user logs on - eg through remote desktop on desktop computers (computers running Windows Server allow local and remote users to log on at the same time, the last time I tried). You need to do one of two things: 1) Ensure that the normal logged-in user is logged back in after your remote session finishes. Depending on your security environment, can you use AutoAdminLogon to boot to your local account directly? If so, set it up, and issue a 'restart' instruction when you end the remote session. 2) Switch to a different remote access solution. Many varieties of VNC are available, including for free (I've used http://www.tightvnc.com/), and don't suffer from the same security restrictions as Microsoft's offering: ie, they don't destroy the GPU driver environment. |
|
3133)
Message boards :
Questions and problems :
What can I do with the following messages about "more disk space needed"?
(Message 73382)
Posted 18 Oct 2016 by Richard Haselgrove Post: Restore points maybe Whenever Windows Update installs updated components, it creates a 'restore point' first so that the system can be rolled back to a working state if things go wrong. Restore points are also created, I think, weekly and when other programs are installed. Since they keep track of the files needed by older versions of system files, they must use significant disk space - probably hidden, and I wouldn't even know where to start looking. |
|
3134)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73352)
Posted 16 Oct 2016 by Richard Haselgrove Post: Cabinets and phone lines are powered from the exchanges by 50V separate from the mains. Mine isn't - not the fibre cabinet, at any rate (the standard copper loom probably is)  That yellow electricity warning tape covers a T-splice into the local distribution cable. |
|
3135)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73349)
Posted 16 Oct 2016 by Richard Haselgrove Post: Landlines can be a lifesaver during an electricity outage. I'm not sure what the battery back-up is for mobile base stations (cellphone masts), but the traditional landline was powered over copper from the telephone exchange, and I think either you or Bernie told me - even showed pictures - of the lead-acid battery stacks in one of your exchanges. That raises a question over your statement about FFTC works. The Fibre cabinet is independently powered, but just from the local supply - I watched mine being installed, and I didn't see any UPS going in there. So broadband goes down during an outage - but your home router will do too, so that's no loss. But if voice calls are digitised in the cabinet and routed over optical - oops. My understanding was that in the standard British BT-style fibre ('Infinity') setup, the landline remained copper back to the exchange, but broadband was separated off at the cabinet and re-routed over the fibre. I also advised small businesses about telecoms during my working days, and a surprising number of them were badly trained by their installer about what to do during a power outage. Small electronic PABXs were becoming affordable, and each had their own style of electronic 'featurefone' powered from the PABX. Without power, the PABX had - under the regulations of the time - to provide failover sockets providing a direct power and signal service back to the exchange. But the feature phones wouldn't work over a direct line, and nobody stressed the need to keep a POTS handset handy where they could find it if the lights went out. |
|
3136)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73290)
Posted 12 Oct 2016 by Richard Haselgrove Post: PS I just looked for at my machines for the first time in a long while and see the 750Ti has had 4 errors in the last two days, coincidentally I just started to use the machine, it didn't have a screen keyboard or mouse till 2 days ago, oh look that is when the errors started. :-) It would be interesting if you could post a link to those errors in Number Crunching at SETI. It might be the "new process initialization error code when the user session is being logged off", mentioned here in the Release notes for Beta BOINC v7.6.33, or it might be something we don't know about yet. |
|
3137)
Message boards :
GPUs :
Guppy rescheduler
(Message 73238)
Posted 11 Oct 2016 by Richard Haselgrove Post: Guppi (with an 'i') is a SETI@Home project-specific class of workunit. Not relevant to BOINC-generic message boards - please discuss at the project instead. |
|
3138)
Message boards :
Questions and problems :
What can I do with the following messages about "more disk space needed"?
(Message 73215)
Posted 10 Oct 2016 by Richard Haselgrove Post: Check all four columns, and if any of them say zero, click the 'edit' (...) That should be fine. The next step would be to check whether one of your projects (the most likely would be Einstein) might actually have filled up the 15 GB space allowed. That can happen. |
|
3139)
Message boards :
Questions and problems :
What can I do with the following messages about "more disk space needed"?
(Message 73165)
Posted 9 Oct 2016 by Richard Haselgrove Post: I suggest you check and review your "BOINC global computing preferences" web account page: for SETI, that would be http://setiathome.berkeley.edu/prefs.php?subset=global&cols=1. Look at the fist line under Disk: Use no more than xxx GB Check all four columns, and if any of them say zero, click the 'edit' link at the bottom of the column and change it to something more reasonable. In these days of cheap, large, hard disks, the default used to be 100 GB. Are you by any chance also attached to the Einstein project? I'm in the middle of trying to troubleshoot a problem at that project, which seems to have zeroed the disk allowance at two of my venues, too. |
|
3140)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73066)
Posted 7 Oct 2016 by Richard Haselgrove Post: We could have debated whether 1989 No. 1796 PART III Regulation 25(2)(e) applied, of course, but I think the train passed before we reached that stage. |
|
3141)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73061)
Posted 7 Oct 2016 by Richard Haselgrove Post: I once pulled up at a railway level / railroad grade crossing, and switched off my front headlights while I waited, as is my custom - because I know that crossing can let you in for a long wait while up to four trains pass. The car in front was a police car, and the copper got out to ask - very politely - if I knew my lights had failed. I explained I was trying not to dazzle him via his rear-view mirror, and we had a little chat about being thoughtful to other road users, before we went our separate ways. |
|
3142)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73059)
Posted 7 Oct 2016 by Richard Haselgrove Post: I always try to remember to take my foot off the footbrake, and use the handbrake instead, when driving in traffic after dark. Most cars over here (including mine) have a third high-level brake light in the top centre of the rear window - and pulling up close behind one of those ruins your night vision for the next few minutes. I don't want to blind the guy behind me in case I need him to stop again at the next obstruction. Conversely, if I'm the last vehicle in an unexpected queue of traffic - say for roadworks - I'll keep my foot in the footbrake until the next vehicle has safely joined the queue, and then switch. |
|
3143)
Message boards :
Questions and problems :
Hide BOINC GUI tray icon
(Message 73050)
Posted 7 Oct 2016 by Richard Haselgrove Post: The easiest way to run BOINC is to start the client. BOINC its main parts are twofold, the client (background program that does most everything) and the graphical user interface called BOINC Manager. Yes. |
|
3144)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 73017)
Posted 6 Oct 2016 by Richard Haselgrove Post: "The engine has fired up but it isn't in gear yet." |
|
3145)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72996)
Posted 6 Oct 2016 by Richard Haselgrove Post: Seti says replica database on-line. I just wanted to see if the i7 chip is faster than the previous i5 chip. But I can wait. At 15:30 UTC (i.e. now), I see: BOINC replica database carolyn Running Replica seconds behind master Offline Please tell me where you see the word "on-line". |
|
3146)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72993)
Posted 6 Oct 2016 by Richard Haselgrove Post: Nope Yup. Replica is offline, but plenty of validators are online. Tasks are being validated - check your total credit (say once per hour) in either BOINC Manager or your account page on the SETI website. It'll be going up (unless you're firing blanks), but you can't see the individual tasks in the lists until the replica is back online and caught up. |
|
3147)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72988)
Posted 6 Oct 2016 by Richard Haselgrove Post: All my rigs are uploading work to Seti OK but neither Boinc or Seti are updating their results since yesterday afternoon UK time on the 5th. Replica offline? |
|
3148)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72749)
Posted 28 Sep 2016 by Richard Haselgrove Post: 11:07 PDT and the board is still down. very curious. SSP shows all services disabled, and Beta has been taken offline too - just like for a normal maintenance. That feels like the first step towards an attempted restart. |
|
3149)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72714)
Posted 28 Sep 2016 by Richard Haselgrove Post: Interesting. The meeting notes are linked from the CW article. Yes, Berkeley is a member of the group, and Gabriel Gonzalez is marked as present. |
|
3150)
Message boards :
Questions and problems :
SETI at Home message board.
(Message 72547)
Posted 19 Sep 2016 by Richard Haselgrove Post: It's listed under 'COMMUNITY' in the dark-blue band across the top of the SETI at Home front page. |
|
3151)
Message boards :
Questions and problems :
Which BOINC questions do you have that you can not get answered?
(Message 72508)
Posted 17 Sep 2016 by Richard Haselgrove Post: It would probably be better to refer to that as https://github.com/BOINC/boinc/blob/master/lib/error_numbers.h - that's the current repository which developers will download from for new projects/applications, and where any future changes will be made. The GIT file was lasted updated five months ago - 3 May 2016 - and your copy is identical. Afterthought - I'm sure that Jord will be explaining that 99% of the time 'Output file absent' simply means that the application crashed before it could create the file - it's a consequence of a previous problem, not a problem in itself, although that's how users present it at help desks. Very, very, rarely it turns out that a project administrator has told BOINC to expect an output file in every case, but the application concerned doesn't create it, or only creates it under certain circumstances. |
|
3152)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72467)
Posted 14 Sep 2016 by Richard Haselgrove Post: I was poking around the Einstein site this morning to look for something (for practically the first time since the 'upgrade' - I don't like it). No sign of an SSL error - but I don't use Firefox. |
|
3153)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72463)
Posted 13 Sep 2016 by Richard Haselgrove Post: It's OK, you can call off the boats. Drove out grateful for the sheet lightning to show me the road ahead, drove back after a good meal by moonlight through a calm, dry, night. I hope the neighbouring valleys fared as well. |
|
3154)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72460)
Posted 13 Sep 2016 by Richard Haselgrove Post: Well, the promised rain, thunder and lightning has arrived - but I'm going out anyway for a better look. If I don't post anywhere tomorrow, please send search parties - preferably with a boat. On second thoughts, check my computers first. If they haven't reported either, I'll simply have been knocked offline. |
|
3155)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72447)
Posted 13 Sep 2016 by Richard Haselgrove Post: Great weather! Let's lock the Cafe and open the patio, to try out the solar BBQ. OK, but best make it a quick one. Locally: Observed at 14:00, Tuesday 13 September Sunny 25°C WIND SPEED 3 MPH EAST NORTH EASTERLY Humidity: 69% Visibility: Very Good Pressure: 1010mb, Steady Observation station: Bingley (Lat: 53.817 | Lon: -1.867) But in the next five hours... Outbreaks of showery rain may turn torrential and thundery across parts of northern England and southeast Scotland. Whilst most places will miss the worst, there is also a risk of large hail, frequent lightning and strong winds. |
|
3156)
Message boards :
Questions and problems :
Which BOINC questions do you have that you can not get answered?
(Message 72400)
Posted 11 Sep 2016 by Richard Haselgrove Post: If it were your FAQs, I suppose you'd do away with the earlier material No, no - I fully agree with giving information about errors that may be encountered, and the form of words that may be used, when running older clients. But I thought you might also be giving advice about what errors might be cured (or, for that matter, caused) by running a later client - when, in this case, it eventually emerges from Alpha testing. I did have a longer list of application errors, but I sent it to Christian by PM at Einstein, before they upgraded their website to the Drupal version that maintains (or so I believe - not used it yet) an outbox. I looked for a local copy today, but couldn't find it - I may have to ask Christian to send it back to me. |
|
3157)
Message boards :
Questions and problems :
Which BOINC questions do you have that you can not get answered?
(Message 72384)
Posted 11 Sep 2016 by Richard Haselgrove Post: please don't post Windows, Linux or OS X errors. I am not looking for those. Not quite sure how those statements go together - I'm assuming that you are aware that the list I've picked out are all Windows application runtime errors, defined in ntstatus.h, found in most Microsoft Visual Studio products? We had a specific recent problem with one of those - exit code -1073741205 (0xc000026b) - earlier this year, resulting in the SETI thread 1073741205 Error Code (Unknown Error). Note that even the thread title is misleading, because the real exit code 0xc000026b is the unsigned long 3221226091, and it was BOINC that was interpreting numbers over two billion as negative. At the time, BOINC was actually displaying the codes as 64-bit negative hex strings (0xffffffffc000026b), which completely defeated google lookups. Christian Beer found that David's attempt at a fix didn't work for Einstein, because it mis-formatted and mis-identified some of their application-specific codes. After working together for a while, we ended up with 572c0e46f0e83564de410dec9d5eb1a0fc321e22 - I provided Christian with a list of the codes I'd found discussed on BOINC Message Boards, and those are the ones he picked out as being of general relevance. 0xc0000094 and 0xc000026b will be reported meaningfully on the websites of projects which have updated their BOINC website sources recently: 0xc0000024 is STATUS_OBJECT_TYPE_MISMATCH, and 0xc0000374 is STATUS_HEAP_CORRUPTION. The one which started us off, 0xc000026b, should be less of a problem in the future: BOINC client v7.6.33 and later contains a bugfix to treat this as a recoverable temporary exit, rather than a hard error. |
|
3158)
Message boards :
Questions and problems :
BAM account manager / Project adding.
(Message 72337)
Posted 8 Sep 2016 by Richard Haselgrove Post: Hmm, noticed that was wrong in the wiki, changed! :) It may have been a very old Wiki page - that wording change was made in 878ec2935307d1ddf31deda784baa51e1fd4f423 |
|
3159)
Message boards :
Questions and problems :
BAM account manager / Project adding.
(Message 72335)
Posted 8 Sep 2016 by Richard Haselgrove Post: You may have forgotten, but BAM is not a project - it's an Account Manager. The menu wording (reading from Windows) is "Use account manager..." - not 'Add'. |
|
3160)
Message boards :
BOINC Manager :
Starting and disconnecting BOINC Manager
(Message 72274)
Posted 5 Sep 2016 by Richard Haselgrove Post: Or from a script: boinccmd --quit |
|
3161)
Message boards :
BOINC Manager :
Starting and disconnecting BOINC Manager
(Message 72256)
Posted 5 Sep 2016 by Richard Haselgrove Post: Thinking about it, why would you want to --detach the Manager? It's a user interface, nothing more. If you want to manage the client, start it: if you just want to resume computation, start the client directly and leave it at that. |
|
3162)
Message boards :
BOINC Manager :
Starting and disconnecting BOINC Manager
(Message 72255)
Posted 5 Sep 2016 by Richard Haselgrove Post: 1) Argh! I wrote the second paragraph (1a) before I'd double-checked that I was answering the right question - and I wasn't! But I'll leave it anyway, in case this thread is useful to anyone else. boincmgr has C:\BOINC>boincmgr /h Usage: boincmgr [/h] [--verbose] [/a] [/s] [/n <str>] [/g <str>] [/p <str>] [/b <str>] [/i] [/c] [/m] /h, --help show this help message --verbose generate verbose log messages /a, --autostart BOINC Manager was started by the operating syste m automatically /s, --systray Startup BOINC so only the system tray icon is vi sible /n, --namehost=<str> Host name or IP address /g, --gui_rpc_port=<str> GUI RPC port number /p, --password=<str> Password /b, --boincargs=<str> Startup BOINC with these optional arguments /i, --insecure disable BOINC security users and permissions /c, --checkskins set skin debugging mode to enable skin manager e rror messages /m, --multiple multiple instances of BOINC Manager allowed and uses boincmgr /a /s for its own autorun at startup. I think you can use start boincmgr / / /whatever in your cmd file to avoid the hang. 1a) What I'm doing at the moment with the SETI@Home optimised application installer is either boinc.exe --detach_phase_two or boincsvcctrl.exe --start - depending, obviously, whether BOINC was running in user mode or service mode when my application had to discover the right way of shutting it down. The ' --detach_phase_two' feels weird, but it was in some early documentation, and it seemed to work (and was the only choice which did work with older clients). They tried to tidy the code up a few years ago, and broke my installer, so I persuaded them to put it back so I didn't have to use different commands for each generation of client. The current code is https://github.com/BOINC/boinc/blob/8604e8e0bdda499694e1040d5f0b86fe5f5fe4e1/client/cs_cmdline.cpp#L144 } else if (ARG(detach_phase_two) || ARG(detach) || ARG(detach_console)) {so those should all be synonyms and work the same way. 2) I do exactly that - close the client but leave the Manager running - in my installer. It works, but I do leave it to the very last possible moment, after all the stages when the user might pause to decide on their next input. I suspect that the Manager might try to restart the Client automatically if it decides it has gone missing for too long - perhaps 30 seconds - so I try to avoid that. Haven't tested it recently though. |
|
3163)
Message boards :
GPUs :
Another "No usable GPUs found" with an AMD HD 7750 on W7
(Message 72237)
Posted 5 Sep 2016 by Richard Haselgrove Post: I don't believe BOINC is installed as a service because I use BoincTasks to start/stop. It's not installed as a service, but that's nothing to do with BoincTasks. You can tell from 04-Sep-2016 15:59:07 [---] Running under account CCEu I see also 04-Sep-2016 15:59:07 [---] Config: don't use coprocessors I'd strip that out of cc_config.xml before going any further. I think it must be <no_gpus>0|1</no_gpus> |
|
3164)
Message boards :
Questions and problems :
Boinc 7.6.31 in Ubuntu 16.04 not getting new work for seti@home
(Message 72143)
Posted 1 Sep 2016 by Richard Haselgrove Post: The event log should tell you what happened to the work - perhaps it was processed and reported when complete (that's what's supposed to happen). The user search turns up nobody with the user name 'Martin Glazer' at SETI, so I assume you registered with a different handle. If you posted one of the ID numbers relating to your user account or your computer account, we could possibly advise better without fishing around in the dark. Jord ('Ageless') has posted some of the ways you can help us to help you in When requesting help on these forums. |
|
3165)
Message boards :
Questions and problems :
Boinc 7.6.31 in Ubuntu 16.04 not getting new work for seti@home
(Message 72136)
Posted 1 Sep 2016 by Richard Haselgrove Post: First question: is your computer asking for new work? Check the Event Log (Ctrl+Shift+E to view it). |
|
3166)
Message boards :
GPUs :
Difficulty with seti@home GPU usage on ubuntu 16.04 LTS under BOINC 7.6.31 (x64)
(Message 72129)
Posted 1 Sep 2016 by Richard Haselgrove Post: Or just click on the link I made in message 72073 I'm afraid it's one of the parts of the BOINC web code which is under-documented, but it can be very useful. Here are some of the current details for your computer, in a less fancy format. SETI@home v8 8.00 x86_64-pc-linux-gnu Number of tasks completed 333 Max tasks per day 37 Number of tasks today 26 Consecutive valid tasks 6 Average processing rate 15.43 GFLOPS Average turnaround time 3.51 days SETI@home v8 8.01 x86_64-pc-linux-gnu (cuda60) Number of tasks completed 0 Max tasks per day 1 Number of tasks today 8 Consecutive valid tasks 0 Average turnaround time 0.00 days Let's try to pick apart some of that. I've picked out the two application types where new tasks have been allocated to your computer today - 26 for processing on your CPU at the time I took that copy, 8 tasks for GPU. The most significant line for assessing the health of your machine is the number 'completed'. These are tasks which ran to a normal end-of-run, and produced validated results, for which you will have received credit. Having zero completed cuda60 tasks is a big alarm signal. Max tasks per day: BOINC rations the distribution of tasks to hosts which exhibit processing errors. 37 per day is a relatively low limit, suggesting there have been problems recently with your CPU application too, but a limit of one per day for your cuda60 application is another big alarm signal. [Quite why BOINC allocated you 8 cuda60 tasks today, while saying your limit is 1, is another of the mysteries of BOINC we haven't been able to fathom yet.] Average processing rate: when a task both validates and runs to the full normal exit point (some SETI tasks exit early, by design), the running time is used to assess the effective speed of the application on your device. This figure is the average over many tasks, and helps BOINC to allocate the correct amount of work to keep your machine busy for the length of time you specify in your preferences. The APR for your CPU - 15.43 GFLOPS - is considerably faster than the "Measured floating point speed 3140.11 million ops/sec" of your computer - a measure of how efficient the SETI applications have become in recent years. But because every one of your cuda60 tasks so far has ended in an error, BOINC hasn't been able to begin estimating a processing speed yet - the final alarm signal. By way of comparison, here's a similar display for one of my machines - Application details for host 3755243 SETI@home v8 (anonymous platform, CPU) Number of tasks completed 837 Max tasks per day 935 Number of tasks today 0 Consecutive valid tasks 903 Average processing rate 36.62 GFLOPS Average turnaround time 1.71 days SETI@home v8 (anonymous platform, NVIDIA GPU) Number of tasks completed 30972 Max tasks per day 22413 Number of tasks today 10 Consecutive valid tasks 22380 Average processing rate 160.57 GFLOPS Average turnaround time 0.43 days (I'm winding down at the moment after a friendly challenge between teams, and starting up other projects again, which accounts for the low number of tasks today. But I think you can see why your figures strike me as being low, and worth investigating further.) |
|
3167)
Message boards :
GPUs :
Difficulty with seti@home GPU usage on ubuntu 16.04 LTS under BOINC 7.6.31 (x64)
(Message 72073)
Posted 31 Aug 2016 by Richard Haselgrove Post: A further indicator can be found in Application details for host 7321608. Neither of these applications: AstroPulse v7 7.08 x86_64-pc-linux-gnu (opencl_nvidia_cc1) SETI@home v8 8.01 x86_64-pc-linux-gnu (cuda60) shows any sign of a validated or completed task during the lifetime of the machine: the line "Average processing rate" is missing from both sections. |
|
3168)
Message boards :
GPUs :
Difficulty with seti@home GPU usage on ubuntu 16.04 LTS under BOINC 7.6.31 (x64)
(Message 72071)
Posted 31 Aug 2016 by Richard Haselgrove Post: Open BOINC Manager in Advanced view, and look down the 'Application' column. GPU applications will have a plan_class name in brackets after the version number. Some of these will start with the word 'cuda', others with the word 'opencl'. Can you correlate whether the acceptable computer behaviour was observed (previously or currently) when one type of application was running the active task, and the interfering behaviour with the other type of application active? If so, which way round are they? If that doesn't provide a clue, is there a correlation with what you see in the 'name' column for the active GPU task? Some task names start with the the identifier ap_, some with a date in the format ddmmyy, and some with the identifier blc and a number. Again, can you associate any task type with the behaviour you're seeing - which are good, and which are unacceptable? |
|
3169)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72063)
Posted 31 Aug 2016 by Richard Haselgrove Post: Most people don't complain about the lack of uploading or downloading work during an outrage. The main bone of contention is about the lack of the forums while data-driven web pages on muarae1 are off air. If they kept that up, I doubt many people would give a monkeys what they do with the other servers for 1/2 a day. Hmmm. "It takes one to know one" ring any bells? |
|
3170)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72032)
Posted 31 Aug 2016 by Richard Haselgrove Post: Nah - they don't tell us anything, they'll have a script for that: "Every %Today% morning we begin a four hour data distribution outage for database and systems maintenance." Seriously - they are so short-staffed that they will have automated everything it's possible to automate. |
|
3171)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 72018)
Posted 31 Aug 2016 by Richard Haselgrove Post: There were traces left behind - splitters working on tapes in a different sequence - suggesting that the project was restarted, implying it was stopped first. No idea how long for - I missed it too. The automated notice (midnight Berkeley time) has just gone up announcing a full maintenance for today, Wednesday 31 August. Next week is Labor day, so I wouldn't be surprised by a Wednesday outage next week too. After that - watch the front page. |
|
3172)
Message boards :
Questions and problems :
BOINC caused small catastrophe
(Message 71879)
Posted 25 Aug 2016 by Richard Haselgrove Post: It would be helpful to post segments of the event logs which cover significant time events during that process - the switches between GPU use suspended and permitted, and the time when the backup program started to run. That might help to narrow down which event went wrong, and either caused GPU processing to restart, or didn't stop it when it should have. Since you've probably rebooted the machine since then, you would have to recover the logs from the file 'stdoutdae.txt' in your BOINC data directory. I have to say the the combination of * SCSI tape backup server * Windows 10 * BOINC GPU computation sounds ambitious to me, and is possibly pushing the boundaries a bit hard. |
|
3173)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 71781)
Posted 22 Aug 2016 by Richard Haselgrove Post: So Internet2 is the modern Arpanet? Or an international JANET |
|
3174)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 71759)
Posted 21 Aug 2016 by Richard Haselgrove Post: Three hours later Einstein remains in la la land. You did see the explanation posted here by a project staff member - early on a Sunday morning, his time? Einstein@Home is currently having network trouble. The main Website and forums are down but upload/download and task reporting is still working. |
|
3175)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 71720)
Posted 20 Aug 2016 by Richard Haselgrove Post: Not just the media. I'm in the middle of an interesting series of conversations with, and about, a UK Public Limited Company - which seems to have adopted the doorstep sales tactics beloved by fraudsters, rather than (as more commonly) the other way round. Nothing to do with science, and nothing to do with computing (distributed or otherwise), so I won't bore you with it now - but if the Guardian makes a story out of the letter of complaint to the company that they (and my MP) will be receiving a copy of, I may get a small glow of satisfaction out of sharing it here. |
|
3176)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 71701)
Posted 19 Aug 2016 by Richard Haselgrove Post: Compromise - the success of the many British teams in these Olympics has been a spectacular vindication of the benefits of public funding - in this case, funds contributed directly by the members of the public you see queuing in front of lottery terminals every Saturday night, not mediated by taxation and Government distribution. Camelot should organise and pay for the homecoming parade, and take a justified share of the credit. (and thus keep the decision well out of the hands of party politicians, of all persuasions. No bandwagon jumpers wanted here.) |
|
3177)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 71681)
Posted 18 Aug 2016 by Richard Haselgrove Post: ... be quite economical in fuel use ... That reminds me - I've got a long trip to plan for next month, so I'd better fill up sometime. I've done round trips to both coasts (Heysham and Whitby), and sundry other local runs, since I last visited a gas station in mid-July. Google maps makes the cross-country work alone about 350 miles, and I'm not even on reserve yet. I can't remember whether your remarks were specifically made in my earlier (orange) or current (silver) model, which to my taste is slightly underpowered in comparison. The orange one I ran for 10 years from new, and I remember deliberately specifying an optional-extra diesel particulate filter when I ordered it: I think they might be standard on the newer model. So I think that diesel engines, especially when not stuck in city traffic, have moved on a long way from their smoky agricultural stereotype. |
|
3178)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 71676)
Posted 18 Aug 2016 by Richard Haselgrove Post: They might have low end pulling power for heavy loads such as towing caravans up hills but generally they are not happy at high speeds. And in recent years the fuel companies cashed in and now Diesel is more expensive than petrol. It is only in recent years that modern diesel fuels don't emulsify in winter causing all sorts of problems. I don't remember you saying that when you were a passenger in my diesel car, and commented on the low revs at high speed (close to the legal maximum in this country), and very slow rate of fuel consumption. The cubic capacity of the engine is really irrelevant - slightly greater use of raw materials during construction, and possibly slightly greater congestion in car parks, but I can't think of any other downsides. The price of fuel is largely manipulated according to changing global fashions. Diesel was cheaper when the major concern was depletion of global oil reserves (because they drink less of the stuff), and again when the major pollutant of concern was carbon dioxide (similarly, less fuel in means less carbon out). Now that the major pollution concern has switched to NOx and particulates, pricing policy has changed too. My major reason for switching to diesel cars in the first place (some 25 years ago) was to avoid starting problems caused by electric spark plug ignition for cars left parked outdoors in a damp climate. They've probably overcome that by now, but it isn't a pressing reason to switch back now I've got used to diesel. The last time I drove a small-capacity petrol engine (courtesy loan while my car was being serviced), it felt like I was riding an angry wasp - had to drop so many gears, and boost the revs so high, just to get up the hills before sunset. |
|
3179)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 71629)
Posted 17 Aug 2016 by Richard Haselgrove Post: So unlike the respect shown by English football fans to other countries' national anthems. |
|
3180)
Message boards :
Questions and problems :
Boinc client wastes resources
(Message 71163)
Posted 3 Aug 2016 by Richard Haselgrove Post: I assume that the GPU resource is idle because no work is downloaded and "Ready to start"? The question then becomes "Why doesn't BOINC request, or possibly receive, GPU work from PrjB or PrjC?"? My suspicion is that it's because your resource shares for those two projects are so low, and may be aggravated if either of those two projects apply limits to the number of tasks you can download or run at any one time. The best way of diagnosing a case like this is to briefly enable the "work_fetch_debug" Event Log option - that's very easy in BOINC v7.6.22, just press Ctrl+Shift+F from BOINC Manager and tick the box. Then open the Event Log window itself (Ctrl+Shift+E), click the line '--- start work fetch state ---', shift-click the line '--- end work fetch state ---' (that should select all lines in between), copy the selected lines and paste them here for us to analyse. Depending how well we know the behaviour of A, B and C, we may also need to ask you to post a full 'scheduler request' and reply from the Event Log for each of them. |
|
3181)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 71150)
Posted 2 Aug 2016 by Richard Haselgrove Post: Do you not have any anti-virus on that machine which normally requires a reboot once a month or so. Everybody knows that Linux is invulnerable to viruses... |
|
3182)
Message boards :
BOINC Manager :
Disable BOINC Manager system tray
(Message 71130)
Posted 1 Aug 2016 by Richard Haselgrove Post: Thinking about Jord's (and my) answer, applications which are designed purely to be 'user-interactive', like a word processor, show their icons on the task bar only. Applications which represent processes which continue in the background - like BOINC - are the only ones which appear in the system tray. |
|
3183)
Message boards :
BOINC Manager :
Disable BOINC Manager system tray
(Message 71129)
Posted 1 Aug 2016 by Richard Haselgrove Post: I'm a Windows user, and I don't know the Gnome interface at all, but... Doesn't every application that you open manually display an icon in the system tray? Isn't that what it's for? Does Gnome have a control equivalent to Windows (ver 7 and up), whereby 'Notification Area Icons' (their new name for the system tray) can be set to 'Only show notifications', as distinct from 'Show icon and notifications'? The former keeps the icons invisible unless there is a need to attract the user's attention, and is normal for a program like BOINC Manager which is normally run minimised. |
|
3184)
Message boards :
BOINC Manager :
Disable BOINC Manager system tray
(Message 71124)
Posted 1 Aug 2016 by Richard Haselgrove Post: I can't find the option you indicated, either in simple view or advanced view... 
|
|
3185)
Message boards :
BOINC client :
How to make boinc client "don't download new task from server when all cores are running"?
(Message 71064)
Posted 28 Jul 2016 by Richard Haselgrove Post: Have you tried simply setting your cache ("Store at least ... days of work") at zero? I believe the client is hard-coded to initiate the next work request 3 minutes before idleness is expected, to allow enough time for a new task to be downloaded and ready to start - it shouldn't take too long to find that in the core work_fetch loop and hack it down to zero, if 3 minutes is too long. Edit - alternative suggestions: 1) Set project Resource Share to zero - backup project, only fetch work when host is idle. 2) Client configuration <fetch_minimal_work>0|1</fetch_minimal_work> |
|
3186)
Message boards :
BOINC client :
"Couldn't resolve proxy name" after every reboot
(Message 71061)
Posted 28 Jul 2016 by Richard Haselgrove Post: I wonder what "<no_proxy>1</no_proxy>" would do? I'll check the Wiki... which says <no_proxy>list of hostnames for which proxy not used</no_proxy> Might be useful in this case - although the list format isn't specified. |
|
3187)
Message boards :
BOINC client :
"Couldn't resolve proxy name" after every reboot
(Message 71059)
Posted 28 Jul 2016 by Richard Haselgrove Post: As Richard said there is now config option for no automatic detection in source code. If you want it you'll need to build the client yourself or if you ask nicely maybe Richard could compile it for you. Ah. I knew I was supposed to be testing something... I can confirm that the proxy section in cc_config.xml now reads <proxy_info>
<socks_server_name></socks_server_name>
<socks_server_port>80</socks_server_port>
<http_server_name></http_server_name>
<http_server_port>80</http_server_port>
<socks5_user_name></socks5_user_name>
<socks5_user_passwd></socks5_user_passwd>
<http_user_name></http_user_name>
<http_user_passwd></http_user_passwd>
<no_proxy></no_proxy>
<no_autodetect>1</no_autodetect>
</proxy_info>With the final line ("<no_autodetect>") set to 1 (as I have done manually in this case), the log entry "automatic proxy check in progress" fails to appear - no other visible sign. I wonder what "<no_proxy>1</no_proxy>" would do? I'll check the Wiki... I have a 64-bit Windows client running now, which you're welcome to test - it would only take a couple of minutes to compile a 32-bit version to match it. |
|
3188)
Message boards :
BOINC client :
"Couldn't resolve proxy name" after every reboot
(Message 71056)
Posted 28 Jul 2016 by Richard Haselgrove Post: I also got "Direct access (no proxy server)". I've compiled a new client and manager from head sources (as of about 40 minutes ago): my startup log now includes 28/07/2016 14:40:51 | | (to change preferences, visit a project web site or select Preferences in the Manager) 28/07/2016 14:40:52 | | [proxy] HTTP_OP::no_proxy_for_url(): https://albertathome.org/rss_main.php 28/07/2016 14:40:52 | | [proxy] returning false 28/07/2016 14:40:52 | Einstein@Home | [cpu_sched] Restarting task PM0147_032C1_318_0 using einsteinbinary_BRP6 version 152 (BRP6-Beta-opencl-intel_gpu) in slot 4 28/07/2016 14:40:52 | NumberFields@home | [cpu_sched] Restarting task wu_sf3_DS-11x271_Grp289150of614400_0 using GetDecics version 211 in slot 1 28/07/2016 14:40:52 | NumberFields@home | [cpu_sched] Restarting task wu_sf3_DS-11x271_Grp291583of614400_0 using GetDecics version 211 in slot 5 28/07/2016 14:40:52 | NumberFields@home | [cpu_sched] Restarting task wu_sf3_DS-11x271_Grp289074of614400_1 using GetDecics version 211 in slot 0 28/07/2016 14:40:52 | GPUGRID | [cpu_sched] Restarting task e25s19_e1s35p0f677-GERARD_FXCXCL12RX_1497613_1-0-1-RND4377_0 using acemdlong version 848 (cuda65) in slot 3 28/07/2016 14:40:52 | NumberFields@home | [cpu_sched] Restarting task wu_sf3_DS-11x271_Grp296499of614400_0 using GetDecics version 211 in slot 2 28/07/2016 14:40:52 | SETI@home | [cpu_sched] Restarting task 26my10aa.32159.14402.3.30.130_0 using setiathome_v8 version 800 (cuda50) in slot 7 28/07/2016 14:40:52 | SETI@home | [cpu_sched] Restarting task 26my10aa.32159.18492.3.30.97_1 using setiathome_v8 version 800 (cuda50) in slot 6 28/07/2016 14:40:53 | | [proxy] automatic proxy check in progress 28/07/2016 14:40:54 | | [proxy] HTTP_OP::no_proxy_for_url(): https://einsteinathome.org/notices.php?userid=144054&auth=144054_96d85d30adf19fcc83a280d8c1575568 28/07/2016 14:40:54 | | [proxy] returning false 28/07/2016 14:40:57 | | [proxy] HTTP_OP::no_proxy_for_url(): http://findah.ucd.ie/notices.php?userid=124464&auth=124464_e433d367c744d5a3863b477f373ea576 28/07/2016 14:40:57 | | [proxy] returning false 28/07/2016 14:40:59 | | [proxy] HTTP_OP::no_proxy_for_url(): http://www.gpugrid.net/notices.php?userid=30277&auth=30277_35c13b5a51da7043408976de34dc6a07 28/07/2016 14:40:59 | | [proxy] returning false 28/07/2016 14:41:01 | | [proxy] HTTP_OP::no_proxy_for_url(): http://numberfields.asu.edu/NumberFields/notices.php?userid=1969&auth=1969_a51674229132bb81e005ca901c1109a5 28/07/2016 14:41:01 | | [proxy] returning false 28/07/2016 14:41:03 | | [proxy] no automatic proxy detected 28/07/2016 14:41:03 | | [proxy] HTTP_OP::no_proxy_for_url(): http://setiathome.berkeley.edu/notices.php?userid=5509&auth=5509_474dbc6bd79c0c382cf0504aa7fc08f2 28/07/2016 14:41:03 | | [proxy] returning false 28/07/2016 14:41:05 | | [proxy] HTTP_OP::no_proxy_for_url(): http://setiweb.ssl.berkeley.edu/beta/notices.php?userid=5135&auth=5135_6d4a11988395ee187d6e4bce89a2d411 28/07/2016 14:41:05 | | [proxy] returning false 28/07/2016 14:46:45 | SETI@home | Computation for task 26my10aa.32159.14402.3.30.130_0 finished 28/07/2016 14:46:45 | SETI@home | Starting task 26my10aa.32159.19719.3.30.229_0 28/07/2016 14:46:45 | SETI@home | [cpu_sched] Starting task 26my10aa.32159.19719.3.30.229_0 using setiathome_v8 version 800 (cuda50) in slot 7 28/07/2016 14:46:48 | | [proxy] HTTP_OP::no_proxy_for_url(): http://setiboincdata.ssl.berkeley.edu/sah_cgi/file_upload_handler 28/07/2016 14:46:48 | | [proxy] returning false 28/07/2016 14:46:48 | SETI@home | Started upload of 26my10aa.32159.14402.3.30.130_0_0 28/07/2016 14:46:50 | | [proxy] HTTP_OP::no_proxy_for_url(): http://setiboincdata.ssl.berkeley.edu/sah_cgi/file_upload_handler 28/07/2016 14:46:50 | | [proxy] returning false 28/07/2016 14:46:52 | SETI@home | Finished upload of 26my10aa.32159.14402.3.30.130_0_0 Strange that the first proxy report appears for a notice check before the 'restart task' messages, and the 'automatic proxy check in progress' after them - but it doesn't appear to affect the outcome. (I'm on wired ethernet behind a NAT router, and from there DSL to a standard domestic ISP) |
|
3189)
Message boards :
Questions and problems :
allocating cycles to BOINC projects
(Message 71046)
Posted 27 Jul 2016 by Richard Haselgrove Post: 3) Very large result file uploading (I've set it to 1 at a time per project on the cc_config to minimize concurrent uploading). Often suffer from poor ISP/Cross continental file transfer crawling from my outback. A good quality (wired) network connection should only require a trivial amount of CPU support from the host device - no matter how much buffering is being caused by a slow upload link. But some - especially WiFi - connections are notorious for introducing latencies via badly-written wireless drivers. In my experience, USB WiFi dongles are particularly bad. |
|
3190)
Message boards :
BOINC client :
"Couldn't resolve proxy name" after every reboot
(Message 71025)
Posted 26 Jul 2016 by Richard Haselgrove Post: David Anderson committed new code overnight: client (Win): add config option to not do proxy autodetect There hasn't been a new test build since early June, but I found it easy (and successful) to follow the 'build your own' instructions for VS2012. Out of curiosity, why would a new project executable need new clients? |
|
3191)
Message boards :
Questions and problems :
Use GPU always, CPU when idle
(Message 70911)
Posted 16 Jul 2016 by Richard Haselgrove Post: The developers have always treated this in a very literal-minded way: if you compute on a GPU, there is always some CPU activity required, so in their view it is impossible to pause the CPU while still running a GPU. I think the rest of us have a more relaxed view of the capabilities of a modern multi-tasking operating system. The machine I'm typing this on is running a GPU task with <2% (of one CPU core out of four) required to provide the GPU support. That <2% is a 30-second average, and most of the time the CPU usage is lower than many Windows background processes. GPU applications vary in the amount of CPU support they require - depending on the project (computational complexity), development tools, driver sophistication, and programmer concerned. But I agree with you - the ability to manage resource usage in this way should be provided for users to activate, or not, as they choose. But don't hold your breath waiting for the developers to change their minds: it's been requested many times before, and rejected every time. |
|
3192)
Message boards :
BOINC Manager :
BOINC Manager dependencies
(Message 70909)
Posted 16 Jul 2016 by Richard Haselgrove Post: Happy crunching! Sounds like a result - happy crunching indeed :-) Since BOINC Manager can only connect to one client at a time (unlike Windows add-ons like BoincTasks), the only other thing I'd suggest is to add -m to your command line, so that you can run two copies (one for each remote). Then I guess you should be able to set up duplicate menu entries, one named for each of your desktops, and not worry about having to exit one when you want to check the other. |
|
3193)
Message boards :
BOINC Manager :
BOINC Manager dependencies
(Message 70907)
Posted 16 Jul 2016 by Richard Haselgrove Post: Sounds like you're ahead of the game, so you're probably familiar with this already, but I'll post it anyway for other readers to refer to. D:\BOINC>D:\BOINC\boincmgr.exe /h |
|
3194)
Message boards :
BOINC Manager :
BOINC Manager dependencies
(Message 70902)
Posted 16 Jul 2016 by Richard Haselgrove Post: In general, BOINC installers are monolithic - they assume you want to run both major components (the client and the manager) on the same computer, and install both components without giving you any freedom of choice. Of course, having the client installed doesn't mean you actually have to run it. Simply not attaching to any projects will prevent it using scarce RPi resources. But again in general, starting the Manager interface will often attempt to start the client on the local machine, and then nag you with 'attach to project' wizards each time you start it. I'm a Windows user, so I can't give you specific RPi advice, but it should be possible to: a) prevent the local client starting at power-up b) invoke the Manager with command-line switches that connect it directly to a remote machine. I'm having difficulty finding the relevant page in the Wiki, but I should be able to find some examples later. In the meantime, you will need to allow remote RPC access to the clients running on your desktops - that's in Controlling BOINC remotely. |
|
3195)
Message boards :
Questions and problems :
Do any projects currently run on Itanium (IA-64) under any OS?
(Message 70871)
Posted 15 Jul 2016 by Richard Haselgrove Post: I think 'host' will probably be determined solely by HostID number - and the second client would get a different HostID. We'd need to dig a lot deeper to get a definitive answer to that. |
|
3196)
Message boards :
Questions and problems :
Do any projects currently run on Itanium (IA-64) under any OS?
(Message 70869)
Posted 15 Jul 2016 by Richard Haselgrove Post: Aside from 'to allow or not to allow private compiles', different physical machines is the requirement on those results needing verification, but if this is still ensured with the 'allow multiple clients' or multiple VMs on one device is doubtful. Does the feeder/scheduler look beyond client ID and any homogeneous redundancy requirements when distributing work? Self-verification is just not a good idea, when there's a higher chance of one class being controlled by one or few users. Self-verification can be allowed by project administrators, if circumstances permit - I've seen it done for Beta testing, where the pool of volunteers is too smaller to ensure timely validation. In that case, you're perhaps looking for the validation failures for further investigation. <one_result_per_user_per_wu/> <one_result_per_host_per_wu/> from http://boinc.berkeley.edu/trac/wiki/ProjectOptions#Joblimits I've also seen at least one project administrator - was it Eric Mcintosh at LHC? - work very diligently to get exact bit-wise matching results from heterogeneous platforms - I don't think I'd be happy contributing cycles to a project that didn't at least pay lip-service to that ideal. |
|
3197)
Message boards :
Questions and problems :
Do any projects currently run on Itanium (IA-64) under any OS?
(Message 70854)
Posted 14 Jul 2016 by Richard Haselgrove Post: If the project requires a second copy for validation, they probably don't want those being used as the odds of having someone validate the work unit would be slim. On the other hand, I suspect very few projects use or require the Homogeneous Redundancy mechanism. In general, and including the projects requiring a second copy for validation, it is assumed that every {OS|CPU|GPU} scientific calculation produces the same numerical outcome, to within the scientific accuracy required by the project. If an Itanium chip routinely produces a different (non-validating) answer when compared with an x86, 68k, PowerPC, RISC or ARM chip, which one of them is right? I'd say that's a false problem. Compile the source code with a reputable IA-64 compiler, don't use excessive optimisation switches, and launch your application under Anonymous platform. If the first few tasks validate, you're up and running. If they don't, go back and check your compiler settings (and any source code mods you needed) very, very, carefully until you find a configuration which does return the same scientific data as the other platforms. |
|
3198)
Message boards :
Android :
No New Tasks after completing several task sets
(Message 70659)
Posted 3 Jul 2016 by Richard Haselgrove Post: No, the SETI server is up and running. The 5 minutes (and 3 seconds) communications deferral is their standard value, allowing other users time to take their turn at the trough. If you ever see any time other than five minutes, it would indicate that something was wrong - most commonly a 1 hour comms deferral during the scheduled maintenance every Tuesday morning, Pacific time. |
|
3199)
Message boards :
GPUs :
Split Work Units on 1 GPU
(Message 70619)
Posted 1 Jul 2016 by Richard Haselgrove Post: Superficially, that looks OK (*). Two questions: Did you put the file in the right place (the POEM@Home project sub-folder of your BOINC data folder)? Did you issue the instruction 'Read config files' from the Options menu in BOINC Manager (Advanced view) afterwards? If so, there should be useful information in the Event Log. * I'm not familiar with POEM, but that <name>Poem@Home</name> looks dodgy. It needs an application name, not a project name. Be prepared to refer to Application configuration for full documentation, if you didn't get your original inspiration from there. |
|
3200)
Message boards :
Questions and problems :
Win 10 remote control using BOINCTasks
(Message 70616)
Posted 1 Jul 2016 by Richard Haselgrove Post: Had to set an 'incoming' exception rule in the W10 firewall for boinc.exe Depends exactly what rule you set there. If it was the minimal ports 80, 443 for http/https, then the remote protocols couldn't (shouldn't) have been able to connect. If you set ports 80, 443, 31416 - or 'all ports' - then it should work. The holes in the swiss cheese are for outgoing connections - BOINC Tasks/View/Manager calling out from the machine they're running on, into the remote machines. |
|
3201)
Message boards :
Questions and problems :
Win 10 remote control using BOINCTasks
(Message 70611)
Posted 1 Jul 2016 by Richard Haselgrove Post: Had to set an 'incoming' exception rule in the W10 firewall for boinc.exe And that will apply to any program that uses the Controlling BOINC remotely protocol - BOINC Tasks, BOINC View, BOINC Manager... |
|
3202)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 70608)
Posted 1 Jul 2016 by Richard Haselgrove Post: Now they say the ozone hole is shrinking? From your link: "The gains have been credited to the long term phasing out of ozone-destroying chemicals." We did that. We acted on the discovery, and the warnings of potential consequences. By 'we', I mean the human race (on all continents), and the politicians we elect to put our wishes into action. Why does that sound so unusual and unlikely? (apologies for using the 'p' word in this thread) |
|
3203)
Message boards :
Questions and problems :
Possible issue: app not responding, force quite, formerly near-complete project has 'no tasks available'
(Message 70603)
Posted 1 Jul 2016 by Richard Haselgrove Post: If someone can help me locate earlier logs, I would appreciate it. Earlier logs should be accessible in a file called 'stdoutdae.txt' (or, on a Mac, possibly simply 'stdoutdae'). One older log rotation called 'stdoutdae.old' could also be available. The log files will be contained in (from your current log) Data directory: /Library/Application Support/BOINC Data |
|
3204)
Message boards :
BOINC Manager :
how to set snooze time for GPU?
(Message 70478)
Posted 26 Jun 2016 by Richard Haselgrove Post: I don't think it's ever been possible to set a 'snooze' time other than 1 hour using standard Manager tools. There used to be a 'Network reminder interval' - slider from 0 to 120 minutes - which survived from the age of dial-up modems until at least v6.10. Something like that could have been adapted to make a default snooze time control, but wasn't. |
|
3205)
Message boards :
Android :
Boinc Android Version 7.4.49 problems
(Message 70403)
Posted 24 Jun 2016 by Richard Haselgrove Post: Also reported at SETI, so that makes four projects. Won't the projects have to update their own local server software, too, before they can service the new type? |
|
3206)
Message boards :
GPUs :
Different projects and PCI-e bandwidth usage
(Message 70322)
Posted 20 Jun 2016 by Richard Haselgrove Post: I suggest that you think about the sort of work each project is doing, and how it fits into your interests and intended construction. GPUs - in the configuation you describe - are probably best suited to integer calculations based on a formula or algorithm. They are probably least suited to projects which search large quantities of pre-recorded data, and perform double-precision floating point arithmetic on it. You can see how bitcoin would fit into the first group. It shares those characteristics with, say, Collatz Conjecture and PrimeGrid. I don't know of any project which falls completely into the second group - Milkyway requires double-precision floating point arithmetic, but I think is doing algorithmic simulations, rather than data searches. Examples of middle-of-the-road single precision data searches include Einstein, GPUGrid, SETI. (edit - GPUGrid is perhaps better described as a simulation based on large volumes of input data - but it still needs that bandwidth) You can probably work out other contenders for yourself by reading down the Category column in Choosing BOINC projects. |
|
3207)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 70287)
Posted 18 Jun 2016 by Richard Haselgrove Post: Let's celebrate Tim Peake, Yuri Malenchenko and Tim Kopra all returning safely together in the same capsule. The International Space Station, and especially the views back from it to planet earth, should remind us how trivial national boundaries and all issues of nationality really are. |
|
3208)
Message boards :
Questions and problems :
Need to reset debt
(Message 70214)
Posted 15 Jun 2016 by Richard Haselgrove Post: Add a project and set it with a decent project weight/resource share and major catch-up ensues. It takes rather less than forever if you set an option like <rec_half_life_days>1.000000</rec_half_life_days> |
|
3209)
Message boards :
The Lounge :
These are no longer with us, may they have peace.
(Message 70002)
Posted 6 Jun 2016 by Richard Haselgrove Post: Dave Swarbrick, 75, musician. Folk-rock fiddler extraordinaire - Fairport Convention and others. |
|
3210)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 69971)
Posted 5 Jun 2016 by Richard Haselgrove Post: Weather-related outages aren't usually clean 'switch off - switch on' events - as those of us who live in rural(-ish) areas know all too well. Winds blow, conductors touch, lightning strikes, branches fall off trees. All accompanied by lights flickering, long before they finally go out. And then the engineers don't always make a clean connection first time when they put it all back together. I'm pretty sure it's all the transients round the edges that kills computers, not the blackout itself. |
|
3211)
Message boards :
BOINC client :
boinccmd --get_cc_status returns "Operation failed: Error 2"
(Message 69946)
Posted 3 Jun 2016 by Richard Haselgrove Post: I liked the "spurious error msg" part in github :) If you liked that, you'll like the prequel: https://github.com/BOINC/boinc/commit/bb2f9339b48df2b170583741a3323616dc119484 |
|
3212)
Message boards :
BOINC client :
boinccmd --get_cc_status returns "Operation failed: Error 2"
(Message 69942)
Posted 3 Jun 2016 by Richard Haselgrove Post: I have a faint recollection that something in that general direction was fixed somewhat recently. I don't think it's a real error anyway. https://github.com/BOINC/boinc/commit/0921a0e7c433f321801b0714f870b8ebf71cf51a Edit - tested with the private build I made to test yesterday's bugfix - no error. It's gone. |
|
3213)
Message boards :
BOINC client :
boinccmd --get_cc_status returns "Operation failed: Error 2"
(Message 69937)
Posted 3 Jun 2016 by Richard Haselgrove Post: I'm on BOINC v7.6.22, Windows 7 Pro/64, and my BOINC programs aren't in a protected area. But I'm getting 'Error 2' every which way I try it. D:\BOINC>boinccmd.exe --get_cc_status
network connection status: don't need connection
CPU status
not suspended
current mode: always
perm mode: always
perm becomes current in 0 sec
GPU status
not suspended
current mode: always
perm mode: always
perm becomes current in 0 sec
Network status
not suspended
current mode: always
perm mode: always
perm becomes current in 0 sec
Operation failed: Error 2
|
|
3214)
Message boards :
Questions and problems :
[Discussion] 4th Generation BOINC credit system
(Message 69903)
Posted 2 Jun 2016 by Richard Haselgrove Post: I suspect the project scientists who commission the research in the first place will do the choosing. I've suspected for a while that many of them would prefer the more controlled environment of a commercial platform, rather than the Wild West of BOINC with all these message boards to worry about. Witness the way SIMAP was at the forefront of BOINC for Android with their HTC 'Power to Give' collaboration - but very rapidly closed down their entire BOINC operation, soon afterwards. |
|
3215)
Message boards :
Questions and problems :
Windows 10 Professional Reboot
(Message 69893)
Posted 1 Jun 2016 by Richard Haselgrove Post: Back to the 'crashes due to reboot' problem. I think I see what may have gone wrong. From https://msdn.microsoft.com/en-us/library/cc231199.aspx: 0x0000026B That sounds close enough to "The device has no drivers installed, or the drivers are configured incorrectly" that it perhaps implies a simple mis-reading between 0000026B C000026B I do wish Microsoft would pay more attention to their typography. |
|
3216)
Message boards :
Questions and problems :
Windows 10 Professional Reboot
(Message 69885)
Posted 1 Jun 2016 by Richard Haselgrove Post: Actually, Windows error 0xC000026B is defined in https://msdn.microsoft.com/en-us/library/cc704588.aspx as 0xC000026B The problem has just been reported at https://github.com/BOINC/boinc/issues/1553, together with a possible solution - use boinctray to monitor windows messaging (it does that), and create a mutex to signal the client (which doesn't). |
|
3217)
Message boards :
BOINC Manager :
Designate disc drive?
(Message 69884)
Posted 1 Jun 2016 by Richard Haselgrove Post: Don't move the program directory ('INSTALLDIR') unless you really need to - which you probably don't just for space reasons. It's about 22 MB and doesn't grow. If you move it, you'd have to change 'run' entries in the user hives too. It would be interesting to hear if project application permissions survive the move, especially if BOINC is installed as a service. Please let us know. |
|
3218)
Message boards :
BOINC Manager :
Designate disc drive?
(Message 69865)
Posted 31 May 2016 by Richard Haselgrove Post: For Windows, there is a single registry entry - or to be pedantic, one entry for the BOINC programs, and the other for all user data. Mine are [HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup] "INSTALLDIR"="D:\\BOINC\\" "DATADIR"="D:\\BOINCdata\\" You could work from that, but it's actually quicker and easier to go the 'reload' (as you call it) route. If you're already running BOINC in the default location, and decide you've changed your mind. 1) Uninstall the old version - that removes the BOINC programs only, and leaves your personal data intact. 2) Move the BOINC Data directory to the new location of your choice. 3) Re-run the BOINC installer, pausing on the file location screen to click the 'Advanced' button and fill in the new location for your data. When you restart BOINC, everything will function - projects will still be attached, preferences will be as you like them, work in progress will start again where it left off. |
|
3219)
Message boards :
Projects :
GPUGRID down?
(Message 69819)
Posted 30 May 2016 by Richard Haselgrove Post: All working at normal speed again. Somebody had a late evening shift... |
|
3220)
Message boards :
Projects :
GPUGRID down?
(Message 69818)
Posted 30 May 2016 by Richard Haselgrove Post: My error message has now become "www.gpugrid.net took too long to respond." - each step sounds like progress, an improvement. More to the point, I got a new task allocated at 30/05/2016 22:13:25 | GPUGRID | Scheduler request completed: got 1 new tasks (timezone UTC+1) and after a bit of kicking it's fully downloaded and ready to run - so the project is alive and doing science, even if it's a bit sickly just now. |
|
3221)
Message boards :
Projects :
GPUGRID down?
(Message 69815)
Posted 30 May 2016 by Richard Haselgrove Post: Just too bad that they let the certificates expire :-( Should not happen Just because somebody is a wizard at protein binding energies, doesn't necessarily mean they're red hot at computer security! That's a problem which affects all small-staff BOINC projects. |
|
3222)
Message boards :
Server programs :
How to catch project reset?
(Message 69813)
Posted 30 May 2016 by Richard Haselgrove Post: I don't think the act of clicking 'reset project' is automatically communicated to the server, either - you would probably first see the next request for work, with no application versions or tasks in progress. |
|
3223)
Message boards :
Projects :
GPUGRID down?
(Message 69811)
Posted 30 May 2016 by Richard Haselgrove Post: Their website security certificate expired (my browser is saying "NET::ERR_CERT_DATE_INVALID"). That was earlier today: I could get through to the message boards using a less secure browser, and I saw that somebody had reported it already. Now, connections are being refused, so I guess they're in the middle of trying to fix things. |
|
3224)
Message boards :
GPUs :
Do not use Nvidia drivers 364.47 through 365.19 on Nvidia GPUs based on the Maxwell architecture
(Message 69744)
Posted 25 May 2016 by Richard Haselgrove Post: Jacob Klein has been pursuing two - now apparently unrelated - driver issues with NVidia. Both started with driver version 364.47: Miscalculations - invalid results - at PrimeGrid. Fixed in v365.19 Driver restarts and BSODs at POEM. Fixed in v368.22 No other BOINC projects are known to be affected. Earlier drivers (up to and including v362.00) are clean for both projects Later drivers (starting with v368.22) are clean - until the next breakage, of course. The advice (and it may possibly apply to Windows only) is to avoid the intermediate drivers if you run either of those two specific projects. |
|
3225)
Message boards :
Questions and problems :
Problem with OS X Mavericks 10.9.5
(Message 69728)
Posted 24 May 2016 by Richard Haselgrove Post: You can also use: Activity menu: Suspend GPU Project web-based preferences: Use GPU (unchecked) - prevents new tasks being downloaded, but allows you to finish off existing ones. |
|
3226)
Message boards :
Questions and problems :
[Discussion] 4th Generation BOINC credit system
(Message 69625)
Posted 16 May 2016 by Richard Haselgrove Post: I don't think there's anything wrong with the RAC calculation, Jord. It's the old gigo problem that applies to all mathematics: Garbage In, Garbage Out. RAC is calculated from credit awarded, 'nuff said. |
|
3227)
Message boards :
Questions and problems :
BOINC on older computers?
(Message 69624)
Posted 16 May 2016 by Richard Haselgrove Post: (... that have shortish tasks.) And do checkpoints. |
|
3228)
Message boards :
Questions and problems :
[Discussion] 4th Generation BOINC credit system
(Message 69565)
Posted 11 May 2016 by Richard Haselgrove Post: There's also a problem with extending even project-generated and server-held averages from CPUs to GPUs. It is reasonable to assume that all CPU cores in a given host run at the same speed. It's less reasonable to assume that all CPUs of the same model number run at the same speed - some might have hyperthreading enabled, others not, even before we mention overclocking. But BOINC made what is now acknowledged as a big design mistake by letting the servers assume that all GPUs (from each manufacturer) in a given host are also comparable. Until and unless the proposed server change allowing tracking of individual GPUs is implemented (and I see no sign of it yet), I fear a lot of extra programming work - much reinventing of many wheels - at the project level. I suspect that most project administrators would much prefer that the credit solution was taken out of their scope and handled centrally. |
|
3229)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 69554)
Posted 10 May 2016 by Richard Haselgrove Post: I guess everybody will be over in the creditnew complaint thread :) Yup. 0.02€ deposited. |
|
3230)
Message boards :
Questions and problems :
[Discussion] 4th Generation BOINC credit system
(Message 69552)
Posted 10 May 2016 by Richard Haselgrove Post: Thanks, Christian. Plenty of food for thought there. First, what are Credits (or Points, or Unicorns) for? For volunteers they are motivational, but for BOINC itself they are a statement of computer science attainment. They're right there on the front page: "24-hour average: 11.562 PetaFLOPS". That figure is reverse-engineered from the total number of BOINC credits awarded, no matter where or how: clicking Top 100 multi-project BOINC participants gives you some idea of where they come from. Speaking personally, I'd like to retain both elements of that: call it 'motivation by scientific attainment', if you like - credits that are in some respect pegged to (hypothetically) measurable quantities. Since there are only a limited number of measurable quantities to pick from, that should also bring with it at least some comparability across projects: I'd like to see that, and I recoil against that "can't be fair" assertion. Maybe if projects were explicit about whether their credits were for single-precision flops, double-precision flops, or integer maths, some of the fog might lift? That's the theory, at least. Now, let's turn to the problem of implementation. A little history. I think we can safely assert that BOINC arose out of the original self-contained SETI@Home Classic project. The same developers worked on both SETI and BOINC, and the expected primary function (signal processing by single-precision floating point maths on general purpose home computer CPUs) probably infuenced the evolving reward mechanism heavily. BOINC credits are also known as cobblestones, and the linkage is explicit. The early days of both SETI and BOINC were perceived to suffer (scientifically speaking) from cheating. Reading CreditNew, specifically the description of the transitions between the first and the second, and the second and the third, credit systems, it's clear that the developers felt the need to remove as much scope for cheating from the volunteers, and as much need for bespoke programming from project developers, as possible. So what happened? SETI was using the second credit system (flopcounting), when GPU computing joined the BOINC framework in the winter of 2008/09. Despite the impression given in CreditNew, flopcounting and GPUs coexisted successfully (-ish - I won't pretend there weren't some difficulties) for over a year. Then, two things happened in quick succession. Initially, BOINC's third generation credit system was trialed on SETI's Beta server on 31 March 2010: the message board thread New credit calculation code still makes informative (and entertaining) reading. Because flopcounters were still present in the SETI applications, the old (second generation) credit claims could be compared with the new (third generation) credit awards. Although the image host I was using at the time is no longer performing, I still have the original graphs from that thread and I can re-post them if they would contribute to the discussion. And the second thing which happened was the public release to sale of the next generation of GPUs: NVidia's Fermi range (that helps to contextualise the dates). It turned out that the new hardware was incompatible with existing software, and much chaos ensued: in the midst of the chaos, a new Fermi application was obtained from NVidia, and when it was deployed to SETI on 7 June 2010, CreditNew was deployed as well. So, CreditNew was tested at SETI Beta for barely more than two months before being declared the new gold standard, and I can find - then or now - no evidence that it was evaluated for 'fitness for purpose'. Certainly, examining the History for boinc/sched/credit.cpp, there are some bug fixes and adaptations to meet new needs, but no revisions that even attempt to address the shortcomings we started to see in those first two months at Beta. And I think that's the biggest problem with the third generation credit system. Nobody owns it - certainly nobody loves it. Volunteers take their grumbles to project administrators: project administrators say 'this isn't what I signed up to deal with - I'm a scientist' (and run to hide behind the nearest bush): and BOINC seems to say 'here you are, we've made this - take it or leave it'. Most people choose to leave it. Of course, other things have happened during the six years that CreditNew has languished in the orphanage. Most significantly, the NSF has withdrawn - strictly, declined to renew - development funding for BOINC. I think that's a positive event - the NSF are saying that BOINC is no longer a computer science experiment: it's grown up, succeeded, left the That's both a challenge and an opportunity. Fundamentally, I think it's time that the projects (re-) took control of the strategic direction of BOINC: they should decide collectively what is needed, and what are merely decorative bells and whistles. After the NSF funding ceased, BOINC set up an ideal structure for this: the Project Management Committee. But it was given absurd terms of reference - looking downwards and inwards towards micro-code, instead of outwards and upwards to the needs of the scientific community. I think it should seize the strategic role instead, and as a first step remove David Anderson from the role of chairperson. David may have been a brilliant experimentalist, but mature, robust, infrastructure needs different skills, and the committee chair needs to be someone who can manage them. Then, I'd ask that the new PMC consider this question of credit, and decide whether it is something they wish to address (I hope they do). Then, if the strategic heads of the organisation are listening, there is more hope that the labourers at the coalface - like the authors of Evaluation Of CreditNew - will feel that their efforts are worthwhile. |
|
3231)
Message boards :
Questions and problems :
[Discussion] 4th Generation BOINC credit system
(Message 69530)
Posted 9 May 2016 by Richard Haselgrove Post: I believe that they are still in the setting-up phase. I did receive an invitation: The Berkeley SETI Research Center cordially invites you to attend a reception and lecture: I believe "The Breakthrough Listen Lab" is a new facility at Berkeley, partially or wholly funded through Yuri Milner's foundation. We may learn more in 10 days' time (but I can't go). SETI@Home has been processing additional data, recorded at the Green Bank Telescope, since Yuri's Night #55, the 55th anniversary of Yuri Gagarin's historic first space flight on April 12, 1961. I think that some of the Milner Money went on that, too - such as the new data recorder at GBT. What I don't know - but I think we should be told - is how broadly the terms of the Milner support have been drafted. In UK philanthropic circles, at least, it is now widely accepted that good practice requires that, if your gift enables a significant enhancement of the work or the receiving organisation, then you should devote a significant - and realistic - proportion of the gift to enhancing the core administration and infrastructure of the hosting body. In this case, it seems as if lab provision has been included, which is a good start, but I have seen no sign of additional staffing, and I would argue that the BOINC codebase should be considered as core infrastructure for the Breakthrough team too - no (public) sign of that either, yet. |
|
3232)
Message boards :
BOINC client :
allow multiple clients
(Message 69481)
Posted 5 May 2016 by Richard Haselgrove Post: When I do Ctrl+Shift+F, a window called "BOINC Diagnostic Log Flags" opens. That's the one. Make a change, 'OK' it. Go and have a look at cc_config.xml It should contain everything described in the first two sections of Client configuration - Logging flags and Options. |
|
3233)
Message boards :
BOINC client :
allow multiple clients
(Message 69478)
Posted 5 May 2016 by Richard Haselgrove Post: I got interrupted by a phone call as I was performing my test, and had to go out for a bit, but I should probably amplify a bit. Machine is Windows 7/64 BOINC was initially installed in user mode (primary instance runs GPUs). Executable files have been updated with self-built versions from head, as at early March this year. I'm pretty sure that doesn't affect the multiple clients issue - no change in that area from 7.6.22/.23/.29/.32 or wherever we're up to now. Both program and data directories are located on a clean data drive: neither are in the protected C:\Program Files\ or ProgramData areas. Having tested that the second Manager instance runs properly when the second client is running, I've tried without launching the client first. Nada - tried (and failed) to connect, no error message. In particular, no message about a bad password. Because this is on a secure private network, I have used BOINC Manager itself, and BoincView, to manage machines remotely for many years with an eight-letter password which is an anagram of wssrpoda - makes it easy to avoid errors in that department. If you're using a modern client, like the recommended v7.6.22, the easiest way to update configurations is to use the Event Log Options tool (Ctrl+Shift+F) - make a trivial change and click 'OK'. That creates a template cc_config.xml file with every possible tag present, and all spelled correctly :-) Then just modify 0 to 1, or whatever, for the tag you need. Did I say I restarted the primary client after making that change? It feels like one which would only take effect at startup, rather than through 'read config files' like some others. |
|
3234)
Message boards :
BOINC client :
allow multiple clients
(Message 69470)
Posted 5 May 2016 by Richard Haselgrove Post: OK, as I suspected, curiosity got the better of me. I created the two batch files BobCat13 suggested in the message I'm replying to. And it failed. Watching the changes in the second data directory, I saw a file stderrdae.txt being created, with the contents 05-May-2016 15:50:59 Another instance of BOINC is running. GLE: Cannot create a file when that file already exists. (0xb7) Armed with that information, I edited the cc_config.xml file for the first (originally running) instance, with the <allow_multiple_clients>1</allow_multiple_clients> option, and restarted the first client. After that, everything worked as advertised. Two different clients running (two instances of boinc.exe in Task Manager), two different instances of boincmgr.exe in Task Manager as well, two different displays on screen with different project lists, etc. etc. |
|
3235)
Message boards :
BOINC client :
allow multiple clients
(Message 69463)
Posted 5 May 2016 by Richard Haselgrove Post: Ah OK, but if he's already having trouble connecting two BOINC Managers, he may as well have trouble connecting boinccmd to the right directory. But the Manager doesn't connect to a directory - and nor does boinccmd. The communication is by TCP/IP (like a network), so it connects to a port. It's the client (the thing that the Manager is connecting to) that reads/writes/manages the directory. So I suppose we ought to consider which user account boinc.exe is trying to run under, what account permissions have been set on the new data directory, and whether they're allowed to get it together. I'm almost getting to the stage of being inquisitive enough to try it myself... |
|
3236)
Message boards :
BOINC client :
allow multiple clients
(Message 69406)
Posted 3 May 2016 by Richard Haselgrove Post: You distinguish the two instances by the port number, not the IP address or name (127.0.0.1 and localhost are synonyms - they're the same machine). Once you have both instances connected, you can sort out the names to suit yourself. |
|
3237)
Message boards :
BOINC client :
allow multiple clients
(Message 69403)
Posted 3 May 2016 by Richard Haselgrove Post: William is the one who's actually done this, but isn't available online right now. I believe you install BoincTasks normally, and go through the manual "add computer" process described in http://efmer.com/b/?q=boinctasks_manual. For the second BOINC instance, use 'localhost' or '127.0.0.1' (you'll have to experiment), and use the "computername:port" format as described for the boinccmd tool. |
|
3238)
Message boards :
BOINC client :
allow multiple clients
(Message 69400)
Posted 3 May 2016 by Richard Haselgrove Post: Well, maybe my problem concerning the password could be solved very easily, if someone could just give me the proper hint. The password required is whatever is contained in the gui_rpc_auth.cfg file in the second client data folder. But that assumes that the second client is running, that it's listening on the correct port, etc. etc. etc. I think you get the same "incorrect password" error message whichever part of your configuration is wrong. Edit - make sure that the password you want to use has been entered properly into gui_rpc_auth.cfg, without a newline character after it. |
|
3239)
Message boards :
BOINC client :
allow multiple clients
(Message 69397)
Posted 3 May 2016 by Richard Haselgrove Post: I've heard many people say that they can't get the Manager to attach to the second client instance (though I'm not certain if the problem is always with running a second Manager instance: it would be worth trying to get the first Manager to attach to the second client). The alternative BOINCTasks manager is apparently much easier to get working with a second client. |
|
3240)
Message boards :
Projects :
News on Project Outages
(Message 69292)
Posted 29 Apr 2016 by Richard Haselgrove Post: Sorry, can't resist: Large Hadron Collider: Weasel causes shutdown |
|
3241)
Message boards :
Questions and problems :
how to set/change interval for requesting new tasks
(Message 69291)
Posted 29 Apr 2016 by Richard Haselgrove Post: However, it's clear that I cannot sit there all the time and push the "update" button in 2-minutes' intervals for hours :-) Ah yes, been there. Still doing that. Haven't got the T-shirt yet (though I do have a task on every eligible computer at the moment). I have realized that by NOT trying to get a task manually, it could take up to 10 or 12 hours to finally receive a new WU :-( No, there isn't a set interval. The client will request, the first time, as soon as it feels the need. But if it requests work, and receives none, the interval between requests will increase: BOINC will back off, longer and longer, between requests. The idea is to avoid projects with large numbers of volunteers and little work creating their own DDOS attack on the server, meaning that nobody can get through and the project gets no work done - a lose-lose situation. So, what can you do? Obviously BOINC itself would say 'let another project run in the gap between GPUGrid tasks' - after all, BOINC was created to make scientific use of idle resources. If you are specifically interested in the *cough* scientific research done by GPUGrid, you can automate the update button using the boinccmd tool: boinccmd --project URL operation --project URL operation and using your operating system's task scheduler to repeat when necessary. I'd suggest every five or ten minutes, rather than two - don't overtax the server. Finally, BOINC is open source - I found the build instructions for Windows using the free 'Community Edition' of Microsoft Visual Studio were remarkably easy to follow and successful. I run a self-compiled client with one of the more irritating features disabled to help me debug some project problems: it would probably be fairly easy to find the increasing backoff code and disable that too, replacing it with a fixed value. That would mean that you wouldn't keep pestering the server after you'd snagged a task, as the boinccmd suggestion would. |
|
3242)
Message boards :
Questions and problems :
how to set/change interval for requesting new tasks
(Message 69289)
Posted 29 Apr 2016 by Richard Haselgrove Post: Not directly. The basic principle is "it asks when (it thinks) it needs new tasks". There are some settings which influence that, but before we dig them all out, could you expand a bit more on what you're hoping to do - what change do you want to see? |
|
3243)
Message boards :
GPUs :
After 2 Collaz only GPU1 is computing
(Message 69133)
Posted 24 Apr 2016 by Richard Haselgrove Post: Because you've left in the [ ... ] which indicate to you, the writer, that sections are optional. Either leave out the sections (if you don't want them), or leave out the brackets (if you do want the contents to be active). |
|
3244)
Message boards :
GPUs :
GPUs recognized, but no work is being assigned to them?
(Message 69102)
Posted 22 Apr 2016 by Richard Haselgrove Post: "BOINC version 7.2.42" - is that standard across your other machines too? |
|
3245)
Message boards :
GPUs :
Quadro FX3700
(Message 69020)
Posted 17 Apr 2016 by Richard Haselgrove Post: http://storage2.static.itmages.com/i/16/0417/h_1460889304_1442221_4743d72e9d.png Yes |
|
3246)
Message boards :
GPUs :
Quadro FX3700
(Message 69016)
Posted 17 Apr 2016 by Richard Haselgrove Post: App_info.xml is very complicated, and it is project specific: you would need a different one for each project you run. But it shouldn't be necessary at all. Why are you looking for one? Nobody here suggested it - if it was mentioned on a project message board, continue the conversation there. cc_config.xml is very simple - all you need is <cc_config>
<options>
<use_all_gpus>1</use_all_gpus>
</options>
</cc_config>
|
|
3247)
Message boards :
GPUs :
Quadro FX3700
(Message 69013)
Posted 17 Apr 2016 by Richard Haselgrove Post: There is no app_info.xml. I searched and is nowhere. app_info.xml is an optional file you can create yourself to explain to BOINC how to run your application if the project you wish to compute for doesn't have a ready-made application for your operating system. It's unlikely to be relevant here. Agentb was referring you to cc_config.xml, which is also an optional file which you create yourself if you need it. I'm afraid I can't help with Linux drivers. |
|
3248)
Message boards :
Questions and problems :
config.xml possible for reducing false GPUGRID 7629MB free disk requirement ?
(Message 68989)
Posted 15 Apr 2016 by Richard Haselgrove Post: Unfortunately, I think not. The disk space requirement is set in the workunit meta-data: the current one on this machine is <workunit>
<name>e24s5_e17s9p1f222-GERARD_A2AR_OSTRUC_NUL2-0-2-RND4822</name>
<app_name>acemdlong</app_name>
<version_num>848</version_num>
<rsc_fpops_est>5000000000000000.000000</rsc_fpops_est>
<rsc_fpops_bound>250000000000000000.000000</rsc_fpops_bound>
<rsc_memory_bound>300000000.000000</rsc_memory_bound>
<rsc_disk_bound>8000000000.000000</rsc_disk_bound>
<file_ref>
<file_name>e24s5_e17s9p1f222-GERARD_A2AR_OSTRUC_NUL2-0-LICENSE</file_name>
<open_name>LICENSE</open_name>
</file_ref>
... many more files
</workunit>The value you would want to over-ride is <rsc_disk_bound> 8,000,000,000 8 (decimal) gigabytes. Per task. If disk usage was associated with the application, you could re-define it in an app_info.xml: but because it's data, it's correctly assigned to the researcher to configure. |
|
3249)
Message boards :
Questions and problems :
[Discussion] 4th Generation BOINC credit system
(Message 68958)
Posted 12 Apr 2016 by Richard Haselgrove Post: The equivalent reference document hosted locally is http://boinc.berkeley.edu/trac/wiki/CreditNew. Because that's held as a Wiki, you can look back through the history, timeline, and (sole) authorship of how it evolved. |
|
3250)
Message boards :
Questions and problems :
Multibrand GPUs BOINC Manager confused
(Message 68840)
Posted 6 Apr 2016 by Richard Haselgrove Post: 06/04/2016 11:04:53 | Collatz Conjecture | Reporting 34 completed tasks BOINC won't request new tasks from a project when too many uploads are stalled (as they were by "File uploads are temporarily disabled."). That's deliberate. 06/04/2016 11:04:58 | Collatz Conjecture | Scheduler request completed: got 34 new tasks Exactly the same number. As with 06/04/2016 11:10:04 | Collatz Conjecture | Reporting 13 completed tasks Strong (but circumstantial) evidence of the "limit on the total number of tasks in progress" that Juha suggested. |
|
3251)
Message boards :
BOINC client :
Suggestion - add GPU task's CPU usage into calculations
(Message 68805)
Posted 5 Apr 2016 by Richard Haselgrove Post: BOINC already has such a calculation. It will over-commit the CPU by, strictly, "less than one CPU-corer". That, in most cases, allows applications that run mostly on GPUs to coexist and make progress alongside applications that run mostly on CPUs. It's a fallacy that a scientific application ever runs exclusively on one CPU, to the exclusion of everything else. Last time I bothered to count, there were something like 100 programs running across the four CPU cores of the Windows 7 machine I was looking at - thanks to the efficient implementation of time-division multiplexing in modern operating systems. It is true that BOINC's automatic algorithm for working out how many fractions of a CPU a GPU application is going to need for its overheads on any particular CPU/GPU hardware combination is pretty rudimentary - but it tends to err on the side of generosity: it sets a higher 'fractional CPU' number than most GPU apps need. That, in turn, brings closer the point at which all the fractions add up to 1.0 or more, and at that point a CPU task is removed from the scheduling, just as you're requesting. And please remember that the scientists who develop the applications we run are primarily interested in finding gravitational waves, binary pulsars, or whatever. You may feel that you don't have enough time to test all combinations on just one machine - a computer you presumably know in some detail: the astrophysicists at Einstein certainly don't have time to test all the combinations which volunteers have brought to their project. |
|
3252)
Message boards :
BOINC client :
Suggestion - add GPU task's CPU usage into calculations
(Message 68802)
Posted 5 Apr 2016 by Richard Haselgrove Post: Thanks, now I'm getting somewhere. But it needs the application name. I thought I could just tell it to give Einstein a certain amount of CPU per GPU task, but surely Einstein is constantly using different applications, so I'd have to manually check up and edit this file all the time. You can put in multiple entries, one for each application you choose run run. And then just leave them alone. Can I put "all" or "*" or something in the application name? No. |
|
3253)
Message boards :
BOINC client :
Suggestion - add GPU task's CPU usage into calculations
(Message 68798)
Posted 5 Apr 2016 by Richard Haselgrove Post: [ ... ] is the convention used in computer documentation to indicate an optional element. If you need to use it, remove the square brackets and supply an appropriate value. If you don't need to use it, remove the brackets and all contents between the opening bracket and the matching closing bracket. |
|
3254)
Message boards :
Questions and problems :
Need to reset debt
(Message 68787)
Posted 4 Apr 2016 by Richard Haselgrove Post: Um. I'm not sure that's true. Checking. Consider: (1) If the server calculates it, it doesn't tell the client. Search "rec" (8 hits in 1 file) D:\BOINCdata\sched_reply_setiathome.berkeley.edu.xml (8 hits) Line 24: <graph_style>Rectangles</graph_style> Line 50: <graph_style>Rectangles</graph_style> Line 77: <graph_style>Rectangles</graph_style> Line 128: <description>View your recent tasks</description> Line 156: <name>arecibo_181.png</name> Line 157: <url>http://boinc2.ssl.berkeley.edu/sg_images/arecibo_181.png</url> Line 189: <file_name>arecibo_181.png</file_name> Line 201: <file_name>arecibo_181.png</file_name> (2) But it does look as if the client updates it locally: https://github.com/BOINC/boinc/blob/71832f73dce34d6a0bc556acabe65bdee026d852/client/cpu_sched.cpp#L557 |
|
3255)
Message boards :
Questions and problems :
Not getting 100% CPU usage
(Message 68759)
Posted 4 Apr 2016 by Richard Haselgrove Post: Multi-threaded tasks like Milkway's very rarely approach 100%, because there's usually one or more threads waiting to synchronise intermediate data with the others. I'd leave that one out of the hunt for now. But the others should be OK, and there are no obvious problems with your prefs files. The new Einstein CPU search is very sensitive to CPU cache size - see Gravitational Wave search O1AS20-100 F and I FAQ - your Intel Q8400 might have been placed in the wrong category, and with four tasks running might be starved of cache memory. |
|
3256)
Message boards :
BOINC client :
How to configure resource share via command line
(Message 68752)
Posted 4 Apr 2016 by Richard Haselgrove Post: "I don't want the client to use 100% of CPU time" isn't actually a Resource Share question - it's in the Global Preferences set. You can set that through the website of any one of the projects the machine is attached to (and you have a choice of four 'venues' - default, home, school, work - if you don't want it to apply to all your computers). Or you can set local preferences just for that machine using a global_prefs_override.xml file - the current documentation describes doing this from the Manager GUI, but you could create the file in a text editor instead. Not directly from boinccmd, though. |
|
3257)
Message boards :
Questions and problems :
Why has BOINC no benchmark for GPUs?
(Message 68732)
Posted 2 Apr 2016 by Richard Haselgrove Post: I think it's SP. Now, if you can upgrade you graphic card without restarting BOINC, I'd like to know how. Otherwise, look at the Event Log after the next restart. |
|
3258)
Message boards :
Questions and problems :
Why has BOINC no benchmark for GPUs?
(Message 68727)
Posted 2 Apr 2016 by Richard Haselgrove Post: In BOINC's Event Log at startup. This machine says 28-Mar-2016 09:23:02 [---] CUDA: NVIDIA GPU 0: GeForce GTX 970 (driver version 350.12, CUDA version 7.0, compute capability 5.2, 4096MB, 3903MB available, 4087 GFLOPS peak) Note that I usually refer to these as theoretical or 'Marketing' FLOPS - they are never achieved in practice by any practical scientific computing program. |
|
3259)
Message boards :
Questions and problems :
Why has BOINC no benchmark for GPUs?
(Message 68722)
Posted 2 Apr 2016 by Richard Haselgrove Post: The relative speed of a GPU is calculated from the card geometry when the device is detected and queried by BOINC at startup. That "GFLOPS peak" value is used instead of a benchmark. |
|
3260)
Message boards :
Questions and problems :
used to work with 2 GPU until today for some reason
(Message 68721)
Posted 2 Apr 2016 by Richard Haselgrove Post: No, I don't - I'm not a fan of the 'sledgehammer and two short planks' school of computer maintenance. You mentioned you had "issues", and something odd obviously happened to that machine before the last restart: 02/04/2016 09:46:08 | | Using state file client_state_next.xml There was no sign of any GPUgrid task (re-)starting in the log you posted - is that new? There were plenty of 02/04/2016 09:48:35 | Collatz Conjecture | Starting task collatz_sieve_2612284211700204306432_6597069766656_0 in the log - are you saying that all Collatz tasks are now showing 'Ready to start'? If you are familiar with BOINC's Event Log Options (Ctrl+Shift+F), it might be worth enabling <cpu_sched_debug> for just one cycle - don't leave it running for ever, it generates a lot of output every minute - and posting any obvious clues. |
|
3261)
Message boards :
Questions and problems :
used to work with 2 GPU until today for some reason
(Message 68718)
Posted 2 Apr 2016 by Richard Haselgrove Post: If it's all (currently) to do with Collatz, I think we just have to wait until the system recovers from 02/04/2016 09:46:19 | Collatz Conjecture | Resetting project You discarded a whole lot of tasks then, and you're slowly getting them back, resent in four blocks of five at a time so far. GPUGrid not running is a simple consequence of 02/04/2016 10:06:14 | GPUGRID | Project has no tasks available - that's rather a common situation at the moment, because they're short of researchers and waiting for the new intake of PhD students to get up to speed. At this stage, I think I'll call off the bug-hunt until your system settles itself down again. |
|
3262)
Message boards :
Questions and problems :
used to work with 2 GPU until today for some reason
(Message 68715)
Posted 2 Apr 2016 by Richard Haselgrove Post: Oh crumbs. Well, I did ask for some details, didn't I? :-) Can we refine that down a bit, please? 1) Which project(s) has tasks running on the GPU which is working? 2) Which project(s) used to run on the GPU which isn't working any more? 3) Has any project ever run tasks on the intel_gpu - if so, which one(s)? In respect of question (3): not many projects have applications for Intel GPUs. The only one I recognise from your list is Einstein, and they are struggling with many incompatibilities between their application and Intel's GPU drivers - the problems are mostly with newer drivers, rather than old ones. They are advising you to read their server log: 02/04/2016 09:46:55 | Einstein@Home | Sending scheduler request: To fetch work. The relevant section of that log is: 2016-04-02 09:23:12.2943 [PID=4616 ] [version] Checking plan class 'opencl-intel_gpu-Beta' To get Intel GPU work from Einstein currently, you must either allow Beta test applications, or use a driver no newer than 10.18.10.3906 - probably not appropriate for your particular GPU (that driver was for HD 4000, HD 4600 etc.). We're working on it.... Einstein Technical News. It might be worth enabling test applications in your project preferences at Einstein, and reporting your results in that thread - but be prepared for validation problems. |
|
3263)
Message boards :
Questions and problems :
used to work with 2 GPU until today for some reason
(Message 68712)
Posted 2 Apr 2016 by Richard Haselgrove Post: Can you give some details of your project attachments, please, and any other configuration changes you've made from default. I have a bug report open with David Anderson for a problem like that which only appears when you have a project exclusion in place for one of the GPUs. But unfortunately, he can't reproduce it and hasn't got time to explore it. Reference: simulator scenario 141 |
|
3264)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 68696)
Posted 1 Apr 2016 by Richard Haselgrove Post: Nice! I hope they're supplying both VHS and Betamax. |
|
3265)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 68634)
Posted 30 Mar 2016 by Richard Haselgrove Post: Isaac is the BOINC server, the one we're using for this message board. Or, I suspect Jord will pop in to say, Isaac is the name that has been given to a succession of BOINC servers down the years. |
|
3266)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 68622)
Posted 30 Mar 2016 by Richard Haselgrove Post: About 10 minutes after the lab opens. Everything's unreachable - web server, scheduler, upload server. Those are usually the easiest ones to fix - maybe not literally plugging one wire back in, but to a tecchie something equivalent. And with SETI, we usually see that everything has been running just fine behind the broken wire, as soon as we can see it. Unless anyone's heard anything about the San Andreas fault? |
|
3267)
Message boards :
GPUs :
Is there a simple way to disable/enable a GPU?
(Message 68603)
Posted 29 Mar 2016 by Richard Haselgrove Post: He'd need to do a boinccmd --quit with some extra command line switches on that final restart. BOINC can't enable/disable GPUs on the fly, it has to be a stop/start. |
|
3268)
Message boards :
Questions and problems :
Boinc tasks completed in a VM don't receive credit; tasks on the host do
(Message 68597)
Posted 29 Mar 2016 by Richard Haselgrove Post: Pick a sample project that you believe you've seen tasks processing for. Go to the website for that project, log in under the the account name used in the VM BOINC instance, find the task records as seen by the project, and verify there that the tasks are completing successfully and being accepted as valid science. |
|
3269)
Message boards :
Questions and problems :
O1AS20 I tasks - Error while downloading
(Message 68590)
Posted 28 Mar 2016 by Richard Haselgrove Post: Einstein has multiple download servers, in different time zones around the world. Sometimes there are problems or delays in distributing the master copies of files to all the mirror servers in time for the first work requests. ChristianB recently wrote new client code so that, if a file is missing from the first server it tries, it tries one of the others instead. But until that new client has been built and is in widespread use, if the file is missing from the first server you try, it's missing, and treated as an error. You could always try moving closer to the master data source in Hannover, Germany... |
|
3270)
Message boards :
Server programs :
How to get clients information in boinc daemons (after starting the project)
(Message 68584)
Posted 28 Mar 2016 by Richard Haselgrove Post: Some projects use <next_rpc_delay>x</next_rpc_delay> to ensure regular updates. See Project Options |
|
3271)
Message boards :
BOINC Manager :
Feature Request: Date Received
(Message 68501)
Posted 24 Mar 2016 by Richard Haselgrove Post: FIFO (date received) is the 'natural' display order for tasks in BOINC Manager, unless a column sort is applied (but no actual datetime is shown - that would be nice, for an indication of turnround time). My wish is for an 'unsort' (remove sort) facility - it's getting tedious diving into the Windows registry every time my mouse finger slips... |
|
3272)
Message boards :
Projects :
which of the projects has high RAM usage?
(Message 68415)
Posted 20 Mar 2016 by Richard Haselgrove Post: Can anyone please tell me a project which is a real RAM-eater? You could try having a browse through http://wuprop.boinc-af.org/results/ram.py?plateforme=all&tri=13&sort=desc (I've sorted that view on the Windows column, though you haven't actually mentioned yet which OS you're running) |
|
3273)
Message boards :
Questions and problems :
GPU task starved because CPU busy with other tasks
(Message 68412)
Posted 20 Mar 2016 by Richard Haselgrove Post: But also be aware that modern multi-tasking operating systems, running on modern multi-core CPUs, are designed to cope with many more active processes than there are CPUs. I'm typing this on a 4-core i5, with somewhere around 95-98 processes active (including 13 browser tabs). One of the active processes is Process Lasso: another is an Einstein intel_gpu task held at real-time priority. I'm not noticing any problematic side-effects from that combination, although my display screen is being drawn by an NVidia GPU running SETI tasks at a mere 'above normal' priority. So I'd say, from personal observation, that it's worth a try, though I completely accept that users should proceed carefully and thoughtfully, and avoid overloading their own hardware: I wouldn't even thinking about running realtime on a single-core computer, or four realtime tasks at once on this quadcore. |
|
3274)
Message boards :
Questions and problems :
GPU task starved because CPU busy with other tasks
(Message 68410)
Posted 20 Mar 2016 by Richard Haselgrove Post: Either approach is possible. I'm doing much the same using Einstein's intel_gpu application using the HD 4000 / 4600 GPU built into various generations of i5 / i7 processors - that's a slightly different question, but it generalises to the discrete AMD GPU too. If your AMD application runs at realtime priority without any unwanted side effects - neither slowing down the pure CPU applications too much by depriving them of resources, nor making the computer's video display jittery or slow to update - then you can lock each succeeding instance of the application to realtime priority by using a program called Process Lasso - the free edition is more than adequate for this task. The user interface is a little daunting at first, but basically you can - - find a running example of the process in the process list - right-click on the current priority class - choose 'priority class --> Always' from the menu And that's it. The program will look after itself from there on, with occasional updates and reminders for a few seconds at startup that donations would be welcome - as nag screens go, it's very polite and non-intrusive. The alternative approach of 'freeing a CPU core' is often advocated, but from your description it sounds to be more drastic than you need. But if you want: BOINC can be instructed to "Use at most 75% of the CPUs" (i.e. three out of four) through your computing preferences - either through a project web site, or locally through the BOINC Manager Advanced View. That will apply for all the projects you're attached to. You may find that Einstein responds best to the Process Lasso approach, SETI to 'Use at most 75% CPUs', and MilkyWay needs no tweaking at all - or some other combination. If you find that just one of your projects requires a completely free core, you can set that up using the Application configuration tool. That's a little more complicated, and I'd suggest you try the other two approaches first. If you need to use an app_config.xml, it might be best to ask on the message boards specific to the project you want to use it for. |
|
3275)
Message boards :
BOINC client :
How is minimum work buffer in days translated to work uits?
(Message 68328)
Posted 16 Mar 2016 by Richard Haselgrove Post: I'm well below that limit. Right now on m 6 core HT (12 threads) I've got 172 WU's total. I'm not running a GPU yet, my card is a NVIDIA and the V8 software isn’t available for it yet It's been available for Windows for a long time, but maybe not for other operating systems. It's helpful to say what you're running when asking questions. I think the hard limit might be 100 CPU tasks in total, rather than per core. The clue would be an Event Log message "This computer has reached a limit on tasks in progress" in response to work fetch requests. |
|
3276)
Message boards :
Questions and problems :
mising gpu in new BOINC....
(Message 68158)
Posted 6 Mar 2016 by Richard Haselgrove Post: I see no sign of a config line like 05-Mar-2016 22:05:39 [---] Config: use all coprocessors Since the two GPUs aren't the same, you need to configure BOINC to use them both, if that's what you want. Option <use_all_gpus> from Client configuration |
|
3277)
Message boards :
Questions and problems :
http transient error in upload: retval -184
(Message 68156)
Posted 6 Mar 2016 by Richard Haselgrove Post: Try setting <http_debug> logging, instead of/as well as <file_xfer> and <http_xfer>. That echos messages received from the remote server, which can be informative. |
|
3278)
Message boards :
The Lounge :
My in-place Windows installation troubles and more failures
(Message 68121)
Posted 4 Mar 2016 by Richard Haselgrove Post: 7-Zip still doesn't show up here, though. No matter what I do. I find that I have to go through the 'Open with', 'Choose default program', 'Browse', 7zFM.exe, route to get 7z files to open properly, with every new 7-zip install. But that doesn't help you with the 7-Zip fly-out context menu (though I can perhaps dig out the registry values, if that would help). |
|
3279)
Message boards :
BOINC client :
Tasks with "Waiting for memory" status don't free up CPU cores
(Message 68110)
Posted 3 Mar 2016 by Richard Haselgrove Post: ... but I can't seem to find a flag that tells how BOINC takes available memory into account when scheduling tasks. Isn't that specified by each project in the workunit definition? <rsc_memory_bound>33554432.000000</rsc_memory_bound> (32 MB) appears in the SETI multibeam spec, for example. The only Einstein task I have immediately to hand is a BRP6 for intel_gpu, which specifies <rsc_memory_bound>260000000.000000</rsc_memory_bound> (<250 MB) |
|
3280)
Message boards :
GPUs :
Should nVidia GeForce FX5200 4x AGP card be able to crunch BOINC
(Message 68043)
Posted 28 Feb 2016 by Richard Haselgrove Post: Sorry, no. The GeForce 5xxx series dates back to 2003. To be a 'usable' GPU under BOINC, the card must be supported by either the CUDA or OpenCL programming languages: for nVidia cards, that support started with GeForce 8xxx cards around 2007 - some three generations and four years later than yours. The best general source that I know for information like that is https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units |
|
3281)
Message boards :
Questions and problems :
This seems to be a bug. Work fetch reporting "no tasks available" as "not highest priority"
(Message 67981)
Posted 24 Feb 2016 by Richard Haselgrove Post: Too late in my time zone for a full discussion, but my understanding is that work is requested for a resource (CPU or GPU) from the project which is 'highest' priority (in the 'closest below zero' sense) for which work fetch is currently allowed. "currently allowed" covers a multitude of sins: Recent scheduler contact, project requested backoff --> disallowed NNT set by user --> disallowed Project has no apps for resource --> disallowed User deselected resource in project preferences --> disallowed Recent work request received 'no work available' response --> disallowed The last line is the killer, and 'recent' can be up to 24 hours ago. It might be helpful if you could post one single, complete [work_fetch_debug] log cycle from your machine, and we could work through it together, project by project, and work out why BOINC didn't request work until it reached NFS. |
|
3282)
Message boards :
Questions and problems :
This seems to be a bug. Work fetch reporting "no tasks available" as "not highest priority"
(Message 67969)
Posted 22 Feb 2016 by Richard Haselgrove Post: So is there a bug if the project's prio is about 2 to 3 orders of magnitude greater (abs()) than it should be? NFS prio should be down around -2.63 to -0.263 as it's already at 45% of maximum potential RAC while MindModeling reports -1.9 prio and is at 3-7% of potential RAC from days on end of no work. How does the NFS REC get such a high magnitude prio of -263 as it's credit payout is NOT 150 times greater than MindModeling? The expected credit per WU is about 3x greater as is the expected RAC over MindmModeling on this particular machine. Prio is not abs() - it's straight maths. The highest possible priority is zero, everything negative is lower. And -263.383 is very much lower indeed. All of this is done on REC, not RAC - the project's actual payout is ignored, instead it's done on something a lot closer to the academic definition of a cobblestone. If MindModeling rarely has work, the chances of hitting it with a random fetch are very low. So most requests will get a response of 'no work available', and will trigger one of the many backoffs - this one per project, per resource. That appears lower down the <work_fetch_debug> log, and looks like 22/02/2016 19:49:52 | FiND@Home | [work_fetch] share 0.000 no applications (resource backoff: 740.17, inc 600.00) |
|
3283)
Message boards :
Questions and problems :
This seems to be a bug. Work fetch reporting "no tasks available" as "not highest priority"
(Message 67958)
Posted 22 Feb 2016 by Richard Haselgrove Post: The log message "not highest priority" comes from your local client, not from the server. It didn't even ask the server for work - so it won't get any reply relating to work availability. |
|
3284)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 67889)
Posted 19 Feb 2016 by Richard Haselgrove Post: As published on the front page. Main web server crash / migration. |
|
3285)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 67833)
Posted 16 Feb 2016 by Richard Haselgrove Post: Copper, because you don't string fiber on poles. It needs to be laid underground and as level as possible, so the light going through the strands follows the straightest path from A to B. I can illustrate that. These are the old copper voice pairs, crossing the canal on the way up to my village.  And the extra cable at the very top is the fibre they installed to bump our broadband speed (for two cabinets, the whole village) from ~5 Mbits to ~40 MBits. 
|
|
3286)
Message boards :
Questions and problems :
Boinc Manager Username
(Message 67593)
Posted 7 Feb 2016 by Richard Haselgrove Post: If your BOINC Manager is running in Simple View, click on 'Project Commands' (bottom right) and select 'Properties'. User name is on the second line. If you are running in Advanced View, look on the 'Projects' tab, in the column headed 'Account'. There is no single username for BOINC Manager itself, because people can (and regrettably do) use different names at different projects. |
|
3287)
Message boards :
Questions and problems :
Why does v7.4.42 need so many supporting programs?
(Message 67590)
Posted 7 Feb 2016 by Richard Haselgrove Post: The extra packages are perhaps specified for a BOINC server build. You should ask the distro maintainer if there are separate packages for 'client only' or 'client + manager' - and if not, why not. |
|
3288)
Message boards :
Questions and problems :
Resource share apparently does nothing at all
(Message 67538)
Posted 5 Feb 2016 by Richard Haselgrove Post: What's the debt formula? Is it based on RAC or total credit? Credit granted per project is wildly different and not useful in inter-project debt calculations. No, it's based on REC (Recent Estimated Credit) - project inflation is excluded. Estimated Credit That paper suggests a delay rate (half life) for REC of 1 month: in the end, a default of 10 days was chosen, but that's configurable. |
|
3289)
Message boards :
GPUs :
Why is BOINC recognizing all four TitanX GPUs but fails to assign to them randomly?
(Message 67403)
Posted 30 Jan 2016 by Richard Haselgrove Post: Thanks, Richard. I will do that. But the question remains: As there are tons of other WUs waiting around with declared use of 0.01 or 0.04 CPUs, how come the scheduler doesn't move on to one of those to keep the valuable GPU resource fully used? Since my "task switch every N minutes" is set to "60", does it simply reserve the GPUs to a particular set of projects for the current 60mins, and doesn't move beyond that? If so, that would seem wasteful. That's a question which I intend to write up fully, with message logs and screen shots, for the boinc_alpha bug list, when I've recovered from the installer launch. I'm seeing it too. |
|
3290)
Message boards :
GPUs :
Why is BOINC recognizing all four TitanX GPUs but fails to assign to them randomly?
(Message 67399)
Posted 30 Jan 2016 by Richard Haselgrove Post: Look at those SETI Beta tasks which are running (or compare the CPU time with the Elapsed time of tasks which are using the same application and have recently completed) to get a realistic measure of how much CPU is actually needed while the task is running. If the SETI Beta tasks are the 'CUDA' type, I expect you'll find that they actually use far less than 47.3% of a CPU - though probably more than the 4% that I've defined for you in app_info.xml Use app_config.xml to define a value for <cpu_usage> which is closer to reality than BOINC's notoriously generous stock estimate. You could do the same for POEM, but I don't have a feel for what that figure would be. |
|
3291)
Message boards :
Questions and problems :
Estimated Computation Size Capped?
(Message 67286)
Posted 26 Jan 2016 by Richard Haselgrove Post: Some project administrators type it in manually, some have a complex formula which derives a value from some characteristic of the task data, and some think of a number and stick with it. |
|
3292)
Message boards :
Questions and problems :
Unable to run BOINC.
(Message 67086)
Posted 22 Jan 2016 by Richard Haselgrove Post: Try the FAQ that Juha linked earlier in this thread. You'll probably want the solution listed as "Step 2". |
|
3293)
Message boards :
Questions and problems :
Win10 - Boinc 7.6.22 - no more tray icon
(Message 67078)
Posted 22 Jan 2016 by Richard Haselgrove Post: Despite the confusion over names, boinctray.exe doesn't, and never has had, a visible icon. The icon you see in the system tray (old terminology) or notification area (new terminology) - or don't see, as the case may be - belongs to BOINC Manager, or boincmgr.exe |
|
3294)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 67048)
Posted 21 Jan 2016 by Richard Haselgrove Post: Or maybe a Platypus - that would find it easier to hide under a settee, wouldn't it? Sea turtles get a bit lumpy when they grow up. |
|
3295)
Message boards :
GPUs :
Why is BOINC recognizing all four TitanX GPUs but fails to assign to them randomly?
(Message 67013)
Posted 20 Jan 2016 by Richard Haselgrove Post: <cpu_sched_debug> covers the whole scheduler process, including both CPUs and GPUs - it simply hasn't been renamed since GPUs were added to the mix. Are you using any app_config.xml files, especially to limit the maximum number of concurrent tasks for any application or project? I think I'm finding a reportable bug in that area, but it's not confirmed yet. |
|
3296)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 66986)
Posted 20 Jan 2016 by Richard Haselgrove Post: People seem unable to read, write and speak their mothertongue these days (ANY mothertongue). Or we'll bang you up in the nearest police station. http://www.bbc.co.uk/news/uk-england-lancashire-35354061 A 10-year-old Muslim boy who mistakenly wrote that he lived in a "terrorist house" during an English lesson at school has been investigated by police. |
|
3297)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 66971)
Posted 19 Jan 2016 by Richard Haselgrove Post: Less to back up and compress when the database is only half full. |
|
3298)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 66969)
Posted 19 Jan 2016 by Richard Haselgrove Post: But no new applications. |
|
3299)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 66958)
Posted 19 Jan 2016 by Richard Haselgrove Post: Anyone need  You can get nasty chafing from twiddling your thumbs all this time... |
|
3300)
Message boards :
Questions and problems :
No Project will download. May have some Windows 10 issues?
(Message 66918)
Posted 17 Jan 2016 by Richard Haselgrove Post: And, although you don't mention installing any anti-virus product on your new machine, check whether you have one, and whether it might be blocking the downloads. |
|
3301)
Message boards :
Questions and problems :
Building an energy-efficient BOINC Computer / Cluster
(Message 66822)
Posted 13 Jan 2016 by Richard Haselgrove Post: It would be a good idea for anyone seeking energy efficiency to consider all those little peripheral extras as well. I've just replaced a (probably 10 year old - maybe more) 16 port 'Fast' (10/100) ethernet switch with an equivalent Gigabit version. The ambient noise level in my crunchery dropped dramatically - the fan in that old switch was louder than anything else. And the new one working at 10 times the speed has a maximum stated power draw of 0.2A, against 0.5A for the old one. I'll probably have a panic attack because of the silence, when I go downstairs tomorrow morning and forget what I've done! |
|
3302)
Message boards :
Questions and problems :
BoincMgr 7.6.22 crashes on start
(Message 66818)
Posted 13 Jan 2016 by Richard Haselgrove Post: I try to keep mine in default format - I'll dump my registry, and try to compare with yours, when I get a moment (bit busy with other testing just now) |
|
3303)
Message boards :
Questions and problems :
BoincMgr 7.6.22 crashes on start
(Message 66735)
Posted 9 Jan 2016 by Richard Haselgrove Post: I'm a little bit worried by that "- BOINC runs from a startup script under the 'Startup' folder of the start menu." That's very non-standard for current versions of BOINC: the 'official' way is for BOINC Manager to launch from the Registry at Windows startup (automatically but silently), snoop around to see if it's supposed to be running at logon, and go away again if it isn't. How did you deploy the new version? Did you allow the Windows installer to do the whole job, or did you simply copy the executable files over the top of your previous, slightly unusual, installation? If you copied the files, that might be your problem. Quite a lot of the supporting technology has been updated between those two versions, and at the very least you'd need to copy the full set of DLLs as well - they may need to be registered in the Windows registry too. (the installer would do that) I'm typing this on a 64-bit Windows 7 computer, with both BOINC and the Manager running - so it can be done. |
|
3304)
Message boards :
Questions and problems :
Unable to run BOINC.
(Message 66632)
Posted 6 Jan 2016 by Richard Haselgrove Post: Yes, Richard, progress at last. It is crunching again, 7.6.6, and suprisingly, all of the wu's that were there are still good and not past their return dates, so perhaps I'll set no new tasks and let them all finish before I start trying to get up to date. All the problems started when I tried to install 7.6.9. Thanks mate. Yes. "Uninstall" meaning remove programs, but leave data untouched, is by design - just as you wouldn't expect updating your word processor to delete your half-written novel. Glad you're back in familiar territory. |
|
3305)
Message boards :
Questions and problems :
Unable to run BOINC.
(Message 66630)
Posted 6 Jan 2016 by Richard Haselgrove Post: That's the classic sign of a corrupted MSI cache. What it means is that it can't find the *previous* boinc.msi file, for a version which it thinks it has to *uninstall* as a part of the (automatic) preparation for installing any different version. You might be able to recover by re-installing exactly the same BOINC version as was most recently running on the machine: Windows (MSI) should detect that, and offer to repair the current installation, instead of going through the uninstall/reinstall cycle. If it does, that should get you past the missing file problem, and you should be able to uninstall fully and install the version you really wanted in the first place. If that doesn't work, the procedures in the FAQ Juha linked are definitely the ones to follow. |
|
3306)
Message boards :
Questions and problems :
Building an energy-efficient BOINC Computer / Cluster
(Message 66615)
Posted 5 Jan 2016 by Richard Haselgrove Post: I calculated that the maximum Wattage my build should have is 100-150W. Could you recommend a BOINC-specialized build (possibly including GPU) for that? Or is such a low-power setup not viable? I'm currently running a Haswell (fourth-generation) Intel i5 based computer, which is drawing under 100 watts from the wall socket (currently fluctuating between 90-95W, varies according to project). That's with all four cores, and the integrated HD 4600 intel GPU, all crunching for BOINC. Power draw excludes monitor, but you'd have that switched off most of the time anyway. The machine is actually an off-the-shelf Dell Optiplex 9020 with mechanical hard drive - power consumption could possibly be reduced even more by using SSD storage, though at added purchase cost. The HD 4600 alone is earning over 10,000 RAC at Einstein - host 8864187. Hope that gives you an idea of what's possible. |
|
3307)
Message boards :
Questions and problems :
Version 8 for Linux Mint?
(Message 66590)
Posted 2 Jan 2016 by Richard Haselgrove Post: This is a SETI-only change, not directly related to BOINC. Since the message refers to an app_info.xml file, you (or your friends) must have installed the previous version by hand. Your best bet is to go back to the friends who helped you install the current version, and refer them to my thread SETI@Home Version 8 - app_info.xml files for upgrading Lunatics and anonymous platform (all OS) SETI is very much in a state of flux at the moment, so it may be better to wait a few days until the new versions become more widely available. Keep an eye on the SETI Message boards (though beware of a likely outbreak of technobabble). |
|
3308)
Message boards :
Questions and problems :
Totally frustrated by the work fetch algorithm that could automate fetching single, temp WU when high priority projects have no work.
(Message 66583)
Posted 2 Jan 2016 by Richard Haselgrove Post: (you'll likely have to create the file) The BOINC v7.6.xx range (you'll be wanting to use v7.6.22 now) have a handy 'Event log options...' picker under the Options menu, or Ctrl-Shift-F. If you change something there, BOINC will write out a complete cc_config.xml file, nicely formatted, with everything in alphabetical order, and pre-populated with default values. It's much easier, and safer from typos, to start from that. There's also a link in the picker direct to Client configuration, where the rest of the documentation lives. |
|
3309)
Message boards :
Questions and problems :
Stuck on "Downloading work from server."
(Message 66332)
Posted 21 Dec 2015 by Richard Haselgrove Post: And if it's not a .png file, go look on the transfers tab (or in the Event Log) and see what file is stuck. If it's an executable file, chances are that it's being 'shielded' by your anti-virus product. [on the simplistic assumption that you're running Windows - you didn't say] |
|
3310)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 66287)
Posted 19 Dec 2015 by Richard Haselgrove Post: I doubt it. Four minutes is one degree of longitude, and one degree is about fifty miles at these latitudes. I'm sure London to Greenwich was a lot less than that when I walked it in a morning, a couple of years ago. Maybe there's something about GMT and UTC that we haven't been told? |
|
3311)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 66284)
Posted 19 Dec 2015 by Richard Haselgrove Post: Also, on the first day of astronomical winter, the sun is at its lowest point in our sky at 12:00 noon and, of course, it is the shortest day of the year. Which is neither the day with the darkest morning (latest sunrise), nor the day with the darkest evening (earliest sunset). The U S Naval Observatory has a handy little gadget which produces a table of sunrise and sunset times for anywhere you like. The darkest morning here is either 29 or 30 December, when sunrise is at 08:26 (it was 08:22 today) - they don't print the seconds, so you can't be exact. Similarly, the evenings are already getting lighter, with sunset at 15:47 today - it's been listed as 15:46 for the last 10 days. The only problem is that I live between two hills, so the times are completely wrong for actual observations..... |
|
3312)
Message boards :
Questions and problems :
Question about web pages reporting invalid credit value "---"
(Message 66209)
Posted 17 Dec 2015 by Richard Haselgrove Post: Back. Personally, I think it's better to distinguish between 'disqualified from credit - nul points' and 'valid, got credit, but only a tiny amount'. I do have one project which awards credit even too low to display as 0.01 - Numberfields, three tasks today alone (pages may need scrolling to see zero credit). |
|
3313)
Message boards :
Questions and problems :
Question about web pages reporting invalid credit value "---"
(Message 66207)
Posted 17 Dec 2015 by Richard Haselgrove Post: Cancel my previous attempt at coding - if "---" is accepted, let's be explicit about it. switch($result->outcome) {
case RESULT_OUTCOME_SUCCESS:
switch ($result->validate_state) {
case VALIDATE_STATE_INIT:
case VALIDATE_STATE_INCONCLUSIVE:
return tra("pending");
case VALIDATE_STATE_INVALID:
case VALIDATE_STATE_NO_CHECK:
return "---";
}
return $string_to_show;
default:
if ($result->granted_credit > 0) {
return $string_to_show;
}
return "---";
}Edit - but then manual credit - "Admin's Benificence" - wouldn't show up. Bother. Shouldn't be trying to solve problems and get ready to go out, at the same time. I'll leave you in peace for a few hours. |
|
3314)
Message boards :
Questions and problems :
Question about web pages reporting invalid credit value "---"
(Message 66205)
Posted 17 Dec 2015 by Richard Haselgrove Post: scientifically speaking, there are the additional options of: We have "pending" for the non-final cases. I'd prefer to reserve 0.00 for "credit was granted, but it was between 0.000001 and 0.004999" |
|
3315)
Message boards :
Questions and problems :
Question about web pages reporting invalid credit value "---"
(Message 66202)
Posted 17 Dec 2015 by Richard Haselgrove Post: ... but there was nothing about that in their news. It was covered in Technical News - Backend upgrade scheduled for Wed Nov 18. |
|
3316)
Message boards :
Questions and problems :
Question about web pages reporting invalid credit value "---"
(Message 66201)
Posted 17 Dec 2015 by Richard Haselgrove Post: What about... As just stated, my personal choice would be the opposite - "---" implies something out of the ordinary, worth looking at. I'd prefer switch($result->outcome) {
case RESULT_OUTCOME_SUCCESS:
switch ($result->validate_state) {
case VALIDATE_STATE_INIT:
case VALIDATE_STATE_INCONCLUSIVE:
return tra("pending");
}
if ($result->granted_credit > 0) {
return $string_to_show;
}
return "---";
default:
if ($result->granted_credit > 0) {
return $string_to_show;
}
return "---";
}
|
|
3317)
Message boards :
Questions and problems :
Question about web pages reporting invalid credit value "---"
(Message 66199)
Posted 17 Dec 2015 by Richard Haselgrove Post: Trouble with the Einstein back-end is its age. While they may have updated the database a little to cover for glaring mistakes, the back-end code filling out the values on the website hasn't been updated since June 2010. Actually, the server code - certainly the web server code, which is what we're talking about here - has been fully updated within the last month. That's how I was able to relate the report to current Github code, and why the same thing happens at MilkyWay. It's also probably why the issue has only just been noticed at Einstein. Personally, I think I'd prefer to see the "---" symbol for all 'no credit' outcomes - it stands out more clearly from 0.01 or 0.09 validations. But that may just be the result of long habituation. It certainly looked odd recently to see (in my localisation) a solid wodge of 8-character times in BOINC Manager, including unstarted tasks - 00:00:00 didn't stand out clearly. |
|
3318)
Message boards :
Questions and problems :
Question about web pages reporting invalid credit value "---"
(Message 66192)
Posted 17 Dec 2015 by Richard Haselgrove Post: In theory, "---" is just a clearer way of representing the value 0.00 (as slipped briefly, but was corrected, in recent alpha builds of the BOINC Manager time displays). If the credit currently awarded, for whatever reason, is zero, the display should be "---". That should be consistent, but a couple of cases were missed. If the credit changes in the future - it changes. Even if the project admins assign credit manually out of the goodness of their hearts. |
|
3319)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 66188)
Posted 16 Dec 2015 by Richard Haselgrove Post: I think the Daily Express is stuck in a timewarp. Around 1947. |
|
3320)
Message boards :
Questions and problems :
Question about web pages reporting invalid credit value "---"
(Message 66186)
Posted 16 Dec 2015 by Richard Haselgrove Post: I think the result database always stores a numeric value for $result->granted_credit. For web display purposes, zero credit is (sometimes) converted to "---" by the function 'result_granted_credit_string' at line 92 of https://github.com/BOINC/boinc/blob/master/html/inc/result.inc There's no conversion from 0 to "---" in the section switch ($result->validate_state) {
case VALIDATE_STATE_INIT:
case VALIDATE_STATE_INCONCLUSIVE:
return tra("pending");
}
return $string_to_show;which means that the cases define('VALIDATE_STATE_INVALID', 2);
define('VALIDATE_STATE_NO_CHECK', 3);aren't converted. The separate case of define('RESULT_OUTCOME_VALIDATE_ERROR', 6);not being outcome 'SUCCESS', gets converted by the default handler at line 105. Or something like that. Unfortunately, I don't currently have any invalid results at any other project to compare their handling of the problem. |
|
3321)
Message boards :
Questions and problems :
Boinc client run all program , why?
(Message 66171)
Posted 16 Dec 2015 by Richard Haselgrove Post: I think we are being held back from advising you fully because of the language barrier. It might be better for you to ask for help in your native language. The SZTAKI Institute in Budapest has a lot of experience in hosting BOINC servers and running BOINC projects: it hosted the most recent BOINC Workshop last year, and apart from Prof. Peter Kacsuk, Jozsef Kovacs, Robert Lovas, Attila Marosii, and Adam Visegradi also contributed. I'm sure that somebody from that list will be able to help you more effectively than we can, or at least put you in touch with somebody who can. |
|
3322)
Message boards :
Questions and problems :
Boinc client run all program , why?
(Message 66138)
Posted 15 Dec 2015 by Richard Haselgrove Post: Is the rendering app multi-threaded (uses all cores on a multi-core CPU)? If so, it should be deployed on the server with an 'MT' plan_class See https://boinc.berkeley.edu/trac/wiki/AppPlanSpec, where the keyword is still 'multicore' instead of 'multithread'. mt is available as a Predefined plan class |
|
3323)
Message boards :
Questions and problems :
Runing issue.
(Message 65916)
Posted 9 Dec 2015 by Richard Haselgrove Post: You would normally have a separate user ID number for each project - for instance, your user ID for this message board is 38312 (but you won't see any tasks here - you'd need to visit each project web site under the ID number appropriate for that site to see the work which has been allocated to you). |
|
3324)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 65887)
Posted 8 Dec 2015 by Richard Haselgrove Post: Going down this early must mean something but the question is what? It means that Jeff is doing the maintenance this week. Jeff has also been working through the last two weekends preparing for Green Bank Telescope data, so there might be more components from that to be installed at Beta. |
|
3325)
Message boards :
GPUs :
Upgrading vido card
(Message 65868)
Posted 7 Dec 2015 by Richard Haselgrove Post: Much more likely to be because the 550Ti is Fermi technology and the 970 is Maxwell technology - two generations later. I don't think MilkyWay has separate applications optimised to cover that difference. |
|
3326)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 65744)
Posted 3 Dec 2015 by Richard Haselgrove Post: If you manage that, can I quietly sidle in beside you, that would be nice. I didn't know you two were baseball fans? |
|
3327)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 65709)
Posted 1 Dec 2015 by Richard Haselgrove Post: My understanding is that GB just showed up at Beta within the past week. I would presume it would need at least a few weeks in Beta to prove itself reliable before coming to Main. Less than a week in beta for something brand new is unlikely...and risky/foolish. And no test workunits until - possibly - tonight. Even then, they won't be the final version, according to email received from Eric during the outrage. |
|
3328)
Message boards :
Questions and problems :
Does Screensaver do any computing (Mac OSX)
(Message 65685)
Posted 1 Dec 2015 by Richard Haselgrove Post: Not quite. BOINC is actually in four parts: BOINC Client BOINC Manager Screensaver Project science applications Of those, the actual computational work is done by the first and the last - the Client and the science applications. They work hand-in-hand - the Client controls the science applications, downloads new work for them, and returns the completed work to the project that needs it. The Manager purely exists so you can set (and if you wish, monitor) the way BOINC works on your computer. You don't need it to be running all the time - but if you close it down, make sure you don't close the Client as well (unless you mean to). There's an option for that. The Screensaver is purely that - decorative effects on screen. It doesn't do any computing - though the good ones will reflect the actual work being done under the hood. All screensavers will suck some computing power from your machine - what's used for display can't be used for science. In extreme cases (depending on the relative strength of your computer and its graphics card), a screensaver may consume such a high proportion of the computer's resources that it significantly slows down the real computing. |
|
3329)
Message boards :
Questions and problems :
this website not secure
(Message 65677)
Posted 30 Nov 2015 by Richard Haselgrove Post: Christian Beer has committed generic code Web: use https urls everywhere, triggered by a similar report of head_20.png (but not other image urls) being served insecurely at Einstein. |
|
3330)
Message boards :
Questions and problems :
Can't recive source code
(Message 65675)
Posted 30 Nov 2015 by Richard Haselgrove Post: https://boinc.berkeley.edu/trac/wiki/SourceCodeGit |
|
3331)
Message boards :
GPUs :
Disabling CPU work without disabling GPU work?
(Message 65644)
Posted 26 Nov 2015 by Richard Haselgrove Post:
? |
|
3332)
Message boards :
GPUs :
Disabling CPU work without disabling GPU work?
(Message 65640)
Posted 26 Nov 2015 by Richard Haselgrove Post: But why can't we get to chose for ourselves? There's a huge difference between running a CPU app full-out (full use of FPU, SIMD registers and all), and the tiny amount of light-weight work that a GPU app will place on it (less than 1% by time, and even less in terms of real work, for a well-chosen GPU app). |
|
3333)
Message boards :
GPUs :
Disabling CPU work without disabling GPU work?
(Message 65619)
Posted 25 Nov 2015 by Richard Haselgrove Post: That will prevent you downloading new work for the CPU, which is ideal if you want to keep the CPU permanently clear for rendering. It won't be ideal if you're happy to let the CPU run BOINC projects when you're not working - sadly, you would have to suspend individual tasks or projects to achieve that, because the BOINC developers haven't ever recognised that some people - like you - might be prepared to donate a small amount of CPU time to support a GPU, but not to donate a full-time CPU: they took an absolute 'all or nothing' view. |
|
3334)
Message boards :
GPUs :
GPU Missing
(Message 65507)
Posted 19 Nov 2015 by Richard Haselgrove Post: No sign of any OpenCL fixes in the release notes, but let's hope it does the trick. http://us.download.nvidia.com/Windows/359.00/359.00-win10-win8-win7-winvista-desktop-release-notes.pdf |
|
3335)
Message boards :
GPUs :
GPU Missing
(Message 65505)
Posted 19 Nov 2015 by Richard Haselgrove Post: Driver version is: 358.91 - and is the latest. The other people encountering the same problem at Milkyway had the same combination. If that's a part of "The big update" (who said they weren't issuing service packs any more - that one was 4 GB, I think I read), then we're going to get a lot of 'Needs OpenCL, hasn't got' reports. |
|
3336)
Message boards :
Questions and problems :
Some problems with domain version of the boinc-client 5.10.45
(Message 65503)
Posted 19 Nov 2015 by Richard Haselgrove Post: Found the reference - #629 |
|
3337)
Message boards :
GPUs :
GPU Missing
(Message 65500)
Posted 19 Nov 2015 by Richard Haselgrove Post: I am wondering if anybody else is having this issue. There seems to be a lot of it about, especially at MilkyWay. See http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=3853 |
|
3338)
Message boards :
Questions and problems :
Some problems with domain version of the boinc-client 5.10.45
(Message 65496)
Posted 19 Nov 2015 by Richard Haselgrove Post: Confirmed both hypotheses - not with a DC, but with two Windows 7 machines, one remotely monitoring the other. Locking the console allows BOINC to continue, logging off kills it. Windows Servers 2003 onwards have a very aggressive 'lock' policy, perhaps precisely because administrators are so lazy about logging off when they should. Maybe some work will get done after all, but it's a crappy solution: servers and service mode were made for each other. |
|
3339)
Message boards :
Questions and problems :
BOINC 7.6.9 downloads but won't launch in order to install
(Message 65494)
Posted 19 Nov 2015 by Richard Haselgrove Post: I've seen (and reported) a marked increase in Task exited with zero status but no 'finished' file messages with v7.6.15 on a machine which is rather over-stressed, plus a few extra Exit status 194 (0xc2) EXIT_ABORTED_BY_CLIENT I've reverted that one to v7.6.9, and I'll check in a few days whether the stats have returned to normal. |
|
3340)
Message boards :
Questions and problems :
Some problems with domain version of the boinc-client 5.10.45
(Message 65492)
Posted 18 Nov 2015 by Richard Haselgrove Post: Which is basically useless. I'd expect a DC to run 95% plus of the time with no user logged in, so why bother installing a BOINC which doesn't run in that scenario? Actually, I suspect it might go on running CPU tasks, provided the exit option "Stop running tasks when exiting the BOINC Manager" is not checked. GPU tasks will stop running, of course, but that's not likely to be a problem on a DC. Volodymyr, could you check that theory, please? And if I'm wrong, maybe 'Locking' the console will allow BOINC to run, even if logging out doesn't - provided your company's security policy allows that. |
|
3341)
Message boards :
Questions and problems :
Some problems with domain version of the boinc-client 5.10.45
(Message 65490)
Posted 18 Nov 2015 by Richard Haselgrove Post: Do check that BOINC continues to work on tasks when no user is logged in to the console - as I presume will normally be the case for a DC. |
|
3342)
Message boards :
Questions and problems :
BOINC 7.6.9 downloads but won't launch in order to install
(Message 65481)
Posted 18 Nov 2015 by Richard Haselgrove Post: Just for fun, I downloaded both v7.6.9 installers and tried them on my (32-bit) XP machine. The 32-bit installer launched fine, but I cancelled it before accepting the licence agreement - I don't want that on here. The 64-bit installer displayed the splash logo while decompressing the msi, but then put this up:  Not the greatest error message in the world, but I think we get the point. |
|
3343)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 65462)
Posted 17 Nov 2015 by Richard Haselgrove Post: Obviously they are not having an outrage today. Their choice I guess. Oh, come on. It's still only 8 am in the lab - they don't usually finish coffee and start work until 9:30. (Except Jeff) |
|
3344)
Message boards :
Questions and problems :
Suspend & Resource Share ??
(Message 65445)
Posted 16 Nov 2015 by Richard Haselgrove Post: They get more CPU cycles. That is, their tasks will be selected to run more frequently, so in the long term the remaining projects will get 33.3333% of the available time each, instead of 25.0000%. But the resource share implementation is 'lumpy' - don't expect CPU cycle allocation to be exact over a timescale of seconds, or even hours. |
|
3345)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 65398)
Posted 12 Nov 2015 by Richard Haselgrove Post: Well it could be GPUZ fibbing, but I'm quite happy to sit back and watch the fun. If and when it tops out we might consider some of this unroll stuff, but we will see. I got mine from Scan Computers in April 2014, but they're now saying they're out of stock: http://www.scan.co.uk/products/2gb-gainward-gtx-750-ti-pcie-30-(x16)-6008mhz-gddr5-gpu-1202mhz-boost-1281mhz-cores-640-hdmi-dvi-d-s But apparently they're still available in Poland... http://www.morele.net/karta-graficzna-gainward-geforce-gtx-750ti-golden-sample-2gb-128bit-dvi-minihdmi-vga-426018336-3071-gtx-750ti-2gb-636767/ |
|
3346)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 65387)
Posted 12 Nov 2015 by Richard Haselgrove Post: Boy, those 750ti's sure seem to be a sweet spot for GPUs. Sounds similar to the Gainward "Golden Sample" I run. They can reach 12,500 RAC on MB only between outages, and don't need supplementary power. |
|
3347)
Message boards :
Projects :
News on Project Outages
(Message 65280)
Posted 5 Nov 2015 by Richard Haselgrove Post: CPDN seems to be making progress. I can access my account info, server status page, forums, etc., and reported a couple of recently completed tasks. However, I don't see any official looking announcement in the forum. There was a moderator briefing today: The work to update the firmware has now completed and the project has now been partially restored. We are aware that a number of project services are still offline and we are working with OeRC support to bring these online as soon as possible. We apologise that the scheduled downtime took longer than expected. |
|
3348)
Message boards :
Questions and problems :
Can't Edit Remote_hosts.cfg On Linux computer to Access Remotely via Boinctasks on Windows Computer
(Message 65261)
Posted 4 Nov 2015 by Richard Haselgrove Post: No - if in doubt, check the documentation: http://boinc.berkeley.edu/wiki/Client_configuration There are separate sections for log flags and options. <cc_config>
<log_flags>
<task>1</task>
<file_xfer>1</file_xfer>
<sched_ops>1</sched_ops>
</log_flags>
<options>
<allow_remote_gui_rpc>1</allow_remote_gui_rpc>
</options>
</cc_config>
|
|
3349)
Message boards :
GPUs :
Want to compute on GPU 760 GTX and on GPU from APU too
(Message 65198)
Posted 2 Nov 2015 by Richard Haselgrove Post: http://www.geeks3d.com/20091230/vga-hack-how-to-make-a-vga-dummy-plug/ |
|
3350)
Message boards :
Questions and problems :
Questions about scheduling, Estimated Credit and Long Term Debt
(Message 65156)
Posted 31 Oct 2015 by Richard Haselgrove Post: That 'Local Scheduling for Volunteer Computing' paper is very old - I can't see an exact publication date, but it has a copyright mark for 2007. All references to 'debt' come from that old scheduling system, which was abandoned in 2010: 'estimated credit' comes from the replacement system, so you will have great difficulty if you try to construct a grand unified theory which encompasses both concepts! I find that the three most useful (and contemporary) pages for trying to understand the current system are http://boinc.berkeley.edu/trac/wiki/CreditNew http://boinc.berkeley.edu/trac/wiki/RuntimeEstimation http://boinc.berkeley.edu/trac/wiki/ClientSchedOctTen |
|
3351)
Message boards :
Projects :
News on Project Outages
(Message 65086)
Posted 26 Oct 2015 by Richard Haselgrove Post: More CPDN progress to report: We have now restarted the majority of services within the CPDN project including all central BOINC services. We are restarting the local (Oxford) upload server shortly in a manner that will allow us to control load into the main infrastructure so those waiting to upload there please be patient. |
|
3352)
Message boards :
GPUs :
WU Completion Time Estimation With GPUs of Different Types
(Message 65052)
Posted 23 Oct 2015 by Richard Haselgrove Post: So far as I know, nobody has found a way of using the existing configuration tools within BOINC to recognise the different capabilities of distinct GPUs within a single project - so you are right to describe this as an enhancement request. David Anderson - the lead developer of BOINC code - has described the decision not to treat each device independently as "something we would have done differently, if we'd known better when we started adding GPU support to BOINC", so it's on his To Do list already - but that list has a habit of moving more slowly than we (or I suspect he) would like. In the meantime, since you mention running tasks 'primarily' from GPUGrid, one possibility is to dedicate just one card full-time to GPUGrid, and the other card to one or more other projects. That can be achieved by using the <exclude_gpu> option in client configuration. |
|
3353)
Message boards :
BOINC Manager :
Boinc not swapping tasks at frequency I requested
(Message 65044)
Posted 22 Oct 2015 by Richard Haselgrove Post: Had to rebuild my Windows 10 OS with latest Insider Build and thus had to create a new set of preferences again. Simply save your entire BOINC data folder - subfolders and all - on something you can preserve: a data partition on your hard drive, a USB memory stick, or whatever you have to hand. After doing whatever you felt the need to do, put the BOINC data folder back where it can be accessed during processing - if you put it on a data partition, it'll be ready to go already. Then run the BOINC installer, and on the 'Advanced' screen tell it where the data folder is. You'll be ready to go - preferences, statistics, applications, work in progress ... the whole kit and caboodle. No waiting, no downloading, no manual faffing about. |
|
3354)
Message boards :
Questions and problems :
I'm having trouble selecting a task in the "Tasks" pane.
(Message 65028)
Posted 21 Oct 2015 by Richard Haselgrove Post: This is how I can reproduce the effect: Both of those reproduce the reported effect for me too, using BOINC v7.6.9 on Windows/64 Note that to reproduce the second effect, Tasks/Name must still be deselected - Tasks/Project deselected by itself doesn't cause the problem. |
|
3355)
Message boards :
Projects :
News on Project Outages
(Message 65012)
Posted 21 Oct 2015 by Richard Haselgrove Post: And Einstein posted this at 12:00 UTC Wednesday 21 October: We need to shut down the project for a few hours to do some urgent database maintenance. Your clients are able to handle that so the interruption should cause no harm. |
|
3356)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64972)
Posted 20 Oct 2015 by Richard Haselgrove Post: If I caught that correctly, they found a bug or two in the forum code causing the replica DB hangs? That was the theory - but since the replica hung again overnight, it may not be the end of the story. |
|
3357)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64969)
Posted 20 Oct 2015 by Richard Haselgrove Post: And down... Well, that saved us going on discussing the difference between a project and a forum. |
|
3358)
Message boards :
Projects :
News on Project Outages
(Message 64953)
Posted 19 Oct 2015 by Richard Haselgrove Post: Update on the CPDN situation, issued 10:00 UTC Monday 19 October 2015 OeRC IT Support has been running a rebuild of the storage array over the weekend. This was initiated on Friday afternoon and is currently at 90%. Once this has been completed he will be running a file check and other checks on the 200TB array to verify the consistency of the rebuilt array, these checks are expected to take the minimum of three quarters of a day. If these are successfully he will restore a single test machine and then will check the consistency of this. If these and all other checks are correct, we in CPDN would be able to start restoring services. |
|
3359)
Message boards :
Questions and problems :
Memory question
(Message 64948)
Posted 19 Oct 2015 by Richard Haselgrove Post: Unfortunately I have no BOINC with a 'Properties' tab... The fields the OP is asking about are shown in the properties dialog, invoked from the command button on the tasks tab. I think you need BOINC v7 to see it, but I don't know exactly where in the development cycle it appeared. |
|
3360)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64933)
Posted 18 Oct 2015 by Richard Haselgrove Post: ... I didn't give it a second thought. And that's exactly your problem, in a nutshell. |
|
3361)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64930)
Posted 18 Oct 2015 by Richard Haselgrove Post: And you got a red-x from me, too. It didn't read like a joke: if you want to make jokes about racism (which I'd advise against), they need to be much better than that. |
|
3362)
Message boards :
Questions and problems :
Invalid tasks, how best to raise an alert.
(Message 64926)
Posted 17 Oct 2015 by Richard Haselgrove Post: This can be solved by having a quota per application (something Einstein e.g. doesn't have yet), so the host stops asking for work for that application it has trouble with, because there the Max tasks per day is 1. For all the applications it doesn't have trouble with the Max is high. AgentB is primarily an Einstein volunteer, so the bracketed exception applies to him. There is also scepticism at projects running the newer forms of the server code as to how effective the 'per application' quota reduction code really is in practice. |
|
3363)
Message boards :
Questions and problems :
Missing MSVCR100.dll
(Message 64885)
Posted 14 Oct 2015 by Richard Haselgrove Post: This may be a red-herring, but browsing through the overnight list of windows security patches on a 32-bit Windows 7 machine, I found I was being offered https://support.microsoft.com/en-us/kb/2999226 The Windows 10 Universal CRT [Universal C Runtime] is a Windows operating system component that enables CRT functionality on the Windows operating system. This update allows Windows desktop applications that depend on the Windows 10 Universal CRT release to run on earlier Windows operating systems. MSVCR100.dll will certainly be a component of the Universal C runtime: this patch may, or may not, correct any problems where a different version of the file may be required because of the way a project has compiled its application programs. |
|
3364)
Message boards :
Questions and problems :
Missed deadline
(Message 64858)
Posted 12 Oct 2015 by Richard Haselgrove Post: If you are still using the old master URL you need to remove and re-add the project. Also, if the tasks you are trying to upload are already removed from the website's list of tasks then just abort the tasks, they are of no use to the project any more. Yes, after 10 consecutive failures - assuming the new scheduler address is properly embedded in the master (home) page html. |
|
3365)
Message boards :
Questions and problems :
How do I know if my GPU is being used?
(Message 64827)
Posted 11 Oct 2015 by Richard Haselgrove Post: Look in the BOINC Manager Event Log (Ctrl-Shift-E), especially the first 20-30 lines after BOINC starts up. If that doesn't answer your questions, copy and post them here, so others can help you analyse them. |
|
3366)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64705)
Posted 5 Oct 2015 by Richard Haselgrove Post: I assume that they would not be allowed to use the Hurricane Electric connection in the CoLo? Therefore the upside of the move had to be balanced against any downside. When the move was first announced I was one who openly said I was 50/50 about the whole thing. I came around after the move when I saw the overall benefits. It is pointless to make any further speculation until we have a viewpoint from Matt, as the expert, and the man who deals with it day by day. On the contrary. I would be prepared to bet that all our bits passed through the Earl Warren building while being tunneled between Palo Alto (HE termination) and the big Cisco router you will have been shown in the SSL rack. Next, the servers were moved to Earl Warren as co-lo tenants, and I'd bet again that the termination router moved there as well. But the bits will still have passed through Campus equipment (such as inr-211, which I still have bookmarked for the Cricket graphs), before being de-tunneled by SETI's counterpart to the Palo Alto router. Then, mid-May, Earl Warren upgraded to 10 gigabit hardware, and our bits - still tunneled - passed through inr-311 instead of inr-211. Finally, in mid-August, the tunnel was dismantled, and all our weird and wonderful IP addresses were exposed directly to the campus routers, probably bloating the routing tables in the process. My guess is that the de-tunneling, coupled with the return of all the students and academic staff for the new semester, is what finally exposed the under-provisioning of inr-311 and related hardware. Or something like that. Edit - in fact, doesn't that thing behind your head in the photo look like  "the recently donated, unconfigured and rather oversized router to ultimately replace the skinnier Hurricane Electric router"? Sourced from Current Closet Status - February 14, 2008 |
|
3367)
Message boards :
Questions and problems :
app_config.xml
(Message 64647)
Posted 3 Oct 2015 by Richard Haselgrove Post: Does the client say it found the app_config.xml file? Juha meant in the Event Log BOINC keeps (and shows from the BOINC Manager - Crtl-Shift-E) on your local machine. If you re-read config files - there's a Manager menu item for that too - the log will re-display the app_config.xml detections (no need to restart BOINC). |
|
3368)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64578)
Posted 1 Oct 2015 by Richard Haselgrove Post: Thanks for the update Richard. One problem is that the Berkeley Campus is, because of it's size, necessarily open and not gated. Therefore they have suffered from cable thefts in the past. Intermittent cable thefts? That's a new one on me. |
|
3369)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64574)
Posted 1 Oct 2015 by Richard Haselgrove Post: Both this BOINC website, and the SETI@Home project, are hosted on the campus of the University of California at Berkeley. For me, network tracert consistently ends up with variations of 19 et3-47.inr-311-ewdc.Berkeley.EDU [128.32.0.103] reports: Destination host unreachable. 19 et3-48.inr-311-ewdc.Berkeley.EDU [128.32.0.101] reports: Destination host unreachable. This suggests an intermittent fault somewhere inside the Berkeley campus network: both the campus network administration team, and our shared BOINC/SETI boffin (Matt Lebofsky) are aware of the problem, but they haven't tracked it down yet. |
|
3370)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64554)
Posted 30 Sep 2015 by Richard Haselgrove Post: edit@Chris: I'm not sure what you mean by day centres? The day centres I know of are not institutionalised - they're voluntary, often funded by charity, and offer all sorts of art, music, excercise, skills based classes. They have almost all been closed, but not by do-gooders with budgets, but by caring people without. Much of the "charitable" provision was created by the vision of volunteers who saw problems close up, in their own communities, and had the energy to do something about it: even the school meals service started that way, here in Bradford. But in recent years, much of the more substantial provision has been funded under contract by public authorities, as an alternative and probably preferable solution to direct, institutional, state provision. But that leads to another of today's headlines: Bradford homeless charity closes today after 37 years helping vulnerable women and children |
|
3371)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64546)
Posted 30 Sep 2015 by Richard Haselgrove Post: I think I'd better to put my reply to that in the politics area. |
|
3372)
Message boards :
Questions and problems :
Screensaver BUG on a Macintosh
(Message 64535)
Posted 29 Sep 2015 by Richard Haselgrove Post: Charlie does listen, honestly. He released BOINC v7.6.12 this morning, specifically to address "SCR: Tweak mac screensaver for slightly better reliability." Try that one, and let him know how you get on. |
|
3373)
Message boards :
Questions and problems :
BOINC 7.6.9 Crashing
(Message 64513)
Posted 29 Sep 2015 by Richard Haselgrove Post: There are two registry keys called "Space Sciences Laboratory, U.C. Berkeley". The one you're quoting is in the 'Current User' hive, but the one you want to look at for setup issues is in the 'Local Machine' hive: HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup The locations in question are INSTALLDIR (for programs) DATADIR (for ... guess what) |
|
3374)
Message boards :
Questions and problems :
BOINC 7.6.9 Crashing
(Message 64500)
Posted 28 Sep 2015 by Richard Haselgrove Post: There's certainly been some heavy customisation going on over the years, if Windows XP boots and runs from drive F: (you said that's the only disk in the system nowadays). F: is well up into the disk letter area usually reserved for removable and/or network drives. I've sometimes run into problems up there with overlapping definitions - plugging in a memory stick to a system where the next available drive letter has previously been mapped as a network drive, for example. The BOINC installer is certainly aware of the differences between local and network drives - an installation in /admin mode will default to the first available network drive, for example. Some of the issues raised seem to be similar to, though not identical with, the problems discussed in https://github.com/BOINC/boinc/issues/792. Check that the registry entries for installation and data directories left behind by the most recent working installation still point to locations which are valid for the current configuration - the installer will try to use them for the new installion locations this time round, as well, and will fall over if it can't find them. |
|
3375)
Message boards :
Questions and problems :
Missing MSVCR100.dll
(Message 64477)
Posted 27 Sep 2015 by Richard Haselgrove Post: Well, I did some searching and found the missing dll answer here: It might be BOINC which has a missing dependency (unlikely), or a science or graphics application from one of the projects you've attached to. That would require further investigation and information. Still would like to know a good way to turn off the screen saver after installation though. Does BOINC only turn on, not off? BOINC adds itself to the list of screensavers known to Windows (if you allow it to at installation time, that is). From that point forwards, you control it using the Windows screen saver settings - you can choose which screensaver to use (or 'none'), how long to wait before it is displayed, and many other things besides. The quickest way is probably to type 'screensaver' into the 'Search programs and files' box, at the foot of the Start button menu. |
|
3376)
Message boards :
Questions and problems :
Detected Trojan
(Message 64393)
Posted 21 Sep 2015 by Richard Haselgrove Post: That file, as it clearly states in the name, belongs to the SETI@Home project. So, deleting it will not affect BOINC - the forum you are writing on here - but it would affect your ability to contribute to the SETI@Home project. In general, you should ask project-specific questions on the forums of the specific project concerned. SETI has a Questions and answers section where new users can post without restriction. But if you post there, you are likely to get the same sort of answer as you get here. It is extremely likely to be a false alarm, but for your own peace of mind, you should check the file with a multi-vendor site like https://www.virustotal.com/. If, as is highly likely, this particular software is being over-senstive, you should tell them to white-list the particular file, and tell the review sites that it has caused you this anxiety and trouble for no purpose. |
|
3377)
Message boards :
Questions and problems :
Increased HDD I/O Activity w/ Version 7.6.(x?)
(Message 64353)
Posted 18 Sep 2015 by Richard Haselgrove Post: Did you install the 'plain vanilla' BOINC v7.6, or the "with added VBox" version? There has been some concern expressed about hard disk activity when VBox-style projects are active, and it's being looked into. But apart from that, I haven't heard of any specific issues related to disk activity. (BTW, the VBox activity complained of consists of re-reading a log file, to see if it's changed. Usually it hasn't, so in most cases - except on the oldest of drives - it's probably being read from cache anyway. Don't most HDD activity indicators signify activity on the interface, rather than in the internal mechanics?) |
|
3378)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64337)
Posted 17 Sep 2015 by Richard Haselgrove Post: Yes, it is, and this is what it looks like:  though personally I prefer the detail in client_state.xml <workunit>
<name>vina_391537_86310_Bre</name>
<app_name>vina</app_name>
<version_num>102</version_num>
<rsc_fpops_est>1155463742373.158900</rsc_fpops_est>
...<result>
<name>vina_391537_86310_Bre_0</name>
<final_cpu_time>160.337800</final_cpu_time>
<final_elapsed_time>167.942581</final_elapsed_time>
<exit_status>0</exit_status>
<state>5</state>
<platform>windows_x86_64</platform>
<version_num>102</version_num>
...
|
|
3379)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64334)
Posted 17 Sep 2015 by Richard Haselgrove Post: (My experience is limited, until now thought there were no specific bit size build work units. Maybe there is, models that were created to cater for 32-bit OS constraints and bigger ones for 64 bit OSses. At any rate I'd think it novel to get full WUs for a platform and the app not being downloaded until actually one is started. E.g. suppose the project is down, what then?) When tasks are allocated to a volunteer for processing, the formal specification for the task is handed to the client in two xml blobs - <workunit> and <result>. Between them they specify, among other things, * Input file(s) * Output file(s) * command line parameters * application to be used (including version number, platform, and perhaps plan_class) At the same time, the formal specification for any application version to be used will also be handed down to the client. Usually, this will just be a re-statement of information already known by the client - but occasionally, a project will issue a new application, or update the version of an application already in use. It's that notification of a new app version which will trigger the downloading of new binaries, libraries and whatever. That will happen at the same time as the downloading of any data files needed for the specific task. There's no question of waiting to download any application related files until the task starts: conversely, no task can be considered 'runnable' (and it won't start) until all related application downloads are complete. |
|
3380)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64326)
Posted 16 Sep 2015 by Richard Haselgrove Post: 'Read config files' isn't enough to re-activate <alt_platform>, it needs a full client restart. But after that, it asked with <alt_platform>, and got the 32-bit application back as before. So the tag works, but use with care - it operates globally, across all projects, and not all of them necessarily supply work flagged as 64-bit. |
|
3381)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64325)
Posted 16 Sep 2015 by Richard Haselgrove Post: OK, it begins to make more sense. After that batch had completed and reported, I reset the project, and fetched a new batch. This time I got the 64-bit app, as <app_version>
<app_name>vina</app_name>
<version_num>102</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>661977170.668584</flops>
<api_version>7.5.0</api_version>
<file_ref>
<file_name>vina_1.2_windows_x86_64.exe</file_name>
<main_program/>
</file_ref>
</app_version>Note that the APR has already risen slightly. The sched_request also showed <platform_name>windows_x86_64</platform_name>, and no alternates. But I'm also getting messages like 16/09/2015 22:20:29 | GPUGRID | Message from server: This project doesn't support computers of type windows_x86_64 so I'd better start putting this computer back together without the <no_alt_platform>, and resetting the projects which were 'adjusted'. |
|
3382)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64323)
Posted 16 Sep 2015 by Richard Haselgrove Post: But... it did work, you wouldn't have had those errors otherwise. There doesn't seem to be any log messages though. My initial quick test of working was "Did it download the 64-bit version of the app at the next work request?", and the answer was "no". It was only as I worked through the logs and other indicators that I saw that it had converted the existing tasks and application_version to run the current workload (including the 32-bit binary) as if it was the 64-bit platform. |
|
3383)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64320)
Posted 16 Sep 2015 by Richard Haselgrove Post: Didn't work for me either, in Windows64. And I'm using BOINC v7.6.9, with cc_config.xml pre-populated with default tags from the GUI - just set the value in the place provided. Didn't work with a simple 'Read config files', didn't work after a full client restart. Edit - I did get a huge number of error messages on startup, including 16/09/2015 21:32:44 | FiND@Home | [error] App version has unsupported platform windows_intelx86; changing to windows_x86_64 and 16/09/2015 21:32:44 | LHC@home 1.0 | [error] App version has unsupported platform windows_intelx86; changing to windows_x86_64 16/09/2015 21:32:44 | LHC@home 1.0 | [error] State file error: duplicate app version: sixtrack windows_x86_64 44401 sse3 Edit 2: So now I have <app_version>
<app_name>vina</app_name>
<version_num>102</version_num>
<platform>windows_x86_64</platform>
<avg_ncpus>1.000000</avg_ncpus>
<max_ncpus>1.000000</max_ncpus>
<flops>618682904.035871</flops>
<api_version>7.5.0</api_version>
<file_ref>
<file_name>vina_1.2_windows_intelx86.exe</file_name>
<main_program/>
</file_ref>
</app_version>- totally muddled, but it's going on fetching new work. Unfortunately, FiND doesn't distinguish the platform work is issued for by plan class, or otherwise display platform or alt_platform in the web lists of tasks issued. Edit 3: but that <flops> value looks like the APR=0.62 from the 64-bit app_ver in Application details for host 105172. Next step is probably a 'flush all tasks' (which will take far less time than is being estimated with that grotty APR), reset project, and see what comes down the pipe next time. |
|
3384)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64312)
Posted 16 Sep 2015 by Richard Haselgrove Post: The decision of which application to use (if both are available choices for the system) is made by the project server, on the basis of the application_version history for the HostID of the computer. If you remove and replace BOINC, but leave the rest of the computer unchanged, a BOINC server will try to recycle any available previous HostID - using your account, computer name, IP address, and probably other things I've forgotten to make the association between hardware and HostID. It does this to avoid unnecessary bloating of the Host table in the database. So, rather than removing and replacing the BOINC client, what you probably want to do is to force a new HostID, and thus start clean with a new host_app_version table. (make sure you don't immediately fill it with zero-runtime-estimate tasks...) The best way of doing that is by keeping the old BOINC installation, and forcing an apparent 'cheat' by tweaking the <rpc_seqno> for the project in client_state.xml downwards, so it appears that two separate computers are trying to contact the project scheduler using the same HostID. Someone may need to remind me whether it's necessary to set <allow_multiple_clients> at the same time for this to work with recent server code - or you could experiment. Flush out all running tasks with NNT before you try this, and update the project 'dry', and inspect for a successful HostID change, before allowing new work. |
|
3385)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64306)
Posted 16 Sep 2015 by Richard Haselgrove Post: My suggested explanation is available for anyone to read at FiNDAH message 2584. It applies to most projects, but the evidence is particularly stark at FiND. |
|
3386)
Message boards :
Questions and problems :
64 bit system suddenly thinks it is 32 bit.
(Message 64303)
Posted 16 Sep 2015 by Richard Haselgrove Post: Is this related to your question 64 bit systems now using 32 bit application for Linux at FiND? Do you have the same observations for other projects? I'll attempt an explanation there, but both you and Ben may find it a little surprising. |
|
3387)
Message boards :
Questions and problems :
Trojan alert at BOINC 7.6.9
(Message 64237)
Posted 14 Sep 2015 by Richard Haselgrove Post: Windows 64 bits without V-Box... Here's a link to the VirusTotal analysis report: https://www.virustotal.com/en/file/a1ea3f31a420dc9db2013ae42320a886e8b001f6f642f056d57438d1d7052d8e/analysis/1441827908/ Looks clean to me. |
|
3388)
Message boards :
Questions and problems :
Trojan alert at BOINC 7.6.9
(Message 64234)
Posted 14 Sep 2015 by Richard Haselgrove Post: It's likely to be a false positive, but we should check it out. I'm assuming Windows, but can you confirm whether 32-bit or 64-bit, with or without VBox? Then we can check it with a service like https://www.virustotal.com/ - or you could reassure yourself by running your own scan there. Many anti-virus programs these days flag a warning for an application which is new and untested, even if it doesn't carry any malicious payload. |
|
3389)
Message boards :
Questions and problems :
Question about reboot.
(Message 64170)
Posted 11 Sep 2015 by Richard Haselgrove Post: That will work, but a service mode installation will disable any GPU that Bill (or anyone else with a similar question) might wish to use for crunching. The alternative is to configure Windows to login automatically after a reboot. You need to think carefully about the security of this - you need to store your password in a way which could potentially be accessed if someone stole the computer, so don't use the same password as you use for important online services - but if your security situation allows it, you would set: All in the registry key HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Winlogon AutoAdminLogon, value "1" DefaultUserName DefaultPassword "DefaultUserName" probably has your user name in it already: "AutoAdminLogon" probably exists but needs the value changing, and "DefaultPassword" would need to be created from scratch and filled in (it can't be blank for this to work). |
|
3390)
Message boards :
BOINC Manager :
FSF endosert Linux Distros are deleting Boinc.
(Message 64144)
Posted 9 Sep 2015 by Richard Haselgrove Post: For distribution, they would simply have to edit (cut down) the copy of all_projects_list.xml distributed with the program. That would remove the "advertisement" when the program first launches. But BOINC itself periodically downloads a fresh copy of the list, with new projects added / old projects removed / changes to existing projects. The maintainers would have to consider whether this contravenes their principles. Also, users can add unlisted projects simply by typing in a url. Is that acceptable? It might be better for the repository maintainers to open a dialog with the BOINC Project Management Committee, and work through their differences to see if a compromise can be reached. |
|
3391)
Message boards :
Questions and problems :
BOINC running way too many processes
(Message 64143)
Posted 9 Sep 2015 by Richard Haselgrove Post: The normal start-up process for current versions of BOINC is to launch an instance of BOINC Manager through the Windows Registry. That instance of the Manager, in turn, launches boinc.exe, and boinc.exe launches the child project processes. Many years ago, it was commoner to launch BOINC from a shortcut in the StartUp folder in Start Menu\Programs. Could you perhaps have a left-over StartUp link launching the second copy? |
|
3392)
Message boards :
Questions and problems :
BOINC running way too many processes
(Message 64045)
Posted 3 Sep 2015 by Richard Haselgrove Post: The OP said very specificslly: Starting the BOINC manager starts 8 SETI processes and 2 boinc.exe processes. The SETI processes each get about 10% of the available CPU. BOINC manager shows 4 running tasks. That's very similar to a bug that Charlie thought he observed, and tried to fix in v7.6.8 But the fix was flawed, and reverted so that v7.6.9 could be released to replace the even more flawed v7.6.6. I thought we might get a second attempt at Charlie's patch (which was related to service mode installations and how to control them - the conversation continued privately after the boinc_alpha discussion went silent), but no sign of it yet. |
|
3393)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64033)
Posted 2 Sep 2015 by Richard Haselgrove Post: They do now. |
|
3394)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 64030)
Posted 2 Sep 2015 by Richard Haselgrove Post: Some assorted test images for Jord and Rom http image  https url https://4.bp.blogspot.com/-GxN5obBzW6Q/VLaEAFzpabI/AAAAAAAAKcE/awOO_nLKPhM/s1600/panic%2Bbutton.jpg https url http img  http url https img Edit - I suppose I'd better add an example of the photobucket problem - https url, http img, same address, different rendering. 
|
|
3395)
Message boards :
GPUs :
Can I run BOINC without the annoying screen saver?
(Message 63883)
Posted 28 Aug 2015 by Richard Haselgrove Post: In fact, it's the opposite - BOINC will actually do more work when the screensaver is *not* running: that's pure eye-candy, and takes away processing power that could be more productively used. If you have no need of it, use the Windows control panel to choose a different screen saver, or just a blank screen - the simpler the better (avoid the 3-D ones), as far as BOINC's concerned. |
|
3396)
Message boards :
Questions and problems :
Where's Git-log gone?
(Message 63800)
Posted 24 Aug 2015 by Richard Haselgrove Post: They've moved the code to GitHub. I don't find the user interface nearly as good for referring to changes: the nearest I've found to the changelog is https://github.com/BOINC/boinc/commits/master |
|
3397)
Message boards :
Questions and problems :
Limiting simultaneously CPU Usage per Project
(Message 63787)
Posted 23 Aug 2015 by Richard Haselgrove Post: Read Application configuration again. <project_max_concurrent> (if available in your version - it sounds as if it probably isn't) would only be valid outside the optional <app> or <app_version> blocks. <avg_ncpu> would be <avg_ncpus>, and only valid inside an <app_version> block. <max_concurrent>, on its own inside an <app> block, is probably what you want with this version, but it won't achieve everything you want if the project supplies more than one application (as CPDN, for example, normally does). Edit - the whole app_version.xml construction was very rough-and-ready back in that early version - it doesn't surprise me that having those syntax errors makes it fail, but without an error message. The later versions are better, but still not entirely foolproof. |
|
3398)
Message boards :
Questions and problems :
Limiting simultaneously CPU Usage per Project
(Message 63785)
Posted 23 Aug 2015 by Richard Haselgrove Post: Yes, it says project_max_concurrent 7.2.42 may be the recommended version for Linux, but it really should have been updated since 28 Feb 2014 - other platforms have. Depending on the Linux distribution you use, a newer supported version might be available from the package maintainer - somebody here might be able to advise. |
|
3399)
Message boards :
Questions and problems :
Limiting simultaneously CPU Usage per Project
(Message 63772)
Posted 22 Aug 2015 by Richard Haselgrove Post: app_config.xml |
|
3400)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63741)
Posted 19 Aug 2015 by Richard Haselgrove Post: Well, we sure had fun last night. One after another thunderstorm passing directly over. Looks like you weren't the only one. From the BBC: Google loses data as lightning strikes Google says data has been wiped from discs at one of its data centres in Belgium - after it was struck by lightning four times. I'm sure the NSA will let them have if back if they ask nicely. |
|
3401)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63721)
Posted 18 Aug 2015 by Richard Haselgrove Post: Seti forums are back. Pretty much everything is back (Matt posted in NC), but I haven't been able to contact the scheduler yet to report completed work. |
|
3402)
Message boards :
Questions and problems :
why credits too low?
(Message 63688)
Posted 17 Aug 2015 by Richard Haselgrove Post: Most of the credits in your second example come from two projects - Collatz and Milkyway. Both of those projects make extensive use of GPUs - 'Graphical Processing Units', or high-powered video cards. They can do much more work than the CPU by itself, but they don't show up in the statistics kept by BOINCstats. |
|
3403)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63660)
Posted 16 Aug 2015 by Richard Haselgrove Post: Looks like they finally got something running to purge that giant pile of results above 2^32 - about six hours ago. Maybe the combination of deleting a couple of million records, at the same time as running the event, was too much for one database to handle. |
|
3404)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63655)
Posted 16 Aug 2015 by Richard Haselgrove Post: And the scheduler went down about half-an-hour before that. Game over for today, I think. |
|
3405)
Message boards :
Questions and problems :
New Tasks Stuck at 0% Downloaded
(Message 63587)
Posted 13 Aug 2015 by Richard Haselgrove Post: What those entries are saying is: "If malware wants to go to a bad place, don't let it out of the box." (redirect to localhost) That blocks the named domain. But in this case we want a particular name to go to a good place - better to give it that positive encouragement, as self-documented in the early part of Microsoft's sample file. People who do this need to remember what they've done, and be ready to undo it as soon as the need has passed. Any server operator on the internet (and not just BOINC download servers) would reserve the right to migrate the service to new hardware with a new IP address. That's what the DNS service is for, and in general it's best to let it do its work automatically - that way, things don't fall over if the addresses change again. |
|
3406)
Message boards :
Questions and problems :
New Tasks Stuck at 0% Downloaded
(Message 63583)
Posted 13 Aug 2015 by Richard Haselgrove Post: The purpose of a hosts file is to associate a domain name with an IP address - I see no names! The hosts file isn't a blocker, it's an enabler. My version is saying "whenever a program wants to visit boinc2.ssl.berkeley.edu, tell it they're living at 208.68.240.21". |
|
3407)
Message boards :
Questions and problems :
New Tasks Stuck at 0% Downloaded
(Message 63568)
Posted 12 Aug 2015 by Richard Haselgrove Post: The purpose of a hosts file is to associate a domain name with an IP address - I see no names! Something like this: # Copyright (c) 1993-1999 Microsoft Corp. # # This is a sample HOSTS file used by Microsoft TCP/IP for Windows. # # This file contains the mappings of IP addresses to host names. Each # entry should be kept on an individual line. The IP address should # be placed in the first column followed by the corresponding host name. # The IP address and the host name should be separated by at least one # space. # # Additionally, comments (such as these) may be inserted on individual # lines or following the machine name denoted by a '#' symbol. # # For example: # # 102.54.94.97 rhino.acme.com # source server # 38.25.63.10 x.acme.com # x client host 127.0.0.1 localhost 208.68.240.21 boinc2.ssl.berkeley.edu |
|
3408)
Message boards :
Questions and problems :
New Tasks Stuck at 0% Downloaded
(Message 63566)
Posted 12 Aug 2015 by Richard Haselgrove Post: From a command line, try ipconfig /flushdns, which will delete the IP resolver cache. If he's using a HOSTS file, there's no resolver cache to flush. There's no resolver in circuit. |
|
3409)
Message boards :
Questions and problems :
New Tasks Stuck at 0% Downloaded
(Message 63561)
Posted 12 Aug 2015 by Richard Haselgrove Post: Thanks, I did see Vader was down just was unsure of it's address. I will try changing my hosts and see what happens. Unfortunately, the other way round. Georgem (working) is 208.68.240.21: Vader (not working) is 208.68.240.13 |
|
3410)
Message boards :
Questions and problems :
New Tasks Stuck at 0% Downloaded
(Message 63560)
Posted 12 Aug 2015 by Richard Haselgrove Post: Setting the <http_debug> Event Log option will show you - in much more detail than you really need - exactly what is going on behind those 'transient HTTP errors'. The detail will include the url of the download server, and the IP address your machine is resolving it to. Check on your own machine, but I would expect to see some combination of boinc2.ssl.berkeley.edu 208.68.240.13 208.68.240.18 208.68.240.21 I've included 208.68.240.18 although it's an obsolete server now decommissioned. If you are attempting to contact it, then you have set a manual IP address somewhere (probably in a hosts file) and you should remove it and use normal dns lookups instead. 208.68.240.13 and 208.68.240.21 are both valid addresses, although 208.68.240.13 is temporarily out of service and the load is being handled by 208.68.240.21: if your machine is trying 208.68.240.13, wait five minutes and try again. This sort of question (relating to one specific project) is probably best asked on the project's own message boards, where you can see that we covered it a few days ago at message 1709176. |
|
3411)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63533)
Posted 11 Aug 2015 by Richard Haselgrove Post: Did anyone think using https guaranteed privacy? Web's random numbers are too weak, researchers warn |
|
3412)
Message boards :
The Lounge :
Grumbles, Glory and Covid-19
(Message 63527)
Posted 11 Aug 2015 by Richard Haselgrove Post: Grumble - all these techies who decided to do their work on the left coast of North America, so there isn't a hope in hell of them finishing before civilised countries have gone to bed. |
|
3413)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63525)
Posted 11 Aug 2015 by Richard Haselgrove Post: Blame Richard. He wanted the uncontrolled database growth fixed, so all those 'cannot find workunit' messages would go away. ;-) And there are still over 11 million of them in there. Just think how long it would have taken if they'd left it another week. |
|
3414)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63518)
Posted 11 Aug 2015 by Richard Haselgrove Post: I see we're missing some CSS on the Server status page. It was all more 'grouped' earlier on. Oh and the Seti menu bar is a bit broken. Looks better if you don't try to use https. Nothing super-secret on there, is there? Edit - removed the half of the quote I'm not referring to. |
|
3415)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63514)
Posted 11 Aug 2015 by Richard Haselgrove Post: David is having fun sorting his millions from his billions: lib: add function for formatting numbers w/ digit grouping |
|
3416)
Message boards :
Questions and problems :
How to suspend a project without starting BOINC Manager
(Message 63435)
Posted 8 Aug 2015 by Richard Haselgrove Post: Too late to edit, but I've confirmed with BOINC v6.12.34 (Manager and Client): if you change the "Run Manager at login?" option - which also affects whether the client starts at login (unless you have a service installation) - it's the local machine which is modified, not the remote client you're connected to. Apart from (at least some of) those general options, and I suspect the 'new in v7.6' Select columns option, most other Manager controls do operate on the connected remote client. So if Bill had a second machine on the same LAN, and could gain control of his laptop for long enough to enable Controlling BOINC remotely, he might be able to work out whether Einstein was locking up the entire machine, or just the user interface. I'm finding that on my four-year-old Windows 7 laptop, the user interface has started becoming very sluggish after ~1 week of continuous running, needing a full restart, even though BOINC science apps are still running normally in the background. The laptop only has 4GB of RAM, which is low for 64-bit Windows: I suspect I may be suffering from "cumulative security update bloat" or "creeping anti-virus feature bloat". Einstein, as a project, also has a peculiarity in that some of its applications shift an awful lot of data off the hard disk in the early stages of startup, and that can lock things up temporarily. Bill, it might be worth, if you have a spare moment when you can spare the machine, letting it start up and leaving it running for five or ten minutes before trying the user interface again - see if it recovers by itself. |
|
3417)
Message boards :
Questions and problems :
How to suspend a project without starting BOINC Manager
(Message 63433)
Posted 8 Aug 2015 by Richard Haselgrove Post: I hit two snags with that approach:  1) The BOINC programs aren't in the environment 'PATH' variable, so you need to be in the program directory first. I keep mine in C:\BOINC, which makes the CD command easy: CD /"C:\Program Files\BOINC" (quotation marks required because of the space) is harder. 2) The https:// protocol isn't part of the project url, although the extra trailing / seems to be acceptable. But as I said, there are several ways to achieve this, once you have control of the computer before or after the lockup. If you have the BOINC Manager GUI running, you can use the Option dialog "Run Manager at login?" to make the same registry change I suggested in my first post (I'm not sure if the client has to be running as well for that to work - I'll test that later). Edit: tested with BOINC v7.6.6 - yes, you can control the "Run Manager at login?" option directly from a bare manager, with no client running. Does that mean it doesn't work for a remote host? If the client isn't running, you can manually insert the line <suspended_via_gui/> into the Einstein project section of client_state.xml - that's what Jord's boinccmd does, except for that to work, the client does have to be running - which may be difficult if the machine locks up so quickly. |
|
3418)
Message boards :
Questions and problems :
How to suspend a project without starting BOINC Manager
(Message 63431)
Posted 8 Aug 2015 by Richard Haselgrove Post: Several ways, some of which you can usefully combine for troubleshooting. 1) Using the Windows registry (run 'regedit') Find the key HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager and set the value DisableAutoStart to 1 2) Using BOINC's Client configuration Edit the file 'cc_config.xml' in the BOINC data directory. Set the value <start_delay>nseconds</start_delay> to a modest value, like 30 or 60. Technique (1) will prevent BOINC starting at all, until you're ready to start it yourself from the start menu. But it will still lockup immediately, unless you also use technique (2) to delay Einstein starting up. You get whatever breathing space you've specified to dive in and suspend either the entire project, or any individual tasks (perhaps GPU tasks) that you suspect may be causing the lockup. |
|
3419)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63387)
Posted 4 Aug 2015 by Richard Haselgrove Post: I usually open it from the link on the front page (second column, under 'ABOUT'). The page source for that column <a href=sah_about.php class=menu>About SETI@home</a></nobr><br> <a href=ap_faq.php class=menu>About Astropulse</a></nobr><br> <a href=sah_sci_newsletters.php class=menu>Science newsletters</a></nobr><br> <a href=http://setiathome.berkeley.edu/tech_news.php class=menu>Technical news</a></nobr><br> <a href=sah_status.html class=menu>Server status</a></nobr><br> <a href=sci_status.html class=menu>Science status</a></nobr><br> <a href=kiosk/ class=menu>World view</a></nobr><br> <a href=sah_about.php class=menu>... more</a></nobr><br> shows a mixture of php and html pages - SETI never did adopt all the BOINC rules ;) |
|
3420)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63385)
Posted 4 Aug 2015 by Richard Haselgrove Post: Oh man, no DB backups. The BOINC server doesn't really do much database work. But I do hope they backed up both User documentation wrote: Forbidden Server documentation wrote: Internal Server Error |
|
3421)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63384)
Posted 4 Aug 2015 by Richard Haselgrove Post: By the way, is it normal that https://setiathome.berkeley.edu/server_status.php is empty? Trying http://setiathome.berkeley.edu/server_status.php instead... Nope, that came back blank too. But the one I opened this morning: http://setiathome.berkeley.edu/sah_status.html works fine. What made you think it was php? |
|
3422)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63381)
Posted 4 Aug 2015 by Richard Haselgrove Post: By the way, is it normal that https://setiathome.berkeley.edu/server_status.php is empty? Maybe David borrowed it when he borrowed the SETI staff yesterday? The BOINC server's file system failed on Friday night, |
|
3423)
Message boards :
Questions and problems :
opencl question
(Message 63357)
Posted 31 Jul 2015 by Richard Haselgrove Post: Don't bother to install it, not needed for CPUs. BOINC projects with applications for CPUs will supply ordinary x86 binary executables. They're well understood, and there's no need for an intermediate language interpreter like OpenCL. OpenCL comes into its own for programming GPUs, but there's still no need to install a separate package: each of the major GPU manufacturers (AMD, Intel, NVidia) supplies OpenCL runtime support in their graphic driver packages. |
|
3424)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63350)
Posted 30 Jul 2015 by Richard Haselgrove Post: So does that mean we should expect opencl_nvidia_sah on Main? Best we take this back to the SETI forums (preferably Beta), but IMO: if Eric restores the status quo policy of the last two years, there's no point: even if you import the application, you still wouldn't get VLAR tasks. |
|
3425)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63346)
Posted 30 Jul 2015 by Richard Haselgrove Post: Quick turn around. Still not quite. The test was made in May 2013: SETI Beta message 46039. As a test I'm sending VLARs to comp-cap 3.0+ NVIDIA cards. Anyone with a Kepler or higher let me know how they work. |
|
3426)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63341)
Posted 30 Jul 2015 by Richard Haselgrove Post: Message boards have just come back up (Main and Beta). |
|
3427)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63333)
Posted 30 Jul 2015 by Richard Haselgrove Post: Edit: I still think that an opt-in/opt-out choice for VLAR's on GPU's would be the best solution, where opt-out should be default. There was a call on the Beta forums recently for a 'VLAR to NV' option, with the suggestion it should be defaulted 'on' at Beta, defaulted 'off' at the main project. I'd support that balance. |
|
3428)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63331)
Posted 30 Jul 2015 by Richard Haselgrove Post: My GTX980 is compute capability 5.2 I got this: 30-Jul-2015 01:28:25 [SETI@home] Sending scheduler request: To fetch work. with a 17-Jul-2015 00:35:14 [---] NVIDIA GPU 0: GeForce GTX 750 Ti (driver version 34475, CUDA version 6050, compute capability 5.0, 2048MB, 410 GFLOPS peak) I think Eric is mis-remembering the policy. We did try VLAR to Kepler (cc 3.0) when the new cards first came out, three years ago: Keplers were released "big ones first", and we thought they might have enough heft to handle VLARs. This cc 5.0 is a Maxwell - they were released with the smaller cards first. VLARs are not good on any NV cards - for some people they cause screen lag to the point of unuseability, for others they just run slower and hence mess up the runtime estimate averages. For at least the last two, maybe 2.5, years the policy has been "VLAR to cc 3.0 and above at the Beta project only, no VLARs to NV at the main project. I can find you the exact quote on the Beta message board once it's back up (I had a hand in testing and reporting at Beta, leading to the policy decision). |
|
3429)
Message boards :
Questions and problems :
Reporting of task results
(Message 63207)
Posted 22 Jul 2015 by Richard Haselgrove Post: Question: Once a task changes status to "Ready to report" does it send its outcome automatically ? or do I need to do something ? (e.g. explicit transfer) It will happen automatically, though not necessarily immediately. There are a number of triggers for task reporting: the commonest is when BOINC needs to request new work from the same project (fetching and reporting use the same communications link). BOINC will normally report tasks anyway after one hour (unless something else has triggered an earlier report). If an attempt to report work fails for any reason (project unreachable, or project servers down for maintenance), the task will remain "Ready to report" and BOINC will retry periodically. If the report is successful, the task will disappear from the BOINC Manager display. |
|
3430)
Message boards :
Questions and problems :
How to access Boinccmd
(Message 63092)
Posted 16 Jul 2015 by Richard Haselgrove Post: When you install BOINC, boinccmd is an additional executable file that is installed as part of the package. The rest of this answer will be applicable to Windows only. You don't say which operating system you use - if it's something else, my apologies for wasting your time. When BOINC is installed on Windows, it's divided into two parts - a program directory and a data diectory. For boinccmd, you need to find the program directory. By default, this will be C:\Program Files\BOINC Navigate to this directory in Windows Explorer and - assuming you are running Windows 7 or later - right-click on the folder name in the left-hand pane while holding the shift key. Holding the shift key while right-clicking will add "Open command windows here" to the context menu. That's the one you want. Click the command window to give it focus (for some reason, Windows doesn't do this automatically), and you're ready to go - type boinccmd followed by whatever commands you need from The boinccmd command-line tool |
|
3431)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63074)
Posted 14 Jul 2015 by Richard Haselgrove Post: Be careful what you wish for. This is, genuinely, the signature on an email I received today. Yes, she'll be the one - location and previous companies are right for the source. Keep an eye on her for me, will you, and let me know when she changes her profile to "Head of Customer Happiness"? That should be marked as some sort of red-letter day in business doublespeak. |
|
3432)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 63065)
Posted 14 Jul 2015 by Richard Haselgrove Post: Be careful what you wish for. This is, genuinely, the signature on an email I received today. ... My spambot antennae are pretty sensitive, but I fear this one might be genuine - the links look right, as far as I can tell without clicking them. They're a credit card issuer, for dog's sake. |
|
3433)
Message boards :
Questions and problems :
Why project schedule priority so high?
(Message 63038)
Posted 14 Jul 2015 by Richard Haselgrove Post: The whitepaper for that is https://boinc.berkeley.edu/trac/wiki/ClientSchedOctTen. Proposal: credit-driven scheduling And that's what the current BOINC v7 clients do. Because REC estimates credit much more closely in alignment with the formal definition of the cobblestone than many projects do, REC (and consequently Resource Share) follow flopcounting much more closely than they follow actual granted credit. I personally suspect that the 'flops' which are 'counted' for REC are those declared by the programmer/deployment administrator in <rsc_fpops_est>. That's my theory (partly) for why SETI's MB and AP tasks award different RAC, and likewise Albert's Arecibo and Perseus Arm surveys (IIRC those were the two apps which trended to different limit values). |
|
3434)
Message boards :
Questions and problems :
GUI RPC request from non-allowed address shows wrong IP addressin event logs
(Message 62928)
Posted 5 Jul 2015 by Richard Haselgrove Post: That one needs re-confirming with a host1 running - ideally - alpha BOINC v7.6.3, or at the very least recommended v7.4.42 I doubt it'll make any difference, but it has to be tested. There's nothing you can do about the retry rate from obsolete v6.10.17 - except note that all Managers of all versions have always updated every second. More than 600 retries in 600 seconds probably implies that the Manager requires multiple RPCs to update all pages. |
|
3435)
Message boards :
Documentation :
app_config.xml page on wiki
(Message 62919)
Posted 5 Jul 2015 by Richard Haselgrove Post: Application_Name and example adjusted. |
|
3436)
Message boards :
Documentation :
app_config.xml page on wiki
(Message 62918)
Posted 5 Jul 2015 by Richard Haselgrove Post: Firstly, thank you to those who document boinc wonderfully well. That's worth thinking about, and I'll maybe implement it when the web site finds its style sheet again, so I can see the layout properly. It would require some adjustment to the example section: name <app_version> questions. I don't think the granularity of app_config goes down as low as the version number level. The hierarchy is Application ... Plan class ... ... Version (number) If you were able to use it, you would take it from <version_num> in the <app_version> section of client_state.xml <plan_class> is not well documented. I'll leave that for more experienced documenters to consider. |
|
3437)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62875)
Posted 2 Jul 2015 by Richard Haselgrove Post: I hope your car/property is ok Richard. We're all OK, thanks - self, car and property. When Chris first posted that BT link, the main part of it was the section that starts Scores of motorists pulled over to watch last night as another storm moved up The Pennines. That's the bit which matched what we saw from the beer garden - with added rain a bit later, though none of the destructive hail or tornados. That part of the BT report was added later. We've had some more rain through here this evening - heading northwards, I'm sorry to say, Jord. But another night of the rough stuff is forecast for tomorrow. |
|
3438)
Message boards :
Questions and problems :
Disk thrashing - system to slow to use.
(Message 62869)
Posted 2 Jul 2015 by Richard Haselgrove Post: I've also had problems with disk thrashing, slow response, and high CPU usage on Windows machines in recent days. In my case, the causes were variously svchost.exe (Host Process for Windows Services) - under Windows 7, two instances ('SYSTEM' and 'LOCAL SERVICE') using over a gigabyte of memory each; Malicious Software Removal Tool; and the Realtime Protection component of Microsoft Security Essentials. I didn't bust a gut looking for contemporary problem reports - just rebooted machines as necessary. My suspicion is that some automatic security updates have had problems recently. Moral: don't assume that all computer problems are caused by BOINC. |
|
3439)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62865)
Posted 2 Jul 2015 by Richard Haselgrove Post: BBC That's BT.com, but it sounds like what passed over me just as they called last orders. |
|
3440)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62859)
Posted 1 Jul 2015 by Richard Haselgrove Post: Where did everyone go??? :-O Yesterday must have been summer - all 24 hours of it. Tonight I walked home (again in shirtsleeves) in pouring rain, to the accompaniment of a 360-degree lightning display. |
|
3441)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62845)
Posted 30 Jun 2015 by Richard Haselgrove Post: Where did everyone go??? :-O I went across the road to the pub - and after an evening meal, I sat outside in the warm, gathering, twilight, and watched first Venus and then Jupiter appear in near-conjunction. And then walked home by the light of an almost-full moon. Not often you can do that comfortably in shirtsleeves, in the north of England. |
|
3442)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62822)
Posted 30 Jun 2015 by Richard Haselgrove Post: Ok, perhaps I am being a little unfair to UCB. But over the last few years I can recall a number of unplanned power outages where manholes exploded, or cable was stolen, and also fibre internet services up the hill to the lab failing. That was one of the drivers to re-locate the Seti closet to the Colo. I've just got the impression rightly or wrongly that its all run a bit haphazardly for a Uni campus. I wonder if those modern ideas of outsourcing come from places like, err, the Haas School of Business :D |
|
3443)
Message boards :
Projects :
News on Project Outages
(Message 62761)
Posted 26 Jun 2015 by Richard Haselgrove Post: Is SEITI@Home not putting out any more tasks? No, that's a local button you've clicked (quite possibly by mistake - it's easy to do while you drag the mouse pointer). Click on the SETI@home line where you read "Won't get new tasks", to highlight it. In the list of commands to the left, the third one down should now read "Allow new tasks". Click it once, and the wording should change to "No new tasks": the "Won't get new tasks" message will also disappear. BOINC should now request new work within a few seconds, possibly up to a minute. If nothing happens after that, click the 'Update' button. But be aware that SETI is planned to be offline tomorrow (Saturday 27 June) for electrical supply maintenance, and seems to be running down the stock of tasks ready to send in anticipation. For further details, visit the SETI website. |
|
3444)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62737)
Posted 25 Jun 2015 by Richard Haselgrove Post: I am about done with some new mod getting their oats on. Mark, I don't know what the mods do... But I do know what you do. And I suspect you also know what you do. Which is why I've obscured the f-bomb, so the Jord doesn't have to hide this post from you, and you can still see it on Monday morning. You drink too much, but you don't keep it private. You get drunk, and you post on the boards while drunk. That annoys everyone, mods and non-mods alike. |
|
3445)
Message boards :
BOINC client :
i686 for macOSX download link broke (404 error)
(Message 62728)
Posted 25 Jun 2015 by Richard Haselgrove Post: Attempting to download the latest test client 7.6.3 for macOSX_i686 and receiving a 404 (not found) error. Which page is that on? They're only making 64-bit builds for Macs these days: http://boinc.berkeley.edu/dl/boinc_7.6.3_macOSX_x86_64.zip |
|
3446)
Message boards :
BOINC client :
"finish file present too long" error
(Message 62717)
Posted 24 Jun 2015 by Richard Haselgrove Post: You probably have to get your application to clean up all its working files and perform whatever post-processing is necessary, before you call boinc_finish(). I think the "too long" message refers to any delay between signalling that the app has finished, and the process finally quitting. |
|
3447)
Message boards :
Questions and problems :
No GPU found after installing separate GPU
(Message 62712)
Posted 24 Jun 2015 by Richard Haselgrove Post: Some motherboards don't allow onboard GPU (VGA) devices to be active at the same time as expansion cards: others allow both to be active, but require dual graphics to be enabled in BIOS. Sometimes, a GPU can only be used if it has a connection to a monitor or dummy load. In short, this is likely to be a computer configuration issue, rather than a BOINC issue. |
|
3448)
Message boards :
Questions and problems :
Elapsed time running slowly
(Message 62668)
Posted 23 Jun 2015 by Richard Haselgrove Post: So you think it actually uses 40% on average? Because what top reports is much lower. With the setting on 60% top also reports around 60% and elapsed time runs pretty much OK (not so much with the setting on 40% as I have said earlier). I don't think anything - I don't know, and I'm as interested as you to find out what is making elapsed time run slowly (in case it needs fixing). I'm just suggesting that in order to get a true understanding, you probably need to understand the characteristics of the measuring tool you're using, too - and maybe compare with the answers given by other measurements. |
|
3449)
Message boards :
Questions and problems :
Elapsed time running slowly
(Message 62662)
Posted 23 Jun 2015 by Richard Haselgrove Post: So 40% will have BOINC allow the science application to run for 4 seconds every 10 seconds, and pause it the other 6 seconds. It will not run science applications at a sustained 40% load only. That possibly depends how long 'top' averages its reported CPU figure over. If it's designed to give a general overview over two or three seconds, you might get a different impression from a profiling tool with millisecond resolution. |
|
3450)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62645)
Posted 22 Jun 2015 by Richard Haselgrove Post: It did go down for a while, but recovered (I think by automatic reboot). |
|
3451)
Message boards :
Questions and problems :
Elapsed time running slowly
(Message 62640)
Posted 22 Jun 2015 by Richard Haselgrove Post: You can see it running at about 1/3 speed of actual time when you look at the counter while it's in 'running' state. It made 10 seconds for 27 actual seconds when I measured it. That possibly suggests that your CPU is 'overcommitted' - running more applications and processes, not necessarily related to BOINC and its projects, than can be handled in real time. In the Windows world, if that can't be accounted for by the applications you know you're running, we'd suggest that you run a virus check to be sure that nothing is stealing CPU cycles. But that can't be a problem with Linux, of course ;) What do your tasks show for CPU time (task 'properties' from BOINC manager while running or completed but not yet reported: on project website after reporting)? Do those figures make sense, when compared with either reported 'elapsed' time or genuine wall-clock time? |
|
3452)
Message boards :
Questions and problems :
Elapsed time running slowly
(Message 62637)
Posted 22 Jun 2015 by Richard Haselgrove Post: Thanks, that'll help with the time tasks take to complete. But it doesn't change the fact that elapsed time is running slowly. 'Elapsed time' is the time spent in the 'running' state. The time spent in the 'suspended' state isn't counted. |
|
3453)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62583)
Posted 16 Jun 2015 by Richard Haselgrove Post: Delight or Coffee? Yes :) |
|
3454)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62581)
Posted 16 Jun 2015 by Richard Haselgrove Post: it was very punctual! it went down exactly when my task was done! ;) Any Turkish on offer? |
|
3455)
Message boards :
Questions and problems :
Permanent HTTP error
(Message 62566)
Posted 15 Jun 2015 by Richard Haselgrove Post: Might be company proxy policy - you'll have to talk to them about that. But it might also be an anti-virus program detecting a false positive, or being suspicious about the application behaviour. Check AV logs. |
|
3456)
Message boards :
Questions and problems :
Private messages on boinc's projects
(Message 62527)
Posted 11 Jun 2015 by Richard Haselgrove Post: If I'm right, it's for PMs, period. Admins don't get any special treatment there. Once you've viewed the PM via the website - so it's no longer 'unread' - then the notice is cancelled. |
|
3457)
Message boards :
Questions and problems :
Private messages on boinc's projects
(Message 62520)
Posted 11 Jun 2015 by Richard Haselgrove Post: I believe that already happens as a 'notice' - and unless you have notice reminders turned off, that means a balloon pop-up too. But BOINC isn't in constant communication with project servers - most projects only check for new notices once every 24 hours. If you want to be notified more quickly than that, you can choose 'notify immediately by email' from the project community preferences, and then you get the full text of the PM in your mailbox (though you have to log in to your project account to reply) |
|
3458)
Message boards :
Questions and problems :
What is the currently recommended version of VirtualBox?
(Message 62464)
Posted 7 Jun 2015 by Richard Haselgrove Post: VirtualBox version recommendations are currently probably best left to the (still very few) projects which use VBox. But it's a good point to bear in mind for when the technology matures. |
|
3459)
Message boards :
Questions and problems :
GPU starts up while computer is in use
(Message 62459)
Posted 6 Jun 2015 by Richard Haselgrove Post: Maybe the 'rectangle comparison' routine from stackoverflow would work then. More than one way to solve a problem, if the will is there. |
|
3460)
Message boards :
Questions and problems :
GPU starts up while computer is in use
(Message 62456)
Posted 6 Jun 2015 by Richard Haselgrove Post: There might be an alternative approach which could be helpful. I run a Windows communications (messaging) program which has this configuration option: Show me as 'Busy' and block my alerts when I'm running a full-screen program or presentation settings are turned on. That's basically an automatic 'do not disturb when I'm busy'. 'Full-screen' should cover both films and gaming, and would be a lot more user-friendly than digging around in the system to locate the image name for each separate game or media player - and it should be a better solution for the timing problem too. Allow BOINC to be configured not to run GPU apps precisely when, and for as long as, the GPU is being called upon to perform at full performance as a graphics device. Although it may not have been expressed on the various message boards as a feature request, there's been low-level grumbling for years from people who say "I keep forgetting to re-enable GPU computing (or un-snooze the GPU) when I stop gaming". MSFullscreenChange "Fires when an element is displayed in full-screen mode, or when it exits full-screen mode." Maybe SHQueryUserNotificationState is what we're looking for. |
|
3461)
Message boards :
Questions and problems :
BOINC does not uses GPU. ATI GPU R600 (R38xx) does not support OpenCL (MilkyWay)
(Message 62439)
Posted 3 Jun 2015 by Richard Haselgrove Post: How can I fix the problem? By persuading the Milkyway project to update their server software. Or else, you can write an app_info.xml file to define the project's own application using the Anonymous platform mechanism. There are examples on the MIlkyway message board for Windows users: I haven't seen one pre-written for Linux, but the principle is the same. Ask at the project. |
|
3462)
Message boards :
Questions and problems :
[Ubuntu] Change location disk usage (SSD/HDD)
(Message 62418)
Posted 1 Jun 2015 by Richard Haselgrove Post: Maybe semantics, yes the tag was eventually 'removed', but was disabled since much much longer. Semantics again, but I think I'd say 'broken' rather than 'disabled'. I don't think the implications were thought through properly when BOINC's directory structure was split from a single combined directory into separate program and data directories: I rather doubt it ever worked properly after that, but not many people tried to use it. |
|
3463)
Message boards :
Questions and problems :
Snooze Bug?
(Message 62404)
Posted 31 May 2015 by Richard Haselgrove Post: I didn't pick up this impression from Gary's initial post, but it turns out that this problem is well known and much discussed at the project message board. The project developer expressed an intention to upgrade the apps by compiling against the new API on 24 March 2015, and reported the work completed on 27 April 2015: I think we can assume that the 11 May 2015 deployment doesn't have last year's bug. Perhaps more significantly, the underlying application doesn't checkpoint - ever. Apparently the BOINC implementation of the code was done by IBM for World Community Grid: is 'not checkpointing' typical of WCG sub-projects? |
|
3464)
Message boards :
Questions and problems :
4cpu process only using one cpu core
(Message 62399)
Posted 30 May 2015 by Richard Haselgrove Post: No button, I'm afraid. But you could paste your original question into New Nbody Version 1.50 |
|
3465)
Message boards :
Questions and problems :
4cpu process only using one cpu core
(Message 62396)
Posted 30 May 2015 by Richard Haselgrove Post: Are you writing as a user, or a project developer? Judging by the volume of complaints about N-Body for Windows on the Milkyway message boards, they are in urgent need of a developer - sadly, this isn't something an end-user can fix. But you could add a "me too" for Linux to the existing bug reports. |
|
3466)
Message boards :
Questions and problems :
Snooze Bug?
(Message 62395)
Posted 30 May 2015 by Richard Haselgrove Post: Thanks for the executables, Gary. Which gives us an earliest possible download date of 10 June 2014 (64dc838626764f9d7b5f104099dd11149b89b61a), but sadly tells us nothing about whether the 31 October 2014 bugfix was applied. |
|
3467)
Message boards :
Questions and problems :
Snooze Bug?
(Message 62385)
Posted 30 May 2015 by Richard Haselgrove Post: This is quite possibly caused by a bit of BOINC code called the 'Applications Programming Interface', or API. BOINC supplies this code, but each project compiles and links it into their science applications: it becomes part of the application, and is responsible for handling all communications between the application and BOINC. Including, in particular, listening for messages telling the application to suspend, and acting on them. There was a bug in the API, causing apps to (mis-)behave in the way you describe. We called them "zombie tasks", and the bug was fixed - very appropriately - on halloween last year: f0c39bdf5117d8f7dd5092033971d7f700bd22dc. So, any science app executable, at any project, deployed before November 2014 is likely to have this bug. It doesn't always show up: it's most likely to be a problem with applications that use GPUs. If those two criteria (date and GPU use) match, ask the project to update their API library and re-compile their application, drawing their attention to the code commit I've linked. |
|
3468)
Message boards :
BOINC Manager :
Newb here, Boinc not showing LHC@Home project I added.
(Message 62376)
Posted 29 May 2015 by Richard Haselgrove Post: Welcome. 'Connected to localhost' is normal and correct: BOINC consists of two parts - an invisible 'client' (which supervises the project work), and a visible 'Manager' (which lets you see, and if you wish control, what's going on). The message says that both components are present, working, and talking to each other. LHC/sixtrack doesn't actually need the VBox add-on to the basic BOINC framework, although other projects based at CERN do. That shouldn't matter. You need to be careful to distinguish between 'Projects' and 'Tasks' when talking about BOINC. Sixtrack should indeed have appeared on the Projects tab immediately after attaching, though there are reports on the sixtrack website at the moment that there may be a project configuration problem, which they're working on. On the other hand, tasks from sixtrack will only appear on your Tasks tab when they have been sent to your computer for processing - and sixtrack, in particular, regularly has no work available for hours or days at a time (and then it has a glut of work which has to be finished in a hurry). In particular, sixtrack has no tasks available to send as I type. While you're waiting for the attachment problem to be sorted out - you can read about it at Cannot Join New Host for Existing Account? - and for new tasks to be made available, you might with to explore the range of other science projects which use the BOINC framework. |
|
3469)
Message boards :
Questions and problems :
Is really Boinc run in idle priority? I think it is not.
(Message 62375)
Posted 29 May 2015 by Richard Haselgrove Post: There's no such thing as a single, "one size fits all", GPU task. For starters, there are two major GPU programming languages in use at BOINC projects - CUDA and OpenCL. Secondly, the nature of the calculation to be performed varies: some projects are largely integer-based, but others require double-precision floating point arithmetic (integer tasks are probably closer to the gaming experience the cards were designed for). Thirdly, different tasks (even within the same project) vary in the amount of data that needs to be transferred to and from the graphics card over the PCIe bus - a major bottleneck in general-purpose computing on GPUs: games only need to transfer data in one direction. And finally, different programmers approach their tasks differently, and can introduce their own variability. GPUGrid - Use CUDA (usually better than OpenCL w.r.t user experience) Have large tasks (potentially bad w.r.t. bus saturation) Seem to know how to program efficiently. Your experience at other projects may vary. |
|
3470)
Message boards :
Questions and problems :
Do projects get done faster with disk checkpoints set to 1 second?
(Message 62361)
Posted 28 May 2015 by Richard Haselgrove Post: You may have the best computer, but unless you have it on the best UPS as well, you're still vulnerable to power outages. Checkpoints are written to enable you to recover with minimal work loss after an outage. They do take (a very small amount of) time to write: projects *might* run slightly faster if you checkpoint less often (longer interval), but you'd lose correspondingly more work every time you reboot for maintenance. |
|
3471)
Message boards :
Questions and problems :
Malware Alert from Avast AV Software on Win 8.1
(Message 62300)
Posted 24 May 2015 by Richard Haselgrove Post: I can't find your report at SETI Beta, but never mind. Looking at the tasks your computer has been allocated recently, I think the file must have been setiathome_7.06_windows_intelx86__opencl_intel_gpu_sah.exe. I've run that file through virustotal, and got a few generic warnings. The Avast! engine check is reporting Win32:Malware-gen, which it is notorious for doing this with new program files (this one was first loaded for testing on 20 May 2015, and there won't be many testers with intel GPUs running it yet). The trouble is that many BOINC project application files behave in ways very similar to malware: they run without a user interface, and they send information from your computer to the internet. Well, the BOINC client does the actual sending, but some anti-virus programs still think they're suspicious. I would judge this to be a false alarm, but look at the whole virustotal report and decide for yourself. If you trust the SETI project to have taken all necessary precautions (and I do), then please continue with your testing. You will probably find you you have to temporarily disable Avast's internet shields while you download a fresh copy of the program, before you can process the tasks you've been allocated. |
|
3472)
Message boards :
Questions and problems :
Windows 10 preview and BOINC 7.4.42 (x64)
(Message 62299)
Posted 24 May 2015 by Richard Haselgrove Post: What does the Event Log say about scheduler work requests? The design behaviour for a freshly installed system (after attaching to a project) is to request 1 second of work for each of the available compute device types - so that's one second of work for CPUs collectively (not 'per CPU'), 1 second for NVidia GPU (if discovered), one second for AMD GPU (ditto), one second for Intel GPU (ditto), and so on. Those one-second requests will, at most projects, result in one task being allocated for each and every device. The project initialisation will also result in user preferences (such as number of CPUs allowed to run concurrently), application speed and task size estimates, etc. etc. being transferred to the new host. Once all this information is available, additional work requests will be issued to utilise any idle resources. At projects like SETI, the 5-minute backoff between scheduler contacts can slow this process down, but the 7-second backoff at Beta should be insignificant. If your report is saying that * multiple CPU tasks have been allocated * all application downloads have completed * all task data downloads have completed * active preferences allow all CPUs to be used * no other preference is inhibiting BOINC resource usage (check 'suspend BOINC when non-BOINC CPU usage is above ...', for example) and there is still an appreciable delay (more than a minute) between tasks being 'Ready to start' and actually starting, then your best bet is to send the event log to boinc_alpha. Otherwise, I think that measured, step-by-step, initialisation procedure is appropriate for 99.999% of users and new computers. |
|
3473)
Message boards :
Documentation :
update app_config.xml page on wiki
(Message 62294)
Posted 23 May 2015 by Richard Haselgrove Post: ow okey sorry didn't know that we used it for comment. I was used to # or & this kind of signs. That's OK, it makes sense when you get used to it. Note that optional sections in app_config.xml may span several lines, and also be nested, like [<app_version>
<app_name>uppercase</app_name>
[<plan_class>mt</plan_class>]
[<avg_ncpus>x</avg_ncpus>]
[<ngpus>x</ngpus>]
[<cmdline>--nthreads 7</cmdline>]
</app_version>]The whole <app_version> structure is optional, but if you want to use it, you must include the opening tag, an <app_name> tag, and the closing tag. You may want to use one or more of the optional lines as well. It's hard to convey all that detail with #, &, ;, ', //, or any of the myriad other ways of indicating a comment line: the [ ] technique is both concise and precise. |
|
3474)
Message boards :
Documentation :
update app_config.xml page on wiki
(Message 62291)
Posted 23 May 2015 by Richard Haselgrove Post: Please read the note at the head of the configuration page: Note: Wherever you see square brackets [ ] used in the examples, these are not used. They show additional options. When using the specific piece of code, delete the square brackets, or the option you try to use will be ignored. This applies to the app_config.xml file, just as it does to the cc_config.xml file documented in the first two major sections of the page - and indeed to all XML files documented in the Wiki. [...] is a conventional way of indicating optional sections when documenting computer code, and is widely used in other contexts as well. |
|
3475)
Message boards :
Questions and problems :
Malware Alert from Avast AV Software on Win 8.1
(Message 62289)
Posted 23 May 2015 by Richard Haselgrove Post: BOINC may have done the downloading - that's its job, that's what it's designed to do. The warning is about the file it was downloading, which appears to have been a recent update to a SETI@Home Beta application. If you participate in a Beta testing programme, please report errors, anomalies, warnings, suggestions for improvement etc. to the project concerned - in this case, the SETI@home/AstroPulse Beta Message boards. [The image only shows part of the file name, so it isn't clear whether it was a SETI@Home application or an Astropulse application that you were testing] |
|
3476)
Message boards :
Questions and problems :
GPU restarts from zero
(Message 62262)
Posted 20 May 2015 by Richard Haselgrove Post: Sounds like a timing issue with checkpoints. Project applications (both POEM and Milkyway) periodically write a 'checkpoint' file - a summary of everything done so far. They need to read this file back in every time they restart, to be able to carry on where they left off: without a checkpoint, they wind back to the beginning. In the early stages of running, before the first checkpoint, BOINC displays an estimate of pseudo-progress, but hasn't actually done enough real work to create a checkpoint and restart file. After Milkyway winds back to zero displayed progress, and starts again on the real work, a checkpoint will have been created and progress should continue. I imagine POEM will do the same, but may take longer to reach the first checkpoint. If you can, adjust your timing preferences so that the GPU tasks aren't interrupted so often - they will make better progress then. Interrupting a CPU task isn't as bad, because you can ask for 'Leave applications in memory' when paused - they can then restart anywhere, without needing the checkpoint file. That's because CPU tasks use the computer's main memory - there's usually plenty of that, and it can be swapped to disk if needed. GPU tasks use the internal memory on the video card: that is more limited, and may be needed by your real work, so it's cleared every time the GPU task pauses, whatever your 'leave in memory' setting. |
|
3477)
Message boards :
Questions and problems :
Resource sharing - Einstein@home hogging the system ...
(Message 62238)
Posted 18 May 2015 by Richard Haselgrove Post: I run all three of those projects, and a few more besides. I don't have any problems with resource shares, and I don't honestly believe that Einstein is deliberately doing anything to provoke this sort of behaviour. Instead, there are a number of factors which all interplay with each other. First, both CPDN and SETI have very relaxed deadlines, unusually so - Einstein's deadlines are more typical of BOINC projects in general. Second, SETI has a reputation for frequent outages, and people who run SETI have a tendency to request the maximum possible cache size - I don't know if this applies to you. And third, SETI has the busiest servers in the business - they are frequently unable to supply work at the precise second it is requested, even though work would have been available a couple of seconds earlier or a couple of seconds later. Einstein, by contrast, almost always has work available on demand - they are one of the most reliable projects. When BOINC decides that it's time to fetch more work, the relative availability of work does tend to skew the mix towards Einstein: BOINC will quite often request SETI work, receive none, and immediately request Einstein work instead. But that doesn't explain the urgent running of the tasks received. That's a combination of cache sizes, deadlines, and the fraction of the time that BOINC is running. So: Einstein deadlines are 14 days, and you want them to run at 10%. That means that they have to fit 33.6 hours of running into that fortnight: if there is ever more than 33.6 hours on Einstein work in the cache, you're in deadline trouble. If your cache settings total more than that, I suggest you turn them down to say 1 day maximum, and see if that fares any better. |
|
3478)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 62207)
Posted 17 May 2015 by Richard Haselgrove Post: SETI probably needed to be shut down manually, because they have multiple inter-dependent servers, and the network connections between them have to be re-started in the right order. This is just a single server, and it can carry on running by itself without a network connection. The first we'll know is when the network cable is pulled - probably while I'm typing this. :P |
|
3479)
Message boards :
Questions and problems :
Is 'deadline' GMT?
(Message 62195)
Posted 16 May 2015 by Richard Haselgrove Post: "If a task is still unstarted on your computer when the deadline is reached, then your local BOINC client will abort it." It's "auto abort" by the client, acting on its own initiative. The web site will still say 'no reply', until a separate, later, scheduler RPC to report the changed state (after all, the task might have started running a few seconds before deadline, and not been aborted at all). |
|
3480)
Message boards :
Questions and problems :
Is 'deadline' GMT?
(Message 62173)
Posted 15 May 2015 by Richard Haselgrove Post: The task deadline shown on a project's web page will always be in UTC (which is effectively the same as GMT). The deadline shown in BOINC Manager will be in the local time zone as set for the computer's clock - which you would normally set to be the local time zone for wherever you're using the computer. When a task is issued to your computer, the deadline is always a fixed number of hours or days from the time it's allocated. So it doesn't matter where you are in the world - Australians get just the same length of time from from allocation to deadline as residents of any other country. If a task is still unstarted on your computer when the deadline is reached, then your local BOINC client will abort it. But if the task has started, and run for even one second before being suspended, it won't be aborted automatically. |
|
3481)
Message boards :
BOINC Manager :
What is this: vbox_windows set to 0
(Message 62100)
Posted 9 May 2015 by Richard Haselgrove Post: Yes, the next VBox task started in full interactive mode under VirtualBox.exe, rather than the usual VBoxHeadless.exe. So that's what it does. And I've added it to Client configuration |
|
3482)
Message boards :
BOINC Manager :
What is this: vbox_windows set to 0
(Message 62099)
Posted 9 May 2015 by Richard Haselgrove Post: I've got one of those too, so I set it to 1 to see what happened. This happened: 09/05/2015 18:53:20 | | Config: open console window for VirtualBox applications But the console window didn't open. Maybe in an hour, when the next VBox task starts up... Edit - I found if (aid.vbox_window || boinc_is_standalone()) {
vboxlog_msg("Detected: Headless Mode Disabled");
|
|
3483)
Message boards :
Questions and problems :
Download problems for cuda stuff
(Message 62084)
Posted 8 May 2015 by Richard Haselgrove Post: This can sometimes be an anti-virus program thinking that the behaviour of the downloaded files is suspicious, and blocking them from running. |
|
3484)
Message boards :
BOINC Manager :
Why does the BOINC Manager window move across boots from where I had left it?
(Message 62039)
Posted 3 May 2015 by Richard Haselgrove Post: I nicely layout my BOINC Manager (snapped to left, almost maximized) and Event Log (snapped to right, with minimum width allowed, giving the rest of the screen to the main window) windows under Win 8.1 (but this has been happening in previous versions of BOINC Manager & Windows; I am just getting around to asking about it here). But, every time my machines get rebooted (or, I exit & restart BOINC Manager for whatever reason), the BOINC Manager window forgets where it was, and starts somewhere else. To achieve a screen layout like that, my starting point (using Windows 7, I don't have 8.1 available to test yet) would be to open both the main Manager window and the Event log, then use Window's "Show windows side by side" tool (right click on task bar and select). That gives two windows of equal width, sharing the entire viewport between them. To obtain the unequal widths, I dragged the left side of the event log as far to the right as it would go, and dragged the right side of the main window to the right until it seemed visually aligned. With a 1920 x 1200 screen, that led to registry values of (main) "WindowIconized"=dword:00000000 "WindowMaximized"=dword:00000000 "Width"=dword:00000529 "Height"=dword:00000488 "XPos"=dword:00000000 "YPos"=dword:00000000 (event log) "XPos"=dword:00000528 "YPos"=dword:00000000 "Width"=dword:00000258 "Height"=dword:00000488 The combination of (main) XPos 0, and (event log) XPos+width 1920, suggests that something is still trying to use one pixel more than I have, but BOINC seems to be happy and re-opens the two windows both where I left them. If I move the main Manager window partially off-screen to the left (either by dragging the entire window to the left, or by extending the left border as far as it can go), BOINC does resize/reposition it on startup so that the entire window is visible - and visible on the primary monitor, if you're using dual monitors. It also sets XPos 30 pixels in from the edge: I think this is sometimes done by Windows progammers to ensure that important elements like the Control box and File menu aren't obscured by the task bar, if the user happens to prefer the task bar docked left. That appears not to be a risk with Windows 7, which doesn't seem to permit a window to extend behind the taskbar under any circumstances, but it was certainly possible with older versions. Having tried to program some of these features myself when I was an early adopter of multiple monitors under Windows 98, I can tell you that it's b****y difficult to account for all permutations without ever losing a window to a position where it can't be dragged back fully onscreen. And Windows 7 doesn't have a Move function (correction: it does on 'Shift right-click'. How many people have found that?). So, on balance, I think the BOINC developers have made a reasonably good stab at adopting a policy which is safe under multiple versions of Windows. And perhaps the major fault is with Microsoft, for not aligning the hard edge when dragging a window border exactly with the edge-of-screen at XPos 0. |
|
3485)
Message boards :
Questions and problems :
CUTS DESKTOP POWER COMPLETELY
(Message 62032)
Posted 3 May 2015 by Richard Haselgrove Post: This can happen if your CPU cooling has failed completely - either a fan is broken/disconnected, or there's a build-up of dust preventing cool air reaching the CPU heatsink. In that case, it isn't BOINC itself shutting down the system, but the extra workload on the CPU when the project science application(s) start up. That can cause the CPU temperature to rise quickly, and - without cooling - trigger a safety shutdown to prevent damage. If the computer runs stably for long enough after restart for you to open BOINC Manager, try suspending computation, and see if it continues to work with BOINC open but no science applications running. That could indicate either a thermal problem or, as Claggy says, insufficient system power. |
|
3486)
Message boards :
Questions and problems :
data usage
(Message 62022)
Posted 2 May 2015 by Richard Haselgrove Post: BOINC does keep records of how much data it has transferred, if you want to check your suspicions, but it doesn't supply any easy tools for inspecting and analysing the history file. The data is in daily_transfer_history.xml: '<when>' is the number of days since 1st. January 1970, so today is day 16557, and I think <up> and <down> are the number of bytes transferred in each direction. Edit - I went and asked one of my machines. Excel 2000 said 'dunno', but Excel 2003 (with a bit of help) said 
|
|
3487)
Message boards :
Questions and problems :
Question about GPU use
(Message 62012)
Posted 1 May 2015 by Richard Haselgrove Post: BTW, all projects do seem to be downloading and running nVidia GPU tasks. However, the messages imply they are also attempting tasks which cannot be run. It's not actually trying to say that, though I can see how you might come to that interpretation. What has happened is that in the past, you were running an ATI card. You have requested work which that ATI card can process: as a result, you were sent both an application (program) and some tasks to process with it. The tasks are long since gone, but your computer still has (or had, until you deleted them manually) the application files, ready for the day when you re-install the ATI card and process more work with it. The messages simply say that BOINC 'remembers' that you have had the capability to run ATI programs: they do not say that BOINC is actually trying to use those programs, or to fetch tasks that would use them. It's untidy, I know: but if you really wanted to get rid of the messages, you would probably have to edit out all references to the former applications from client_state.xml: that's a difficult and delicate process, and I'd advise you not even to attempt it until you've got more experience with BOINC and how it manages its internal affairs. |
|
3488)
Message boards :
Questions and problems :
New box, R9-280x, Catalyst 14.12 Not using GPU
(Message 62001)
Posted 30 Apr 2015 by Richard Haselgrove Post: Oh, and the same thread does contain all the resources you need to bypass the dodgy server software - I think with all the typos sorted out in the end. |
|
3489)
Message boards :
Questions and problems :
New box, R9-280x, Catalyst 14.12 Not using GPU
(Message 62000)
Posted 30 Apr 2015 by Richard Haselgrove Post: Ill do that, although I got the R9 specifically to run MW@H. Wasn't aware of any issues. Scan the thread where you have a moment. The main resource they seem to be lacking seems to be human - BOINC administrators. The project 'scientists' mostly seem to be students, and I don't think they're allowed to have the keys to the server. |
|
3490)
Message boards :
Questions and problems :
New box, R9-280x, Catalyst 14.12 Not using GPU
(Message 61996)
Posted 30 Apr 2015 by Richard Haselgrove Post: Try asking at the Milkyway forum - there's a recent thread called AMD R9 290X does not receive any GPU work. I believe they haven't updated their server software in a while, and the current code can't handle the detection strings from the R9 range. |
|
3491)
Message boards :
Projects :
News on Project Outages
(Message 61984)
Posted 29 Apr 2015 by Richard Haselgrove Post: CPDN (Climate Prediction) Database downtime 30 April 2015 07:00 - 12:00 GMT |
|
3492)
Message boards :
GPUs :
GPU Missing
(Message 61891)
Posted 25 Apr 2015 by Richard Haselgrove Post: I can't remember which one's are often used by Boinc users. One of the free variants of VNC, I think. I was recommended to try TightVNC. http://sourceforge.net/projects/vnc-tight/ |
|
3493)
Message boards :
GPUs :
GPU Missing
(Message 61863)
Posted 25 Apr 2015 by Richard Haselgrove Post: You've answered your own question. The GPU will go missing every time you use RDP. |
|
3494)
Message boards :
Questions and problems :
Rules for specific GPUs
(Message 61824)
Posted 23 Apr 2015 by Richard Haselgrove Post: This is an acknowledged weakness in the original coprocessor design for BOINC (https://boinc.berkeley.edu/trac/attachment/wiki/WorkShop14/workshop_14.pdf, slide 52), which is going to be changed. Sometime, we don't know when. Until then, there's not much you can do, except set up two (or more) sets of configuration files, and write a script to switch between them. |
|
3495)
Message boards :
BOINC client :
CPU affinity mask, pretty please?..
(Message 61806)
Posted 21 Apr 2015 by Richard Haselgrove Post: BOINC doesn't do it, and probably never will do it. But Process Lasso can (I think). (BTW, I'm pretty sure my Core 2 Quads - two Q6600 and a Q9300 - are true 4-core chips, no HT involved) |
|
3496)
Message boards :
Questions and problems :
Why does one of my hosts have some projects' priorities stuck at their past values
(Message 61655)
Posted 17 Apr 2015 by Richard Haselgrove Post: So, the problem seems to be in processing/applying an RS change received from some projects. I doubt it matters whether the project change was initiated by BAM!, or directly by the user at the website. I'm going to be offline until tomorrow night, but I run three of the projects on your list - Einstein (which uses very old server code), and SETI, LHC (both of which use recent code) - so I can look into it when I get back. |
|
3497)
Message boards :
Questions and problems :
Why does one of my hosts have some projects' priorities stuck at their past values
(Message 61646)
Posted 16 Apr 2015 by Richard Haselgrove Post: Also, if somebody knows the local XML file this is stored, maybe I can edit it manually? I think the information flow should be: 1) Receive "sched_reply_[project url].xml". This file is hard to read in Windows - use WordPad instead of NotePad. You get a completely new one with every update - timestamp should be current. Not worth editing, but have a look to see that the contents 'look right' (compare with a project which has updated properly). <resource_share> should be 15-20 lines down, after <project_preferences> 2) Info copied to "account_[project url].xml". This one is only changed when it needs to be - look at the datestamp. This one would be worth editing if you can't get RS to transfer any other way - use Notepad. If you see something odd (and if it still looks odd when you compare it with a good one), post it here - but remove the line with <authenticator> for security purposes. 3) All info for all projects is combined into "client_state.xml" - but I think we should find what the problem is before getting into that one. The only other thing I can think of is to uninstall BOINC, remove all the data, and reinstall. But, I have so many really long PrimeGrid jobs that are finished 90%, yet will take weeks to finish the next 10%. And, PrimeGrid is one of those projects on this machine that is stuck... :-( I would hate to lose all that work... That's what I call the "sledgehammer and two short planks" approach to computer maintenance. Could we try and work out what's going wrong first, please? There might be a bug here we can get corrected - or at least learn something to add to the reply toolbox for the next person who encounters it. After that, you're welcome to take your hard disk out into the yard and re-educate it with extreme prejudice, if you still want to.... ;) |
|
3498)
Message boards :
Questions and problems :
Why does one of my hosts have some projects' priorities stuck at their past values
(Message 61638)
Posted 16 Apr 2015 by Richard Haselgrove Post: Because Resource Share can vary from project to project (it's the only one which is defined by BOINC centrally which is set 'per project': the others are all either part of the global set which BOINC itself keeps synchronised, or project-specific definitions), I'd guess you'd have to update every project before the RS transfer is complete. |
|
3499)
Message boards :
Server programs :
Description of the database
(Message 61635)
Posted 16 Apr 2015 by Richard Haselgrove Post: I don't know of any documentation, but the 'schema' files in http://boinc.berkeley.edu/gitweb/?p=boinc-v2.git;a=tree;f=db have the formal code needed to generate the tables and indexes. Be aware that some fields have been recycled and are no longer used for the purposes their names suggest... |
|
3500)
Message boards :
Questions and problems :
Why does one of my hosts have some projects' priorities stuck at their past values
(Message 61629)
Posted 16 Apr 2015 by Richard Haselgrove Post: OK, I can go along with that explanation - I'm not a BAM! user (as you can probably tell), so I'm not clear on the finer points. So, the end result with the problematic projects is that the project web sites display one value (100) for resource share, and BOINC Manager still displays a different value, even after you do an 'Update' so that the client contacts the project and gets any updated information? Similar trouble-shooting process. Taking only the problematic projects: do you have any other 'Venues' (groups of preference settings - default, home, school, work - aka 'host locations') set up? If so, is the RS set at 100 for all of them? Can you match the number shown in BOINC Manager with anywhere else? If you call up the 'Properties' page for the project in BOINC Manager, does the 'Host location' show what you expect? (Note that you can change the host location project-by-project, but only via the project website or possibly BAM! - there's no local override) |
|
3501)
Message boards :
Questions and problems :
Why does one of my hosts have some projects' priorities stuck at their past values
(Message 61627)
Posted 16 Apr 2015 by Richard Haselgrove Post: I think anything you set to 100 via BAM! would be called a "Resource Share" in BOINC terminology. Priorites (especially in the context of Event Log debug flags) are something different entirely. It sounds as if something is failing in the communication between BAM! and your computer(s). I suggest you draw up a complete list of the projects which have not updated their Resource Share as you expect, and submit it to Willy de Zutter via the BAM! forum: he might be able to identify a common factor (age of server software?) between the projects affected. |
|
3502)
Message boards :
Questions and problems :
Settings to not report tasks
(Message 61571)
Posted 14 Apr 2015 by Richard Haselgrove Post: Suspending network activity doesn't block remote management - it only stops internet activity such as uploads, downloads and contacting project servers. Perhaps the wording is misleading. There's no way to request new work from the project you're researching without reporting completed (i.e. uploaded) work - 'request' and 'report' are combined into a single scheduler interaction. The only way I know of is to suspend network activity - having first downloaded as much work as the project allows. You could possibly do something sneaky with the host network settings (re-direct the upload url to a non-existent IP address, perhaps), which would allow work requests to continue for a while - but even then, BOINC would notice that the uploads weren't getting through, and stop requesting new work after a while (a high ncpus, and a low max_concurrent, might put off that day as long as possible). |
|
3503)
Message boards :
GPUs :
Windows 8.1, ATI R9 270, no usable GPU
(Message 61560)
Posted 13 Apr 2015 by Richard Haselgrove Post: I am getting no usable GPUs found. I will try to give enough info to help. I've been crunching for 16 years and this is one of the first problems I have had. The usual, I'm afraid. That line says you made a service install: uninstall BOINC (you won't lose any data), and re-install it using the same folder locations - but visiting the 'Advanced' page on the way through, to uncheck the service install option. |
|
3504)
Message boards :
Questions and problems :
How does BOINC decide to ask for tasks???
(Message 61548)
Posted 12 Apr 2015 by Richard Haselgrove Post: With that many projects attached, I'd only enable that one flag to answer your opening question. It's very verbose - you'll have a lot to work through. Then, turn it off again, and you can play with the others if you have other questions later. Since you're running v7.4.42, you have the handy little 'Event Log diagnostic flags' dialog on Ctrl+Shift+F, so it's easy to pick and choose. |
|
3505)
Message boards :
Questions and problems :
How does BOINC decide to ask for tasks???
(Message 61544)
Posted 11 Apr 2015 by Richard Haselgrove Post: Well, there's a debug log flag that answers the question in the title: <work_fetch_debug> Work Fetch runs after a maximum pf 60 seconds (some events trigger an earlier cycle, but 'Allow new tasks' isn't one of them). You'll notice that all the projects contacted in the first few cycles have been at the request of the projects - they request the periodic callback, but it doesn't guarantee that the client needs more work from that project at that moment. If we had the debug data available, we could work through all the projects in turn, seeing where they came in priority order, whether they had applications for your hardware, whether they were allowed to fetch by your project preferences, whether they were subject to any of the various backoffs, etc. etc. - but the mass manual update has consigned that to history. Maybe next time? |
|
3506)
Message boards :
BOINC client :
Problem on starting client 7.5.0 x64
(Message 61477)
Posted 8 Apr 2015 by Richard Haselgrove Post: * Is BOINC Manager allowed through your firewall? Er, that would be 31416 (pi), unless they've changed it recently? |
|
3507)
Message boards :
BOINC client :
Problem on starting client 7.5.0 x64
(Message 61474)
Posted 8 Apr 2015 by Richard Haselgrove Post: Well, I downloaded and installed v7.5.0 (x64 version) over an hour ago, and it's run smooth as silk ever since. It might be worth Richie hunting through the .txt files in the BOINC directories, especially any with 'err' in the name. |
|
3508)
Message boards :
BOINC client :
Problem on starting client 7.5.0 x64
(Message 61473)
Posted 8 Apr 2015 by Richard Haselgrove Post: The icon you see in the tray - err, "notification area" - when BOINC is running doesn't come from boinctray.exe - silly name - but from boincmgr.exe |
|
3509)
Message boards :
GPUs :
[Linux] Intel GPU Ivy Bridge OpenCL
(Message 61368)
Posted 3 Apr 2015 by Richard Haselgrove Post: There have been some recent (but unsuccessful) attempts to run beigenet at Einstein@Home, where the project administrator is game to help with the attempt. BRP4 Intel GPU app feedback thread Read from that post to the end of the thread before you start. |
|
3510)
Message boards :
GPUs :
[Linux] Intel GPU Ivy Bridge OpenCL
(Message 61364)
Posted 3 Apr 2015 by Richard Haselgrove Post: Did anyone check through the rest of the Collatz pages? I see: On the front page, "For GPU processing: Intel HD 2500 or later embedded GPU with Intel OpenCL driver" On the applications page, plan class "(opencl_intel_gpu)" listed for three Linux platforms - 32-bit for Collatz Mini and Solo, and 64-bit for Mini only. Unfortunately, their apps page doesn't have the enhancement (like SETI Beta) which enables us to see which application versions are completely unproductive - zero GigaFLOPS. Finally, the server status page lets us see that tasks are available for all four task types, so - providing you have the right sub-projects enabled in your preferences - it should be possible to fetch work and find out whether the listed application types are in fact viable. |
|
3511)
Message boards :
BOINC client :
How to add project on headless Ubuntu?
(Message 61337)
Posted 1 Apr 2015 by Richard Haselgrove Post: You could use either The boinccmd command-line tool Controlling BOINC remotely |
|
3512)
Message boards :
Questions and problems :
Boinc not shutting down on exit from tray icon
(Message 61261)
Posted 29 Mar 2015 by Richard Haselgrove Post: BOINC consists of two programs - a Manager and a Worker. The tray icon represents the Manager. When you choose exit from the tray icon, the Manager will always shut down, but it's optional whether the worker shuts down as well. To change the behaviour to what you want, open the Manager and 1) From the Tools menu, select Options and check the lower box, 'Enable Manager exit dialog'. 2) From the File menu, select 'Exit BOINC Manager'. In the exit confirmation dialog that shows up, check both options. |
|
3513)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 61110)
Posted 20 Mar 2015 by Richard Haselgrove Post: While we're here (and off-topic, while SETI is still up), may I refer you to Some sad news I need to pass on... |
|
3514)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 61109)
Posted 20 Mar 2015 by Richard Haselgrove Post: Must have been - I've just seen it too, and reported it to boinc_alpha. David says he's fixed it, so I'm replying to test - and indeed, the error message has gone away. |
|
3515)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 61105)
Posted 20 Mar 2015 by Richard Haselgrove Post: Must have been - I've just seen it too, and reported it to boinc_alpha. It's visible here while I'm typing this reply. |
|
3516)
Message boards :
Projects :
News on Project Outages
(Message 61104)
Posted 20 Mar 2015 by Richard Haselgrove Post: It might still be worth doing. Prior to my testing the workround, my older computers failed both to contact the server to be allocated new work (which would include reporting old work), and to upload data files for completed work. With the workround in place, both functions are operating normally - a task is uploading as I type, and will probably report before I've finished. [especially since I just broke off to report a bug here] Back to GPUGrid - it's a recent change which has caused the problem, and it may still be a work-in-progress on the server: be prepared to the situation to be somewhat fluid. Yes, automatic report complete - though it was from one of my newer BOINC installations. |
|
3517)
Message boards :
Projects :
News on Project Outages
(Message 61086)
Posted 19 Mar 2015 by Richard Haselgrove Post: Can't upload files to GPUGrid...website is accessible There's a verified workround at https://www.gpugrid.net/forum_thread.php?id=3846&nowrap=true#40528 - use the certificate bundle from v7.4.36 Or simply update to the current recommended BOINC version for your operating system. |
|
3518)
Message boards :
Questions and problems :
BOINC doesn't utilize all CPUs on NUMA multi-node systems (>64 threads)
(Message 61074)
Posted 19 Mar 2015 by Richard Haselgrove Post: I have to disagree with you on the conclusion, because: There is device management at the BOINC client level, for GPUs and other co-processors, but even that is very rudimentary - for example, it isn't (yet) possible for the client to manage tasks automatically if they have different requirements, e.g. if a computer has a mixture of single-precision-only and double-precision-capable GPUs from the same manufacturer. I'm pretty sure that for the time being, the BOINC client treats all CPUs as identical replicas of each other, and delegates all core management to the operating system. I believe this discussion is worth to be checked by the BOINC devs. Agreed. They'll probably need to consider it as a feature request. |
|
3519)
Message boards :
Questions and problems :
BOINC WU on LinuxMint with NVIDIA GPU
(Message 60956)
Posted 14 Mar 2015 by Richard Haselgrove Post: The mystery, then, is why the tasks were allocated and downloaded in the first place. I've always thought that there's something strange about Milkyway's scheduler. |
|
3520)
Message boards :
Questions and problems :
BOINC WU on LinuxMint with NVIDIA GPU
(Message 60920)
Posted 13 Mar 2015 by Richard Haselgrove Post: Richard - Sorry to bust your bubble, but Milkyway and Collatz are both downloading WUs now. I have them both set to 0.25C + 0.5NV. Don't worry about bursting my bubble - I have grave doubts about MilkyWay's ability to comprehend and manage the plan_class facility, so it doesn't really surprise me that their 'opencl_nvidia' or 'opencl_nvidia_101' plan classes don't do what they say on the tin. I had a long discussion with them while they were trying to develop their multi-threaded CPU application (nbody), culminating in the revelation When I started releasing the format was to use gpu classes and I kept the convention. Since the idea is to get a gpu version of nbody out I thought it was a place holder for turning on that option. My comment was more of a warning not to be too concerned if your machine continued to suck on empty from that project - but if it works, that's great. Given the level of project support, it might have been a long slog tracking down any remaining problems. |
|
3521)
Message boards :
Questions and problems :
BOINC WU on LinuxMint with NVIDIA GPU
(Message 60917)
Posted 13 Mar 2015 by Richard Haselgrove Post: I doubt you'll ever get any GPU tasks from Milkyway. Their applications page indicates that their applications require support for the OpenCL programming language, and your message log implies that BOINC hasn't detected OpenCL support in your NVidia drivers. Roland might have some ideas on that, but if you are satisfied by running GPU projects with CUDA applications, I'd suggest you rest on your laurels before embarking on another round of driver updates. |
|
3522)
Message boards :
GPUs :
GPUs Working?
(Message 60916)
Posted 13 Mar 2015 by Richard Haselgrove Post: How 'remote' is your server, in networking terms? If you access it across a WAN, then RDP/VNC is probably your only option. But if you have LAN access, you could use RDP to enable Controlling BOINC remotely, and then BOINC's own inbuilt networking capability to verify that it's working. |
|
3523)
Message boards :
GPUs :
Using GPU without being logged (windows) - possible?
(Message 60908)
Posted 13 Mar 2015 by Richard Haselgrove Post: It is possible to use GPUs in service mode, provided both these conditions are true: 1) the Windows version is XP (nothing later) 2) the BOINC version is 6.12.34 (nothing later) You can always control (2), but I doubt that your "University's new PC lab" is still running XP. So, no, it's not possible, and that's the Microsoft security restriction Jord was describing. |
|
3524)
Message boards :
Questions and problems :
Desire new option for WU: PRIORITY in addition to SUSPEND and ABORT
(Message 60748)
Posted 8 Mar 2015 by Richard Haselgrove Post: Unfortunately, project administrators are only human. <rsc_fpops_est> can go wrong for a number of reasons, but the two commonest ones are 1) An administrator in a hurry - especially when testing new applications/tasks - can forget to make the necessary adjustment. 2) Some tasks are non-deterministic in nature. I'm approaching the finishing line after 324 hours, on a task which was (reasonably on the basis of the average for the app) initially estimated at 5 hours. It also had a deadline of 28 February, but hey - who's in a hurry? But my resource shares are such that BOINC showed no tendency to switch away from the task once started. I think if I was trying to nurse a long task to completion, I'd use a combination of Resource Share and NNT for other projects to ensure that BOINC gave that project sufficient priority. |
|
3525)
Message boards :
Projects :
News on Project Outages
(Message 60703)
Posted 6 Mar 2015 by Richard Haselgrove Post: Climate Prediction (CPDN) It's not currently possible to download the data files for new CPDN tasks. Staff are investigating a possible hardware failure. More news later. |
|
3526)
Message boards :
Projects :
News on Project Outages
(Message 60567)
Posted 28 Feb 2015 by Richard Haselgrove Post: And this BOINC-dev site was practically inaccessible this morning. I wonder what the common factor is. Linux servers falling over? All running BOINC server software? External DDOS attack? |
|
3527)
Message boards :
Questions and problems :
BOINC not downloading tasks as directed
(Message 60538)
Posted 25 Feb 2015 by Richard Haselgrove Post: I also discovered that using the web interface to set the cache to 3 and 2 will not be picked up by BOINC until I hit the clear button again. Apparently when the 5,5 setting was picked up it became another local override. The setting to 3,2 was never pick up even after the project it was set on did an update AND after I forced an update several times. Is this the intended behavior? It seems wrong. "Intended" might be a bit wishful, but yes - that's how it's been written to work. If you make any change via the Manager graphical interface, the complete current set of values becomes fossilised in amber and completely overrides the web-based preferences. Use the web to control BOINC, or use the Manager to control BOINC. Don't even try to mix'n'match. |
|
3528)
Message boards :
Projects :
News on Project Outages
(Message 60507)
Posted 24 Feb 2015 by Richard Haselgrove Post: Climate is downloading files but I can't upload completed work to them Yes, the current disk space at the British Atmospheric Data Centre (BADC) is full. CPDN project staff have been notified, but they will need to persuade BADC staff to do some housekeeping before the storage facility can accept more uploads. Meanwhile, the CPDN regional models should upload results to their respective regional upload servers without problems. |
|
3529)
Message boards :
Questions and problems :
Incorrect CPU threshold
(Message 60469)
Posted 22 Feb 2015 by Richard Haselgrove Post: That was a couple of days later, into December, so I assumed he meant January. But it hasn't happened yet. Thanks - I knew about the migration from trac tickets to git issues, but didn't know the urls (they may not have been there the last time I tried to look - it was a while ago, when the announcement was made). But I wasn't looking there for a resolution, just to show Yacob the state of play. I'd have seen the code being checked into git itself if it had been - but it hasn't. |
|
3530)
Message boards :
Questions and problems :
Incorrect CPU threshold
(Message 60462)
Posted 22 Feb 2015 by Richard Haselgrove Post: I wrote to the boinc_alpha bug reporting mailing list on 30 November 2014, under the title "Max_concurrent, work fetch, and idle resources": ... some users might set max_concurrent for a project which was known to place heavy resource (memory, disk?) demands on the system, and intend to donate the remaining resources to light-weight projects like, as it happens, NumberFields and SIMAP. Should there be a special work_fetch for the idle resource under these circumstances? David Anderson replied: I understand the problem. That was a couple of days later, into December, so I assumed he meant January. But it hasn't happened yet. https://boinc.berkeley.edu/trac/ticket/1373 |
|
3531)
Message boards :
Questions and problems :
combine Cross Project IDentifiers
(Message 60438)
Posted 20 Feb 2015 by Richard Haselgrove Post: Nobody else can combine the statistics for you, but you can combine them for yourself. First, the display name 'Brett Collins' is ignored. What matters is your email address. You have to use the same email address at each project for the CPIDs to line up. If you used a different email address at any project in the past, you can change it to your current preferred email, using your account page at that project - that won't lose any of your old statistics, or need any changes in your BOINC installations. Second, you have to have at least one computer which does work for multiple projects. It can be one computer on all projects, or multiple computers each of which work on different pairs of projects. Just so long as there's no project isolated with a computer that doesn't link anywhere else. If you do that, the CPIDs should converge over the next few days, but you'll have to be contacting the project servers over that time - that might be difficult with some of the projects you've listed, which have repeated downtime. If any project server is inactive, you won't be able to line up that project's CPID until it comes back online. |
|
3532)
Message boards :
Documentation :
Some broken/unupdated links
(Message 60327)
Posted 15 Feb 2015 by Richard Haselgrove Post: All your links are badly formed: I presume you mean Installation: http://boinc.berkeley.edu/wiki/Installing_BOINC Web resources: http://boinc.berkeley.edu/ So, what exactly were your questions? |
|
3533)
Message boards :
Questions and problems :
CPU is 2% and not 100%
(Message 60211)
Posted 10 Feb 2015 by Richard Haselgrove Post: is there a command to force a connection or a download of new files? The boinccmd command-line tool |
|
3534)
Message boards :
Projects :
News on Project Outages
(Message 60165)
Posted 7 Feb 2015 by Richard Haselgrove Post: The GPUGrid Nividia work units are again available. Though scarce, coming available one at a time as previous crunchers complete their 'step' in the calculation and return their results for regeneration as a new WU. |
|
3535)
Message boards :
BOINC Manager :
Console commands for boincmgr
(Message 60137)
Posted 5 Feb 2015 by Richard Haselgrove Post: Having said that (and agreeing that the first step has to be enabling remote_hosts.cfg or equivalent on the fast host to be monitored, and restarting the BOINC client to pick it up - maybe just reading config files is enough), I've just confirmed that the command line "C:\Program Files\BOINC\boincmgr.exe" /m /n Fast /p password (invoked via a Windows shortcut) opens a second instance of the Manager on my slow machine, and displays the activity of the Fast machine, obviously using the default RPC port number. The original Manager instance, monitoring the local BOINC client, is still running. |
|
3536)
Message boards :
Android :
BOINC client not working on Android 5.0
(Message 60134)
Posted 5 Feb 2015 by Richard Haselgrove Post: Reposting from Einstein@Home's Technical News: We just added a new Android app version 1.46 for BRP4 tasks (search for binary radio pulsars in Arecibo data), specifically for Android 5 devices. |
|
3537)
Message boards :
Android :
BOINC client not working on Android 5.0
(Message 60129)
Posted 5 Feb 2015 by Richard Haselgrove Post: The 7.4.41.apk was built on 22 January, but I'm not familiar with Android distribution channels. |
|
3538)
Message boards :
Questions and problems :
Win 7 + BOINC + seti@home
(Message 60054)
Posted 30 Jan 2015 by Richard Haselgrove Post: Then I must be the 0.0016257099% case.... I couldn't do an upgrade, because I was already at v7.4.36 and there's nothing newer yet, so I: Checked boinctray.exe was running Stopped and exited BOINC, Client and Manager (tray still running) Uninstalled BOINC (tray not running) Installed BOINC (tray not running) Exited Installer (tray not running) Started Manager (tray not running) Verified Client and apps running Still no sign of boinctray. Note - (1) I didn't install as service, and (2) I didn't allow the installer to start the Manager. Either of those might have changed the outcome - in fact, I'll try 2 now. BRB Edit - yup, (2) makes a difference. The installer starts boinctray.exe IFF you ask the installer to start BOINC Manager. 99.9983742901% of users must do that. |
|
3539)
Message boards :
Questions and problems :
Win 7 + BOINC + seti@home
(Message 60047)
Posted 29 Jan 2015 by Richard Haselgrove Post: Have you done a full reboot (not sleep or hibernate) since installing v7.4.36, to get boinctray.exe running again? |
|
3540)
Message boards :
Projects :
News on Project Outages
(Message 60040)
Posted 29 Jan 2015 by Richard Haselgrove Post: LHC@Home have got their upload problems sorted out. |
|
3541)
Message boards :
Projects :
News on Project Outages
(Message 60039)
Posted 29 Jan 2015 by Richard Haselgrove Post: LHC@Home 1.0 ('LHC classic') have filled their upload server and can't accept any more completed work. Since work already completed can't be validated and cleared from the server until wingmates can upload and report, they're probably stuck in gridlock until somebody at CERN notices and takes the server offline to clean it up or add extra storage capacity. I've sent a PM to the project scientist, but this is an administration problem, rather than a science problem - and the CERN IT administration team don't make themselves available via BOINC tools. If anybody knows a backroom contact at CERN, please pass on the message. |
|
3542)
Message boards :
Projects :
News on Project Outages
(Message 60029)
Posted 28 Jan 2015 by Richard Haselgrove Post: Einstein@Home is back up - Oliver says: I updated 19 servers and gateways in two data centers in the past seven hours, fixing the Ghost CVE, upgraded kernels, updated our master/slave databases and deployed pending package updates. |
|
3543)
Message boards :
Projects :
News on Project Outages
(Message 60025)
Posted 28 Jan 2015 by Richard Haselgrove Post: Einstein@Home "[Later] today we're going to shutdown the project for at most 1-2 hours to deploy important security fixes on our infrastructure. 28 Jan 2015" Looks like the shutdown has already taken place, but they're having difficulty bringing it back up - all services are disabled or showing errors. |
|
3544)
Message boards :
The Lounge :
What software does this forum run on?
(Message 60024)
Posted 28 Jan 2015 by Richard Haselgrove Post: Is this forum custom made or is it well known forum software, maybe with some tweaks? In other words, it's bespoke software, specially written to support BOINC projects. It has some special features not found elsewhere, like links to the BOINC databases of scientific research undertaken, and - while it has some features similar to social media sites - it probably has fewer social features than most other modern forum software. It's free and open source, so there is nothing to stop you using it separately from the rest of the BOINC server software (though you would still need the BOINC MySQL database layer) - but you might find it was as hard to remove the science linkages as it was to configure an alternative forum. |
|
3545)
Message boards :
Projects :
News on Project Outages
(Message 60004)
Posted 27 Jan 2015 by Richard Haselgrove Post: As in the CPDN thread: The scheduled [CPDN] downtime has been pushed back a week and will now be on Thursday 5th and Friday 6th February. |
|
3546)
Message boards :
Projects :
News on Project Outages
(Message 59986)
Posted 27 Jan 2015 by Richard Haselgrove Post: CPDN Scheduled downtime 29-30 January 2015 Jonathan Miller says: I have to schedule some project downtime this week, 29 & 30 Jan 2015. |
|
3547)
Message boards :
Questions and problems :
"Use GPU when the computer is used" not working anymore
(Message 59935)
Posted 23 Jan 2015 by Richard Haselgrove Post: So I suspect it might be a problem with the PrimeGrid or Collatz project applications not responding to a suspend request from BOINC. Jerome has mentioned that the applications he is experiencing difficulty with are Collatz OpenCL and PrimeGrid PPS. These applications (OS X, OpenCL, AMD GPU) were deployed: Collatz Solo: 11 Feb 2014 Collatz Mini: 27 Apr 2014 PrimeGrid PPS: 16 Aug 2014 These apps may have been compiled during the period when the API bug fixed in commit f0c39bdf5117d8f7dd5092033971d7f700bd22dc (31 Oct 2014) was active - IIRC, that commit was made in response to a similar 'not stopping when directed' bug in the SETI Astropulse OpenCL applications. I can't remember whether projects were formally advised to re-compile their applications against the updated API after that commit, but it might be worth considering. |
|
3548)
Message boards :
Questions and problems :
No tasks being drawn from newly added projects
(Message 59929)
Posted 23 Jan 2015 by Richard Haselgrove Post: Because we Lunatics recommend that users free a CPU core when running OpenCL GPU apps He's talking about SETI/Astropulse. There's no way he can be "having a WU running on the CPU to assist the GPU", to quote him exactly. Even if he had an application running on the CPU, it couldn't "help" the GPU - I don't know of anyone who has (yet) written that sort of bi-modal application to make separate use of CPU and GPU so that one resource 'helps' the other resource. The nearest we know of is the common advice for OpenCL apps (like SETI/AP) to limit the number of apps running on the CPU by one, or even to one on a dual-core CPU, to leave some resources spare to support the GPU. He might be meaning that. |
|
3549)
Message boards :
BOINC client :
BM Not Adhering to app_config.xml Settings for CPU and GPU Usage
(Message 59918)
Posted 22 Jan 2015 by Richard Haselgrove Post: I think we've had this one before. '<user_friendly_name>' is not a valid tag specified in Application configuration. |
|
3550)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 59883)
Posted 20 Jan 2015 by Richard Haselgrove Post: Green buttons do an action. I thought we'd been through all that some years ago, and got the distinction between 'buttons' and 'links' sorted out. With that in place, do we need a colo(u)r distinction on top? Can anyone dream up a situation where a green link, or a blue button, might be called for? |
|
3551)
Message boards :
BOINC Manager :
Suggestions For New Or Updated Features
(Message 59666)
Posted 15 Jan 2015 by Richard Haselgrove Post: Otherwise I'll look forward to BM v7.4 coming out - and let's hope they get a new designer to code something a bit better for next year... :) BOINC Manager v7.4 has been out for a while now - v7.4.26 released 11 November 2014, v7.4.37 released 1 January 2015 (BOINC version history) And with it, you get a shiny new BOINC client as well - that's the bit that does all the work and processes all those XML files (if needed). |
|
3552)
Message boards :
Questions and problems :
stop one specific project on schedule, keep others running
(Message 59642)
Posted 14 Jan 2015 by Richard Haselgrove Post: I'm pretty sure QuakeCatcher is implemented as a 'non-CPU-intensive task' - in which case, a new feature introduced last week will do what you request. <dont_suspend_nci>0|1</dont_suspend_nci> But do note that the feature isn't implemented yet in current versions of BOINC - you will need to wait for the next round of test releases, and your feedback will be useful. |
|
3553)
Message boards :
Questions and problems :
Just Received Virus Warning About BOINC.EXE
(Message 59615)
Posted 13 Jan 2015 by Richard Haselgrove Post: BOINC.exe just tried to download trojan.gen2. This particular file has already been discussed in a Climate Prediction message board thread, although unfortunately the poster mis-judged which thread to discuss it in. Have a look in Scheduler process down? |
|
3554)
Message boards :
Questions and problems :
Resource shares aren't updated
(Message 59538)
Posted 10 Jan 2015 by Richard Haselgrove Post: It looks pointless, repeating the same sum for each separate project: 10/01/2015 23:03:38 | AQUA@home | [prio] -0.000000 rsf 1.000000 rt 0.000000 rs 9974.614504 Many of those are dormant (those with prio zero). rsf look to be in the ratio of my resource shares, but if they're supposed to be "resource share fraction" (as the variable name suggests), shouldn't they add up to unity? |
|
3555)
Message boards :
Questions and problems :
Resource shares aren't updated
(Message 59536)
Posted 10 Jan 2015 by Richard Haselgrove Post: Any idea then what the "rs" stands for in priority_debug? It seems to be 'rec_sum' 642 if (log_flags.priority_debug) {
643 msg_printf(p, MSG_INFO, "[prio] %f rsf %f rt %f rs %f",
644 p->sched_priority, p->resource_share_frac,
645 p->pwf.rec_temp, rec_sum
|
|
3556)
Message boards :
Questions and problems :
Resource shares aren't updated
(Message 59533)
Posted 10 Jan 2015 by Richard Haselgrove Post: I'm still looking for which debug flag shows the RS values. Anyone want to chime in? I couldn't find one either, but the output from 'boinccmd --get_project_status' gets it (to six decimal places, plus a lot of other guff besides. It would take a lot of faffing around with findstr or grep to get both the RS and the project name in one go) |
|
3557)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 59398)
Posted 7 Jan 2015 by Richard Haselgrove Post: ... that storage space has filled up, and that is the primary reason that the AP splitters are working so slowly...nowhere to go with the work. Since MB work occupies the same fanout directories as AP work, that will affect MB splitting as well. But at least we can crack on with returning MB work for assimilation, and thus recycle the limited space available. |
|
3558)
Message boards :
Questions and problems :
7.4.27 not downloading tasks if resource share 0
(Message 59304)
Posted 5 Jan 2015 by Richard Haselgrove Post: v7.4.36 has been promoted to 'recommended' - the standard download page will show it. |
|
3559)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 59280)
Posted 3 Jan 2015 by Richard Haselgrove Post: Got word back from Matt that he believes these are datasets which have already had the AP work split and sent, but not the MB work. He may be mistaken. I'll send him a copy of my processing records. Tape First processed Last processed JobCount 01se12ab 06-Nov-2012 23-Nov-2012 169 03ja12ai 21-Apr-2012 11-Jun-2013 72 04jl12ac 17-Sep-2012 30-Oct-2012 237 06my12ab 03-Aug-2012 19-Aug-2012 240 15fe13ad 28-Apr-2013 18-May-2013 213 20au12af 18-Nov-2012 26-Nov-2012 95 24no12aa 15-Dec-2012 11-Jun-2013 220 26fe13ae 29-Apr-2013 13-May-2013 143 28jn12ab 20-Sep-2012 28-Oct-2012 136 30jn12ad 04-Apr-2013 06-Jun-2013 270 03mr13aa 02-May-2013 18-May-2013 190 04ja13ad 13-Feb-2013 01-Apr-2013 183 17dc12ad 31-Jan-2013 15-Feb-2013 157 18oc12ac 02-Mar-2013 17-Mar-2013 208 21my12aa 23-Jan-2013 31-Jan-2013 118 22oc12aa 14-Mar-2013 28-Apr-2013 123 24jn12ae 30-Mar-2013 20-Apr-2013 124 27au12ab 12-Nov-2012 22-Dec-2012 204 27fe13aa 17-Apr-2013 19-Apr-2013 142 29ja13aa 19-Apr-2013 29-Apr-2013 153 |
|
3560)
Message boards :
News :
BOINC 7.4.36 released to the public
(Message 59275)
Posted 3 Jan 2015 by Richard Haselgrove Post: What happened on the next cycle? 1/3/2015 4:32:45 AM | Milkyway@Home | [work_fetch] REC 29009.084 prio -1004.307 can't request work: scheduler RPC backoff (15.66 sec) [1] You still have work at this point [2] Not ready to re-contact MilkyWay yet [3] Should be ready to fetch the next task now |
|
3561)
Message boards :
The Lounge :
Part of linux in BOINC
(Message 59259)
Posted 2 Jan 2015 by Richard Haselgrove Post: Your best bet would be one of the statistics aggregation sites like BOINCstats. Try this for starters: http://boincstats.com/en/stats/-1/host/breakdown/os/ Note that most of these sites will just give you cumulative figures to date: it may be harder to spot trends over time. |
|
3562)
Message boards :
Questions and problems :
How can I find out whether the graphic card is being used for computation?
(Message 59250)
Posted 2 Jan 2015 by Richard Haselgrove Post: One thing I would check, is to make sure that you have the latest drivers for both your video cards ... "The latest" video driver is not always good advice. In the specific case of the NVidia 8600 GTS running under Windows Vista, the latest driver currently available is 340.52, which is known to have an OpenCL bug: we're waiting for a promised hotfix. Erish, don't upgrade the NVidia machine: what you have already is plenty good enough. |
|
3563)
Message boards :
The Lounge :
That thread there that you want to post in but really hate it when other people do it in other threads..
(Message 59093)
Posted 30 Dec 2014 by Richard Haselgrove Post: I made this thread because of the likes of this. Two people in one thread continually quoting the whole shebang and adding their ditty to the bottom. OK, as 50% of the people doing that, I can take a hint. Artistic Impression: nul points Technical Merit: not assessed, because I've failed already. There now appear to be about one and a half people with a brain engaged with either BOINC or SETI, and I was trying to bring one of them up to speed. Clearly not wanted. I think the general conclusion is clear: SETI has completed its task of searching for intelligent life on planet earth, and proved conclusively that there is none. So, time to euthenise the project, and wind up all the message boards. Goodbye. |
|
3564)
Message boards :
Questions and problems :
BOINC credit across projects not consistent?
(Message 59091)
Posted 30 Dec 2014 by Richard Haselgrove Post: The http://boincstats.com/en/stats/-1/cpcs shows GPUGrid giving out WU credits per second at the highest rate of participating projects. Sorry, didn't see your edit. Forget it - it's meaningless. The value 'credit_per_cpu_second' is (was) calculated by the project servers, and exported in the stats tables I was writing about this morning. It was removed from the server code in April 2010 (changeset 2536797068de70d37b1d5b5d64e160720b3f8430), and as Joe Segur wrote at SETI in January 2012 (message 1189741), Unfortunately, BOINC is no longer updating the credit per CPU second value. The update was removed from BOINC source code in April 2010, (changeset as above). The change would have been implemented quickly [at SETI], so any cpcs values for this project reflect processing done before then. There may be a few projects where the servers are still using older code so the values are current, but can only be meaningful at a project which has no GPU apps. |
|
3565)
Message boards :
Questions and problems :
Questions about "Show host average" on Statistics page of BOINC Manager and 3rd party reporting
(Message 59089)
Posted 30 Dec 2014 by Richard Haselgrove Post: OK, here you go. http://boinc.berkeley.edu/gitweb/?p=boinc-v2.git;a=blob;f=lib/util.cpp;h=6ee94fb745b866bf4821c3a62fe52a583e404f01;hb=HEAD#l229 That's the current code - function update_average() returns an intermediate value called 'weight' - that's your RAC. Note that this code is executed entirely on the server: most commonly when a task is returned and validated. If no tasks are being validated, then - to a first approximation - credit and RAC are not updated. Note also that validation is carried out asynchronously: * You report a completed task and drop the connection * The server validates the task (if possible - requires a running server and a wingmate) * The server calculates the new values for credit and RAC * You contact the server again, to request or report work. * the server reply contains the updated values calculated while you were disconnected * Your client retrieves these updated values from the reply, and updates the statistics graphs. The transfer of credit and RAC values to external stats sites is also asynchronous. The server (assuming it's running properly) periodically generates a complete database dump of all (host, user, team) credits: usually once every day, although projects can set their own interval. You can check the current state of the Asteroids stats export at http://asteroidsathome.net/boinc/stats/ - it should match what your account pages showed for your credit and RAC at or about 29-Dec-2014 14:50 (or up to 24 hours before you read this message). There are the files which are used to generate the tables and graphs used by every external stats site - but since it takes a finite amount of time to notice the availability of a new set of files, download them, decompress them, and process them into their own statistics database. During this process, a slow site may still show yesterday's values, while a fast one shows today's - but that should be the only difference (E & OE). the client doesn't properly calculate decay That's where the confusion has arisen. The client doesn't ever calculate credit, RAC, or decay - that's only done on the server. The client only calculates REC, as we've discussed in another place. |
|
3566)
Message boards :
Questions and problems :
Problems caused by BOINC + Virtual Box installation failure
(Message 59084)
Posted 30 Dec 2014 by Richard Haselgrove Post: There are two (separate) Windows commands which have helped me in situations like this. netsh winsock reset catalog netsh int ip reset Enter either one at a command prompt with administrative rights (no special procedure for Win XP: requires 'elevated privileges' or 'run as administrator' on Vista and later). They also require the machine to be re-booted after use. IP addresses starting 169 are Windows' free-floating autoconfiguration addresses: they imply the TCP/IP stack has failed to acquire an IP address from the DHCP server responsible for the network segment - usually your router on a home network, or a server on a business network. Do check the obvious (network cable unplugged) before starting on any time-consuming tests. |
|
3567)
Message boards :
Questions and problems :
Questions about "Show host average" on Statistics page of BOINC Manager and 3rd party reporting
(Message 59056)
Posted 28 Dec 2014 by Richard Haselgrove Post: The expansion of the acronym RAC is Recent Average Credit. So 'averages' in BOINC-speak are the same as RAC, but possibly reported (and recorded) at different times by the various aggregation sites. Boinc Stats-RAC is also an average, but with a different decay function. The standard BOINC RAC decays with a half-life of 7 days. |
|
3568)
Message boards :
Questions and problems :
Ubuntu Linux with GeForce GT 630 - No usable GPUs found
(Message 58901)
Posted 23 Dec 2014 by Richard Haselgrove Post: There's what sounds like an authoritative walk-through of a solution at Einstein: BOINC doesn't detect my GPU [Solved] |
|
3569)
Message boards :
Questions and problems :
BOINC credit across projects not consistent?
(Message 58883)
Posted 23 Dec 2014 by Richard Haselgrove Post: Try here http://boincstats.com/en/stats/-1/cpcs BOINCstats is an independent third-party stats site. The official BOINC record - the site that generates the "projects you are participating in" table on the account page at every project - is http://boinc.netsoft-online.com/e107_plugins/boinc/get_cpcs.php Note the weasel words in the rubric at the top: "If a value in Row A and Column B is greater than 1, that indicates that project A is awarding more credit per cpu second than project B." "Per cpu second" is a meaningless measure for any application which runs on a GPU or other coprocessor, and at the point the whole pack of cards falls over. |
|
3570)
Message boards :
Questions and problems :
BOINC 7.4.27 install problem
(Message 58810)
Posted 21 Dec 2014 by Richard Haselgrove Post: Also check stderrgui.txt in the same place. |
|
3571)
Message boards :
Questions and problems :
BOINC 7.4.27 install problem
(Message 58808)
Posted 21 Dec 2014 by Richard Haselgrove Post: Yes, it worked for me too - leaving this screenshot behind for future reference, while you and Jord work on the wxWidgets issue. Have fun.  (note the 'target' box is just too short to show the whole line, and can't be resized. It starts like the other one) "C:\Program |
|
3572)
Message boards :
Questions and problems :
BOINC 7.4.27 install problem
(Message 58804)
Posted 21 Dec 2014 by Richard Haselgrove Post: "c:\program files\boinc\boinc.exe" --detach_console I have mine in C:\BOINC and C:\BOINCdata to avoid spaces, but I can shift the programs at least. Hold on while I test. |
|
3573)
Message boards :
Questions and problems :
BOINC 7.4.27 install problem
(Message 58801)
Posted 21 Dec 2014 by Richard Haselgrove Post: I tried including the quotation marks; that leads to Windows complaining about the file name being invalid Could you copy and paste the entire contents of the 'target' box here, please, on a line by itself, so we can read it the same way that Windows does without the surrounding narrative? |
|
3574)
Message boards :
Questions and problems :
BOINC 7.4.27 install problem
(Message 58798)
Posted 21 Dec 2014 by Richard Haselgrove Post: I take it you're stuck on the "c:\program files\boinc\boinc.exe --detach_console" instruction. As you can see, it runs from Program Files, not from Program. So check the link you added to the shortcut to the start up of BOINC. Right-click on the BOINC shortcut and check the "Target" and "Start in" lines. Under these circumstances, I suspect that the 'Target' and 'Start in' may need to be "c:\program files\boinc\boinc.exe" --detach_console "c:\program files\boinc" respectively, including the quotation marks as shown. There are bits of DOS-like behaviour hidden under the Windows hood, and they don't like that space in the path name. The quotation marks bind the whole string together into a command line - so include the (x86) inside the quotation marks if that's where your boinc files live. An alternative way to create a desktop shortcut is to locate the BOINC application on disk using Windows Explorer. Select (highlight) just the BOINC application (will be shown as boinc.exe if you have have file extension display turned on - but just boinc by default). Then, drag the file to the desktop while holding down both the SHIFT key and the CTRL key. The mouse pointer will display a curly arrow, and the tooltip "Create link in Desktop". Once the basic shortcut has been created, use the 'properties' dialog to add the [space] --detach_console parameter at the end of the target box. |
|
3575)
Message boards :
Questions and problems :
Can't set proxy
(Message 58711)
Posted 16 Dec 2014 by Richard Haselgrove Post: See When asking for help This sounds as if the BOINC Core Client isn't running (I've just tried shutting down a Windows client but leaving the Manager running - the 'Connect via HTTP proxy server' checkbox is disabled in Windows too under those circumstances, but is available once the manager has connected to a running client). Neither BOINC Manager nor boinccmd sets the proxy server configuration directly - they simply pass a message to tell the client to adjust the configuration files. The message is passed via TCP/IP, but to localhost, so shouldn't require the internet proxy to be active in advance. |
|
3576)
Message boards :
Projects :
News on Project Outages
(Message 58622)
Posted 12 Dec 2014 by Richard Haselgrove Post: SIMAP has distributed all workunits from its final batch of BOINC work. Once they (and any necessary resends) have been returned, SIMAP will cease to be a BOINC project. |
|
3577)
Message boards :
Questions and problems :
[Windows] BOINC & AMD TurboCore
(Message 58604)
Posted 10 Dec 2014 by Richard Haselgrove Post: Is this a bug or is this supposed to happen intentionally? It might be a total power consumption restriction, rather than a temperature limitation. BOINC itself places very little demand on the CPU, but the science applications that BOINC manages will, if allowed, drive all possible CPU cores at full power. Running a CPU core at 3.5GHz instead of 2.5GHz will draw significantly more power, and the system designers may have deliberately prevented the battery and charging systems from being over-driven. I'm not familiar with the AMD TurboCore specifications, but the equivalent Intel TurboBoost system allows the highest clock speed only when a single core is loaded: the maximum operating speed drops as loads are added to additional cores. |
|
3578)
Message boards :
Questions and problems :
boinccmd how to check if it works?
(Message 58496)
Posted 7 Dec 2014 by Richard Haselgrove Post: this information helps! 'Strong' would be normal if it is a computer under your direct control. 'Weak' would be better for a shared or public computer, or if you have permission to install BOINC on a machine belonging to somebody else. |
|
3579)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 58396)
Posted 2 Dec 2014 by Richard Haselgrove Post: David Anderson wrote:There was a bug involving zero-share projects. v7.4.32 is now available on the 'download all' page. It installs and runs OK (64-bit Windows), but I don't use the backup project setting. Any testers? |
|
3580)
Message boards :
Questions and problems :
You've been hacked and I've been hacked and I'm in a lot of trouble can you help?
(Message 58295)
Posted 29 Nov 2014 by Richard Haselgrove Post: The OP was reassured by Ray Murray in this thread at LHC Classic. Panic over. http://lhcathomeclassic.cern.ch/sixtrack/forum_thread.php?id=3899 |
|
3581)
Message boards :
Projects :
What project can I run on my Macs GPU?
(Message 58274)
Posted 27 Nov 2014 by Richard Haselgrove Post: The projects list lists only three projects that support Intel GPU and only two of them Mac+Intel GPU: Collatz and Seti. The projects list only lists production projects. Beta test projects sometimes have a wider range: the Albert applications page (Beta project for Einstein@home) lists the following plan_class application: 'FGRPopencl-intel_gpu-lion' - Gamma-ray pulsar search #3 but like all Beta projects, they may not generate work for all applications all the time. |
|
3582)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 58273)
Posted 27 Nov 2014 by Richard Haselgrove Post: There is also a difference in available memory Specifically, when v7.2.42 was developed, the newest NVidia architecture for which specifications were known was the Kepler generation, with 48 CUDA cores per multiplexer. No details for the next generation were released in advance, but when the Maxwell generation (9xx series, compute capability 5) arrived, it had 32 cores per multiplexer. That is properly accounted for in the v7.4.27 client - hence the change in calculated 'Peak GFLOPS'. I suspect that the v7.4 client range has better handling of the 4GB (32-bit) VRAM limit, but I don't know the details. |
|
3583)
Message boards :
BOINC client :
My Wish List
(Message 58089)
Posted 23 Nov 2014 by Richard Haselgrove Post: If the multi-threaded app in question can have its core usage controlled via a command-line parameter (most Windows MT apps can, using the --nthreads parameter), you should be able to use an app_config.xml to achieve this behaviour. Client configuration - Application configuration You would need to set both <avg_ncpus> (so BOINC can schedule other projects around the MT task properly), and <cmdline>--nthreads (to stop the MT app itself stepping outside its allotted boundary). |
|
3584)
Message boards :
Questions and problems :
Error in "Tools for OS X" wiki? Moving data folder doesn't work for external drives
(Message 58031)
Posted 22 Nov 2014 by Richard Haselgrove Post: I do not believe with Apple's sandbox in their latest versions that it is easy to move the application from the root drive. Apple intentionally does not mount any drive except the root drive with permissions to execute files. I believe with the proper commands it is possible to remount a drive with those permissions enabled, but it is not something I have done. You should be able to move the data to any place you like though. Don't forget that BOINC treats project executable files as data, and stores/executes them in/from the data folder. |
|
3585)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57823)
Posted 18 Nov 2014 by Richard Haselgrove Post: I hardly ever visit Einstein anymore, not since their server started getting hiccups in loading the forums. Website performance problems resolved! (and they have been). Agree about Drupal though - another triumph of style over function. Who did they get to implement it - a first-grader? (don't bother answering that - I'm off to bed) |
|
3586)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57820)
Posted 17 Nov 2014 by Richard Haselgrove Post: Did Richard H. fall victim to the flu, the bird-flu, ebola or something else bad? I haven't seen a post of his in ages. No, neither bird-flu nor ebola - you just haven't been looking in the right places (and I've been in the wrong places - a bloody boring weekend meeting 200 miles away from home). http://einstein.phys.uwm.edu/forum_thread.php?id=10991&nowrap=true#135044 http://albertathome.org/content/project-server-code-update?page=6 http://climateapps2.oerc.ox.ac.uk/cpdnboinc/forum_thread.php?id=7910&nowrap=true#50837 If only SETI would give me some decent intellectual stimulus..... |
|
3587)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57585)
Posted 11 Nov 2014 by Richard Haselgrove Post: Great, I see they have released 7.4.27 to the public and notified no one. Well, Windows 10 is on the feature list, but it's not the only added support. http://boinc.berkeley.edu/trac/wiki/VersionHistory |
|
3588)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57537)
Posted 11 Nov 2014 by Richard Haselgrove Post: Joe posted that it's a holiday in the US of A - we might have to do it all again tomorrow... |
|
3589)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57487)
Posted 9 Nov 2014 by Richard Haselgrove Post: And played the funding card again. That makes it sound like you're the one with the vendetta, Mark. Read more carefully. There's only one mention of funding in his post: Second, because of funding, Astronomy isn't top dog at Arecibo anymore, so Astronomers get a smaller fraction of the observing time. SETI@Home has always got its data effectively for free, essentially by watching over the Astonomers' shoulders while they work. Fewer astronomers, less astronomy, less data for SETI to process. Nothing that SETI's finances can do about that. |
|
3590)
Message boards :
Questions and problems :
BOINC Scheduler broken badly
(Message 57467)
Posted 8 Nov 2014 by Richard Haselgrove Post: Every single one of your recent GPUGrid tasks, on both machines, has failed with an error. http://www.gpugrid.net/results.php?hostid=185997 http://www.gpugrid.net/results.php?hostid=183557 There's something more seriously wrong than just a scheduling problem: both machines refer to a driver error: SWAN : FATAL : Cuda driver error 3 in file 'swanlibnv2.cpp' in line 446. Cuda drivers in Linux are outside my competence: I suggest you post on the GPUGrid message boards to ask for help. |
|
3591)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57357)
Posted 5 Nov 2014 by Richard Haselgrove Post: That is all I will say in a forum not designated for politics. And that's why I won't say anything else on the subject, here at any rate. Sorry about bringing it up. |
|
3592)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57348)
Posted 5 Nov 2014 by Richard Haselgrove Post: Oh, I agree - in fact, I'd say it was almost exclusively a political (or governance, to avoid party-politics labels) problem. And I get the feeling that those who self-select to put themselves forward for election are far more interested in being "in Government", than they are in doing any actual governing for the common good. |
|
3593)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57326)
Posted 4 Nov 2014 by Richard Haselgrove Post: One last thought before the beer leads me off to bed. Britain - aided and abetted by internet technology - seems to be reverting to a Dickensian prototype of spivs and con-artists. And half a century of Consumer Protection and Fair Trade is powerless to stop it. OK, that's worked it out of my system. G'night all. |
|
3594)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57324)
Posted 4 Nov 2014 by Richard Haselgrove Post: Try BT Call Barring feature or this No, I'm not falling for that one. This isn't *my* problem, it's a system-wide problem, and needs tackling at source. Just as every time I buy a new product or service, why does the modern world expect *me* to do all my own research, whether it's getting my roof fixed ('ask a friend to recommend a reliable contractor'), buying electricity ('use a price comparison website, but not one funded by commission from the limited range of companies they recommend'), or finding a lawyer to sue the pants off the last shyster lawyer who I used to sue... - 'nuff said. |
|
3595)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57315)
Posted 4 Nov 2014 by Richard Haselgrove Post: Well, we've got a Server Status Page with a standing outage notice - that's progress, it wasn't there when I went out to dinner. Hi Annie - is it my imagination (I don't think so), or have the number of spam recorded calls gone up exponentially over the last couple of months? They keep calling my fax line too, which must be fun for them. My CallerID phone died while I was in Budapest, so I got a new one - which seems to be capable of displaying a much greater range of numbers. Everything from apparently genuine local numbers, to 001441255814185 - all spam. Leaving them on speakerphone while you get on with life is fun too. |
|
3596)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57281)
Posted 4 Nov 2014 by Richard Haselgrove Post: This is BOINC, where you can thank David and Jord for the pages. We've had them at SETI for, er, ages. I guess they went into SVN whenever that happened, and it just took rather a long time for this site to be brought up to current standards. Maybe - web: improve forum pagination Git 490b74068287c4240bf1e5def51968fcddd33f30, 11 Jul 2012 |
|
3597)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57274)
Posted 4 Nov 2014 by Richard Haselgrove Post: Wait a minute! Pages! We have pages! Thank you, Bernd!!! Wrong site! Bernd is responsible for Einstein, where they still don't have pages. This is BOINC, where you can thank David and Jord for the pages. |
|
3598)
Message boards :
BOINC client :
BOINC 7.2.42 - Project resource sharing issue ?
(Message 57252)
Posted 1 Nov 2014 by Richard Haselgrove Post: This is a known and anticipated problem with the revised scheduling mechanism - see the notes at the very end of http://boinc.berkeley.edu/trac/wiki/ClientSchedOctTen. The way to test whether this is indeed the cause of the particular problem you're seeing would be to enable <work_fetch_debug> in cc_config.xml, and observe the movement in the REC and prio values - are they (slowly) converging between the projects concerned? If so, BOINC will eventually return to the old behaviour by itself. But that's hard work, and causes much eyestrain.... Another useful option from cc_config.xml is <rec_half_life_days>X</rec_half_life_days> I personally find the default 10 days to be too high, with the effects you describe: I set it to 1 on my own hosts a long time ago. |
|
3599)
Message boards :
Questions and problems :
CPU load only 2%
(Message 57218)
Posted 30 Oct 2014 by Richard Haselgrove Post: The GPUGrid 'acemd' tasks are designed - as the project name suggests - to run on GPUs - in your case, the "- Video: NVIDIA Geforce GTX 660". It's not usually necessary or desirable for the CPU to be busy as well. If you want to offer your CPU for scientific research as well, you can attach to another BOINC project and work on that at the same time. |
|
3600)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 57107)
Posted 28 Oct 2014 by Richard Haselgrove Post: I am getting seriously pissed off with seti@home. Some weeks they can't be bothered to shut down until 10.30 in the morning on Tuesdays their time. Today they have just done it at 7.30am. about time they got their act together I think. I think we established that it's Jeff Cobb who's the early riser, but I don't know about the other two. We could probably reverse-engineer the staff roster if we tried hard enough... I was somewhat surprised to get a reply from Matt to a recent email which started Hi Eric, Jeff, Matt - I don't know who'll be on duty for maintenance today, so "to whoever it may concern"... Put him down for 23 September ;) |
|
3601)
Message boards :
Questions and problems :
New client on Windows Server 2011 not getting data
(Message 57051)
Posted 24 Oct 2014 by Richard Haselgrove Post: Well, service mode and usage of the Nvidia GTX 580 are incompatible - the GPU drivers are inaccessible to services, a Microsoft restriction. If you want to use the GPU (and only the GPU), you'll need to run BOINC/SETI in user mode - and with a logged-in user as well. |
|
3602)
Message boards :
BOINC Manager :
OS X v. 7.4.23 - project URL in "Add project" in unreadable
(Message 57008)
Posted 23 Oct 2014 by Richard Haselgrove Post: Exactly the same behaviour - wasting space by mis-aligning the topmost row - can be seen in the task list in the main BOINC Manager screen. |
|
3603)
Message boards :
BOINC Manager :
OS X v. 7.4.23 - project URL in "Add project" in unreadable
(Message 57006)
Posted 23 Oct 2014 by Richard Haselgrove Post: This is probably caused by the top entry in the event log not being exactly aligned with the top of the scrolling window. As you gradually enlarge the event log dialog (vertically), blank space slowly grows at the bottom until there's enough room for another complete line to be displayed. But because the top line is a little too low, the bottom line overlaps the bottom margin, but only while there's exactly enough space to display an integer number of lines. |
|
3604)
Message boards :
BOINC client :
Selecting Disk Drive
(Message 56982)
Posted 22 Oct 2014 by Richard Haselgrove Post: It does for me, but only partly, after jumping through several hoops, and with some quirky bits. I'll write to boinc_alpha as well, but let you go first. |
|
3605)
Message boards :
BOINC client :
Selecting Disk Drive
(Message 56975)
Posted 22 Oct 2014 by Richard Haselgrove Post: Unfortunately not. BOINC would need to know the location of the data directory, in order to read the cc_config.xml file, in order to.... Yes, but that's BOINC 5, before sandboxing and Windows 7 protected special folders and all that jazz. So, let me think about this. On a typical modern-ish Windows setup - Windows 7 is what I'm familiar with - we'd default to: BOINC executables in C:\Program Files\boinc ... (need administrator permission to write data here) BOINC data in C:\ProgramData\boinc Registry HKLM...INSTALLDIR points to the first of these Registry HKLM...DATADIR points to the second of these So, we could put a 'stub' cc_config.xml into C:\ProgramData\boinc (and have it found by the existing registry directive), and use it to redirect all subsequent reads into drive M:, or wherever? Would one put any other cc_config values into that stub file, or into a second cc_config.xml file in the redirected data folder? It just feels unclear in the multi-folder environment which has evolved since the original Wiki was written. I might feel moved to test it sometime. |
|
3606)
Message boards :
BOINC client :
Selecting Disk Drive
(Message 56969)
Posted 22 Oct 2014 by Richard Haselgrove Post: I was reading through the BOINC wiki earlier. If I understand correctly you can create a cc_config.xml file and use the <data_dir> tag to specify a custom path without reinstalling? Unfortunately not. BOINC would need to know the location of the data directory, in order to read the cc_config.xml file, in order to.... Er. I don't know why that's still there. Possibly a left-over from when cc_config.xml lived in the same directory as the executable programs? But that wouldn't be possible on modern versions of Windows, at least. |
|
3607)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56679)
Posted 13 Oct 2014 by Richard Haselgrove Post: Oh, and the border around the forum is really cute. All of the above. Just some over-inflated web designer showing off. But he forgot to put a border round the preview pane. |
|
3608)
Message boards :
BOINC client :
My Wish List
(Message 56644)
Posted 12 Oct 2014 by Richard Haselgrove Post: Hi! +1 I've absent-mindedly double-clicked on a task before now, expecting that would be the natural thing to happen - and been mildly irritated when it didn't. |
|
3609)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56599)
Posted 10 Oct 2014 by Richard Haselgrove Post: The blue/gray can be disabled by going into your forum preferences and unchecking "Highlight special users". Joined the aristocracy, eh? Be careful next time you visit France. |
|
3610)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56558)
Posted 8 Oct 2014 by Richard Haselgrove Post: Yep, full moon tonight alright, and a big ass moon too:) Well, it's shining big and bright over my view, to light my way to the pub later. There's some in there that might be howling a bit too.... |
|
3611)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56556)
Posted 8 Oct 2014 by Richard Haselgrove Post: Yep, full moon tonight alright, and a big ass moon too:) I think we're all under the influence.... |
|
3612)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56552)
Posted 8 Oct 2014 by Richard Haselgrove Post: Oops, now my home page at SETI is garbage, I don't have permission to view the forum_index, and several of my posts have disappeared. There wasn't a full moon over Berkeley tonight, was there, by any chance? |
|
3613)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56541)
Posted 8 Oct 2014 by Richard Haselgrove Post: David Anderson, head-honcho BOINC and such wrote:We're fiddling with Apache issues; avatars don't work for the moment.Well, that explains that. :) boinc_alpha mailing list seems to be broken too. |
|
3614)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56538)
Posted 8 Oct 2014 by Richard Haselgrove Post: I can only see Jord's and Zappy's avatars.It's because we use the external http://Gravatar.com website. The avatars there aren't stored on the BOINC server. It works, but you have to get the new url first. 4-5 clicks on the 'update' button should do it. |
|
3615)
Message boards :
Projects :
News on Project Outages
(Message 56537)
Posted 8 Oct 2014 by Richard Haselgrove Post: Einstein @ Home was down for planned database maintenance today. It's back up now, but they've changed their scheduler url. You may have to update the project several times (up to 10) before the new url is transferred, and you can report completed work or fetch new work. Else, wait several hours for BOINC to do this by itself. |
|
3616)
Message boards :
Questions and problems :
Stopped making contact with server
(Message 56418)
Posted 4 Oct 2014 by Richard Haselgrove Post: Then look at the event log on your local computer and see if it has asked for any recently. |
|
3617)
Message boards :
Questions and problems :
Warning message on my account page
(Message 56412)
Posted 4 Oct 2014 by Richard Haselgrove Post: We had a spate of these at SETI@Home, because BTW, I enabled some PHP flags that show warning and error messages This one appears to be a 'long-standing' (but previously hidden) 'problem', now visible to team administrators only - it's not on my account page. Errors like this need to be gathered up into clear, precise bundles (what you saw, where you saw it, and any special circumstances) and emailed to David Anderson. |
|
3618)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56292)
Posted 27 Sep 2014 by Richard Haselgrove Post: Maybe I'm misunderstanding, but hopefully I'm not. Multi-threaded apps (tasks running on more than 1 core) use OpenCL to do so. This basically helps the task complete faster by using parallel processing instead of serial. That is only one possible example, out of many. The BOINC programming manual refers to You may be able to use OpenCL, MPI, OpenMP, CUDA, languages like Titanium or Cilk, or libraries of multi-threaded numerical "kernels", to develop a multi-threaded app. In general on this forum, we refer to OpenCL when we mean OpenCL, and multi-threaded when we mean one of the others - most commonly OpenMP. |
|
3619)
Message boards :
Questions and problems :
Trojan Detected by Bitdefender
(Message 56290)
Posted 27 Sep 2014 by Richard Haselgrove Post: Further discussion and analysis of this same application file in the thread Checksum Errors at ClimatePrediction.Net |
|
3620)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56259)
Posted 25 Sep 2014 by Richard Haselgrove Post: Perhaps the good Dr. should do his web code tinkering here before foisting it on actual projects like Seti. Interesting feedback from David. I don't think this was a deliberately-planned exercise, but I think it's had beneficial spin-offs. BTW, I enabled some PHP flags that show warning and error messages |
|
3621)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56251)
Posted 25 Sep 2014 by Richard Haselgrove Post: Perhaps the good Dr. should do his web code tinkering here before foisting it on actual projects like Seti. He wouldn't get the feedback nearly so quickly. Nothing like a rampant SETI crowd to give him a brown-trouser day - and at least one of the fixes is working here now, where we needed it. |
|
3622)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56248)
Posted 25 Sep 2014 by Richard Haselgrove Post: Yay! [ pre ] shows spaces and XML tags properly now! |
|
3623)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56229)
Posted 24 Sep 2014 by Richard Haselgrove Post: I like the mini-Number Crunching forums at Seti. Just one thread! I liked this one even better: 
|
|
3624)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 56220)
Posted 24 Sep 2014 by Richard Haselgrove Post: Wow, did the BOINC database crash? I get server errors on both the forums and in BOINC. Can upload, not report. The SETI scheduler and database were throwing wobblies for several hours either side of Tuesday maintenance - from about midnight Monday to 6 pm Tuesday, Pacific time. And that was before David mungled the web server code at midnight Tuesday... And just moments ago, ALL of the US sites were unreachable for me. Not just Seti and BOINC, but also CNN, Google.com, the USGS, Achievement Hunter. Many of the problems were internal to the SETI system (limited work allocation in response to an otherwise successful request), which I don't think could have been caused by an external attack. Unless their security is a lot weaker than I expect... |
|
3625)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56214)
Posted 24 Sep 2014 by Richard Haselgrove Post: Jord was wrestling with that a few days ago... Talking of BOINC code not working first time, the broken [pre/] and [code/] containers have now migrated to SETI and (I presume) SETI Beta. |
|
3626)
Message boards :
Questions and problems :
change location of disk usage
(Message 56212)
Posted 24 Sep 2014 by Richard Haselgrove Post: Which operating system are you using? The reference to 'kernel power' suggests possibly Linux to me, but best to be explicit. The generic answer is 'shut down BOINC, move the Data directory/folder to your new preferred location, tell BOINC the new location, restart BOINC'. In Windows, the easiest way is to uninstall BOINC (that removes the programs, but none of your data): move the previous data folder to the new location: and reinstall BOINC, using the 'Advanced' button on the options screen to point BOINC to the new data location. |
|
3627)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56178)
Posted 22 Sep 2014 by Richard Haselgrove Post: Still 100% of all processors <sigh> Well, I said in my first reply that I hadn't seen anybody really testing the new config feature yet. And new code from BOINC has a habit of not working first time. Sigh. OK, I'll find myself an MT project one of these fine months and see what I can make it do. FWIW, I did try BURP years ago, going down the anonymous platform route, and it responded properly to the --nthreads command - but I detached again because I wanted the disk space back. So I'll have to start from scratch. Good thing I'm on unlimited downloads. |
|
3628)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56174)
Posted 22 Sep 2014 by Richard Haselgrove Post: <app_config>
<app>
<name>Blender</name>
<avg_ncpus>4</avg_ncpus>
<max_concurrent>1</max_concurrent>
</app>
<app_version>
<app_name>Blender</app_name>
<plan_class>mt</plan_class>
<cmdline>--nthreads 4</cmdline>
</app_version>
</app_config>You're supposed to be able to display them with [ pre ], but it's not working. Jord was wrestling with that a few days ago... |
|
3629)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56171)
Posted 22 Sep 2014 by Richard Haselgrove Post: here is the new one I don't see a </app> tag to close off the first section, before the beginning of the <app_version> section. It is easiest to spot if you are absolutely consistent and rigid with your indenting, four spaces each level. And wasn't avg_ncpus in the other section? <app_config> <app> <name>Blender</name> <max_concurrent>1</max_concurrent> </app> <app_version> <app_name>Blender</app_name> <plan_class>mt</plan_class> <avg_ncpus>4</avg_ncpus> <cmdline>--nthreads 4</cmdline> </app_version> </app_config> |
|
3630)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56168)
Posted 22 Sep 2014 by Richard Haselgrove Post: It is a quad core hyper threaded = 8 processors 1) You have to maintain the nested structure of an xml file - inner tags surrounding by middling tags surrounded by outer tags - like a set of russian dolls. </app_config> will be right at the end, to match <app_config> at the beginning, and so on. 2) Square brackets represent optional components. Don't leave the brackets there - put in the option, or take it out entirely. 3) <app_name> isn't the name of an executable file, but the tag for the 'application' used by BOINC to identify the processing group in question. You probably need to be looking at - BUT NOT CHANGING - client_state.xml to find the exact form of all the components in use - yes, the spelling and capitalisation has to be right. You'll rapidly learn that computers are STUPID - they can't work anything out for themselves, they have to be told what to do, in excruciatingly precise detail. |
|
3631)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56166)
Posted 22 Sep 2014 by Richard Haselgrove Post:
You didn't place the settings inside an app_version block which specified which version of which application they referred to. How many CPUs does that computer have? |
|
3632)
Message boards :
Questions and problems :
Multi-Threaded Apps
(Message 56163)
Posted 22 Sep 2014 by Richard Haselgrove Post: Is there an app_config operand that will tell a mt app to only use xx% or yy total processors, rather than allowing it to grab EVERY processor on a box Yes, there is. http://boinc.berkeley.edu/wiki/Client_configuration#Application_configuration Look at the app_version section (which means that you have to be using v7.2.42 or later). Set the same figure - number of cores/threads you want the app to use - in both <avg_ncpus> (controls scheduling the other cores) and after <cmdline>--nthreads (controls the behaviour of the MT app itself). That rather assumes that the particular app you're interested in follows the standard rules of behaviour, and I don't think many people have tested it yet. I'd assume from the use of cmdline that the application behaviour will only change when a new task starts, or at least re-starts. |
|
3633)
Message boards :
Questions and problems :
Trojan Detected by Bitdefender
(Message 56159)
Posted 22 Sep 2014 by Richard Haselgrove Post: I've run several models with that application, on a machine protected by Norton Anti-virus. No warning has been shown. Provisional judgement: it's a false alarm. But I'd run it through https://www.virustotal.com/ to be certain. Edit: hmmm. Detection ratio: 10 / 54 - that's more than I would like, but none of the really big players see it as a problem. https://www.virustotal.com/en/file/d6db1b0463eb8b362d5e689102866d098c519482576c6bac1ce2f61b4c062f91/analysis/1411403613/ |
|
3634)
Message boards :
Questions and problems :
Program hangs both my PCs; screens go black, happens nearly immediately... clean OS installs!
(Message 56114)
Posted 19 Sep 2014 by Richard Haselgrove Post: It's most unlikely to be BOINC causing all that by itself, but it might be related to one or other of the SETI@Home applications. Might be best to post at the SETI@Home forums (if you are a new member there too, you can still post in the 'questions and answers' area) - with access to your account data and computer records at the project, including what tasks have been allocated, other members will be better placed to help you. |
|
3635)
Message boards :
Questions and problems :
app_config.xml - < cmdline > tag - what is wrong?
(Message 56089)
Posted 17 Sep 2014 by Richard Haselgrove Post: EDIT: from Wiki page: Cmdline is a sub-component of app_version. The app_version construct is labelled as 'new in 7.2.39'. |
|
3636)
Message boards :
BOINC Manager :
Feature request: computing preferences in schedule
(Message 56033)
Posted 14 Sep 2014 by Richard Haselgrove Post: See http://boinc.berkeley.edu/trac/wiki/PrefsRemodel for how the developers think about remodeling the preferences. Version 1 was created nearly 4 years ago (8 December 2010). Has there been any practical progress since then? |
|
3637)
Message boards :
BOINC Manager :
Feature request: computing preferences in schedule
(Message 56030)
Posted 14 Sep 2014 by Richard Haselgrove Post: A related feature would be the existence of computing preferences profiles where you could change from one to another with only a couple of clicks. But I don't think it's possible to set any of the day-of-week preferences via a web preferences page: if you set any of those via the GUI in BOINC Manager, you'll create a global_prefs_override.xml file - and there's only one of those: then switching web-based venues has little effect. Maybe we need global_prefs_override.xml global_prefs_override_home.xml global_prefs_override_school.xml global_prefs_override_work.xml files, or even global_prefs_override_weekday.xml global_prefs_override_weekend.xml global_prefs_override_holiday.xml - then a file-picker to switch between them? |
|
3638)
Message boards :
Questions and problems :
BOINC 7.2.42 win-x86: manager can't connect to client in service install on Windows XP
(Message 55888)
Posted 8 Sep 2014 by Richard Haselgrove Post: May not be necessary to download anything. Try netsh winsock reset
Usage: reset
Remarks: Resets Winsock Catalog to a clean state.
All Winsock Layered Service Providers which were previously
installed must be reinstalled.
This command does not affect Winsock Name Space Provider
entries.and/or netsh int ip reset
Usage: reset [name=]<string>
Parameters:
Tag Value
name - The name of a file to which to append information
regarding what settings were reset.
Remarks: Resets TCP/IP and related components to a clean state.
Examples:
reset resetlog.txtBoth require a reboot after use. But glad you got it sorted eventually, anyway. |
|
3639)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 55837)
Posted 7 Sep 2014 by Richard Haselgrove Post: There have never in the history of mankind been so few words so misimderstood. Let me be very clear, and very blunt, about this. The words were very well understood. They said: Mark is drunk. Again. The words that you write when you are drunk - mot only on the message boards, but in PMs and emails as well - are what has led to this ban. The ban will stay in place at least until you have demonstrated that you have your addiction to alcohol under control. Seek professional help, my friend. Proper, medical, counselling and treatment for alcoholism. The only words that Angela and Eric want to hear from you right now are the name of your treatment center, and the date you enter their programme. |
|
3640)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 55831)
Posted 6 Sep 2014 by Richard Haselgrove Post: The project administrator posted on the subject: SETI message 1568071 The ban was intentional, and means what it says. |
|
3641)
Message boards :
Questions and problems :
6.2.19 having problems getting work
(Message 55791)
Posted 3 Sep 2014 by Richard Haselgrove Post: And another user who contacted me privately/anonymously, with even more serious problems - which I have reported privately to the project team. |
|
3642)
Message boards :
Questions and problems :
Estimated remaining time vs. percentage of processor use
(Message 55666)
Posted 28 Aug 2014 by Richard Haselgrove Post: I get all these arguments. Just two quick observations that you may wish to feed into that theoretical assumption. 1) Uninstalling the BOINC programs doesn't remove the user data that is used to drive the calculating algorithm. 2) Even if you take the additional (manual) step of removing the user data from the machine, much of the runtime estimation (depending on project) nowadays relies on data held on the project server. You won't have deleted that. |
|
3643)
Message boards :
Questions and problems :
BOINC saving project files to every HD connected to PC
(Message 55565)
Posted 24 Aug 2014 by Richard Haselgrove Post: Great, so it only affects things in the C:\Programdata\ directory, meaning that if the original user moves his BOINC data directory to some place else, it no longer happens. Yes, I tend to put my stuff into c:\BOINCdata or D:\BOINCdata, and there's only a single address for locations like that. (unless you give them a network share name, which is a whole 'nother kettle of fish) |
|
3644)
Message boards :
Questions and problems :
BOINC saving project files to every HD connected to PC
(Message 55561)
Posted 24 Aug 2014 by Richard Haselgrove Post: You both might find it helpful to refer to some of the available reference material on "special folders", such as http://en.wikipedia.org/wiki/Special_folder http://technet.microsoft.com/en-us/library/ee176604.aspx In short, these are two different addresses for the same place, rather than two different places. |
|
3645)
Message boards :
Questions and problems :
How long are stats kept?
(Message 55510)
Posted 21 Aug 2014 by Richard Haselgrove Post: In Advanced View of the BOINC Manger, how long are stats tracked in the statistics tab? Does it only show so much, or just keep building the graph? As it says in Client configuration: <save_stats_days>N</save_stats_days> |
|
3646)
Message boards :
Documentation :
Boinc ServerUpdates
(Message 55461)
Posted 19 Aug 2014 by Richard Haselgrove Post: Page updating was broken when they switched from SVN to GIT as the primary source code repository. I don't know how you'd get a simple summary like that now - but there have been many, many incremental updates to the server code since then. |
|
3647)
Message boards :
Questions and problems :
BOINC not reading app_config.xml files
(Message 55408)
Posted 15 Aug 2014 by Richard Haselgrove Post: BOINC is not very good at updating the (displayed, cosmetic) GPU fraction when you add or change app_config.xml - the facility was created in a rough-and-ready way, with the intention of coming back and completing it later, but as always other new ideas are more exciting to work on... If you see the right number of tasks running, then it's working, and the display will sort itself out eventually - either after a restart, or after more tasks are fetched, I still haven't worked out exactly what the trigger is. But it'll get there, and you don't have to worry in the meantime. |
|
3648)
Message boards :
Questions and problems :
BOINC not reading app_config.xml files
(Message 55393)
Posted 14 Aug 2014 by Richard Haselgrove Post: You shouldn't have any square [ or ] brackets in there. They are a conventional way of indicating - in the documentation - that some part of the file is optional. If you don't want to use that element, leave the line(s) out entirely. If you do want to use it, set the value you want, and remove the square brackets. |
|
3649)
Message boards :
Projects :
Projects are offline. Often.
(Message 55387)
Posted 13 Aug 2014 by Richard Haselgrove Post: If you're new to BOINC, read about the philosophy behind the platform in Why Use Boinc? - in particular the opening section on the relative costs. BOINC is designed from the ground up to be inherently tolerant of the failures to be expected when using cheap hardware, and without access to dedicated data centres and the associated 24/7 engineering staff with an unlimited supply of spare parts. |
|
3650)
Message boards :
Projects :
News on Project Outages
(Message 55385)
Posted 13 Aug 2014 by Richard Haselgrove Post: Update on CPDN: I have run in a new database server, and she is currently rebuilding the databases from an SQL dump. |
|
3651)
Message boards :
Projects :
ClimatePrediction.Net (AKA CPDN) NEWS
(Message 55384)
Posted 13 Aug 2014 by Richard Haselgrove Post: I just sent their support an e-mail to let them know their project is down. The transitioner, feeder, and both upload servers are "not running." Anything BOINC related gives the "Database offline" error. They know perfectly well that their project is down, and they've been working on it since Friday evening - as I posted on the project outages thread yesterday morning. The last thing they probably need at this time is a deluge of unsolicited email. The latest news on the outage, 10 minutes ago, is: I have run in a new database server, and she is currently rebuilding the databases from an SQL dump. |
|
3652)
Message boards :
Projects :
News on Project Outages
(Message 55340)
Posted 12 Aug 2014 by Richard Haselgrove Post: Extended outage at CPDN: I am sorry to say that the CPDN project has had to be taken offline while we deal with a problem with the RAID array on our main database server. |
|
3653)
Message boards :
Questions and problems :
GPU stopped working yesterday
(Message 55310)
Posted 10 Aug 2014 by Richard Haselgrove Post: Get your driver updates from the Nvidia website, not Microsoft or Windows Update. |
|
3654)
Message boards :
GPUs :
Run only GPU tasks
(Message 55291)
Posted 9 Aug 2014 by Richard Haselgrove Post: Visit the project preferences page for your account, and deselect 'Use CPU'. |
|
3655)
Message boards :
Questions and problems :
Best builds for BOINC machines? How did mine turn out?
(Message 55290)
Posted 9 Aug 2014 by Richard Haselgrove Post: In general, dual-cpu motherboards will only be found in servers, and in high-spec business workstations designed for CAD and similar applications. I have an elderly Dell Precision 490 with dual quad-core Xeon E5320 CPUs, and at the time I bought it, Mac Pro computers with similar specifications were available. Given the intended applications, and the cost, these machines are designed primarily for rock-solid business reliability. Overclocking would be a considered a serious drawback in such an environment. I am only aware of one counter-example: the IBM Skulltrail. This was designed as a fearsome gaming machine with massive overclocking capability (Google suggests somebody got one up to 6 GHz), but unfortunately it was released just as GPUs took over the gaming and distributed computing communities, and it sank without trace. |
|
3656)
Message boards :
BOINC client :
Does anyone know what is.
(Message 55255)
Posted 7 Aug 2014 by Richard Haselgrove Post: It's probably the Host (computer) CPID, which is not the same thing as the User CPID you're thinking of. I don't think it works very well - it's implicated in the batches of 'abandoned tasks' we see sometimes. |
|
3657)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 55233)
Posted 5 Aug 2014 by Richard Haselgrove Post: I'm finding that my packets from the UK are taking a (s)cenic - and very slow - route to Berkeley via Paris, Washington and Los Angeles. But at least my SETI uploads are getting through now. |
|
3658)
Message boards :
Projects :
News on Project Outages
(Message 55232)
Posted 5 Aug 2014 by Richard Haselgrove Post: CPDN will remain offline overnight 05-06 August 2014 (at least). I have taken the project offline for at least 24 hours, this to investigate the issues reported with the scheduler, specifically the error message: |
|
3659)
Message boards :
Web interfaces :
Minor issue with RSS feed (News)?
(Message 55188)
Posted 31 Jul 2014 by Richard Haselgrove Post: I wonder if this could be related to the BOINC Manager Notice being refreshed every time a comment is added to a notified news item. SETI@Home News |
|
3660)
Message boards :
Projects :
Albert@Home
(Message 55138)
Posted 29 Jul 2014 by Richard Haselgrove Post: Ugh. Why couldn't I simply copy and paste that from the BOINC Manager display? Yup, something like that. Unfortunately, the 'archive' file doesn't contain the previous announcement from thread 9009. What sort of an archive is that? |
|
3661)
Message boards :
Projects :
Albert@Home
(Message 55135)
Posted 29 Jul 2014 by Richard Haselgrove Post: Yes, it was pre-announced that the long-delayed Drupal conversion would be taking place this week - but unfortunately it was only pre-announced on the pages you can't now see. They did say that they would try to keep the underlying BOINC processes running throughout the upgrade - after all, the Drupal layer is only a GUI - but I've taken my personal load off the server, so I can't confirm how well they've succeeded. Edit - found this in the notice feed on one of my machines: Here we go again: Drupal migration now scheduled for July 28th Ugh. Why couldn't I simply copy and paste that from the BOINC Manager display? |
|
3662)
Message boards :
Projects :
News on Project Outages
(Message 54927)
Posted 17 Jul 2014 by Richard Haselgrove Post: Advance warning, from SETI's front page: Planned Power Outage - July 23rd-24th, 2014 |
|
3663)
Message boards :
News :
BOINC servers down Thursday 17 July
(Message 54926)
Posted 17 Jul 2014 by Richard Haselgrove Post: Maybe it is the co-lo after all? This has appeared on SETI's front page: Planned Power Outage - July 23rd-24th, 2014 But silently - no BOINC notice feed, no message board thread. |
|
3664)
Message boards :
Projects :
Bitcoin Utopia not science project
(Message 54921)
Posted 16 Jul 2014 by Richard Haselgrove Post: All of those might be diluting the value Richard points out, I don't know. Well, since I'm helping to look at various aspects of the credit system at the moment, I thought I'd have a scout around. I looked at three of the major statistics sites: BOINCstats (which tends to be regarded as authoritative by David Anderson); BOINC Combined Statistics (the in-house stats site - the power behind those 'Projects in which you are participating' lists on your account pages at BOINC projects); and Free-DC (the source of the graphs on BOINC's Top 100 multi-project BOINC participants page). The three sites all collect and aggregate their data at different times of the day, and monitor a different number of projects (between 67 and 145). At the time I collected the figures, the grand total computing power was reported to be between 13 PetaFlops and 32 PetaFlops on the different sites, using that RAC --> GFlops reverse calculation. But all of that was wildly different interpretations of the Bitcoin Utopia phenomenon, which the three sites were displaying as 6.2, 18.1, and 25.3 PetaFlops respectively. Strip out that one project (which seems to be accounting for 80% of all credit awarded, these days), and the residual figure is a much more consistent 6.8 PetaFlops across all three sites - and pretty close to the 6.99 PetaFlops we started with here. Looking in particular at the 'Top 100' page, I thing we can identify visually that Collatz, DRTG and PrimeGrid are other high-paying integer math projects (I'm not sure why DRTG shows so strongly - it seems to have been offline for the last couple of months). Other high-paying projects are SETI, Einstein, and GPUGrid. SETI and Einstein have large numbers of computers attached: GPUGrid certainly does a lot of floating point maths, but I think rather overstate the work done by their credit awards. I could see no sign of any large-scale input from the sort of Uppercase project that Jord describes. If we leave out those three big integer projects, then BOINC's overall floating-point computing power comes out at 4.0 - 4.2 PetaFlops. That still puts it (if the methodology stands up) on a respectable par with ~#11 on www.top500.org - but no way is Bitcoin Utopia the second-most powerful supercomputer in the world, as its daily credit output would have you believe. |
|
3665)
Message boards :
News :
BOINC servers down Thursday 17 July
(Message 54915)
Posted 16 Jul 2014 by Richard Haselgrove Post: The spammers will be mortified, for where should they post their stuff then that day? :) I thought isaac had been moved down to the co-lo, same as the SETI servers? But there's no sign of a planned outage on http://ucbsystems.org/. Is there anything that SETI should know? |
|
3666)
Message boards :
Projects :
Bitcoin Utopia not science project
(Message 54912)
Posted 16 Jul 2014 by Richard Haselgrove Post: Shrug. It's not as if the credits are worth anything, so until they do, it doesn't really matter to the powers that be. There's one place where the BOINC credits do mean something scientific (or at least they're supposed to): the front page of this website. There's a little box in the top right-hand corner labelled 'Computing Power', and underneath it says (currently) "24-hour average: 6.990 PetaFLOPS". BOINC doesn't record or pass out statistics on how many floating point operations have been performed, so how does that get there? Answer - it's reverse engineered, calculated from the total user RAC at each project, on the assumption that every credit is worth the same. When Bitcoin Utopia first went ballistic, that 24-hour average reached 17.063 PetaFLOPs - which, given that Bitcoin ASICs perform integer operations only, was getting perilously close to scientific fraud. At least some common sense has prevailed now. |
|
3667)
Message boards :
Questions and problems :
question about seticoin, how knows about it ???
(Message 54882)
Posted 14 Jul 2014 by Richard Haselgrove Post: There has been no mention, AFAIK, of anything called seticoin at the BOINC project 'SETI@Home'. But there are many different searches for extra-terrestrial intelligence under the generic title of SETI, and some even invite volunteers to participate in the search from home. Apart from one small clip from a SETI@Home screensaver, there is nothing in the (very sparse) information supplied by the originator to link this concept to BOINC. |
|
3668)
Message boards :
Questions and problems :
Intel i7 Turbo Boost not boosting at 100%
(Message 54869)
Posted 13 Jul 2014 by Richard Haselgrove Post: It sounds like it is working exactly the way Intel designed it. It's not just the current temperature which is taken into account, but workload and power consumption as well. Wikipedia has a quite accessible article, with examples. Intel Turbo Boost |
|
3669)
Message boards :
GPUs :
Configurable limit for compute units
(Message 54812)
Posted 9 Jul 2014 by Richard Haselgrove Post: Answer from David Anderson: I don't think David understands GPU programming. As you said yourself, there's no API for that. BOINC can schedule fractional GPUs, but once an application finds itself running on a GPU, all alone, it is free to expand to fill the space available: BOINC does not constrain the app to run in half a GPU, a third of a GPU etc. This is different from the multithreaded CPU application case, where we have both <ncpus> x [to define the scheduling] <cmdline>--nthreads y [to control the app behaviour while running] David likewise said (at first) that he wouldn't let us have direct access to nthreads: it took a long time, but he was eventually persuaded to add it to the app_config.xml specification. Please keep prodding for an equivalent in the GPU case - even though it might need cooperation with the maintainers of CUDA, OpenCL etc. |
|
3670)
Message boards :
Questions and problems :
Bad things happen on loss of network connectivity
(Message 54744)
Posted 5 Jul 2014 by Richard Haselgrove Post: The policy is currently "try every new upload, just once, even during a project backoff". That was introduced to cover the case where a project has multiple application types, uploading results to different servers, and just one of the servers is down. This allows data to be uploaded to the working servers, even though attempts to upload files to the dead server will result in individual file upload backoffs - which still exist, in addition to the project-wide backoff. I'd like to see the 'try once' policy kept, please, although I'd obviously like to see the bug which causes tasks to fail on network loss fixed as well. |
|
3671)
Message boards :
Questions and problems :
Client not uploading results
(Message 54655)
Posted 30 Jun 2014 by Richard Haselgrove Post: Thanks Claggy, will do. There's a ready-made thread at http://climateapps2.oerc.ox.ac.uk/cpdnboinc/forum_thread.php?id=7705. This problem affects all users, so it's easiest to keep all the reports together. |
|
3672)
Message boards :
Questions and problems :
Run Manager at login?
(Message 54651)
Posted 30 Jun 2014 by Richard Haselgrove Post: The normal process under Windows is that Windows starts the Manager at login, and in turn the Manager starts the Client. When the Manager is started automatically like that, it runs minimised in the Notification Area (the old 'system tray', by the clock). Under Windows 7, it's completely hidden down there, unless you customise the notification area BOINC icon to 'Show Icon and notifications'. But it's running, and can pop up balloon notifications if needed. I don't know if Windows 8.1 behaves exactly the same, but it's worth looking. |
|
3673)
Message boards :
Questions and problems :
Tasks with GPU never complete - they get to 88.5% or thereabouts and just hang forever?
(Message 54599)
Posted 26 Jun 2014 by Richard Haselgrove Post: And SETI has no user with the screen name 'Decklin', according to user search. |
|
3674)
Message boards :
Questions and problems :
current problem... doesn't stay on a unit.
(Message 54571)
Posted 24 Jun 2014 by Richard Haselgrove Post: Let me hazard a guess: the screensaver bug? |
|
3675)
Message boards :
GPUs :
New Monster in the making.
(Message 54564)
Posted 23 Jun 2014 by Richard Haselgrove Post: As far as the project goes I believe YoYo. Here is a thread where my team mate mentions benchmarking concerns. He has a lot of high end servers in his home. YoYo replies in that thread. I've emailed my team about it, but it's a cross-continent and hence cross-timezone effort, and things move slowly. Another useful thread would be http://yafu.myfirewall.org/yafu/forum_thread.php?id=147 |
|
3676)
Message boards :
GPUs :
New Monster in the making.
(Message 54560)
Posted 23 Jun 2014 by Richard Haselgrove Post: YAFU is the test bed. Excellent. I'll pass that back to the team, and we may have some patches for you to test. Who should we get in touch with? |
|
3677)
Message boards :
GPUs :
New Monster in the making.
(Message 54558)
Posted 23 Jun 2014 by Richard Haselgrove Post: With the number of cores found in today's crunching machines going up and up, hopefully more projects will adopt multi-threaded applications, those can really be a memory saver. I don't know the YAFU project, but I have done MT-app testing at both AQUA (now ceased operating) and MilkyWay. If YAFU are using straight CreditNew code (i.e. BOINC server source code, without local modification), then I'd say it's 99% sure that CreditNew is to blame. There are intensive studies of CreditNew under way at the moment on the 'Albert' project (Einstein@Home's testbed) - see the News thread Project server code update. At the moment we're looking at CreditNew for GPU apps: MT should be studied too, but unfortunately MT is about the only class of apps which Einstein/Albert can't provide a testing environment for. Does YAFU have a beta or similar test environment where server patches could be tried out in a few weeks' time, under observation in a controlled environment? |
|
3678)
Message boards :
Projects :
News on Project Outages
(Message 54452)
Posted 13 Jun 2014 by Richard Haselgrove Post: Seti dropped off it seems? http://www.downforeveryoneorjustme.com/setiathome.berkeley.edu says it's not just me. Can't connect to scheduler or upload server either. Nothing scheduled by way of co-lo maintenance. |
|
3679)
Message boards :
Questions and problems :
How to make Boinc ignore mouse input for idling purposes
(Message 54430)
Posted 11 Jun 2014 by Richard Haselgrove Post: My understanding is that for Windows, the actual monitoring of the mouse and keyboard hardware is done by a little helper application called 'boinctray.exe' (72 KB on my 64-bit system). It should be relatively simple (!) for somebody with the relevant compiler to comment out the source lines relating to mouse activity and produce a replacement which only fires when the keyboard is in use - or vice versa. But I've never taken my BOINC knowledge that far, though. I'm pretty sure it can't be done any easier way. |
|
3680)
Message boards :
GPUs :
BOINC not using GTX 780 Ti
(Message 54364)
Posted 7 Jun 2014 by Richard Haselgrove Post: 6/6/2014 6:36:33 PM | | Running as a daemon There you go. BOINC can't detect or use any GPU in that mode. "Running as a daemon" equates to the 'service mode' or 'Protected Application Execution' mode listed as an option during installation. To change that, you'll have to uninstall BOINC (less drastic than it sounds, user settings and work in progress are preserved), and re-run the installer, this time making sure that PAE is unchecked (you have to enter the 'Advanced' page to unlock the checkboxes). Make sure you re-install using the same Data directory: D:\ProgramData\BOINC Work should be continued from where you left off, but this time with added GPU support when running under your personal logon account. |
|
3681)
Message boards :
Projects :
ATLAS
(Message 54361)
Posted 7 Jun 2014 by Richard Haselgrove Post: Unlikely, with a Beijing domain name. Atlas is a well-used name - Einstein have one in Hannover, too. |
|
3682)
Message boards :
Questions and problems :
when update the BOINC software,please DONOT set up screen saver.
(Message 54358)
Posted 7 Jun 2014 by Richard Haselgrove Post: You don't say which operating system you are referring to, but on Windows (at least), whether to use the screen saver or not is an option presented to the user during installation. If the user chooses not to install the BOINC screensaver, that choice is remembered for subsequent upgrade installations (though it returns to default if BOINC is uninstalled before the new version is installed - not usually necessary). I agree it might be better if the default setting was to uncheck the 'use screensaver' option, but there are occasions when a screensaver is still useful - it can attract attention and spark an interest in science and distributed computing, if the computer is running in a school, shop front, or conference/exhibition setting. |
|
3683)
Message boards :
GPUs :
BOINC not using GTX 780 Ti
(Message 54356)
Posted 7 Jun 2014 by Richard Haselgrove Post: Can anyone help? A GTX 780 Ti card can only crunch for BOINC if it's plugged into a computer. We might be able to help if you told us something about your computer. When requesting help on these forums |
|
3684)
Message boards :
GPUs :
Cuda initialisation failed after replacing nVidia card
(Message 54222)
Posted 21 May 2014 by Richard Haselgrove Post: Which Quadro drivers did you install, specifically? I ask this as only the Quadro ODE Graphics Driver supports CUDA. We'll have to wait and see on that. He's only returned CPU tasks from GPUGrid so far, no CUDA jobs have been reported. All tasks for host 175175. |
|
3685)
Message boards :
GPUs :
Cuda initialisation failed after replacing nVidia card
(Message 54221)
Posted 21 May 2014 by Richard Haselgrove Post: After some chasing round the houses, I found a user called Marcin Dobija at SETI with a computer like that. Windows 7, 64-bit NVidia driver 334.95 BOINC version 7.2.42 (always nice to be given information in the opening post) The K2000 is based on a Kepler GK107 chip: the SETI applications (I see you've been allocated cuda23 and cuda32 versions) are compatible with the consumer 'GeForce' graphics cards based on that chip, so I don't see why they shouldn't work with the professional variants too. I'll see if I can get a message passed to the developer - but it might be better to continue the conversation on SETI's own message boards. |
|
3686)
Message boards :
Questions and problems :
Geforce 8400GS vs I5 750 for boinc, which would be better?
(Message 54160)
Posted 18 May 2014 by Richard Haselgrove Post: Neither matters for BOINC - BOINC itself uses the minimum resources possible. Which is better for the science project(s) you have chosen will depend on the project programmers. It's better to choose which area of science you're interested in first, and then read the forums belonging to projects active in that kind of science. Most will have efficiency discussions already, and most will also have a "Question and Answer" area where you can post before you've done any work for the project. Once you have done some work you can ask the project specialists directly. It isn't all about speed. I doubt a 8400GS with Compute Capability 1.1 will be able even to run the current applications at GPUGrid. |
|
3687)
Message boards :
BOINC Manager :
Sceduling system
(Message 54100)
Posted 13 May 2014 by Richard Haselgrove Post: Might be easier to refer to one of mine, with one complete 'cycle' of the log - the sequence in which the lines are shown is relevant. 13/05/2014 15:23:19 | | [work_fetch] ------- start work fetch state ------- 13/05/2014 15:23:19 | | [work_fetch] target work buffer: 25920.00 + 3456.00 sec 13/05/2014 15:23:19 | | [work_fetch] --- project states --- 13/05/2014 15:23:19 | Albert@Home | [work_fetch] REC 0.000 prio -0.000000 can't req work: "no new tasks" requested via Manager 13/05/2014 15:23:19 | Test4Theory@Home | [work_fetch] REC 0.000 prio -0.000000 can't req work: "no new tasks" requested via Manager 13/05/2014 15:23:19 | LHC@home 1.0 | [work_fetch] REC 0.000 prio -0.000000 can't req work: "no new tasks" requested via Manager 13/05/2014 15:23:19 | Milkyway@Home | [work_fetch] REC 0.000 prio -0.000000 can't req work: "no new tasks" requested via Manager 13/05/2014 15:23:19 | SETI@home Beta Test | [work_fetch] REC 0.159 prio -0.000000 can't req work: "no new tasks" requested via Manager 13/05/2014 15:23:19 | boincsimap | [work_fetch] REC 2344.479 prio -0.036650 can req work 13/05/2014 15:23:19 | NumberFields@home | [work_fetch] REC 2319.814 prio -0.037250 can req work 13/05/2014 15:23:19 | Einstein@Home | [work_fetch] REC 2263.506 prio -0.038438 can req work 13/05/2014 15:23:19 | GPUGRID | [work_fetch] REC 233705.920 prio -1.750106 can req work 13/05/2014 15:23:19 | SETI@home | [work_fetch] REC 241905.082 prio -1.775359 can't req work: some task is suspended via Manager 13/05/2014 15:23:19 | | [work_fetch] --- state for CPU --- 13/05/2014 15:23:19 | | [work_fetch] shortfall 6613.57 nidle 0.00 saturated 28004.59 busy 0.00 13/05/2014 15:23:19 | Albert@Home | [work_fetch] fetch share 0.000 (blocked by prefs) 13/05/2014 15:23:19 | Test4Theory@Home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | LHC@home 1.0 | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | Milkyway@Home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | SETI@home Beta Test | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | boincsimap | [work_fetch] fetch share 0.500 13/05/2014 15:23:19 | NumberFields@home | [work_fetch] fetch share 0.500 13/05/2014 15:23:19 | Einstein@Home | [work_fetch] fetch share 0.000 (blocked by prefs) 13/05/2014 15:23:19 | GPUGRID | [work_fetch] fetch share 0.000 (blocked by prefs) 13/05/2014 15:23:19 | SETI@home | [work_fetch] fetch share 0.000 (blocked by prefs) (no apps) 13/05/2014 15:23:19 | | [work_fetch] --- state for NVIDIA --- 13/05/2014 15:23:19 | | [work_fetch] shortfall 32410.61 nidle 0.00 saturated 2211.14 busy 0.00 13/05/2014 15:23:19 | Albert@Home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | Test4Theory@Home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | LHC@home 1.0 | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | Milkyway@Home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | SETI@home Beta Test | [work_fetch] fetch share 0.000 (blocked by prefs) 13/05/2014 15:23:19 | boincsimap | [work_fetch] fetch share 0.000 (no apps) 13/05/2014 15:23:19 | NumberFields@home | [work_fetch] fetch share 0.000 (no apps) 13/05/2014 15:23:19 | Einstein@Home | [work_fetch] fetch share 0.000 (blocked by prefs) 13/05/2014 15:23:19 | GPUGRID | [work_fetch] fetch share 1.000 13/05/2014 15:23:19 | SETI@home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | | [work_fetch] --- state for intel_gpu --- 13/05/2014 15:23:19 | | [work_fetch] shortfall 509.64 nidle 0.00 saturated 28866.36 busy 0.00 13/05/2014 15:23:19 | Albert@Home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | Test4Theory@Home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | LHC@home 1.0 | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | Milkyway@Home | [work_fetch] fetch share 0.000 13/05/2014 15:23:19 | SETI@home Beta Test | [work_fetch] fetch share 0.000 (blocked by prefs) 13/05/2014 15:23:19 | boincsimap | [work_fetch] fetch share 0.000 (no apps) 13/05/2014 15:23:19 | NumberFields@home | [work_fetch] fetch share 0.000 (no apps) 13/05/2014 15:23:19 | Einstein@Home | [work_fetch] fetch share 0.333 13/05/2014 15:23:19 | GPUGRID | [work_fetch] fetch share 0.667 13/05/2014 15:23:19 | SETI@home | [work_fetch] fetch share 0.000 (no apps) 13/05/2014 15:23:19 | | [work_fetch] ------- end work fetch state ------- Points to watch for: target work buffer: 25920.00 Work fetch will be triggered when saturation (we'll come to this later) falls below this value. Until then, nothing. --- project states --- Each attached project follows. You have Asteroids@home | can req work, which is all you need at this point. --- state for CPU --- (And similarly for AMD/ATI GPU, NVIDIA GPU, intel_gpu etc.) Everything from here on relates to that named resource, until the next 'state for' line. From your first message, you have "don't need (NVIDIA GPU: project preferences; AMD/ATI GPU: no applications; Intel GPU: no applications)", and those reasons are restated in the resource blocks. So, for CPU? Read it left to right: shortfall 0.00 nidle 0.00 saturated 984.67 busy 0.00 That means your CPU(s) have enough work on hand (from all attached projects, combined) to keep them busy for 984 seconds - a bit over a quarter of an hour. That's the number which is compared with 'target work buffer' at the top - and yours must be even smaller. If 'saturated' was smaller than 'target', there would be a shortfall figure greater than zero. Also, we see that your CPU(s) is/are active - "n(umber) idle zero". Work fetch should occur - assuming 'can req work' still applies - when either 'saturated' falls below 'target', or there's an idle CPU. When work is needed, the project with highest priority (negative numbers, so 'highest' means closest to zero) will be contacted first - assuming it's not blocked for some reason. The project in the first block are listed in priority order, so it's easy to see in my list that boincsimap will be the next to be contacted with a request for CPU work. Clear as mud? |
|
3688)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 54099)
Posted 13 May 2014 by Richard Haselgrove Post: v7.3.18 won't run OpenCL GPU tasks - device is detected, but 'unavailable'. Two reports, one for NVIDIA, one for Intel GPU, both Windows. Best to enable <coproc_debug> logging and report that - with evidence - to the boinc_alpha mailing list. Especially when you're running pre-announcement test builds (Rom hasn't emailed availability yet, or added it to the download_all web page). You didn't have to abort the tasks - they sit there, waiting to run, until you change back to the previous version, or Rom builds a replacement for us to test. |
|
3689)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 54097)
Posted 13 May 2014 by Richard Haselgrove Post: v7.3.18 won't run OpenCL GPU tasks - device is detected, but 'unavailable'. Two reports, one for NVIDIA, one for Intel GPU, both Windows. Bypass this one - how did it get out of the door? |
|
3690)
Message boards :
BOINC Manager :
Sceduling system
(Message 54077)
Posted 12 May 2014 by Richard Haselgrove Post: It is possible to set a verbose logging option: <work_fetch_debug> but be warned that it generates a great deal of highly technical output. For details see Client configuration. |
|
3691)
Message boards :
Questions and problems :
Error al limitar el número de procesadores con tareas multiprocesador (Failed to limit the number of processors with multiprocessor tasks)
(Message 54071)
Posted 12 May 2014 by Richard Haselgrove Post: collatz opencl cpu apps keep running if the "leave in memory" box is checked in the computer pref. and the way the app is it will take or try to take 100% of the cpus (cores) no matter what the setting, in otherwords it is an mt app. Well, it functions by using multiple cores, certainly, but not in the way that BOINC defines an MT app (from that description). What does Process Explorer say while the app is running - does it show additional threads spawned (one per core), the way AQUA and MilkyWay MT apps do? Or does it operate by launching microkernels, like a GPU? so collatz has a little bit of work to do. And possibly BOINC as well, to add a predefined opencl_cpu plan class which supports processor %age in the way that MT's nthreads does. |
|
3692)
Message boards :
Questions and problems :
Error al limitar el número de procesadores con tareas multiprocesador (Failed to limit the number of processors with multiprocessor tasks)
(Message 54069)
Posted 12 May 2014 by Richard Haselgrove Post: Juan talks of receiving "a multiprocessor task", and that corresponds with "Executing ( 4 CPUs )". BOINC refers to these as either 'multicore' or 'multithread' applications. But looking at the Collatz Applications page, they have no apps defined with an 'MT' plan class. It's possible that Jon hasn't quite got the hang of sending/handling the --nthreads N parameter from http://boinc.berkeley.edu/trac/wiki/AppPlan. There are very few MT projects around, so not much experience for project admins to draw on - but it sounds as if Yafu have got it right, and could maybe help (except their website is unreachable, because they haven't renewed their domain name). Edit - it would be best to stick with limiting the number of processors, and work with Jon to debug the plan_class problem. |
|
3693)
Message boards :
GPUs :
No usable GPUs found - Ubuntu 14.04
(Message 53939)
Posted 3 May 2014 by Richard Haselgrove Post: I was wondering why neither GPUGRID, nor PrimeGrid were downloading any work units... You'll never get any work from GPUGrid with that configuration, because the project supports NVidia cards only. I can't speak for PrimeGrid. |
|
3694)
Message boards :
Questions and problems :
Linux and ATI GPUs
(Message 53924)
Posted 2 May 2014 by Richard Haselgrove Post: btw We had a report at GPUGrid of an NVidia knockoff sold through ebay, too. A nice-looking GTX 650, apparently, but it reported compute capability 1.1 ... |
|
3695)
Message boards :
GPUs :
No usable GPUs found - Ubuntu 14.04
(Message 53896)
Posted 1 May 2014 by Richard Haselgrove Post: It's often been noted that display drivers in Linux take a long time to initialise, and it is possible for the BOINC client to have started (and passed the GPU detection phase) before they're ready. You just need to add a delay to the BOINC startup script. |
|
3696)
Message boards :
Questions and problems :
Benchmarks
(Message 53890)
Posted 1 May 2014 by Richard Haselgrove Post: On an other matter: 'Uploading' is a simple file transfer, from one hard disk (yours) to another hard disk (theirs). It's going to take the same length of time whenever you do it, so you might as well get it over and done with quickly. 'Reporting' involves connecting to a database, finding the records for the user and her computer, and updating several fields in them. In days of yore, when servers were expensive and slow, it was found that 'connecting to a database' was computationally expensive, taking - oooooh, milliseconds, at least. By comparison, once the connection is open, database engines are optimised to do the searching and updating quickly. So, save up the reports for later, and send them all in together. One report, one database connection, but lots of work done during the time it's open. Result - one happy (unstressed) server. |
|
3697)
Message boards :
Questions and problems :
Benchmarks
(Message 53887)
Posted 1 May 2014 by Richard Haselgrove Post: Richard, LOL - I'm not upset (see the wink) - and I wasn't checking the other projects deliberately, NumberFields just happens to be one that I follow closely anyway. And your basic premise is quite correct - from what I'm told (by people who have walked the code), benchmarks continue to have a much greater influence than you would expect from a cursory reading of CreditNew. But I invite you to read the third occurrence of "benchmark" in that document. It would be interesting if you could report, as your experiment continues, whether (and over what time scale) host_app_version.pfc_avg kicks in as described. |
|
3698)
Message boards :
Questions and problems :
Benchmarks
(Message 53882)
Posted 1 May 2014 by Richard Haselgrove Post: Thanks,Richard. No, still not quite what I said. BOINC (the current buggy code, not the alpha) will do a benchmark in every session of five days or longer. It might do it right at the beginning, or it might do it later, but within five days. If your BOINC sessions last for ten days or longer, the second half (after the benchmark) could well - in fact, will always - be a stretch of longer than five days without a benchmark. You'll just have to find a way of turning down your "overclock the living daylights out of the system" without restarting BOINC... ;) |
|
3699)
Message boards :
Questions and problems :
Benchmarks
(Message 53877)
Posted 1 May 2014 by Richard Haselgrove Post: So, if i leave boinc running forever(no up/downgrade), it will never do a benchmark, if the last benchmark was less than 5 days, before i started the endless boinc? No, I don't think that's how it works (worked) - have a read of that old thread, there are some example logs. I think BOINC would run benchmarks once per session, however long that session was (see my reference to a static variable, not saved and restored between sessions). If there are more than five days between restarts, BOINC will do that one automatic benchmark at the start of the session. If you restart BOINC after less than five days, it won't do a benchmark at the start of the new session, but it will do a benchmark on the 5-day anniversary of the previous one, whether the timing is convenient or not (beware long-running jobs which don't checkpoint). |
|
3700)
Message boards :
Questions and problems :
Benchmarks
(Message 53875)
Posted 1 May 2014 by Richard Haselgrove Post: What about , if i do a manual benchmark. You can do a manual benchmark any time you want. Case 3) is undefined - it was a bug, and the devs didn't believe it would happen (there's a thread about that someplace here). Feel free to try it out. http://boinc.berkeley.edu/dev/forum_thread.php?id=8943 |
|
3701)
Message boards :
Questions and problems :
Benchmarks
(Message 53872)
Posted 1 May 2014 by Richard Haselgrove Post: We're in the middle of some changes. The theory is 1) On restart, any time the installed version of BOINC has changed (upwards or downwards) 2) On restart, if more than five days since benchmarks were run. In practice, also 3) Five days after benchmarks were last run, if they weren't run at the start of the current session. Case (3) is in process of being removed - I think it's gone from the latest alpha builds. |
|
3702)
Message boards :
Questions and problems :
Use HDD for BOINC work files instead of SSD
(Message 53813)
Posted 27 Apr 2014 by Richard Haselgrove Post: ThatCould not open directory 'slots' from '/media/el_gallo_azul/Storage/var-lib-boinc-client/boinc-client'is pretty suspicious. I've seen it several times before. I don't understand it because there's no directory 'slots' in /media/el_gallo_azul/Storage/var-lib-boinc-client/boinc-client. BOINC operates with three sub-directories below the main data directory. You have the main directory listed there: 27-Apr-2014 11:20:12 [---] Data directory: /media/el_gallo_azul/Storage/var-lib-boinc-client/boinc-client so create these three directories within it: notices projects slots if they're not already there. Make sure that BOINC has permission to read and write to all three sub-directories, and in the case of 'projects' and 'slots' to execute files - including files in daughter directories. |
|
3703)
Message boards :
BOINC client :
Make BOINC reconfigure for new GPU
(Message 53643)
Posted 15 Apr 2014 by Richard Haselgrove Post: If it's any consolation, I've got a Quadro FX 1500, and it doesn't support CUDA or OpenCL either. Mind you, I did buy the computer in November 2006, a couple of years before BOINC supported CUDA - so I didn't know what I was going to be missing. The computer is still going strong, and has had a replacement CUDA card running for five of those years: it's about to get its second transplant. |
|
3704)
Message boards :
BOINC client :
Make BOINC reconfigure for new GPU
(Message 53584)
Posted 11 Apr 2014 by Richard Haselgrove Post: 1) BOINC itself will silently forget about any GPU it can no longer find. The new GPU - provided you've installed a valid driver for it - should show up, but if it is significantly different from the remaining old one (faster, slower, bigger. smaller) BOINC won't use one of them by default (that will be marked in the Event Log at startup). You can enable <use_all_gpus> via a cc_config.xml file - see Client configuration 2) As above - BOINC won't be trying to find the missing GPU. But some project might be telling you that "app_version xxx refers to missing GPU yyy", or words to that effect. It would be better to copy the relevant lines from the event log window, and paste them here - that way, we get to see exactly what you are seeing. If it is an application report from a project, the best way to attempt a first fix would be to set 'No New Tasks' for the project in question, and - when all active tasks for the project in question have been completed and reported - reset the project. |
|
3705)
Message boards :
BOINC Manager :
Time elapsed ?
(Message 53578)
Posted 11 Apr 2014 by Richard Haselgrove Post: BOINC Manager should be displaying a value for "Elapsed" (fourth column on the Tasks tab in Advanced view). That is the time taken so far, and for a completed task with status 'Ready to report' will be the total time for the task. If that column isn't visible, you might possibly have set the column width to zero with a mouse drag. |
|
3706)
Message boards :
Questions and problems :
How platform selected with app_info config?
(Message 53504)
Posted 3 Apr 2014 by Richard Haselgrove Post: I answered on this question experimentally. Could we refine that request, please? By all means give highest priority to primary platform when alternative platforms are available. But please retain order of appearance as a secondary discriminant after the platform has been selected. To remove this code feature would break useful functionality. |
|
3707)
Message boards :
Questions and problems :
Seti needs more diskspace notice now plaguing me
(Message 53483)
Posted 2 Apr 2014 by Richard Haselgrove Post: How do I know where the data directory is? I just always default to how it is downloaded and set up. In case anyone asks you again, say if you come back with another question or suggestion, it's shown in the log you just posted: 4/2/2014 7:49:43 AM | | Data directory: D:\ProgramData\BOINC That's (one of the reasons) why we often ask people to post their start-up logs. |
|
3708)
Message boards :
Android :
Where can I find the implementation code of the projects running on my android?
(Message 53462)
Posted 1 Apr 2014 by Richard Haselgrove Post: You would have to ask at the websites of the individual projects whose tasks you are running. Several of the projects are open source and make their code available under the GPL or similar terms. You will be able to find source code links to whichever code repository system they use (svn, git or similar) somewhere on their websites - though you may have to dig for it, or ask an experienced hand for help via the project message boards. |
|
3709)
Message boards :
Questions and problems :
Tasks all enter state: uploading suspended user request
(Message 53384)
Posted 28 Mar 2014 by Richard Haselgrove Post: You mention the Tasks pane, the Projects pane, and the Disk pane. Did you happen to pass by the Transfers pane? That's where information about uploads will be found. |
|
3710)
Message boards :
Questions and problems :
GTX Titan Z
(Message 53330)
Posted 26 Mar 2014 by Richard Haselgrove Post: No card would be optimal for BOINC: BOINC doesn't compute. Different projects would find different cards optimal - it depends on the nature of the science they do, and the skillsets of their programmers. Some developers get the best out of AMD/ATI cards, other developers work better with NVidia cards. A large card like the TITAN-Z might be efficiently used by a project like GPUGrid: it could be utterly wasted on a project with much smaller tasks. |
|
3711)
Message boards :
Web interfaces :
Password changes
(Message 53239)
Posted 20 Mar 2014 by Richard Haselgrove Post: No, it won't be a problem. A password is needed for the initial attachment only. Once the project is up and running, security is handled by a 32-character 'authenticator'. That won't change. |
|
3712)
Message boards :
Questions and problems :
Boinc stuck on updating
(Message 53177)
Posted 15 Mar 2014 by Richard Haselgrove Post: I having the same issue. It looks like in version 7.2.36 there was an addition of the a ACCEPT_LANGUAGE header in all requests. http://wiki.bc-team.org/index.php?title=7.2.36 http://lists.ssl.berkeley.edu/pipermail/boinc_cvs/2013-December/030104.html The underscore character is invalid in the name of http headers. It should be Accept-Language. My http proxy (squid) is blocking all BOINC requests because of this. Rewriting this header works for me. RFC 3282 (section 3) contains the formal specification, if anyone (still) needs convincing. The change was made in http://boinc.berkeley.edu/trac/changeset/32995b78c5848fe7748f1622d6e8cc5c3bd593cd/boinc-v2 and should be active in versions built from the v7.2.42 tag and later. |
|
3713)
Message boards :
Questions and problems :
Full of "Server is out of disk space" errors since last night
(Message 53138)
Posted 12 Mar 2014 by Richard Haselgrove Post: Also, visit the World Community Grid message boards and see if anybody has reported the problem there already. If nobody else has reported it yet (I'd be surprised), then post your event log messages there as you did here. But if the problem has already been reported - it'll apply to everybody, not just to you - there's no need to repeat it. |
|
3714)
Message boards :
Questions and problems :
Full of "Server is out of disk space" errors since last night
(Message 53134)
Posted 12 Mar 2014 by Richard Haselgrove Post: What can I do? At the very least, tell us which project these tasks are supposed to be uploading to. |
|
3715)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 53115)
Posted 11 Mar 2014 by Richard Haselgrove Post: The outrage began early this week. Someone who couldn't sleep I guess.... They were kicked out of bed early. Daylight Saving Time started this weekend - everyone's out of step until the end of the month. |
|
3716)
Message boards :
Questions and problems :
us.intellitxt.com Why does bonic connect to this well know adware server?
(Message 53104)
Posted 10 Mar 2014 by Richard Haselgrove Post: Sometimes anti-malware products can use that technique to 'innoculate' your system, to prevent it accessing known bad sites. Going back to your original question, did you actually see that extra traffic go down your DSL line, either as a router traffic log or by a usage report/measurement by your ISP? If you were relying on measurement tools internal to your computer, they, too, may have been fooled by all that loopback traffic. We used to get regular reports from users of Zone Alarm, which indicated network activity as if it was external, where nothing was actually happening. |
|
3717)
Message boards :
Questions and problems :
us.intellitxt.com Why does bonic connect to this well know adware server?
(Message 53095)
Posted 10 Mar 2014 by Richard Haselgrove Post: Hmmm... comparing our two Resource Monitor dumps, your 5K/s is going to ip 127.0.0.1 which sounds legit, but mine reports it as us.intellitxt.com. I wonder if my ISP's or my DNS table is just screwed up. Check the 'hosts' file in C:\WINDOWS\system32\drivers\etc Sometimes malware puts rogue entries in there. |
|
3718)
Message boards :
GPUs :
Work failing on GTX 6xx 'Kepler' GPUs.
(Message 53068)
Posted 9 Mar 2014 by Richard Haselgrove Post: This thread was about an older SETI application called 'setiathome_enhanced' - specifically, version 6.10 This application was replaced in the spring of 2013 with 'setiathome_v7', v7.00 Both these applications are fully compatible with your GTX 460 card without any special configuration - the environment variable fix doesn't apply to you. I suggest you visit us at the SETI@home Message Boards, where people will be able to see your results and the type of errors you are getting: the advice you can expect there will be much more specific than it can be here. @ Jord: I suggest this thread has outlived its usefulness and can be de-stickied once trev has read this reply and moved to the other forum. |
|
3719)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 53009)
Posted 6 Mar 2014 by Richard Haselgrove Post: Getting this from the scheduler, which is better than nothing. 06/03/2014 22:13:27 | SETI@home | Project is temporarily shut down for maintenance 06/03/2014 22:13:27 | SETI@home | Project requested delay of 3600 seconds And this site is very slow to respond, too. |
|
3720)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 52998)
Posted 6 Mar 2014 by Richard Haselgrove Post: Yes I know, we are not down, but I was called to the help desk here and now I can't find a place that would serve me a coffee... I'm reminded of the help desk for the last Lunatics installer launch: 
|
|
3721)
Message boards :
Questions and problems :
BOINC 7.2.42 scheduling issue
(Message 52984)
Posted 5 Mar 2014 by Richard Haselgrove Post: Yeah, well, on a normal day explaining what goes on in such logs make my head spin already. But tonight my mind's a total blank. I can ask for the logs, but deconstruct them? No. So I asked Richard to come take a look. Hope he has some explanation. I got the call, but it's pub closing time and bedtime in my timezone. I'll take a look sometime tomorrow. |
|
3722)
Message boards :
BOINC client :
possible problem task switching every hour??
(Message 52948)
Posted 3 Mar 2014 by Richard Haselgrove Post: Its better to use the cc_config.xml: It would be better if one developer fixed the bugs, rather than 200,000 users worked round them one by one. |
|
3723)
Message boards :
BOINC client :
possible problem task switching every hour??
(Message 52945)
Posted 3 Mar 2014 by Richard Haselgrove Post: Well, then it's simple: You found a bug. Automated benchmarks should not happen, unless after a client restart, or after changing the number of CPU cores to use from a low amount to a higher one. Right. For that second one, I've found 28-Jan-2014 15:09:56 [---] Starting BOINC client version 7.2.39 for windows_x86_64 ... 28-Jan-2014 15:10:30 [---] Benchmark results: Note that's exactly five days (to the second) before the benchmark I logged before. So I think an automatic benchmark will run if you: * Boot and run benchmarks * Reboot once, within 5 days and without changing version (or doing anything to trigger a manual benchmark) * Leave running until 5 days are up. Edit: that's what I read from the code change in http://boinc.berkeley.edu/trac/changeset/2985faec3ec529dd18b3ec5ef2755508e18999f3/boinc-v2 The variable declaration static bool did_benchmarks = false; may say 'static', but I doubt its value is preserved across a restart unless saved in client_state.xml - and I don't think it is. Feel free to tell him - he's ignoring my reports, and I don't feel strongly enough about this one to divert his attention. |
|
3724)
Message boards :
BOINC client :
possible problem task switching every hour??
(Message 52942)
Posted 3 Mar 2014 by Richard Haselgrove Post: CPU benchmarks won't be run automatically. Haven't done so since alpha version 6.11.8. They're now only started once every 5 days after a client restart. Since a client restart will remove all remnants of tasks out of memory anyway, they can't interfere. Hmmm. That may be what the documentation says (and I remember seeing it the first time round), but we don't necessarily believe everything we read, do we? 09-Dec-2013 19:03:48 [---] Starting BOINC client version 7.2.34 for windows_x86_64 10-Dec-2013 16:25:59 [---] Running CPU benchmarks 10-Dec-2013 16:25:59 [---] Suspending computation - CPU benchmarks in progress 10-Dec-2013 16:26:30 [---] Benchmark results: 10-Dec-2013 16:26:30 [---] Number of CPUs: 4 10-Dec-2013 16:26:30 [---] 2258 floating point MIPS (Whetstone) per CPU 10-Dec-2013 16:26:30 [---] 7475 integer MIPS (Dhrystone) per CPU 10-Dec-2013 16:26:31 [---] Resuming computation 29-Jan-2014 20:48:31 [---] Starting BOINC client version 7.2.39 for windows_x86_64 02-Feb-2014 15:10:30 [---] Running CPU benchmarks 02-Feb-2014 15:10:30 [---] Suspending computation - CPU benchmarks in progress 02-Feb-2014 15:10:30 [NumberFields@home] [cpu_sched] Preempting wu_sf2_DS-24x13_Grp509343of819200_0 (left in memory) 02-Feb-2014 15:10:30 [NumberFields@home] [cpu_sched] Preempting wu_sf2_DS-24x13_Grp492690of819200_2 (left in memory) 02-Feb-2014 15:10:30 [boincsimap] [cpu_sched] Preempting 20130921.931155_0 (left in memory) 02-Feb-2014 15:10:30 [SETI@home] [cpu_sched] Preempting 17au13ae.5760.10701.438086664203.12.190_0 (left in memory) 02-Feb-2014 15:10:30 [boincsimap] [cpu_sched] Preempting 20130921.932392_0 (left in memory) 02-Feb-2014 15:11:01 [---] Benchmark results: 02-Feb-2014 15:11:01 [---] Number of CPUs: 4 02-Feb-2014 15:11:01 [---] 2159 floating point MIPS (Whetstone) per CPU 02-Feb-2014 15:11:01 [---] 6831 integer MIPS (Dhrystone) per CPU 02-Feb-2014 15:11:02 [---] Resuming computation I've got plenty of examples of benchmarks run immediately after restart, but those two weren't associated with a client startup or any configuration change. Edit - that "[SETI@home] ... Preempting ... (left in memory)" line would have been a GPU task. |
|
3725)
Message boards :
BOINC client :
possible problem task switching every hour??
(Message 52937)
Posted 3 Mar 2014 by Richard Haselgrove Post: GPUGrid updated their Windows 'Long Run' application on 23 Jan 2014 to fix that bug - I don't think Linux was affected by that particular problem (you don't mention an OS, but for some reason I had you mentally flagged as a Linux user?). I don't think DistRT and MilkyWay should be linked by "code left in the GPU" - if they are, BOINC has another problem that needs to be fixed. But then, I don't regard MilkyWay as having very high-quality code (measured by robustness to unexpected circumstances), judging by the number of unanswered error reports on their message boards. |
|
3726)
Message boards :
BOINC client :
possible problem task switching every hour??
(Message 52935)
Posted 3 Mar 2014 by Richard Haselgrove Post: ok, so there is no preemptive switching of tasks when the gpu is involved. BC must ask the app to give up and never gets a response. GPU tasks can be and are preemptively switched, under the same rules as CPU tasks. The difference is that GPU tasks are completely removed from GPU memory when preempted (whatever the setting for 'leave applications in memory'), and restarted from the checkpoint file - does DistRT checkpoint? There is one exception to the 'remove from memory' rule - when benchmarks are being run. Then, the GPU tasks are suspended, but kept in GPU memory - the same as happens with CPU tasks when LAIM is set. One or two projects had problems when that policy first started, but all that I know of have updated their code successfully. I suggest you ask DistRT to explore the difference between "exit/restart" (full preemption) and suspend/resume (quick pause for benchmarking), and adapt their code to support the latter as well. Let's hope they are as responsive as Matt Harvey was to a similar discussion at GPUGrid. |
|
3727)
Message boards :
Questions and problems :
BOINC 7.2.42 scheduling issue
(Message 52934)
Posted 3 Mar 2014 by Richard Haselgrove Post: If you want to analyse the High Priority (aka EDF or 'Panic Mode') decision-making process, add the <rr_simulation> log flag to the list Jord gave you earlier. But be warned, it generates a lot of output and is quite hard to interpret. |
|
3728)
Message boards :
Questions and problems :
Multi-CPU on Linux
(Message 52904)
Posted 1 Mar 2014 by Richard Haselgrove Post: Yes, it should run out of the box with no special setting up - exactly the same as any single-CPU installation. |
|
3729)
Message boards :
Questions and problems :
Boinc stuck on updating
(Message 52852)
Posted 27 Feb 2014 by Richard Haselgrove Post: OK, thanks both, got it now. |
|
3730)
Message boards :
Questions and problems :
Boinc stuck on updating
(Message 52845)
Posted 27 Feb 2014 by Richard Haselgrove Post: No, I'm pretty sure it's the Manager issue. But surely the Manager has no involvement in result file uploads (if we've cleared up the terminology)? They are uploaded by the Client working on its own. The hotfix (as described) does indeed enclose a Manager only, so it makes sense that it addresses the GUI wizard problems - but I'm not so sure about other transfers. Unless you're saying that a failed Manager wizard use might poison libcurl/http, so subsequent file transfer requests fail as well? In which specific case, a client restart should be enough to clear it? |
|
3731)
Message boards :
Documentation :
Please document what installer setting "allow all users ... to control BOINC" does
(Message 52825)
Posted 27 Feb 2014 by Richard Haselgrove Post: Yes, I saw that too. I think it made it into the Charity Engine build, but I don't think it's crossed over into generic BOINC yet. Remember that we're still struggling under [trac]#1025[/trac]: I have a strong suspicion that whatever is written in the documentation doesn't work under Windows 7. I'm still waiting for Rom to emerge from under the Android Avalanche, so we can get back to fixing three year old bugs. |
|
3732)
Message boards :
BOINC client :
BOINC doesn't write in-progress downloads to disk?
(Message 52693)
Posted 21 Feb 2014 by Richard Haselgrove Post: I think they are, and can be resumed from part-way. There is a possible problem with compressed (zipped) files, and I suspect (without knowledge) that the BURP files are probably zipped. If the files are zipped in advance, and stored on the project server in their compressed state, downloads can be interrupted and resumed normally. But - initially at least - files which were compressed 'on the fly' as they were downloaded couldn't be resumed: they had to be restarted from the beginning. So, if they use that technique, it slightly depends if both you and the project are using recent enough BOINC software to handle resumptions of compressed-on-the-fly files. One further possibility is BOINC's display of the size and progress of resumed downloads. I've never summoned up the energy to investigate in detail, but I have a suspicion the BOINC treats a resumed file as a complete, smaller, file for progress display purposes. In other words, instead of showing you a 300MB file resuming at 50%, it treats it as a 150MB file starting from 0%. You might be able to confirm that observation? If true, I agree it's confusing. |
|
3733)
Message boards :
Questions and problems :
Communicating with BOINC Client Problem
(Message 52663)
Posted 19 Feb 2014 by Richard Haselgrove Post: Under Firewall, there is a Tab called Internet Connections for Programs. I added every BOINC application and provided each with FULL internet access and have NetGuard turned OFF for each program. It isn't an internet connection you're looking for. It's an internal loopback connection - entirely within your own computer - on IP 127.0.0.1 or "localhost", port 31416. That's what the BOINC Manager uses to communicate with the (running) BOINC client. |
|
3734)
Message boards :
Questions and problems :
Communicating with BOINC Client Problem
(Message 52661)
Posted 19 Feb 2014 by Richard Haselgrove Post: OK, Jord was on the right lines, then. I'll leave you in his capable hands. |
|
3735)
Message boards :
News :
HTC/BOINC partnership
(Message 52631)
Posted 18 Feb 2014 by Richard Haselgrove Post: Did you also see that SIMAP has been working with Samsung to the same end - except this time including the project application, too? Many smartphones are a supercomputer! http://www.samsung.com/at/microsite/powersleep/ |
|
3736)
Message boards :
Questions and problems :
No file found
(Message 52626)
Posted 18 Feb 2014 by Richard Haselgrove Post: Specifically, "the correct boinc.msi installer file" that it's looking for is the *old* version, needed to uninstall the previous version of BOINC before the new one can be installed. Windows has responsibility for keeping that old file safe: it's nothing to do with the new version of boinc.msi that you've just downloaded. |
|
3737)
Message boards :
Questions and problems :
Communicating with BOINC Client Problem
(Message 52583)
Posted 16 Feb 2014 by Richard Haselgrove Post: Taurus, you didn't answer whether or not you can see boinc.exe running, using a separate tool like Task Manager. The original problem ("communicating with BOINC Client") can mean there's a problem communicating, as Jord is trying to talk you through: or that there's no client running to be communicated with. Did you install BOINC while logged in under the same Windows user account as you're now trying to use to run BOINC? |
|
3738)
Message boards :
Questions and problems :
Is it normal to fulfill all requests but from idle device?
(Message 52567)
Posted 15 Feb 2014 by Richard Haselgrove Post: BOINC 7.2.33, recommended one. Well, in the message I linked, you reported 36 tasks, and received 36 tasks in return. Later in the same thread, you reported 43 tasks, and got 43 tasks in return - I commented on that one, as potential evidence that the 'work in progress' limits were coming into play. The only difference this time is that you didn't include the section of the log that would have told us how many (if any) tasks were being reported. "Requesting new tasks for CPU and NVIDIA and intel_gpu" is identical in both cases. it's plain wrong fetch decision. I see nothing wrong with the fetch decision (the client request). I think it would be less ambiguous in English to describe it as a 'wrong supply decision', to emphasise that it's the server response that needs investigating in detail - as I think you accepted in your later post. |
|
3739)
Message boards :
Questions and problems :
Is it normal to fulfill all requests but from idle device?
(Message 52560)
Posted 15 Feb 2014 by Richard Haselgrove Post: Yes, we did discuss and analyse this situation in some detail at SETI: message 1471467 and following. There are three elements to the problem: 1) Human psychology 2) SETI-specific project management 3) BOINC code objectives With the current server code, (1) and (2) combine to produce the effect Raistmer has illustrated. SETI, as a project, has chosen to limit the number of 'tasks in progress' for any one host: human psychology seems to dictate that people ask for as much work as possible. We started the discussion at SETI with the work request set at 15 days: we're now down to 2.1 days, but that's probably still more than the project 'in progress' limits will allow. Let's leave the rest of that conversation for the project board. But BOINC code objectives are appropriate for discussion here. As we all know, BOINC comes in two parts. a) A client, which requests work b) A scheduler (on the server) which allocates work In the client code, there's a strong emphasis (among many others) on finding work for an idle resource. In the server code, the emphasis seems to be on filling requests from the fastest device first. [I say 'seems': I haven't walked the code. It may be simply that the scheduler plods through the available applications in sequential, index, order, and fills the request dumbly as it encounters tasks and applications which match] But it certainly appears that the server does nothing to fulfill the client priority of avoiding idle resources: I see no indication - even at Einstein, where the server logs are accessible - that the "1.00 devices" in the work request is used to prioritise the server actions. In an integrated client/server system with shared objectives, surely it should? |
|
3740)
Message boards :
Questions and problems :
Latest git seems to be missing stddef.h?
(Message 52333)
Posted 4 Feb 2014 by Richard Haselgrove Post: The link you gave fails, and takes me to this. I think Jord meant http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev - that version is showing for me. |
|
3741)
Message boards :
Questions and problems :
Unable to get BOINC to connect/work for non-admin users
(Message 52322)
Posted 4 Feb 2014 by Richard Haselgrove Post: It's a known problem with the way user permissions are set up by the installer - see ticket [trac]#1025[/trac]. We keep trying to drag the devs back to fix it, but something more urgent always intervenes. |
|
3742)
Message boards :
Questions and problems :
1,000,000 WORK UNITS!!!!! now how do I sign on to Einstein?
(Message 52314)
Posted 4 Feb 2014 by Richard Haselgrove Post: That's the theory, but note that the documentation states that they found it impracticable in practice. So REC stands for 'Recent Estimated Credit', and the estimation is based on speed and time, same as it always was. The actual credit awarded by SETI - lousy or otherwise - doesn't come into it, despite what some paranoid project admins (and users) think. ... Hence we will use a surrogate called estimated credit, maintained by the client. |
|
3743)
Message boards :
GPUs :
ATI GPU R600 (R38xx) does not support OpenCL
(Message 52311)
Posted 3 Feb 2014 by Richard Haselgrove Post: That link is not usable by other people, unfortunately. Try http://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=557409 instead. You might like to revisit your parallel thread on the MilkyWay forum. |
|
3744)
Message boards :
Questions and problems :
mining with an APU (es:AMD A8-6600K)
(Message 52306)
Posted 3 Feb 2014 by Richard Haselgrove Post: It always helps to give questioners a link to the documentation, and sometimes to refer to it yourself. You need a closing tag too, as shown. You need to create a cc_config.xml if you have not and add the line |
|
3745)
Message boards :
Questions and problems :
Incorrect "Total disk space" detection
(Message 52267)
Posted 1 Feb 2014 by Richard Haselgrove Post: Yeah I didn't think it would. I just figured 6 GB ought to be enough... Does this large volume span multiple physical disks? Edit - BOINC does talk about 'disk' space, rather than 'volume' space - though I don't know whether that distinction was intended to be semantically significant. |
|
3746)
Message boards :
Questions and problems :
Incorrect "Total disk space" detection
(Message 52266)
Posted 1 Feb 2014 by Richard Haselgrove Post: 6.12.22 is quite old. Could you try a newer version in case this is fixed already? The OP needs to think through the client-manager duality of BOINC. The client always runs in the background, with no user interface whatsoever - neither X nor command line (except for initial startup). So at that level, every client release should meet his requirements: if any dependencies on libX.. have crept into the client, they should be found and eradicated. And the design philosophy of BOINC has always been that any management tool - command line or graphical, from BOINC or a third party - will be compatible with any version of the client. Some quirks have crept in over the years, and (perhaps inevitably) older managers can't cope fully with newer features: but the basic functions have remained intact. So updating the client to bring over the bug fixes shouldn't be a problem, even if he retains his old command line management tool. |
|
3747)
Message boards :
GPUs :
ATI GPU R600 (R38xx) does not support OpenCL
(Message 52255)
Posted 1 Feb 2014 by Richard Haselgrove Post: Do check with your project - Milkyway? - that the tasks you completed in 36 seconds have been accepted as valid results. |
|
3748)
Message boards :
Questions and problems :
boinc - tasks information inb command line?
(Message 52254)
Posted 1 Feb 2014 by Richard Haselgrove Post: Drago, try boinccmd utility in BOINC application directory. Full details of what boinccmd can do are in the Wiki: Boinccmd tool |
|
3749)
Message boards :
Questions and problems :
Blue Screen every time I run Boinc
(Message 52253)
Posted 1 Feb 2014 by Richard Haselgrove Post: I have almost exactly the same machine as the OP. Intel 4770K, 16 GB RAM, NVidia GTX 780 3 GB. I hope your power supply is adequate to provide the 250W that card needs. It appears to be able to run on the CPU just fine. But, within a minute or two of running on the GPU, it crashes. Which is too bad, because the GPU is about 50-60 time faster. Brand new machine, so, obviously, the latest drivers. Not obvious at all. The drivers will have started to go out of date the moment the computer left the factory gates - and you won't know how long it spent on the road or in the warehouse. Also, many manufacturers prepare an 'OS OEM image' when they first launch a model, and don't update it during the product lifetime. This isn't just for you, and we've asked it of many other users before you: please always give actual version numbers, rather than just saying "the latest". |
|
3750)
Message boards :
Questions and problems :
No Usable GPU GTX 760
(Message 52213)
Posted 29 Jan 2014 by Richard Haselgrove Post: I finally heard back from MSI through the support ticket I submitted. They suggested that perhaps the latest Forceware drivers aren't supporting my GPU somehow... and that downgrading the drivers might solve the issue? It's possible, I suppose, but if you did the driver search carefully, there's not much to go wrong. Following the information you (and BOINC) have given us - GTX 760, Windows 7, 64-bit (and that's all you really need) - I ended up at http://www.nvidia.com/download/driverResults.aspx/71704/en-us |
|
3751)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 52206)
Posted 28 Jan 2014 by Richard Haselgrove Post: The 28-Jan-2014 10:27 timestamp version (standard, not VBOX - untested) installs cleanly and shows v7.2.39 in all the right places. |
|
3752)
Message boards :
GPUs :
ATI GPU R600 (R38xx) does not support OpenCL
(Message 52204)
Posted 28 Jan 2014 by Richard Haselgrove Post: Don't ask me - I'm exclusively NVidia. But a late-model driver sounds right for a late-model card. |
|
3753)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 52202)
Posted 28 Jan 2014 by Richard Haselgrove Post: There are reports - mine being one of them - that the 32-bit version of v7.2.39 for Windows doesn't install cleanly - but it leaves v7.2.38 in working order. |
|
3754)
Message boards :
GPUs :
ATI GPU R600 (R38xx) does not support OpenCL
(Message 52196)
Posted 28 Jan 2014 by Richard Haselgrove Post: I uninstalled my first driver in Safe mode, and then I restarted my computer to install a brand new (freshly downloaded) video driver. That's what we were trying to explore with those clinfo logs, and suchlike. It's beginning to look a little like that, isn't it? Though perhaps we should say "BOINC does not support R9 280X GPUs yet" |
|
3755)
Message boards :
Questions and problems :
No Usable GPU GTX 760
(Message 52193)
Posted 28 Jan 2014 by Richard Haselgrove Post: Can you see a NVidia control panel, which should have been installed with the driver? That has a 'system information' link (bottom left corner), which will give you an NVidia's eye view of the system. |
|
3756)
Message boards :
API :
Running only one process but let it use multiple CPUs
(Message 52148)
Posted 26 Jan 2014 by Richard Haselgrove Post: I think you'll find that all of that is handled by the MT plan_class mechanism. From my experience (purely as a Windows user), the necessary information is parsed by the client from the user's preferences, and passed to the server as part of the work request. Thus, if I have an 8 core (8 thread) CPU, and I have set my project preference to "On multiprocessors, use at most 75% of the processors", then my computer will ask for, and be allocated, work as if it is a 6-core computer. If the project is issuing single-threaded work, I'll get a collection of tasks, and the client will schedule them 6 at a time, to use my preferred number of cores. If the project is issuing MT work, the tasks I receive will already be configured to use my preferred number of cores - 6, in this case. But that will take up as much of my computer as I have said I'm prepared to make available to BOINC: so no other project's task will be able to share the machine. The problem I was mentioning before only arises, and programming is only needed, if you as a project don't want to monopolise all the resources I offer to BOINC, whether that is 100% of my physical hardware or some lower proportion expressed through my preferences. |
|
3757)
Message boards :
GPUs :
ATI GPU R600 (R38xx) does not support OpenCL
(Message 52145)
Posted 26 Jan 2014 by Richard Haselgrove Post: That's great - you certainly have OpenCL, and a reasonably recent one (version 1.2) configured for your GPU - and your CPU as well, though that's less useful. You mentioned installing a new driver, too. What does BOINC say at start-up now? |
|
3758)
Message boards :
GPUs :
ATI GPU R600 (R38xx) does not support OpenCL
(Message 52139)
Posted 25 Jan 2014 by Richard Haselgrove Post: It might help if you could download and run the application contained in http://boinc.berkeley.edu/dl/clinfo.zip and post the output here. That would help to show whether the problem is with your driver/runtime, or with BOINC's detection of the GPU's capabilities. |
|
3759)
Message boards :
API :
Running only one process but let it use multiple CPUs
(Message 52130)
Posted 24 Jan 2014 by Richard Haselgrove Post: Hello I'm new to this forum. The links at the bottom of that page take you to a discussion of multi-threaded applications, and the default "mt" plan_class. That seems to go part way towards what you want. 'mt' schedules a single application instance, and configures it to use every available CPU core. I'm familiar with running the OpenMP type of application (mentioned in the Wiki) under Windows. In that particular case, the configuration is in two parts: a directive to BOINC (ncpus) telling the client how many cores the application is going to need (all of them, by default, but it's made explicit); and a directive to the application (cmdline --nthreads) telling it how many threads to start (likewise, by default explicitly enumerating the number of available cores). Given those two degrees of configuration, it would be possible to copy and adapt the default 'mt' plan_class code to specify some function other than 100% of available cores as the number of threads to be scheduled and used (keep them the same). You might deploy the application for 75% of available cores, for example, or for (n-2) cores. Extending the concept beyond that, you could envisage adding a project custom preference to the web page, allowing users to specify the number of cores they are willing to make available to your application. Then, that number could be passed via a custom plan_class based on 'mt' with the work allocated to the user. Again, I'd recommend that you pass the user's preferred number both as ncpus to control scheduling, and as --nthreads to control the application - and again, keep them the same: that should be possible without too much extra work. But I think you'd have to ask users to configure it via the website, rather than directly in BOINC Manager. |
|
3760)
Message boards :
API :
Can close-coupled workflow type application be deployed on a BOINC server?
(Message 52129)
Posted 24 Jan 2014 by Richard Haselgrove Post: I think the biggest problem is those 3GB upload files. This sort of project might work well in a closed community like a university campus, where you can rely on all users having high-speed (ideally gigabit) bi-directional ethernet connections. Some city-states with high fibre-optic penetration might also be a possibility. But the general volunteer community around the world would struggle with limited upload speeds. I suppose one extra question that needs to be asked is - roughly how many of these multipart jobs are you intending to process (and how quickly)? Do you think recruitment from a closed community pool could supply you with enough participants? |
|
3761)
Message boards :
Questions and problems :
Best remote access tool now logmein free is shutting?
(Message 52082)
Posted 22 Jan 2014 by Richard Haselgrove Post: A similar discussion thread at SETI seems to be coming down in favour of Teamviewer: http://setiathome.berkeley.edu/forum_thread.php?id=73874 Yes, Windows Remote Desktop is fundamentally incompatible with BOINC tasks running on GPUs. Other solutions like VNC work for remote access over a LAN, but are more complicated to set up for fully remote use. |
|
3762)
Message boards :
Questions and problems :
Use HDD for BOINC work files instead of SSD
(Message 52046)
Posted 20 Jan 2014 by Richard Haselgrove Post: You can leave the BOINC program directory on the SSD and place the BOINC data directory on your HDD. However, that's no configuration item but you'll have to uninstall BOINC, move (copy) the data directory to the desired path and reinstall BOINC pointing the installer at the new location via the admin option (third screen?). Sorry Gundolf - Linux doesn't have the same installer screens as the Windows version. But the principle is right. BOINC keeps all its files in a single 'DATA' directory (plus subdirectories), and in principle that directory can be placed on any drive. The current location should be reported in the startup messages in the BOINC event log. You should be able to move it from there to your HDD - make sure you set write and execute permissions as needed: 'data' in this context includes project application files. Then, you need to tell BOINC where to find the files you've moved. I *think* you should find BOINC configurations in a hidden .BOINC file in your home directory, but you may need further help from a Linux specialist to find that - I think the location may vary from distro to distro, |
|
3763)
Message boards :
Questions and problems :
Reporting Tasks Immediately - cc_config file
(Message 52027)
Posted 18 Jan 2014 by Richard Haselgrove Post: As far as I know setting the immediately flag also involves you won’t get new work while reporting. Not by conscious design. You request new work when you need new work - but yes, the chances that that will exactly coincide with a finished task completing uploading are pretty low. So, you'll probably end up making at least twice as many demands on project servers (all the projects you're attached to - you can't set this for individual projects) as you needed to before. |
|
3764)
Message boards :
Questions and problems :
Reporting Tasks Immediately - cc_config file
(Message 52021)
Posted 18 Jan 2014 by Richard Haselgrove Post: For details of how to write your own cc_config.xml file, refer to the Wiki: Client configuration Current versions of BOINC will report tasks a maximum of 24 hours after they were completed - usually much earlier, because a need for new work prompts a server connection sooner. A change currently in testing (in version 7.2.36 and after) reduces the maximum delay to 1 hour. There are few cases where reporting more often than once per hour is important, and over-frequent reports can put BOINC project servers under strain. When the next 'recommended' version is released, you might consider upgrading to that, and removing your 'immediate' flag: see if that is adequate for your needs. |
|
3765)
Message boards :
Questions and problems :
BOINC client 7.2.33 woes
(Message 52006)
Posted 16 Jan 2014 by Richard Haselgrove Post: For the (most recent) previous discussion, see Client has exited 3 times in last ... |
|
3766)
Message boards :
Questions and problems :
BOINC client 7.2.33 woes
(Message 52003)
Posted 16 Jan 2014 by Richard Haselgrove Post: This appears to be the problem described in trac ticket [trac]#1025[/trac]. As you can see from the final comment ("Problem appears to continue into v7.2.23 and beyond."), I re-raised the question when version 7.2.xx was being prepared for release three months ago - we had some brief discussion with the developers (there are threads on this boards which go into further detail), but they concluded it couldn't be fixed in the time-scale they wanted for the v7.2.23 release. It looks like we're gearing up for v7.2.38 sometime in the next 24 hours: too late for that one too, but I'll send the developers a reminder email. |
|
3767)
Message boards :
Questions and problems :
Blue Screen every time I run Boinc
(Message 51983)
Posted 13 Jan 2014 by Richard Haselgrove Post: Try googling for 'BCCode: 124'. Most of the chatter seems to revolve around RAM problems. It would be wise to rule out faulty RAM by running memtest or suchlike. But assuming it passes that, I wonder if running BOINC stresses your hardware in other ways. Some of the Tom's Hardware threads that Google finds link your BSOD with random computer freezes. I had an episode of that recently, also on a 3.5 GHz i7 overclocked to 4.5 GHz. With specialist help, that was tracked down to memory voltage settings - the 'auto' setting for VTT hadn't picked up the XMP profile value. If you do your own overclocking, you might like to review things like that - use something like a full CPU-Z text report to confirm that your BIOS is giving you exactly the settings you asked for. |
|
3768)
Message boards :
BOINC client :
Is there (a plan for) an official version for ARM Linux (not Android) ?
(Message 51965)
Posted 10 Jan 2014 by Richard Haselgrove Post: Projects don't make special WUs for different processors - they're just data, and (in principle) could be processed on any platform. What would need to be different is the science application. I'm not sure how much would have to be done to convert 'ARM for Android' to 'ARM for Linux' - hopefully just a new cross-compilation with different compiler switches, but each project would have to do at least that much for each of their applications. And test them, of course. |
|
3769)
Message boards :
GPUs :
Intel Iris Pro: computation error/output file absent
(Message 51957)
Posted 9 Jan 2014 by Richard Haselgrove Post: This subject came up again in a related discussion at another message board. Prompted me to look and see what was available: https://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=23106&lang=eng&OSVersion=Windows%207%20(64-bit)*&DownloadType=Drivers says that it installs driver 9.18.10.3257 for Intel® IrisTM Pro graphics 5200 running Windows 7* and Windows 8* 64 - that sounds like the right combination for Einstein and your setup. |
|
3770)
Message boards :
News :
New BOINC installer includes VirtualBox
(Message 51955)
Posted 9 Jan 2014 by Richard Haselgrove Post: The advantage of a Wiki is that you can still see what was written, and what it was replaced by. http://boinc.berkeley.edu/w/?title=VirtualBox&diff=5465&oldid=5463 Or Jord's version, as he wrote it. Edit: See also http://boinc.berkeley.edu/w/?title=Release_Notes_for_BOINC_7.2&diff=5479&oldid=5477 and Jord's version. |
|
3771)
Message boards :
GPUs :
Intel Iris Pro: computation error/output file absent
(Message 51943)
Posted 8 Jan 2014 by Richard Haselgrove Post: Hi guys, 'Output file absent' means that the application crashed before it completed its work, so there are no results to send back to the project. Since it's an Einstein problem, please report it on the Einstein message boards - specifically in the thread Errors running Einstein@home on Intel 4000 graphics: you'll see that other people have been having problems with that 10.18.10.3345 driver. I don't know whether any earlier driver, like the 9.18.10.3071 that works for 4xxx series GPUs, is available for your 5200: if not, knowing that there are chips out there which can't currently be used might spur the project into researching what the problem is more quickly. |
|
3772)
Message boards :
Questions and problems :
Blue Screen every time I run Boinc
(Message 51942)
Posted 8 Jan 2014 by Richard Haselgrove Post: You can disable the automatic restart after BSOD, to give you time to read and record the stop messages for later research. Right-click 'Computer', properties. Advanced system settings. Startup and recovery settings. Uncheck 'Automatically restart' in System Failure. Alternatively, for BSODs which have already happened, you can download and run BlueScreenView (though this doesn't always work - it depends whether the computer successfully saved a dump file). |
|
3773)
Message boards :
Questions and problems :
Freeze on Samsung ATIV Book 9 Plus with latest Win8.1 and BOINC 7.2.33 when GPU is not suspended
(Message 51802)
Posted 23 Dec 2013 by Richard Haselgrove Post: You mention that you are running the Einstein project, and that you are running driver 10.18.10.3282 on your HD 4400 graphics. That combination has caused problems for other people too - read Errors running Einstein@home on Intel 4000 graphics at Einstein. The general consensus seems to be that downgrading to the 9.18.10.3071 driver might help. |
|
3774)
Message boards :
BOINC Manager :
error 1402 on repair or remove installation
(Message 51792)
Posted 19 Dec 2013 by Richard Haselgrove Post: There's an independent FAQ service (not connected to the official BOINC developers) usually available at http://boincfaq.mundayweb.com/, but the site doesn't appear to be available tonight. This problem is usually caused by problems accessing the old boinc.msi file - used by the Windows msiexec to install the old version, and required again to UNinstall the old version, as part of the upgrade process. A number of alternatives are covered in the FAQ - two of the simplest to describe are: 1) Try to re-install the previous version, to re-make the missing keys and re-cache the msi file. 2) Perform an msi cleanup to remove traces of the old version of BOINC from the msi database. |
|
3775)
Message boards :
Projects :
Strange total credits in BOINC Stats
(Message 51778)
Posted 18 Dec 2013 by Richard Haselgrove Post: BOINC Stats has an FAQ on the subject: http://boincstats.com/en/page/faq#14 |
|
3776)
Message boards :
Questions and problems :
Boinc performance
(Message 51747)
Posted 16 Dec 2013 by Richard Haselgrove Post: Sorry, I got distracted. 'Flops' is a slippery concept. Not all floating point operations are equal: addition/subtraction is quicker and easier to do than multiplication/division, and they in turn are easier than trigonometry. Which operations do you count and time? BOINC is contradictory on the subject. CPU operation speed is measured by the Whetstone benchmark, which "... was designed to defeat compiler optimizations". GPU speeds, on the other hand, are given as 'peak' flops, which I would take to mean the fastest the transistors could possibly switch, if they were only asked to do simple addition, and there were no inconvenient annoyances like memory accesses or bus transfers to get in the way. My guess is that manufacturers' figures - especially if distributed via the advertising department - are likely to be closer to 'peak' flops: they find some number and make it as big as possible, without necessarily being able to translate that into real work. That was particularly common in the early years of the last decade, when the Pentium 4 range was wound up to absurd speeds in competition with AMD. Now, Intel have the 'Core' and 'Core2' ranges, with lower rated frequencies, but much higher productivity. Think of the difference between a tuned moped, and a low, heavy, motorbike like a Harley or a BMW. The moped's engine will probably be rated with a higher RPM, but which would you prefer to ride across a continent? |
|
3777)
Message boards :
Questions and problems :
Is it possible to return results on a regular port?
(Message 51746)
Posted 16 Dec 2013 by Richard Haselgrove Post: Um. Not an easy question. You are aware how BOINC works - the project you've chosen sends you data, you process it, you return the answers to the project? Well, the "Project file upload handler" is the process that runs on the project's servers to receive and store the files which are returned (uploaded) by the volunteers like you. I'm not sure that there's much more that can be said about it. SETI's lives at http://setiboincdata.ssl.berkeley.edu/sah_cgi/file_upload_handler If you paste that into a browser address bar, you should expect to see a human-readable reply like <data_server_reply>
<status>1</status>
<message>no command</message>
</data_server_reply>When a BOINC client contacts a server, saying it has a file to upload, it should receive a fuller reply, saying if it's OK to upload or not. BOINC will try to contact the domain address above. It will expect your local network infrastructure to resolve the domain to 208.68.240.16 - you could check whether the domain resolves correctly, and whether the IP address is reachable, over your network. But in your opening post, you did say "My company is now restricting outbound ports". You are aware, I presume, that BOINC projects remind you of your personal responsibilities - SETI's version says: Run SETI@home only on computers that you own, or for which you have obtained the owner's permission. I would interpret the word 'computers' in that sentence to include servers, routers, and any other paraphenalia associated with the network. A company being cautious, for example, might think that computer files being sent outside the company to an unknown 'file upload handler' looks like a competitor attempting industrial espionage. If you are clear to go on the 'authorized computer use' question, you should consult your company's technical staff, specifically their network technicians, to see if a block is in place and if it could be modified. If you don't feel comfortable about approaching the network techs, then you probably shouldn't be running SETI in the first place. |
|
3778)
Message boards :
Web interfaces :
Suggestion: Current server time shown on web pages.
(Message 51726)
Posted 14 Dec 2013 by Richard Haselgrove Post: Actually, there would be some merit in displaying the server's local time, whether in UTC or converted to the viewer's local time. Most experienced server admins will instinctively install and configure NTP synchronisation tools on their servers, but on smaller 'one man' (or one woman) projects that may get overlooked: I reckon that the NumberFields@Home server is currently running about five and a half minutes ahead of UTC (estimated by a discrepancy between deadlines and the local arrival time of tasks). I've seen it rise to about 20 minutes, before being reset by a reboot. That's just mildly amusing, and causes no immediate harm to anyone: but more seriously, if a server happened to have its clock reset while tasks were in progress, clients with accurate clocks might miss deadlines. Having time visible would allow users to alert project staff to problems more easily. And even more amusingly, BOINC is designed to cope properly when even the most experienced sysadmin occasionally forgets (as they do) to configure NTP on a new server. For a while, one very famous multi-server project was displaying message board posts as having been made up to a minute in the future, because of a time discrepancy between their BOINC and web servers... |
|
3779)
Message boards :
Questions and problems :
Is it possible to return results on a regular port?
(Message 51715)
Posted 13 Dec 2013 by Richard Haselgrove Post: 31416 is only used for internal (localhost) loopback communications between the 'client' and 'manager' components of the local BOINC installation. Most projects report back on port 80, though a few (WCG is a case in point) are moving to https on 443. 31416 can also be used between local machines if a single 'aggregating' manager like BoincTasks is used to monitor multiple local machines - but that is not a default configuration, and would only apply if you set it up manually. |
|
3780)
Message boards :
Questions and problems :
Need disk space
(Message 51708)
Posted 12 Dec 2013 by Richard Haselgrove Post: Well, I'm out of my depth now, but these three lines stand out for attention: Thu Dec 12 05:32:51 EST 2013||Reading preferences override file |
|
3781)
Message boards :
Questions and problems :
Need disk space
(Message 51706)
Posted 12 Dec 2013 by Richard Haselgrove Post: If you look around here for examples of other people's start-up logs, you'll see things like 11/20/13 16:39:54 | | Memory: 3.41 GB physical, 6.82 GB virtual That section is printed every time BOINC starts up, but only at startup. You need to restart BOINC, and post the equivalent - the whole 20/30 lines, not just the bit I've picked out - from your own machine. |
|
3782)
Message boards :
BOINC Manager :
Windows 64-bit BOINC version 7.2.33 setup Crashed!
(Message 51698)
Posted 11 Dec 2013 by Richard Haselgrove Post: To explain: there is no problem with the v7.2.33 copy of boinc.msi But what the BOINC installer does is to UNinstall the previous version of BOINC, and for that to work, it needs access to the previous boinc.msi - which Windows should have cached on your computer. The FAQ gives various ways of recovering the missing msi, or working round the problems caused by it being missing. |
|
3783)
Message boards :
Questions and problems :
Boinc performance
(Message 51686)
Posted 9 Dec 2013 by Richard Haselgrove Post: Hello everybody. Look at two figures on that page: Recent average credit RAC 1,680,274,919 Average floating point operations per second 8,401,374.6 GigaFLOPS The first number is exactly 200 times the second number - well, 199.9999998809718590574451947423 times it, but near enough. That's not a coincidence - that's arithmetic. BOINCstats doesn't collect (BOINC doesn't record) the number of FLOPs performed, so instead the combined performance of all BOINC projects is estimated from the combined RAC (average credit awarded). That's all written up in the official Wiki (Computation credit). It has a wonderful side effect: all our computers can be doubled in speed (or be speeded up by any arbitrary amount) simply by doubling credit..... Of course, the definition page also says that the credit awarded by a project should also use the same exchange rate against the FLOP standard, but casual observation suggests that they don't. |
|
3784)
Message boards :
BOINC client :
astropulse_6 STILL won't stop computing!
(Message 51676)
Posted 8 Dec 2013 by Richard Haselgrove Post: BOINC is a management application - I've described it as a 'command and control module' in the past. BOINC allows you to choose between dozens of 'projects' you may want to volunteer for - you can choose as few, or as many, as you wish. One of those projects is called 'SETI@Home', and - like many projects - it has more than one type of application. Both the original setiathome and Astropulse are applications supplied by the SETI@Home project. Yes, like you, I've seen other reports that Astropulse runs non-stop (or, occasionally, doesn't run at all - which seems to be the other side of the same coin). But, given the number of other projects and applications out there which do behave as expected, I'm firmly of the opinion that the problem lies in the application (whether Astropulse, as in your thread title, or SETI@Home as in your screen-shot) not doing what it's told, rather than BOINC failing to give the appropriate management instructions. |
|
3785)
Message boards :
BOINC client :
astropulse_6 STILL won't stop computing!
(Message 51672)
Posted 8 Dec 2013 by Richard Haselgrove Post: And - as has been pointed out to you at SETI - the task you have highlighted is NOT an astropulse task. This is NOT a BOINC bug, but something specific to the way your computer handles (or fails to handle) one or more of the SETI applications. It'll be better for you to continue exploring the problem on the SETI message boards. |
|
3786)
Message boards :
Questions and problems :
Trying to get the IntelGPU to run apps
(Message 51525)
Posted 25 Nov 2013 by Richard Haselgrove Post: What project? Does it have any intel_gpu applications? Have you chosen appropriate application types (in preferences) at whichever project it is? |
|
3787)
Message boards :
Projects :
Collatz Conjecture Credit
(Message 51513)
Posted 24 Nov 2013 by Richard Haselgrove Post: There is a user MossyRock at SETI who has a computer just like that. Significantly, it is also fitted with a NVIDIA GeForce GTX 650 graphics card. The timing and credit awarded match Task 3253370173. That task ran in total (Run time) for 1,259.83 seconds, and the computation was done on the graphics card. I suspect you will find that something similar applies to the Collatz task: in these cases the run time (also described as elapsed time) is the significant value to compare, rather than the CPU time. There is a formal definition for the value of a BOINC credit (in terms of the number of floating point calculations performed), but as a statement of fact: each project is independent, and is free to award credits as it sees fit. SETI tries (but doesn't always succeed) to follow the defined standard: I'll leave others to answer your question about 'why' the figures seem different. |
|
3788)
Message boards :
BOINC Manager :
Saving passwords for remote computer - is it possible?
(Message 51503)
Posted 24 Nov 2013 by Richard Haselgrove Post: Here's a little trick which has worked for me for many years, across many (Windows) versions of BOINC Manager. All my remote clients are on the same private LAN, behind a firewall, and nobody else has access. So I'm not concerned about BOINC security at all: and I've set every machine (via gui_rpc_auth.xml) to use the same, very simple, password. That includes the 'Manager' machine I'm using to access the remotes. When I want to view/manage a remote machine, I open the 'Select computer...' dialog, type a random character into the 'Host name' field (space will do), and backspace/delete it again. That populates the password box with an obfuscated (blobbed) version of the simple password. Then, I can type the name of the remote computer, or select it from the memorised list - and it lets me in. |
|
3789)
Message boards :
News :
New BOINC installer includes VirtualBox
(Message 51489)
Posted 23 Nov 2013 by Richard Haselgrove Post: ... which requires BOINC 7.0.28 or above ... That is indeed what I read at http://climateathome.com/climateathome/. But you said LinuxMint ... still have version 7.2.27 of BOINC in their App Store. Surely 7.2.27 is 'above' 7.0.28? |
|
3790)
Message boards :
News :
New BOINC installer includes VirtualBox
(Message 51487)
Posted 23 Nov 2013 by Richard Haselgrove Post: The bundling of BOINC and VirtualBox installers applies to Windows only. They are two different applications, and (if the Windows experience can be followed - I have no Linux experience) an installation of VirtualBox by itself will be detected and utilised by BOINC v7.2.27 VirtualBox availability for Linux is described at https://www.virtualbox.org/wiki/Linux_Downloads. You should check that the version of VirtualBox you install is compatible with the BOINC project you wish to run - some, like Test4Theory, are having problems with the latest releases. |
|
3791)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 51465)
Posted 22 Nov 2013 by Richard Haselgrove Post: Clarification from Rom Walton on the email lists: Just to clarify what has happened the last couple of days. |
|
3792)
Message boards :
Questions and problems :
Boinc Notices and 7.2.28
(Message 51449)
Posted 22 Nov 2013 by Richard Haselgrove Post: Looks to me like an antivirus program might have quarantined it? Many AV companies, these days, keep a 'reputation' database and will block a new program until enough users have installed it without problems. If you're trying to download something within hours of its first release to the internet, it's common to get caught up in that 'chicken and egg' cycle and to need to actively approve the file in your AV console. |
|
3793)
Message boards :
Questions and problems :
Issue with 7.2.28 and Notices
(Message 51411)
Posted 21 Nov 2013 by Richard Haselgrove Post: Rom has posted a test version of BOINC which may alleviate some of these problems. Use at your own risk. Available installers: Windows 7.2.29 - boinc_7.2.29_windows_intelx86.exe - boinc_7.2.29_windows_x86_64.exe |
|
3794)
Message boards :
Questions and problems :
No usable GPUs found after installing 7.2.28
(Message 51393)
Posted 20 Nov 2013 by Richard Haselgrove Post: There appears to be some undocumented change in the 331.82 driver - I tested "331.82-desktop-win8-win7-winvista-64bit-international-whql" - which causes GPU detection to fail with 20-Nov-2013 14:48:31 [---] Caught SIGSEGV in NVIDIA GPU detection if BOINC is run without restarting the computer first. After restarting the computer (and on a sample of one machine only, so far), all seems normal. |
|
3795)
Message boards :
Questions and problems :
No usable GPUs found after installing 7.2.28
(Message 51391)
Posted 20 Nov 2013 by Richard Haselgrove Post: WHQL The 'recommended' Windows drivers from the NVidia site also carry the WHQL mark, to indicate that they have been tested by Microsoft. |
|
3796)
Message boards :
Questions and problems :
Issue with 7.2.28 and Notices
(Message 51390)
Posted 20 Nov 2013 by Richard Haselgrove Post: Could you both check whether you have a file called 'stderrgui.txt' in your BOINC data directory, please? And if so, is there a full runtime debugger log file in there? Don't (please don't :P) post the whole thing, but just check if it matches - Unhandled Exception Record - (are both addresses identical?) If so, the devs are aware, and your reports will give some useful extra routes by which the problem can be triggered. If your addresses are different, please tell us and hang on to the file - it might be needed for additional diagnosis. Thanks. |
|
3797)
Message boards :
News :
BOINC 7.2.28 released to the public
(Message 51272)
Posted 11 Nov 2013 by Richard Haselgrove Post: I wonder if this could be because the (temporary, erroneous) application of CPU throttling to GPU applications has been reversed? Something to cover in that 'additional testing', perhaps? |
|
3798)
Message boards :
Questions and problems :
multiple user accounts on one mac
(Message 51242)
Posted 10 Nov 2013 by Richard Haselgrove Post: This thread got me thinking - always a dangerous thing to do on a Sunday morning! As Jord has indicated, there are two major components of the BOINC program suite. BOINC itself, often referred to as the 'core client' or just 'client', concentrates on managing the resources inside you computer - starting and stopping jobs, deciding what job to do next, fetching new work, reporting the results of completed work, keeping records and general housekeeping. Because the client is focused inwards on the computer, it seem right that for ordinary 'in the field' work there is just one copy running per computer. [There are esoteric situations and reasons where more than one copy might be needed, but for general use - no, just the one] BOINC Manager, or 'The Manager', is a different program with a different purpose. It is focused outwards, on the (current) user, and is used as a two-way communications channel between that user and the inner workings of the client - what's going on, what work is being done, how long before it finishes, is there anything else I need to know or do? With modern computers allowing multiple users to be logged in at the same time, and fast switching between users - whether for family or business convenience - it would seem to be more appropriate if the 'Manager' was accessible to all users, if that was the security model chosen at installation time. The current 'one copy open at once' policy for the Manager was instituted because the Manager spends a lot of its time minimised or even hidden in the dock or notification area. People were opening additional copies from the menu, because they didn't know about, or couldn't find, the existing one - so now a re-launch brings any running instance to the foreground instead. Except that doesn't work between different user accounts - and it's really not user friendly to have to go round rootling in somebody else's private user space to close down their copy (which might have been started automatically without them even knowing), before you can see what's going on in your own area - even to pause a GPU so you can watch a film. So, I think we need to do a little bit more thinking around this area of multiple user information display and control. And while we think about it, that means that trac ticket [trac]#1025[/trac] has to be fixed in parallel. Back to Zoran's problem. There's a little-known screen which I can't find described in any of the user wikis (though I expect Jord will tell me he has an FAQ for it):  (quite why that is a graphical screen, when I can only access it from the command line, I don't know) The option of interest here is /m --multiple for allowing multiple instances of the Manager - that would allow Zoran to work as he wishes. It would be a great improvement if that was automatically applied for the first BOINC Manager running in each user space. |
|
3799)
Message boards :
Questions and problems :
No usable GPUs found. AMD Radeon HD5850 on Win7 x64
(Message 51201)
Posted 7 Nov 2013 by Richard Haselgrove Post: That sort of solution seems to be commonly needed under Linux, but I've never found any problem like that with Windows. I run some older (Windows XP, BOINC v6) machines with GPUs and BOINC service installs: other newer machines with an AutoAdminLogin so there's no startup delay. Neither group has ever missed GPU detection on startup. |
|
3800)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 51189)
Posted 6 Nov 2013 by Richard Haselgrove Post: Rom Walton wrote:Richard, Jacob, et al., Tested already, it works. |
|
3801)
Message boards :
Projects :
News on Project Outages
(Message 51159)
Posted 5 Nov 2013 by Richard Haselgrove Post: CPDN BOINC websites are inaccessible, with a 'forbidden' access error message. Administrators are aware, and say: There is a wider problem with the vm infrastructure here. Edit - they seem to be back, for now. |
|
3802)
Message boards :
Questions and problems :
DSP chips?
(Message 51089)
Posted 29 Oct 2013 by Richard Haselgrove Post: Try it and let us know how you get on? |
|
3803)
Message boards :
Questions and problems :
one more feature to make perfect
(Message 51068)
Posted 28 Oct 2013 by Richard Haselgrove Post: Nothing like this? 
|
|
3804)
Message boards :
Questions and problems :
one more feature to make perfect
(Message 51038)
Posted 26 Oct 2013 by Richard Haselgrove Post: Advanced view, activity menu, 'Network activity suspended'. Or 'based on preferences', with a timetable set on the 'network usage' page of Computing preferences. |
|
3805)
Message boards :
Server programs :
download full packages of server_stable
(Message 51019)
Posted 25 Oct 2013 by Richard Haselgrove Post: Hi, Without meaning to hijack your very relevant question, it's likely that BOINC's old system was copied to many, many other websites. My first search turned up https://code.google.com/p/openmalaria/wiki/UnixBuildBoinc https://wiki.debian.org/BOINC/ServerGuide/PreparationOld - "Old" is the only hint of deprecation. I think BOINC's own pages were pretty much updated (though we await Falcon's reply), but I don't know how much effort was put into promulgating the change to project webmasters, to name but one group. |
|
3806)
Message boards :
Projects :
Gridcoin -- First Cryptocurrency to Utilize BOINC
(Message 50957)
Posted 21 Oct 2013 by Richard Haselgrove Post: Yes. The limit is one hour. Best you check that everything is working right after you post it - better yet, use the 'preview' facility. |
|
3807)
Message boards :
Questions and problems :
Client has exited 3 times in last ...
(Message 50948)
Posted 21 Oct 2013 by Richard Haselgrove Post: Rom Walton has just posted: Whatever the 'all users' issue is, it'll have to wait until the next 7.2 Sorry folks, we tried. See [trac]#1025[/trac] And I got a nice new machine to play with :-) |
|
3808)
Message boards :
News :
Client configuration add-on
(Message 50944)
Posted 21 Oct 2013 by Richard Haselgrove Post: Agreed totally. The master reference point is [HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup] Within that, the key search order is DATADIR (for BOINC v6 and BOINC v7) INSTALLDIR (for legacy BOINC v5, and possibly even older) The default locations listed in Jord's FAQ for the various OS versions. |
|
3809)
Message boards :
Questions and problems :
Intel 4000 GPU + nvidia 660 GPU ?
(Message 50930)
Posted 20 Oct 2013 by Richard Haselgrove Post: Which suggests that at some point in the past, the intel_gpu must have been detected (otherwise there wouldn't have been an application to use the now-missing intel_gpu) Einstein host 8688998 still shows a task which you successfully completed on 25-26 September. What changes did you make after that? Do you still have a stdoutdae.txt or stdoutdae.old file which contains log messages as far back as that? |
|
3810)
Message boards :
Questions and problems :
Controlling BOINC when running under unprivileged account
(Message 50926)
Posted 20 Oct 2013 by Richard Haselgrove Post: The whole point of 'Protection application execution' (aka 'service mode') installation is that it causes BOINC to run 'as a service' - i.e. with any, every, or no user logged in to the console. If you want to have 'per user' control of BOINC's behaviour, you would need to install it in user mode. Even then, an un-priviledged user account will have difficulty managing settings like this: if that user never wishes BOINC to run when he or she is busy, the surest way would be for an administrator to set [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager] "DisableAutoStart"=dword:00000001 in the registry for that user account. Be aware that your CPDN models would never make further progress except when that user logs out and another user logs in. |
|
3811)
Message boards :
News :
Client configuration add-on
(Message 50923)
Posted 20 Oct 2013 by Richard Haselgrove Post: Sorry for the past delay, just updated the program (some options like coproc and exclude_gpu still not supported). This is actually a very well-timed debate. Last week, Rom Walton emailed the alpha testing list to say that the current v7.2.23 alpha client was being considered as a release candidate. It shouldn't be released quite yet, because at least one bug has been partially fixed since then, and there are others still to do, but it's getting very close. I'd suggest that it would be really cool to work to v7.2.23++ as your design specification, and release your next version to coincide with the public update of BOINC. v7.2.23 is available from the 'all versions' BOINC download page (for all platforms except Mac OS X, so far). A little-known, but very handy little trick is that you can generate a fully-populated cc_config.xml framework with all currently-defined tags (and current or default values) by using the 'exclusive applications' pane of the 'computing preferences' dialog in the GUI. It's probably better to use that for your master reference list, rather than the Wiki - they have been known to slip out of synch, and any discrepancies you find could usefully be fed back into the Wiki. Having said that, it does seem odd to have found an <android_debug> tag in the list generated by my Windows client. The developers might want to look into that. |
|
3812)
Message boards :
Questions and problems :
All project with CUDA capabilities are freeze and Display driver stopped responding and has recovered..
(Message 50908)
Posted 19 Oct 2013 by Richard Haselgrove Post: Well, only after you've confirmed that the application from that particular project has a problem. MilkyWay doesn't have a CUDA application, only OpenCL. |
|
3813)
Message boards :
Questions and problems :
All project with CUDA capabilities are freeze and Display driver stopped responding and has recovered..
(Message 50905)
Posted 19 Oct 2013 by Richard Haselgrove Post: If the crash happens between checkpoints, that should already happen. The risk is that the crash may happen during checkpointing, when the files are inconsistent. That can be mitigated by not removing the old checkpoint until the new one is complete - then a restart would go two steps back instead of just one - but there's still a moment of risk while the file names are being switched over. The checkpoint files themselves are the responsibility of the individual projects. Better to report any identified problems with them on the project's own message board, not here. As I said (and linked) in my first reply, GPUGrid are aware they have a problem, and have diagnosed the cause. They're just waiting on some (scarce) developer time to fix it. |
|
3814)
Message boards :
News :
Client configuration add-on
(Message 50900)
Posted 18 Oct 2013 by Richard Haselgrove Post: BOINC Client Configuration is a Windows GUI for editing the BOINC client configuration. Does anyone know whether this application is still under active development? The most recent version is dated June 2012 (16 months ago), and there seem to be significant discrepancies from the current state of the BOINC code. For example, No sign of <android_debug> Still contains <debt_debug> It would have been a nice idea if this news had been posted in a timely fashion, but times change.... |
|
3815)
Message boards :
Questions and problems :
All project with CUDA capabilities are freeze and Display driver stopped responding and has recovered..
(Message 50895)
Posted 18 Oct 2013 by Richard Haselgrove Post: You don't say which project(s) you're running, but it might be similar to a problem we're seeing at GPUGrid - http://www.gpugrid.net/forum_thread.php?id=3491. With a BSOD or other sudden, unplanned, closedown, BOINC doesn't get a chance to finish writing the files for work done so far on jobs in progress - the 'checkpoint' files. If they're in a mixed-up, partially-written, state, it's likely that the project application won't be able to use them. Most times, the job just exits with an unrecoverable error, and BOINC moves on to the next one. But the current GPUGrid app behaves just as you describe, with continual driver restarts. They've just found out what the problem is, apparently, but haven't had time to fix it yet. Since you've found a way of getting control of the machine and letting BOINC run, the next move might be to abort any partially-run CUDA tasks that will be waiting to run next time the GPU is available. Tasks that haven't been started yet should be OK. Make sure you exit BOINC fully after aborting the tasks - BOINC doesn't complete its clean-up until the next restart. Give that a try and report back, please. |
|
3816)
Message boards :
Questions and problems :
Intel 4000 GPU + nvidia 660 GPU ?
(Message 50890)
Posted 18 Oct 2013 by Richard Haselgrove Post: You don't need to do anything special in cc_config to enable HD 4000 - see my host 5744895. But for Einstein, you do need an up-to-date Intel OpenCL driver with v1.2 support. |
|
3817)
Message boards :
BOINC client :
BOINC 7 development discussion thread.
(Message 50825)
Posted 10 Oct 2013 by Richard Haselgrove Post: 7.2.21 has an installer bug, which causes the installer to tell you after the installation that 'a component' did not update and that you'll have to run Repair Installation. It'll do this after the repair installation as well. It'll also do it on a clean installation, But BOINC itself will still run if you start it manually, so you can do any further testing you had planned. Looks like OpenCL on CPU still has a few rough edges that need smoothing out. |
|
3818)
Message boards :
Projects :
News on Project Outages
(Message 50699)
Posted 1 Oct 2013 by Richard Haselgrove Post: SETI and SETI Beta are both down Same location, same power supply (it was down earlier), but a different server. And without such a complicated back-office network of linked database and storage servers, much easier to re-start cleanly without human intervention. |
|
3819)
Message boards :
Questions and problems :
Inability to function/connect after a bounce.
(Message 50690)
Posted 30 Sep 2013 by Richard Haselgrove Post: If the OP looks at his own previous thread (Client has exited 3 times in last ... ) - which seems to be about the same problem as this one - he will see that we are ready to examine his problem in some depth. It will be most helpful if he could indicate whether, when he logs back in after this mysterious bounce, he uses the same login account as he used for the original installation. Does the work-round that you reported success with last time not work on this computer? There has been some discussion of this issue on the boinc_alpha email list. I am hopeful that there will soon be new Windows builds available to test for solutions, but they have been delayed because the developer responsible for installation issues has had to replace a failed computer. |
|
3820)
Message boards :
Questions and problems :
Intel HD 4600 or AMD 5450 GoGreen!
(Message 50619)
Posted 23 Sep 2013 by Richard Haselgrove Post: I have just purchased an i5 (Haswell) - the i5-4570 CPU @ 3.20GHz [Family 6 Model 60 Stepping 3] model. That was in a full retail factory build - just a consumer item (I deliberately wanted a factory OEM Windows installation, for other testing). I find that for the HD 4600 to work efficiently, I have to limit CPU usage to 3 out of four cores. In that mode (3 CPU tasks plus one Intel GPU task), the entire system unit is drawing 88W from the wall. There are a limited number of projects with Intel GPU applications as yet, but the two I've tried so far worked flawlessly at the first attempt. |
|
3821)
Message boards :
BOINC client :
Boinc v7.2.11 for over a month...
(Message 50594)
Posted 21 Sep 2013 by Richard Haselgrove Post: There's now a v7.2.16 available to test... (Windows only, at the moment) |
|
3822)
Message boards :
Questions and problems :
10 projects and no new tasks for weeks..
(Message 50580)
Posted 19 Sep 2013 by Richard Haselgrove Post: Since we have 18/09/2013 22.30.24 | Einstein@Home | URL http://einstein.phys.uwm.edu/; Computer ID 6508825; resource share 100 we can track that back to http://einstein.phys.uwm.edu/hosts_user.php?userid=572711 If you wait until BOINC requests work by itself (rather than clicking update), and then follow that link, you (and we) can click on the date/time in the far right-hand column and see what the server thought of your request - that can be easier for less-experienced users than setting up local debug logging. The lack of allowed disk space does seem like something to check first, though. |
|
3823)
Message boards :
Questions and problems :
Client has exited 3 times in last ...
(Message 50567)
Posted 18 Sep 2013 by Richard Haselgrove Post: I now have a 'clean', factory-fresh, Windows 7 Pro machine, and have reproduced this problem. Just waiting for my report to re-appear from the boinc_alpha mail server. |
|
3824)
Message boards :
Questions and problems :
When time is making no sense...
(Message 50495)
Posted 14 Sep 2013 by Richard Haselgrove Post: BOINC Manager seems to set the time and date format according to the language setting within the BOINC Manager itself. Showing  in a UK-localised copy of Windows7/64 showing 24-hour format for the system clock, for a Dutch language setting for BOINC Manager (options) Now, can anyone tell me the Dutch for 'restore default settings'? :P Edit - never mind, found it. English:  American:  Note how the DMY - MDY order has switched in the last one, too. All via language setting in BOINC Manager itself, didn't touch Windows settings. Testing done with Manager v7.2.11 |
|
3825)
Message boards :
Questions and problems :
Client has exited 3 times in last ...
(Message 50487)
Posted 13 Sep 2013 by Richard Haselgrove Post: Primarily writing @ Jord... We seem to get enough stories like this that I really fear there's some unidentified weakness in the 'permissions and groups' setup phase of the BOINC installer, when the 'allow all users to control' option is selected. It would be ideal if Rom, or somebody, could get hold of an absolutely pristine computer with a completely fresh Windows installation that could be wiped clean between attempts, and walk it through. There's a temptation to use a VM for this kind of work, but it probably should be repeated with a typical mall-bought home OEM machine - preferably one with a 'restore to factory state' recovery partition. I say that because of warnings like this from Microsoft http://support.microsoft.com/kb/307545 (this is the XP/FAT32 version, but I think similar warnings apply for other versions, and other file-systems, too) Warning Do not use the procedure that is described in this article if your computer has an OEM-installed operating system. The system hive on OEM installations creates passwords and user accounts that did not exist previously. If you use the procedure that is described in this article, you may not be able to log back into the recovery console to restore the original registry hives. |
|
3826)
Message boards :
BOINC client :
BOINC has become a total mess
(Message 50332)
Posted 27 Aug 2013 by Richard Haselgrove Post: I'm also getting a lot of "not requesting tasks: project is not highest priority", my question is where can you see (and hopefully edit) these priorities? You can see (but not edit) all the work fetch metrics by enabling the <work_fetch_debug> log flag in cc_config.xml. For all questions like this, it's worth looking through client configuration to see if there's a suitable tool. [WFD] (as we call it for short) generates a lot of output very quickly, so you may wish to increase the default log file size. Details are in the same configuration documentation. |
|
3827)
Message boards :
GPUs :
Nvidia GPUs Hot Hours After Running SETI
(Message 50310)
Posted 25 Aug 2013 by Richard Haselgrove Post: Thanks. That does point to a specific issue with SETI's CUDA app - cuda42 in this case, though there are other cuda generations built from the same code base. I'll keep passing messages back. The clock increases when running SETI are to be expected. The clock increase for GPU 1 (the inactive one) is probably a function of SLI - in my experience, if you disable SLI in the NVidia control panel (leaving the SLI strap connector in place), the clock speeds of the two cards vary independently according to load. The curious issues - and potential problems - are the failure to downclock when idle, and the failure to release GPU RAM. Did you wait and allow the SETI tasks to finish normally, or did you suspend them manually? That shouldn't make any difference, but it would be nice to know. I'm still not seeing any reported results on your machine's results page at SETI. If you have any tasks 'Ready to report', could you please update the project, so that we can read the output generated by the application? |
|
3828)
Message boards :
Projects :
i am trying to find an active boinc project that is going to be completed..
(Message 50304)
Posted 24 Aug 2013 by Richard Haselgrove Post: NumberFields at Home is doing an exhaustive search over a finite number of fields in the realm of algebraic number theory. So, it has a fixed end point, and a plan to publish complete tables when they've finished - they recently published the first batch: Results of first decic search posted. I've no idea how long the remaining searches are expected to take, but I'm sure Eric Driver would welcome such an enquiry. |
|
3829)
Message boards :
GPUs :
Nvidia GPUs Hot Hours After Running SETI
(Message 50303)
Posted 24 Aug 2013 by Richard Haselgrove Post: Thanks - that gets us a bit further down the road. Unfortunately, it doesn't look at if SETI has been running on that machine, even during the overnight scheduled hours: All tasks for computer 7034435 All my most relevant experience comes from Windows 7 (not 8), but some may help. What state do you leave the machine in when you leave in the evening? If you log off from your Windows account, neither BOINC nor SETI will have access to the NVidia driver, and cuda applications can't run. [In fact, I suspect BOINC will close completely when you log off]. For CPU-based applications, the solution is to install BOINC in 'service' or 'Protected Application Execution' mode - but unfortunately the protection is so strong that, again, applications can't communicate with the graphics driver in that mode. So, that's a no-go. If you (or your employer) are worried about the security aspects of leaving the machine logged on while you are away from your desk, you could either a) lock the machine when you leave - Ctrl+Alt+Del, or Windows+L: I think this can only be done manually. b) use a blank screensaver with a password required to resume. But that doesn't explain the high temperatures. Since you're using MSI Afterburner anyway, could you leave that running overnight behind the lock or screensaver? It might give some clues in the other traces (GPU usage, memory usage, clock speeds etc.) which you could compare with the 'cool' state after reboot. There have been problems reported with the Mac OS X version of BOINC on some Macbooks - even the simple BOINC start-up action of querying the GPU device capabilities switched the GPU into high-speed, high-power, mode, and it never reverts to low power - even if no actual computation is being carried out. Maybe something similar is happening in Win8? Again, Afterburner should reveal it if so. BTW, I assume you've adjusted the Windows default power management settings? You can blank the monitor while BOINC runs, but nothing else should be shut down when the machine is idle - use the 'never' settings. |
|
3830)
Message boards :
GPUs :
Nvidia GPUs Hot Hours After Running SETI
(Message 50290)
Posted 23 Aug 2013 by Richard Haselgrove Post: I've never heard of behaviour like that, although not many people who post about BOINC use that sort of scheduling. We also don't have a lot of experience of Windows 8 yet, either. I've passed a message to the developer who contributed that particular cuda42 application to SETI. He might need more information, like the NVidia driver version you're using. Could you post your SETI account ID here, or a direct link to the host ID for that computer? There may be clues buried in the stderrtxt for the tasks you've completed. Thanks in advance. |
|
3831)
Message boards :
Questions and problems :
BOINC Sometimes Stops Processing (Milkyway@home Units)
(Message 50225)
Posted 18 Aug 2013 by Richard Haselgrove Post: Do you ever use the BOINC Manager 'Tools|preferences' menu to change that 'max CPUs used' value? My experience with the version of multithreading being used at Milkyway is that: The number of cores that N-Body wants to use is set - for all cached tasks - each time work is downloaded from their server. If BOINC is using two cores at that moment, then the Milkyway work is set to require two cores: if BOINC is using all four cores when work is downloaded, then the work is set to require four cores. The problem might occur if you download new work when four cores are available, but subsequently change your preferences to only allow two cores to run. The work which expects all four cores to be available might then be stalled indefinitely. For more details, see my post on the subject at Milkyway. There are reports at Milkyway that multithreading works differently on different operating systems - Alinator has had problems with WinXP/64 - but I'm using Win7/64 same as you. However, I did have problems with BOINC v7.0.64 (they were to do with GPUs, so they won't affect you): it's working better with the alpha v7.2.10 client, which may or may not be promoted to 'recommended' sometime this fall. |
|
3832)
Message boards :
Questions and problems :
Cannot attach to Climateprediction.net
(Message 50149)
Posted 8 Aug 2013 by Richard Haselgrove Post: At 17:45 UTC, 8 Aug 2013, Andy Bowery wrote: So you will see that we have the majority of services restored now! climateapps2 has now been replaced completely. However I am aware of a number of issues which still need to be solved these are: From previous experience, the new server will be hit by a very heavy workload of users uploading and reporting work done during the outage - it's like to respond very slowly for a few days to come. Edit - there are more comments (and more copies of that email) in the 'Projects' section of this message board. |
|
3833)
Message boards :
GPUs :
GPU works, then is missing after every restart. Completely removing BOINC and reinstalling fixes. Ubuntu Linux 13.04, NVIDIA GeForce 8400GS
(Message 50138)
Posted 8 Aug 2013 by Richard Haselgrove Post: Make sure that Linux has time to fully initialise the NVidia drivers before you start BOINC (or before BOINC auto-starts) - I think this requires the X-server to initialise the monitor and desktop, too (I'm on the Windows side, so please excuse sloppy *nix terminology). I think people have published various delay lines for the BOINC start-up scripts over the years, but I don't have any links. Alternatively, and as a temporary work-round, try simply closing BOINC down after the desktop initialisation is complete, and restarting it - I think BOINC should find the GPU at the second attempt, after a decent interval has elapsed. That would be less trouble than uninstalling and reinstalling. |
|
3834)
Message boards :
Projects :
News on Project Outages
(Message 50124)
Posted 7 Aug 2013 by Richard Haselgrove Post: Seti forum appears to be down... There appears to be some maintenance work scheduled on backup servers at the CoLo facility. http://systemstatus.berkeley.edu/ (CMR: 2278) |
|
3835)
Message boards :
Questions and problems :
Install BOINC both as a Service and a Non-Service?
(Message 50112)
Posted 7 Aug 2013 by Richard Haselgrove Post: And it doesn't matter if the password is identical for all BOINC client instances, and for all machines on the LAN. |
|
3836)
Message boards :
Questions and problems :
Install BOINC both as a Service and a Non-Service?
(Message 50110)
Posted 7 Aug 2013 by Richard Haselgrove Post: Is there any reason that wouldn't work over the LAN with BoincTasks? I have no direct experience with this either, but I was told by a reliable tester who tried it: Running two BOINC core clients at once worked fine. BOINC's own Manager always connected to the first instance of the client, no matter how hard she tried to get it to use the second RPC port. BOINC Tasks was able to connect to, monitor, and manage both client instances at once. |
|
3837)
Message boards :
GPUs :
seti@home + linux +cuda
(Message 50092)
Posted 6 Aug 2013 by Richard Haselgrove Post: You'll find most of the relevant people hanging out in http://setiathome.berkeley.edu/forum_thread.php?id=72467 Or maybe you have already. Is 'randomaussie' the same thing as 'BlackDeath'? |
|
3838)
Message boards :
GPUs :
seti@home + linux +cuda
(Message 50090)
Posted 6 Aug 2013 by Richard Haselgrove Post: Ask at SETI's own message board. The project does not (yet) support Linux + Cuda natively, but there are third-party applications available. You could even help test and refine them, in the hope that they can eventually be promoted to stock project applications. |
|
3839)
Message boards :
Questions and problems :
please update version history file
(Message 50083)
Posted 5 Aug 2013 by Richard Haselgrove Post: While 7.2.7 was released to the public for Android, it did not see light of day on any other platform, other than for testing. We didn't even get that far. Even testers are back at v7.2.5 for Linux, OS X, and Windows, unless you can compile your own. |
|
3840)
Message boards :
Questions and problems :
Statistics page shows no data. Date is 01.Jan.70
(Message 50068)
Posted 3 Aug 2013 by Richard Haselgrove Post: First thing to do is to check that your computer's system clock is set accurately, and remains accurate through a complete power cycle (shut down, remove power input, wait a couple of minutes, power up again). If it fails that test, replace the CMOS battery on the motherboard, and repeat the test until you're sure that the system time is preserved through a restart. Then, navigate to your BOINC data directory, and delete the file statistics_[project url].xml That should fix it. |
|
3841)
Message boards :
GPUs :
How do I type in flops for an Nvidia 660 Ti?
(Message 50060)
Posted 2 Aug 2013 by Richard Haselgrove Post: Look up the documentation on app_info.xml: http://boinc.berkeley.edu/wiki/Anonymous_platform The <flops> entry belongs outside the <coproc></coproc> block where you have it now. |
|
3842)
Message boards :
Projects :
News on Project Outages
(Message 49645)
Posted 18 Jun 2013 by Richard Haselgrove Post: Who coughed wrong at the Seti database? Dunno, but it affects SETI Beta too. Got a 'feeder not running' error at 12:25:47 UTC - too early for regular maintenance, even if it's Jeff's day on the roster. |
|
3843)
Message boards :
BOINC client :
My Wish List
(Message 49606)
Posted 14 Jun 2013 by Richard Haselgrove Post: Actually that brings up another wish...separate CPUs from GPUs in terms of scheduling and such. To be honest, I find that the v7.0 scheduler is much more tightly integrated between CPU and GPU than the old one was. Up to and including v6.12, there was separate recording and accounting for 'debt' between projects, for the different compute units. A lot of time spent crunching with a GPU for a project didn't affect its CPU resource share. But with v7, all crunching (whichever compute unit) goes into a 'common pot' of REC - and if you crunch GPU for a project, BOINC vary rarely schedules CPU work for that same project. They're interlinked. |
|
3844)
Message boards :
Questions and problems :
Cannot install new Bonics!!!!!!!!
(Message 49587)
Posted 12 Jun 2013 by Richard Haselgrove Post: FAQ |
|
3845)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 49576)
Posted 11 Jun 2013 by Richard Haselgrove Post: Did I forget the candles? And what scent you used in the candles... |
|
3846)
Message boards :
Questions and problems :
BOINC causing PC performance problems since new version
(Message 49537)
Posted 8 Jun 2013 by Richard Haselgrove Post: HostID 3080266 belonging to Richard Gregory matches the description, if you want to start the diagnostics before he gets back to us with confirmation. |
|
3847)
Message boards :
Questions and problems :
BOINC causing PC performance problems since new version
(Message 49534)
Posted 8 Jun 2013 by Richard Haselgrove Post: Two separate issues. 1) Milkyway are currently sending out some WUs with some very dubious estimates. I have one running at the moment which started with an estimate of over two years. It reached 10% after 100 minutes, so I'm confident that it will finish in less than a day. But BOINC - particularly the new v7.0.64 - seems inordinantly slow to accept the evidence of its own eyes when a unit is over-estimated: the task is still showing 17,913 hours remaining to run, and I suspect will remain in high priority for many hours to come. All you can do is pester the project to be more careful in their workunit preparation (you'll find me there already). 2) I'm guessing, but the screen slowness and laggy mouse pointer might be related to teething problems with the new applications launched at SETI last week. [It's most unlikely to be BOINC itself to be causing this] Could you help to narrow it down, please, by suspending first one project at a time, then one task at a time, until you find a single task which clears the problem when you suspend it? Then, go to the message boards for the project responsible for the task, and post your experiences there - quote the full names of the Application and (task) Name that seemed to be the culprit. Project volunteers will be better placed to help you on that one. |
|
3848)
Message boards :
Questions and problems :
Suddenly no CPU tasks are crunched anymore, only GPU tasks
(Message 49531)
Posted 8 Jun 2013 by Richard Haselgrove Post: Are you by any chance related to 'The Great Sun Jester' at SETI? I am not getting work units for my CPUs |
|
3849)
Message boards :
Questions and problems :
boinc 6.12.34 question re:0 not working
(Message 49513)
Posted 5 Jun 2013 by Richard Haselgrove Post: Hummmmm Go to your account page http://setiathome.berkeley.edu/home.php At 'Computers on this account' (middle of third group), click 'View' On turn, for each computer that you want to configure, click 'details' Scroll right down, almost to the bottom of the page. See Location: choose one, and click 'update'. |
|
3850)
Message boards :
Questions and problems :
boinc 6.12.34 question re:0 not working
(Message 49505)
Posted 4 Jun 2013 by Richard Haselgrove Post: On your SETI@home preferences page - http://setiathome.berkeley.edu/prefs.php?subset=project - edit the second item to read: Use CPU Any CPU work you currently have will be completed, but your computer will stop fetching any new work for the CPU - then you'll be running GPU only as you wish. |
|
3851)
Message boards :
Questions and problems :
UBUNTU 11 - Boinc v7 not available, 6 doesn't work anymore. Bye.
(Message 49479)
Posted 2 Jun 2013 by Richard Haselgrove Post: Hello, I suspect that the first message you saw said that you needed SETI v7, not BOINC v7. |
|
3852)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 49463)
Posted 1 Jun 2013 by Richard Haselgrove Post: Thanks from me too. The url to check for updates is http://ucbsystems.org/ |
|
3853)
Message boards :
Questions and problems :
BOINC Running while PC Sleeps
(Message 49449)
Posted 31 May 2013 by Richard Haselgrove Post: "Know thine laptop". Mine has air inlet grilles on the base, and fan exhaust on the left-hand side. No sign of air movement on the keyboard side, except when I turn the speakers up loud. So, positive air pressure from the cooler stand underneath, and keeping desk clutter clear of the left side, works for me. All-night GPU crunching with the lid closed (and hence the screen drawing less power) works fine. |
|
3854)
Message boards :
Questions and problems :
Woffor
(Message 49441)
Posted 31 May 2013 by Richard Haselgrove Post: It's the GUFF announced in the front page news at SETI@home. SETI@home Version 7 has been released You signed up for that, remember? |
|
3855)
Message boards :
BOINC client :
7.1.1 manager seti 7.7 opencl-ati running out of order
(Message 49427)
Posted 30 May 2013 by Richard Haselgrove Post: Tasks always run in the order they were issued, not in deadline order. Sometimes if a block of tasks are allocated at the same time, a task in the middle of the block can complete its download first and start running (if a device is idle). But once all downloads are complete, the tasks should be processed in order from the top down (if you have the 'native' BOINC Manager display, with no column sorting applied). |
|
3856)
Message boards :
Questions and problems :
Installed Boinc on debian but can't use
(Message 49409)
Posted 29 May 2013 by Richard Haselgrove Post: Explain to me why I've never been able to run the BM when there is a password in the gui_rpc_auth.cfg [think I'm repeating]. Remove the password again by editing the file, restart BOINC and the BM connects without a wrong password warning. I think, on Windows, I've occasionally seen a 'wrong password' popup dialog when there's actually nothing wrong with the installation, configuration or anything. Try again a few seconds later, and it works. My working assumption is: based on other observations, some project applications go through a heavy start-up phase when called upon by the BOINC client. Einstein gravity wave and CPDN seem to be particular cases in point. When the core client starts one of these apps, it becomes unresponsive for a few seconds, until whatever heavy work the app has to do is finished, and it has settled into normal crunching (in the Einstein case, the workload seems to be loading and possibly decompressing several large binary data files). It seems to me that if the Manager attempts to connect with the core client during this unresponsive phase, then 'wrong password' is at least one possible response, where 'client busy' might be a more accurate description. Suggestions: 1) code-walk the source to see if 'wrong password' requires an actual message to that effect from the client, and isn't just a catch-all last chance 'oops, something went wrong' type of message. 2) try your damnedest to watch the behaviour, activity, and CPU usage of the core client, while trying to start the Manager and connect to said client. See if BM has any more luck connecting after the client has been running a little while. |
|
3857)
Message boards :
Server programs :
10 tasks ready to be sent but wont send even though boinc client is attached to project
(Message 49336)
Posted 26 May 2013 by Richard Haselgrove Post: 25/05/2013 10:25:22 PM | cplan | Scheduler request failed: Couldn't resolve host name Surely "Couldn't resolve host name" indicates a DNS error? "Resolve" isn't a firewall function. |
|
3858)
Message boards :
Questions and problems :
BOINC credit system
(Message 49308)
Posted 25 May 2013 by Richard Haselgrove Post: Thanks for letting my mind wander back to the alpha mail list discussion and the mechanism that had to be there to account for not-yet awarded credit [waiting on validation]. Also the REC doc is good reference. Thank you too. A project which has been running for a while will have accumulated a substantial REC. A newly-attached project, or a project which has been idle (not supplying work) for a while, will start with a REC at or near zero. That means that the new project will be given a higher priority for work fetch and running tasks, until its REC rises (and the other project's REC decays) to the point at which their respective REC values are in proportion to Resource Share. I think. And don't tell me about the problems trying to balance projects with GPU applications (REC in the 100,000 range) with CPU-only projects (REC in the 1,000 range)... You can accelerate the balancing process with <rec_half_life_days>X</rec_half_life_days> Smaller is good. |
|
3859)
Message boards :
Questions and problems :
BOINC credit system
(Message 49306)
Posted 25 May 2013 by Richard Haselgrove Post: ... Given that v7 clients are credit share driven, the more credit is given, the less computing time is accorded... at least this is how I interpret the new ways... to counteract over-awarding projects... Yes, that's how the main CreditNew document is written. But in fact, the current operational document is ClientSchedOctTen: rather than being credit share driven, v7 clients are REC driven - "There are problems with using project-granted credit as a basis for this approach". You bet. The estimated credit is scaled from speed*time, so effectively we're using a simple flop-counting scheduler: there's no resource-share punishment for projects granting high actual credit. Mind you, what the server does with the returned data, and how it turns it into credit, is a closed book to me. It's frequently been likened to a random number generator. |
|
3860)
Message boards :
Questions and problems :
Screensaver disable
(Message 49268)
Posted 23 May 2013 by Richard Haselgrove Post: The option to use, or not use, a screensaver has been present in the BOINC installer program for each version as far back as I can remember. If you select not to use a screensaver (via the 'advanced' button in the installer), that choice is remembered for subsequent version upgrades - as long as you do upgrade, that is, rather than uninstalling manually and starting again from scratch: then you have to re-make your default choices. |
|
3861)
Message boards :
BOINC client :
BM .64 cc_config.xml changed...
(Message 49265)
Posted 23 May 2013 by Richard Haselgrove Post: That's not new in v7.0.64 - for some time now, the Manager (and I think it is a Manager thing, not the core client) has created a fully-populated template for cc_config.xml in alphabetical order, every time you use the GUI tool for adding an exclusive application. It's a very useful feature, IMHO. |
|
3862)
Message boards :
Projects :
LHC sixtrack - no work?
(Message 49234)
Posted 22 May 2013 by Richard Haselgrove Post: Always check the Sixtrack Server Status page (or the equivalent at other projects) when you have a question like this. If there are no (or very few) 'Tasks ready to send', then it's unlikely that you will receive any work. Sixtrack, in particular, tends to release work in batches at unpredictable intervals: you may get some, or you may not. |
|
3863)
Message boards :
Questions and problems :
How to clean up an excessively large BOINC folder?
(Message 49213)
Posted 19 May 2013 by Richard Haselgrove Post: What about a project reset? Try it and see. I doubt it will be as comprehensive as a project detach. |
|
3864)
Message boards :
Questions and problems :
How to clean up an excessively large BOINC folder?
(Message 49207)
Posted 19 May 2013 by Richard Haselgrove Post: Look at the project sub-folders, one at a time, to find out which project(s) are taking up the bulk of the space - from my experience, BURP, CPDN and Einstein are potential candidates. Once you've identified where the space has gone, you can decide what to do. One tactic would be to set 'No New Tasks' for that project, finish (and report) all work in progress, and then detach/remove that project from your BOINC installation. That will remove the project folder completely: if you re-join the project, a new folder will be made, and only the currently needed files will be downloaded into it. |
|
3865)
Message boards :
Questions and problems :
BOINC causes random freezing of OS
(Message 49182)
Posted 17 May 2013 by Richard Haselgrove Post: the pc crashes regularly Can your PSU still supply all voltages within specification when under full load? |
|
3866)
Message boards :
Questions and problems :
BOINC upgrade download does not recognize external drive..
(Message 49127)
Posted 14 May 2013 by Richard Haselgrove Post: Jord, Please refer to trac ticket [trac]#815[/trac]: if the previous installation was on a drive letter which doesn't exist on the system any more, the installer (as of v6.4.5, four years ago) doesn't even get started. Bruce, There's a workround under my name at comment #6 in that ticket. Or see if you can set up a mapped drive using the drive letter of the old 500 GB drive, pointing somewhere - anywhere - that has some genuine online storage available. Judging by the ticket comments, it has to be a local drive, not network, but maybe a simple USB stick would be enough to let the setup program start? |
|
3867)
Message boards :
Questions and problems :
upgrade gone wrong,
(Message 49038)
Posted 7 May 2013 by Richard Haselgrove Post: cool, but how do i know what version i had before something deleted it? the shortcut on my desktop is no help... Look at the "Computers on this account" link on your account page at the website of any project that computer has been attached to. Since you haven't upgraded yet, the old version number will still be displayed. |
|
3868)
Message boards :
Projects :
News on Project Outages
(Message 48977)
Posted 4 May 2013 by Richard Haselgrove Post: NumberFields@home Thanks to both the project administrator and David Anderson working/debugging until after midnight, the project is up and running again. Users may need to retry 'update' a few times, until a master file download supplies a new scheduler address. Report any other problems on the project message board (not here), as usual. |
|
3869)
Message boards :
Projects :
News on Project Outages
(Message 48976)
Posted 4 May 2013 by Richard Haselgrove Post: NumberFields@home The project was taken offline around 18:00 UTC 3 May 2013 for a planned migration to new server hardware. The migration was expected to take ~4 hours, but the project remains offline while they wrestle with minor incompatibilities between BOINC and the new server's operating software. |
|
3870)
Message boards :
Questions and problems :
Computation Error
(Message 48955)
Posted 3 May 2013 by Richard Haselgrove Post: At least you left enough clues in your message log to find your computer (and you used the same user name for confirmation, bless you) You are using a NVIDIA GeForce GT 640 'Kepler' GPU, and the stock SETI@home Enhanced v6.10 (cuda_fermi) application. Use the environment variable workround from the second section of SETI message 1296126. Sorry, I'm not equipped to post a Windows 8 version of the screenshots. |
|
3871)
Message boards :
Questions and problems :
No useable GPUs found
(Message 48950)
Posted 3 May 2013 by Richard Haselgrove Post: There you go: 03/05/2013 08:52:40 | | Starting BOINC client version 7.0.64 for windows_x86_64 You installed in 'Protected Application Execution' or 'service' mode. GPU crunching is not possible, because of the new driver security features in Windows 7/8. Uninstall BOINC (your data, like CPU tasks, is kept safe), then reinstall but uncheck the PAE option: you need to click through to the 'Advanced' page in the installer to do this. |
|
3872)
Message boards :
Questions and problems :
BOINC & SpyBot Search & Destroy Immunization
(Message 48909)
Posted 1 May 2013 by Richard Haselgrove Post: 'Immuni(s/z)ation' presumably requires writing something into BOINC's program folder - and that's probably forbidden by BOINC's own sandbox security measures. That sounds like BOINC is doing its job properly. |
|
3873)
Message boards :
Questions and problems :
Latest Boinc Version Will not Install on Either of my Computers
(Message 48836)
Posted 28 Apr 2013 by Richard Haselgrove Post: There's a BOINC FAQ about this one. |
|
3874)
Message boards :
BOINC client :
Feature request: multiple project tasks on a single GPU
(Message 48810)
Posted 24 Apr 2013 by Richard Haselgrove Post: I suggest to add a Boinc feature to run multiple project tasks on a single GPU for stock project executables without the use of app_info.xml (<count>.33</count>). They beat you to it - already implemented! You need one (or more) app_config.xml file(s) - details are in the final section of application configuration. You can even load the app_config files and change the GPU loadings on-the-fly without exiting BOINC - though you are recommended to upgrade to BOINC v7.0.64 first - the recommended version since yesterday. |
|
3875)
Message boards :
Questions and problems :
Download Failures
(Message 48788)
Posted 23 Apr 2013 by Richard Haselgrove Post: What sort of file suffers the errors, and what sort of internet connection does that machine use? Some mobile connections compress image files on-the-fly to save bandwidth, and that would cause the checksum test to fail. Your 'computing preferences' (which are shared across all BOINC projects you participate in - you only have to set them once) should contain an option Skip image file verification? Try setting that. |
|
3876)
Message boards :
API :
How to make the WrapperApp to work?
(Message 48690)
Posted 19 Apr 2013 by Richard Haselgrove Post: This is the latest error message: That error message is typical if the file has been compressed with a newer version or zip utility, using features - typically a better compression ratio - that an older unzip utility doesn't recognise. Try making the utopia1_26012_windows_x86_64__opencl_nvidia_100.zip file again from the original sources, using less aggressive/more compatible compression settings, or an older utility. |
|
3877)
Message boards :
Questions and problems :
Limiting CPUs per task?
(Message 48668)
Posted 18 Apr 2013 by Richard Haselgrove Post: Depending on the project/application and the multi-threading technology used, it can be done through anonymous platform (app_info.xml) - though that needs some care and knowledge, and isn't recommended for a general user. I did ask David Anderson to extend the app_config.xml tool-set to include this, but he refused. |
|
3878)
Message boards :
BOINC Manager :
Screen saver!!!!!!!!!!!
(Message 48645)
Posted 16 Apr 2013 by Richard Haselgrove Post: Sorry not the answer. This is not associated with the Operating system. I have already checked that. The Boinc program is running in the background and when it senses inactivity it starts to display its currently running project graphics. I want to know how to stop Boinc from doing this! Humor us, please, and have a look in your operating system's Display/Screen Saver control panel anyway. If you see the word 'BOINC' there (and it might be, believe us), try changing it. |
|
3879)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 48582)
Posted 10 Apr 2013 by Richard Haselgrove Post: Uploading no, that is merely saving data from your hard drive to a hard drive on the server. Sure, it needs a scheduler contact to get the work allocated in the first place, but post-allocation, the download is free-standing and can continue after the scheduler and database are disabled. With the network access at the old server closet, and before we were told about RFC1323, downloads - especially 8MB Astropulse downloads - could keep freewheeling for literally hours after the supply faucet was turned off. That may be what gave the impression that workflow was continuous over the last few maintenance outages. |
|
3880)
Message boards :
BOINC client :
Bm counters stops for about 1min and then resumes. Any ideas?
(Message 48575)
Posted 10 Apr 2013 by Richard Haselgrove Post: I've been seeing that, on and off, for two years at least. I put it down to the new Windows laptop I bought then, rather than any particular version of BOINC. |
|
3881)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 48536)
Posted 4 Apr 2013 by Richard Haselgrove Post: I think SETI must be offsetting their new co-lo costs by sub-letting their old closet to other BOINC projects. SIMAP's download, upload and dropped connection rates this morning have been as bad as SETI's ever were. :-D |
|
3882)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 48522)
Posted 3 Apr 2013 by Richard Haselgrove Post: You had better not give too detailed directions in case ALL the SETI crunchers suddenly turn up on the doorstep and they haven't got enough cups to give everybody a coffee. Mark forgot to mention that Berkeley Espresso is between the Alehouse and the Hotdogs. I don't think we'll go short of coffee. |
|
3883)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 48518)
Posted 3 Apr 2013 by Richard Haselgrove Post: And another guess, is that this is where they relocated: It's the first place I've visited in Street View that has cycling directions. 14 minutes down the hill from the lab (if you go the steep way), 48 minutes to get back up. |
|
3884)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 48516)
Posted 3 Apr 2013 by Richard Haselgrove Post: And another guess, is that this is where they relocated: That building looks very similar to Warren Hall, 2195 Hearst Avenue (Google Street View) - the home of IST. |
|
3885)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 48500)
Posted 3 Apr 2013 by Richard Haselgrove Post: Not to spoil all the drinking and fun and eveything, but I just looked at the Seti Cricket graph and IT'S ALIVE! <maniacal laugh suppressed> Have some of the servers come back online, or has that graph been reassigned to something else? Did you read the figures, as well as looking at the pretty pictures? During normal running, the top figure om the Y- (left hand vertical) axis is 100 Megabits, and we use ~93% of that. Currently, the axis is scaled to 400 bits, and 93% of 400 bits is about three hours per WU. So 8 WUs per day, per world (and that's MB - I'll leave somebody else to work out the AP rate). Back to the beer... |
|
3886)
Message boards :
The Lounge :
The Seti is Slumbering Cafe
(Message 48489)
Posted 2 Apr 2013 by Richard Haselgrove Post: Hello Bill! So long as you do it quietly... |
|
3887)
Message boards :
Questions and problems :
I need a way to fine-tune download/upload back-off period
(Message 48395)
Posted 29 Mar 2013 by Richard Haselgrove Post: When the upload time limit for files was 14 days I opened a track ticket asking for more time: Your r18845 became Revision: 511637c9b6402b9275af6f86e83f0323fa5b893d
Author: David Anderson <davea@ssl.berkeley.edu>
Date: 14/08/2009 20:00:29
Message:
- client: in the final stage of CPU scheduling,
give preference to multi-threaded jobs.
Avoid running N-1 1-thread jobs and 1 N-thread job on N CPUs
- client: change file transfer giveup time from 14 to 90 days
svn path=/trunk/boinc/; revision=18845but it seems to have vanished again with all the messing around from SVN to GIT and now GIT-v2. |
|
3888)
Message boards :
GPUs :
PCIe Bus Saturation
(Message 48294)
Posted 21 Mar 2013 by Richard Haselgrove Post: Check before you buy.  If you only have a PCIe x1 slot, it's likely to be the very short physical connector. Most GPUs will require the full x16 connector, even if they're running at a lower speed. |
|
3889)
Message boards :
Questions and problems :
GTX Titan windows 8 problems
(Message 48221)
Posted 15 Mar 2013 by Richard Haselgrove Post: User 35589 at Primegrid. Computers hidden. Not possible to help. |
|
3890)
Message boards :
Questions and problems :
GTX Titan windows 8 problems
(Message 48216)
Posted 15 Mar 2013 by Richard Haselgrove Post: Nvidia Gtx Titan don`t run BOINC tasks when using windows 8 That's hardly surprising. BOINC is infrastructure, it doesn't supply any tasks for crunching: that's the job of the project science applications - which you haven't told us about. Any suggestions ? Yes. Read When requesting and giving help on these forums..., and think about what information we might need to answer your question. Then post it. |
|
3891)
Message boards :
Questions and problems :
can i turn off the floating, wavy, 3D effect?
(Message 48196)
Posted 13 Mar 2013 by Richard Haselgrove Post: There are instructions in this FAQ: Disable the Seti screen saver from moving |
|
3892)
Message boards :
Questions and problems :
Transferts OK with rosetta and KO for Seti (transient http error)
(Message 48149)
Posted 11 Mar 2013 by Richard Haselgrove Post: SETI's download link is just very, very busy. If you are running Windows (you don't say), enabling RFC 1323 on your network connection has been found to be a big help. Various tools (running an optimiser, adding a registry entry directly) are discussed in Windows TCP Settings - Help with server communication at the SETI message boards. |
|
3893)
Message boards :
BOINC client :
astropulse_6 won't stop computing!
(Message 48132)
Posted 10 Mar 2013 by Richard Haselgrove Post: I see. Astropulse is related, but different than SETI. It would probably be discussed - at some length - on the SETI message boards. And any fixed application would be downloaded automatically to your computer when you requested work, the same way the original one was. |
|
3894)
Message boards :
Questions and problems :
Cuda Computation error caused by unknown Nvidia driver version
(Message 48127)
Posted 9 Mar 2013 by Richard Haselgrove Post: We came across a similar problem with Windows versions of BOINC in August 2011: http://boinc.berkeley.edu/trac/changeset/32f4f68614a9e740423d38d33c0172703b9a6ed4/boinc One of the discovery machines was my Optimus 420M laptop: it has dual graphics adapters, a low-power Intel chip for extended battery life, and a high-power NVidia chip for demanding applications. The code as originally written in coproc_detect.cpp checked the first graphics adapter, found it couldn't get an NVidia driver version from it, and didn't check any further - it broke out of the test and returned "driver version unknown". We later heard that desktop computers whith multiple graphics cards, and no monitor attched to the NVidia card, were affected the same way, and cured by that fix. For Windows machines, and the projects we were aware of at that time, the missing driver version was a cosmetic irritation only, and didn't affect compuation - evidently WCG is more touchy. Going by dates alone, Windows versions from 6.12.37 onwards, and all versions in the 7.0.xx line, should be fixed - I can't help with Linux, I'm afraid. Edit - looking at the top hosts lists for Einstein, there are plenty of both ATI and NVidia GPUs, running both Linux and Windows. Driver versions are listed for every combo, except Linux/NVidia. And looking at what is now http://boinc.berkeley.edu/trac/browser/boinc/client/gpu_nvidia.cpp, I see no attempt to retrieve an NVidia driver version, except within an #ifdef _WIN32 directive. If WCG science apps require driver information (they seem to be the only project which does), they'll have to ask for it to be added to BOINC. |
|
3895)
Message boards :
Questions and problems :
Something wrong with BOINC Manager?
(Message 48120)
Posted 9 Mar 2013 by Richard Haselgrove Post: Ah yes, stupid me. Jord, You do need quotes round the program name, because of the space in the path name: "C:\Program Files\BOINC\boincmgr.exe" --gui_rpc_port=31416 --detach_console otherwise Windows treats it as a separator. |
|
3896)
Message boards :
BOINC client :
I can't believe I am the only one.....
(Message 48082)
Posted 6 Mar 2013 by Richard Haselgrove Post: Another thought. Your post at SETI Beta (which is what brought me here), showed your account with just one computer, running BOINC v7.0.52: Computers belonging to TRuEQ & TuVaLu Yet, your work fetch debug log, as posted here, bears no relation to a v7.0.52 log: 06/03/2013 19:35:02 | | [work_fetch] ------- start work fetch state ------- So, what version are you logging here? |
|
3897)
Message boards :
BOINC client :
I can't believe I am the only one.....
(Message 48077)
Posted 6 Mar 2013 by Richard Haselgrove Post: You have projects excluded from GPUs. There is active debugging work going on in this area, but as David said on Saturday, "Work fetch with GPU exclusions is proving to be tricky." On Monday (this week), Rom Walton added I've fallen behind on debugging some manager bugs, I have posted a more So you are not alone, but the work isn't finished yet. |
|
3898)
Message boards :
BOINC client :
I can't believe I am the only one.....
(Message 48074)
Posted 6 Mar 2013 by Richard Haselgrove Post: I hope you can make some sence from this. 2013-03-06 19:19:51 | Moo! Wrapper | Scheduler request completed Yes, that you don't know the meaning of "one iteration only". OK, I'll have a think. |
|
3899)
Message boards :
BOINC client :
I can't believe I am the only one.....
(Message 48069)
Posted 6 Mar 2013 by Richard Haselgrove Post: But I did click on the update button on both projects and none tried to fetch new work... They will do that as soon as they are out of work. How can you demonstrated that they didn't request work, if you don't post the work fetch log - which these days is copiously annotated with the reason why work isn't being requested, for each separate project? Edit - I see you posted again while I was typing. Good, that'll help you work it out. |
|
3900)
Message boards :
BOINC client :
I can't believe I am the only one.....
(Message 48066)
Posted 6 Mar 2013 by Richard Haselgrove Post: It would be a great deal more help to show how (one iteration only) of [Work_fetch_debug] looks in your message log. From that screen, it looks as if Moo! Wrapper is restricted by a 'maximum tasks in progress' setting, and SETI@home is restricted by their notoriously overloaded download link. Both of these are project issues, nothing to do with BOINC. On the subject of the SETI download link: somebody with a better knowledge of networking protocols than me has finally found the relevant page in the manual: RFC 1323. There's an active dicussion with examples and HowTo going on at the moment in Windows TCP Settings - Help with server communication. We'll write it up in a sticky over the next few days, once all the wrinkles have been tested and ironed out. |
|
3901)
Message boards :
BOINC client :
astropulse_6 won't stop computing!
(Message 48054)
Posted 5 Mar 2013 by Richard Haselgrove Post: Is Astropulse just there to draw the 3D spectrum analyzer display during screen saver mode, or is it doing useful calculations for the project? Have a look at the Astropulse FAQ page. Some of the details are horribly dated, but the basic description of what the project is about is still sound. |
|
3902)
Message boards :
BOINC client :
astropulse_6 won't stop computing!
(Message 48028)
Posted 3 Mar 2013 by Richard Haselgrove Post: I'm using Windows XP. What directory has the event log in it? The current event log can be displayed on screen by pressing Ctrl+Shift+E, or from the 'Advanced' menu in BOINC Manager. Older logs are in the BOINC Data directory. The current location for the data directory is re-displayed in the current log every time BOINC restarts, in the first three or four lines. The file name to look for is stdoutdae.txt, or for older messages stdoutdae.old |
|
3903)
Message boards :
GPUs :
GeForce TITAN and BOINC
(Message 48021)
Posted 2 Mar 2013 by Richard Haselgrove Post: User 'ExchangeMan' has one installed and running at SETI - see SETI message 1342424. Very early days yet, still being used primarily for testing. |
|
3904)
Message boards :
BOINC client :
astropulse_6 won't stop computing!
(Message 48018)
Posted 2 Mar 2013 by Richard Haselgrove Post: Is this a known bug? It's slowly becoming better known, as more people report it. We'll try and get a message back to the programmer. Edit - it would help if you could identify exactly which application seems to be exhibiting this problem. When you go into Task Manager to end the process(es), you will see the full program names: if you could note them down, and close them one at a time - then tell us at which point the jerkiness stops. That would be a great help, thanks. |
|
3905)
Message boards :
Questions and problems :
Preferences when to use GPU - problem
(Message 48016)
Posted 2 Mar 2013 by Richard Haselgrove Post:
That's curious, because I was looking for your account at SETI, and you appear to be the owner of computer 6922055. That computer has tried to run over 60 "SETI@home Enhanced v6.10 (cuda_fermi)" jobs on the GPU, but has failed with an error every time so far. It looks to me like the error you get without that environment variable: the card and driver look OK, so I can't think of any other obvious reason for the failures. I hope you used copy and paste to take the variable name from my post: people have come to grief, even when they were sure they'd re-keyed it correctly. |
|
3906)
Message boards :
Questions and problems :
Preferences when to use GPU - problem
(Message 48011)
Posted 2 Mar 2013 by Richard Haselgrove Post: Hey Everyone--I have a related, but opposite, issue: I'm wondering if my GPU is underused. No, the SETI application will be using a lot more than 1/768th of your new GPU. With that combination, please make sure that you read my thread NVidia driver problems which cause computation errors - once you set the environment variable described (and illustrated) there, you should quickly earn enough credit to achieve posting rights. |
|
3907)
Message boards :
BOINC client :
project_attach error :invalid url:URL
(Message 47901)
Posted 21 Feb 2013 by Richard Haselgrove Post: how can i do this? Ask your institution's network management/administration team, or a fellow student on a networking course - depends on your circumstances, you haven't told us anything about that. |
|
3908)
Message boards :
BOINC client :
project_attach error :invalid url:URL
(Message 47899)
Posted 21 Feb 2013 by Richard Haselgrove Post: You need to look at your active DNS server, and check that it returning the appropriate IP address to match the URL and server IP address you are trying to use. |
|
3909)
Message boards :
Projects :
News on Project Outages
(Message 47891)
Posted 21 Feb 2013 by Richard Haselgrove Post: Seti's website is unavailable at this time (11:21pm EST) Only the upload server has failed this time (plus the web pages, of course). Scheduler and download server are still functional, so new work can be fetched - although BOINC will stop requesting it once a backlog of completed tasks has built up on the computer. |
|
3910)
Message boards :
Projects :
News on Project Outages
(Message 47826)
Posted 18 Feb 2013 by Richard Haselgrove Post: SETI@home Project web pages are down, but the actual crunching (scheduler, uploads, downloads) are working normally. |
|
3911)
Message boards :
BOINC client :
GPU starved and client responds "don't need"
(Message 47823)
Posted 17 Feb 2013 by Richard Haselgrove Post: Any more ideas before i downgrade to 7.0.28 again? Learn to read your own logs. Here's a single work fetch cycle from lower down, and its outcome: 2013-02-17 08:45:38 | | [work_fetch] work fetch start At that time, OProject had the highest priority (note that all numbers are negative): SETI@home had middle priority: and Moo! Wrapper had lowest priority. Both OP and SETI were being delayed before attempting to pester their servers again ('RPC backoff' - Remote Procedure Call). Moo was fetchable - work was requested and allocated. That's how it works. If you try to bypass normal scheduling by clicking the update button, BOINC will only request work from the highest priority project. We have had bugs with that: it should be the highest priority fetchable project, and I think it's fixed now in v7.0.52. You would have to match up the work fetch cycle in the log with the time you clicked the button. Please do that in the peace and comfort of your own home: we don't need a new log snippet for every twist and turn in your search. |
|
3912)
Message boards :
BOINC Manager :
boinccmd problem: can't connect to local host
(Message 47816)
Posted 17 Feb 2013 by Richard Haselgrove Post: If it's a local server, consider how your network is (or isn't) establishing the correct connection between your client computer and your server. See my recent message 47709 for some pointers. |
|
3913)
Message boards :
BOINC client :
GPU starved and client responds "don't need"
(Message 47805)
Posted 17 Feb 2013 by Richard Haselgrove Post: Where do I find a moderator here to change thread name to " GPU starved and client gives answer "don't need"??? You can change it yourself by editing that last post. |
|
3914)
Message boards :
BOINC client :
GPU starved and client responds "don't need"
(Message 47790)
Posted 17 Feb 2013 by Richard Haselgrove Post: Regarding the thread title: "don't need" is not a server response. It is a client-generated information message, explaining why the client didn't request new work: as is clear from the context 2013-02-17 08:36:04 | Moo! Wrapper | [sched_op] Starting scheduler request If the OP could amend the thread title, and focus on the real issue, there might be something there we can look into: I'm not quite sure why no work would be needed, from that mess of a log file. |
|
3915)
Message boards :
BOINC client :
GPU starved and client responds "don't need"
(Message 47788)
Posted 17 Feb 2013 by Richard Haselgrove Post: Not sure about the wisdom of controlling operations with having both a app_config.xml and app_info.xml. The app_config is a replacement control file to app_info, so maybe there's a conflict between these 2. They work fine together. But if I'm using both, I'd put everything in app_info that is defined there, and only spill over into app_config for the tags which can't be defined anywhere else - basically, just max_concurrent at this stage. |
|
3916)
Message boards :
Server programs :
couldn't resolve host name
(Message 47709)
Posted 10 Feb 2013 by Richard Haselgrove Post: 192.168.0.12 is a private, non-routable IP address. You don't say anything about two important factors: 1) Whether the Windows/UBNT clients are trying to contact your server by name (I guess so, because BOINC is trying to resolve a hostname in a url) or by IP address. What <scheduler_url> appears in the client_state.xml file of your clients? 2) Whether your test clients are on the same physical 192.168.0.xyz local subnet, or whether they are attempting to connect to the server over a wider intranet or even the public internet. To address (1): if the clients are trying to contact the server by name, you will have to be using a local DNS server on the private network (which the client machines will have to use as their primary DNS server - use a forwarder to resolve external names outside the local scope). To address (2): if you are expecting to allow access to the server beyond the local subnet, you will have to register a public url with the global DNS server network, and configure it to resolve to a public (routeable) IP address on your internet connection. Then, configure your peripheral firewall/router to forward BOINC traffic to your server's private IP address. If you are working within an organisation that has a system administrator / network manager, consult them. |
|
3917)
Message boards :
Questions and problems :
How can I increase the amount of memory my projects use?
(Message 47701)
Posted 9 Feb 2013 by Richard Haselgrove Post: Do you really mean that you have 100 GB of (available) RAM in that machine? Or might you be saying that you have 100 GB of available disk space? Over the years, I've heard the word 'memory' used to describe either or both of those. |
|
3918)
Message boards :
Questions and problems :
WHS2011 - will not use GPU
(Message 47659)
Posted 6 Feb 2013 by Richard Haselgrove Post: I've never seen that error before, and can't find it in the source code. So I'll have to ask the developers. You'll have to ask Raistmer. It's an application (nVidia OpenCL) error, not a BOINC error. http://setiathome.berkeley.edu/result.php?resultid=2823247964 |
|
3919)
Message boards :
Questions and problems :
Boinc Install Exclusive service only
(Message 47471)
Posted 22 Jan 2013 by Richard Haselgrove Post: Don't think you can do that on Windows, just Linux. You can use GPUs in service mode under Windows XP, up to and including BOINC v6.12.34 But you can't use GPUs in service mode under Vista or Windows 7, whichever version of BOINC you choose. |
|
3920)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 47411)
Posted 20 Jan 2013 by Richard Haselgrove Post: Even if there's no GUI (I very rarely use the GUI for cc_config.xml, I prefer to edit it manually), it would be nice if app_config.xml could be read and acted upon while BOINC is running - the same as (most tags in) cc_config.xml, and global_prefs_override.xml I appreciate that some changes introduced via cc_config.xml (use/don't use all GPUs, for instance) are significant enough to justify a full restart, but I think the changes which can be controlled by app_config.xml (so far) are simple enough that the core client could accept them dynamically. |
|
3921)
Message boards :
Documentation :
fedora 64 bit considerations
(Message 47383)
Posted 18 Jan 2013 by Richard Haselgrove Post: Done. |
|
3922)
Message boards :
Questions and problems :
Domain controller recommended BOINC version?
(Message 47378)
Posted 18 Jan 2013 by Richard Haselgrove Post: The warning about Domain Controllers is at http://boinc.berkeley.edu/wiki/Release_Notes#BOINC_7_incompatible_with_Domain_Controllers. As was BOINC v6. When are they going to get round to [trac]#652[/trac]? |
|
3923)
Message boards :
Questions and problems :
Multi-core Computing
(Message 47370)
Posted 18 Jan 2013 by Richard Haselgrove Post: Might you be able to specify the number of CPUs used using the new app_config.xml feature of 7.0.4x? The <max_ncpus> - probably yes. The <cmdline>--nthreads - almost certainly not. So you would probably find that the application used a different number of threads from the number of cores that BOINC had allocated. I can't think that that would be a good idea, in general. |
|
3924)
Message boards :
Questions and problems :
Multi-core Computing
(Message 47359)
Posted 17 Jan 2013 by Richard Haselgrove Post: Hi Richard, If Sim1 is set up the same way as AQUA used to be (which is the same as the way the new Milkyway app appears to want to run under app_info), the number of CPU cores the task is going to use is 'locked in' to that task at the time the task is allocated by the server. In other words, to perform the sort of tests you're describing, you need to * Set NNT, and finish existing work * Change to the CPU setting you want to test * Allow new work to be allocated at that CPU setting Clumsy and clunky, but the only way I know to perform a valid test - unless you want to dive into client_state.xml every time you change a CPU setting, and edit * Both <avg_ncpus> and <max_ncpus> for the application * <cmdline>--nthreads 8</cmdline> The first determines whether or not BOINC will schedule the task with the number of cores you've chosen: the second determines how many threads the application actually spawns, irrespective of the number of cores BOINC is expecting it to use. The other complication is the number of GPU(s) and GPU tasks you're trying to run at the same time. If you add up all the fractional CPU demands for the GPU tasks (the fictional numbers displayed by BOINC, not the actual amount you might calculate for yourself), and if the total is equal to or greater than 1.000000, BOINC probably won't schedule everything you want to run concurrently. At AQUA, I got round that by wrapping the MT AQUA apps in an app_info.xml file specifying <....ncpus> as 2.1 or 3.1, for apps I wanted to run 3-threaded or 4-threaded respectively with matching <cmdline> settings. That allowed an extra 0.9 CPUs for the GPU apps I was running: but that experiment long predated the changeset I pointed you to, and I haven't got far enough into Milkyway's N-body apps to re-test yet (we're still waiting for the v1.06 bug-fix release). |
|
3925)
Message boards :
Questions and problems :
Multi-core Computing
(Message 47353)
Posted 17 Jan 2013 by Richard Haselgrove Post: If an mt-app, such as Sim1, runs using all the CPU cores/threads, and there is a GPU in the system, no work will be done on the GPU while the mt-app is running. I remember that one. I think it was a client bug that only lasted a few versions, and I thought it had been fixed. Does it still happen with the v7.0.44 current alpha? Edit - ah, yes, there it is. Changeset f2c38486, July 2012. |
|
3926)
Message boards :
Questions and problems :
Task exited with zero status but no 'finished' file... (almost) all of them
(Message 47352)
Posted 17 Jan 2013 by Richard Haselgrove Post: Be careful messing around with the system clock - far worse things that an 'exit 0' can happen. Have a look at trac ticket [trac]#588[/trac] - from 2008 and BOINC v5.10.45, but still open (although I think they put in fixes for some of the grosser errors since then). |
|
3927)
Message boards :
Questions and problems :
Multi-core Computing
(Message 47345)
Posted 17 Jan 2013 by Richard Haselgrove Post: You can read our fun and games with a true multi-threaded application at MilkyWay Nbody 1.04. The app is MT, BOINC fully schedules and supports it - but the project admins haven't quite mastered the art of loading it on their servers with the required parameters yet. |
|
3928)
Message boards :
Questions and problems :
Task exited with zero status but no 'finished' file... (almost) all of them
(Message 47309)
Posted 16 Jan 2013 by Richard Haselgrove Post: The BOINC service can be stopped with the net command, not sure if it's "net BOINC stop" or "net stop BOINC" but you'll figure it out. And it might be boinc-client instead of BOINC. That only works if BOINC is running in 'daemon' or service mode. The OP has told us it's running in 'user' mode: 13/1/2013 12:00:48 | | Running under account User The stop/restart can still be achieved with a batch file, but you'd need to find the boinccmd tool and invoke that - I probably have some code I could dig out. |
|
3929)
Message boards :
Questions and problems :
Task not paused when computer in use
(Message 47172)
Posted 11 Jan 2013 by Richard Haselgrove Post: It would only log the GPU switch-off if there was an actual GPU task running at the time. |
|
3930)
Message boards :
Questions and problems :
Task not paused when computer in use
(Message 47159)
Posted 10 Jan 2013 by Richard Haselgrove Post: I've changed the preference on my SETI page and on the BOINC manager, so I'll see how the PC copes. It's relatively old, and can be RAM-bound as I run 4Gb on Vista 32-bit - I understand I'd need to change to 64-bit to use any more RAM. (Reminds me of having to use QEMM to access above 640K on DOS!) Yes, from the (local) activity menu - that takes effect immediately, whereas the website change might not be noticed for several hours (at the next update/work fetch/work report). |
|
3931)
Message boards :
Questions and problems :
Task not paused when computer in use
(Message 47156)
Posted 10 Jan 2013 by Richard Haselgrove Post: @RichardH Happy to do some testing on this as I'm based at home for the next week or so. Just let me know what you want me to change... I was just thinking that if you weren't doing intensive work on the computer, you might try turning off the 'suspend computation when user is active' preference, and instead try suspending the GPU manually at random intervals when you happen to be passing. See if you get the same occasional run-ons. |
|
3932)
Message boards :
Questions and problems :
Task not paused when computer in use
(Message 47149)
Posted 9 Jan 2013 by Richard Haselgrove Post: Would turning off the GPU processing show if that's causing the problem? I'm pretty sure that turning off GPU processing would send exactly the same 'stop it' message to the application as the 'user active' or 'CPU busy' cases - with, presumably, the same chance of success. It might be a quicker and more manageable way of getting to the bottom of the problem, if you had a few spare minutes to devote to testing sometime? |
|
3933)
Message boards :
Questions and problems :
Task not paused when computer in use
(Message 47141)
Posted 9 Jan 2013 by Richard Haselgrove Post: Sorry, Holmes is still suffering from a caffeine under-dose this morning. Dr Watson has just pointed out the last few lines - that AP task managed to finish while computing was suspended and the task itself pre-empted. Could you add one more line to your cc_config logging section, please? <checkpoint_debug>1</checkpoint_debug> That will provide more evidence if an application is active when it shouldn't be, without adding too much to the clutter. Normal SETI apps checkpoint regularly, but Astropulse more rarely - IIRC, only 111 times in the course of a run. CPDN can also run for maybe 15 minutes between checkpoints. Ah - and another check I should have made: All tasks for computer 4499642. You're not using the standard project application as I assumed, but a '(opencl_nvidia_100)' contributed by a third-party developer - and only deployed on 21 Aug 2012. Does that sound like the right sort of starting date for your problems? OpenCL indicates that this is a GPU app: your log only indicates suspending CPU activity, although it does attempt to preempt the AP task too. Just to be on the safe side, it would be good to ensure 'Use GPU based on preferences' is selected in the middle of the three sections of the Activity menu (BOINC Manager, advanced view). The task in your log (2786568256) did manage to suspend/resume several times during the run: Termination request detected. GPU device synched, awaiting termination... but it seems to have missed the final call. I'll have a think about that one. |
|
3934)
Message boards :
Questions and problems :
Task not paused when computer in use
(Message 47140)
Posted 9 Jan 2013 by Richard Haselgrove Post: OK, the monitoring seems to be in place - we can see the "preempting" at 08:48:30 (that's BOINC giving the instruction for the apps to get out of the way). Now, we just have to wait and catch it in the act, and see whether (just possibly) Astropulse gets sent the 'stop it' message, but fails to act on it. That's certainly one possibility - I've documented an Einstein application behaving that way - but it's not been reported as a common problem. The current Astropulse application (v6) has been operational since 13 Mar 2012, and I'm pretty sure that it's not a deliberate "dark side" operation as Dagorath suggests. But I wouldn't rule out bugs in the BOINC API library that handles messaging between the BOINC client and the science applications. |
|
3935)
Message boards :
Questions and problems :
Task not paused when computer in use
(Message 47135)
Posted 8 Jan 2013 by Richard Haselgrove Post: Does that look right? I wasn't sure if you meant cpu_sched or cpu_sched_debug, so I put them both in. That looks fine (without full proof-reading) I just meant <cpu_sched>. That should be enough for this hunt, and doesn't flood the log too much. Many moons (and many versions) ago, preempting would be reported at the default logging level, but that was taken out in one of the dumbing-down cycles - now it has to be invoked specifically. <cpu_sched_debug> - as you've probably found by now - is more verbose, and probably more than you need at the moment. |
|
3936)
Message boards :
Questions and problems :
Task not paused when computer in use
(Message 47132)
Posted 8 Jan 2013 by Richard Haselgrove Post: Best to set the set the <cpu_sched> debug logging flag while testing this, so that you can see explicit messages like 08/01/2013 13:32:46 | GPUGRID | [cpu_sched] Preempting 34px4-NOELIA_hfXA-21-30-RND8601_0 (removed from memory) when they do - or don't - happen. |
|
3937)
Message boards :
Questions and problems :
Problems with BOINC under Linux / Ubuntu 12.04 (BOINC 7.0.27 / 42)
(Message 47011)
Posted 2 Jan 2013 by Richard Haselgrove Post: Regarding idle detection: There was an active discussion on this subject over the developer mailing list yesterday, leading to trac changeset [20daf38]. That won't be in any pre-built client yet, but if you have the ability to compile your own from a git trunk checkout, it might be worth a test. |
|
3938)
Message boards :
Questions and problems :
BOINC on ARM devices (round 2)
(Message 46789)
Posted 13 Dec 2012 by Richard Haselgrove Post: This is more of a passing interest, so, although I'm not involved in this, I found the subject title of this thread intriguing and would like to indicate that if you are looking to try SETI on an ARM, you'll need to downgrade to BOINC 6.x and similarly SETI 6.x for now. Note that SETI v_7 is still in (official) Beta testing, and subject to change before release. SETI v_6 is all you need for running SETI on their main project for now - although that's been liable to change at short notice "real soon now" for well over a year. |
|
3939)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 46772)
Posted 13 Dec 2012 by Richard Haselgrove Post: At any rate, tried "Post to Thread", but it would not either, so not sure what "standalone-post" is, but will let it be.[/Off-topic] Try editing your message 46771 - that was what I meant by stand-alone. If you can't edit the title at the same time as editing the post, then the message board software has developed a bug since last time I tried it (some years ago). |
|
3940)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 46769)
Posted 13 Dec 2012 by Richard Haselgrove Post: [Off-topic], thought the thread owner could edit the title and am sure I saw some instruction [to e.g. insert [RESOLVED], but beat me, either editing a post or appending a post, can't get the title to turn into edit mode.[/Off-topic] Make a new stand-alone post in the thread (not a reply or quote): then edit the new post. |
|
3941)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 46765)
Posted 13 Dec 2012 by Richard Haselgrove Post: People, you can stop PMing me with requests that I forward all your enhancement request information to the developers. I already did that mostly based on the contents of this thread. Could you summarise what requests you've received, please? I do wish people would post in public on the boards - when everything is done through secret back-channels, the rest of us don't know what would be a duplicate request/report, and what might be new. (That isn't a criticism of you, Jord - I appreciate your role as a filter/firewall protecting the staff from the raw undigested flood of user feedback: but I do think bugs/problems/wish lists should be aired for peer comment and review). For my part, I suspect that the new features - especially fractional GPU usage - still need to be hooked up to the runtime estimation and work fetch processes, but I need to study them in more detail before I can make a proper report. |
|
3942)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 46728)
Posted 12 Dec 2012 by Richard Haselgrove Post: I haven't noticed any problems with CPU apps, but for GPU apps there seems to be perhaps a '<' test where there should be a '<=' test. <gpu_usage>0.50</gpu_usage> only allowed one task to run, but <gpu_usage>0.49</gpu_usage> allowed two tasks. |
|
3943)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 46718)
Posted 11 Dec 2012 by Richard Haselgrove Post: Ooops, my bad. I had looked back at an earlier version of the documentation, and forgot to come back forward to the present day before copying and posting. Try these instead - and let's hope I don't get duplicated </app> tags this time. 1) For CPU-only apps. <app_config> 2) For GPU-only apps <app_config> From the changesets, it looks as if you may have to use BOINC v7.0.41 (or v7.0.42, just available) for the GPU version to work consistently. |
|
3944)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 46716)
Posted 11 Dec 2012 by Richard Haselgrove Post: I've been playing around with this feature, and it seems to be very useful in two quite different forms. 1) For CPU-only apps. <app_config> As many users will have noticed, when a project is underworked in BOINC v7 (either a newly attached project, or following a server outage), the client has a tendency to both fetch and schedule that project exclusively until REC and priority start to balance out with other pre-existing projects. This usage of app_config can help to even out the balance between projects, though I fear boinc may starve low-priority projects while the re-balancing process finishes - I'm testing this. 2) For GPU-only apps. <app_config> has the effect of supplying a <count> value, and allowing multiple copies of a GPU app to run on a single GPU resource - something popular with GPU users. This is much easier and safer than using a full app_info.xml file, because it doesn't disable the automatic downloading of a new application version if the project server issues one. |
|
3945)
Message boards :
Questions and problems :
BOINC 'suspened' but progress bar keeps moving
(Message 46713)
Posted 11 Dec 2012 by Richard Haselgrove Post: And that's with which BOINC version and on which operating system? Provided both activity settings are set correctly: * Run based on preferences * Use GPU based on preferences there's no way that idle detection could work for the CPU but not work for the GPU. I've submitted two documented examples of the Einstein Windows GPU application (BRP4 v1.32) failing to respond to a 'suspend' instruction from BOINC (BRP4 1.31/1.32 GPU app release: feedback thread) In one case, two tasks were running on the same card at the same time: one noticed that it was supposed to suspend, the other carried on regardless. I'm beginning to think this is primarily an application problem, rather than a BOINC problem - it would be helpful if future posters to this thread could indicate which project/application is active when they observe the problem. |
|
3946)
Message boards :
BOINC Manager :
System indefinitely slow after deactivating GPU
(Message 46700)
Posted 10 Dec 2012 by Richard Haselgrove Post: If all is running as it should be, the BOINC client itself will place a vanishingly small demand on your GPU - certainly undetectable at your desktop. All the BOINC client does is to start and - crucially for your question - stop a science application supplied by one or other of the projects you are attached to. If your general preference is "Don't use GPU when computer is in use", then BOINC should detect your usage, and send a message to the science application to 1) stop GPU processing 2) unload itself from GPU memory There are occasional reports from the Windows world of science applications which fail to notice the 'stop' message, and continue processing when BOINC thinks they are inactive. From what you say about temperatures, it sounds as if your science app has stopped, but maybe it hasn't released the card's video memory? It might be worth discussing your problems with other users of that application on the project's message board. |
|
3947)
Message boards :
Questions and problems :
SETI@home Enhanced needs 32.00MB more disk space
(Message 46690)
Posted 9 Dec 2012 by Richard Haselgrove Post: A quick search at SETI@home finds no users called 'NukeSmith' - so we can't see anything about your account or your computer, how many other projects you might have joined (and might be using some of BOINC's allocation of disk space), etc. It's far better to post this sort of question at the project concerned, from the account you use to crunch. In this case, either Number crunching or Questions and answers would be appropriate. |
|
3948)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 46688)
Posted 9 Dec 2012 by Richard Haselgrove Post: Well, on documentation, apparently we're not to use that one, but the 'official' one at http://boinc.berkeley.edu/wiki/Client_configuration#Application_configuration .. yeah, I don't get it either. ;-) That's the one which starts "There are three configuration mechanisms...", numbered 1 Configuration file ? ;-) Edit - the document we've all been editing is directly linked from the front page of Software Development (http://boinc.berkeley.edu/trac/wiki/SoftwareDevelopment?version=127#Currentdevelopment), so I think it's as well that we keep it accurate, too. |
|
3949)
Message boards :
BOINC client :
BOINC 7.0.40-42 and new app_config.xml
(Message 46683)
Posted 9 Dec 2012 by Richard Haselgrove Post: A user at WCG pointed out that the wiki from which I copied the template has a closing tag error, a missing s in </gpu_versions>. He's got it to work as expected, to the point that if max per setting is running and no other is buffered, threads will go idle. In his example had 2 SIMAP running, 4 CEP2 buffered, and app_config set to allow only 1 CEP2 to run at a time. His quad had 1 idle core. Wiki updated - thanks. |
|
3950)
Message boards :
Questions and problems :
CPU Non-intensive projects not considered with limited network connection
(Message 46664)
Posted 8 Dec 2012 by Richard Haselgrove Post: It rather depends what the non-CPU intensive applications are supposed to do. The one I know best is QCN (Quake Catcher Network). It's supposed to do nothing much most of the time, until an earthquake happens: then, it's supposed to pass on details as quickly as possible so that the epicentre can be calculated and appropriate warnings issued. There's not a great deal of value in running that application, if lack of network access prevents the warning being issued for up to 23 hours. Considerations like that may have influenced the designers of BOINC. What is the work that would be accomplished (non-intensively) if the app in question was allowed to run continuously without network access? |
|
3951)
Message boards :
Questions and problems :
burp running on 8 CPU and no GUP WU running
(Message 46661)
Posted 8 Dec 2012 by Richard Haselgrove Post: It probably depends which version of BOINC you are using - you did remember to tell us that, didn't you? There was a bug, introduced around v7.0.22 (while fixing another bug) and corrected in v7.0.31. It had the sort of effect you are describing. BTW, it's (some Intel) CPUs which are hyper-threaded. An application like BURP's will be multi-threaded, as Jord says. That means you have to look for the acronym MT, rather than HT, in the version change log thread I've linked. |
|
3952)
Message boards :
BOINC client :
BOINC 7.0.40 Intel GPU test project?
(Message 46648)
Posted 8 Dec 2012 by Richard Haselgrove Post: Hello, I think you'll find that v7.0.40 is very much a 'work in progress' as far as Intel GPU (OpenCL) support is concerned. For example, changeset [1a76e1a4] seems to be a critical component, and it was only added 12 hours after v7.0.40 was made available for download. As far as projects are concerned, I think the answer so far will be 'no' - unless you feel able to write one yourself? But from what I've read, it shouldn't be too difficult to convert existing OpenCL code to run on the Intel platform. Look for projects which have existing OpenCL applications, adequate resources, a wide user base, and a history of open Beta testing. They will probably be the first to appear. |
|
3953)
Message boards :
BOINC client :
Where did the developer "Check-In" list go?
(Message 46531)
Posted 1 Dec 2012 by Richard Haselgrove Post: Not a hub, still a local BOINC repository - but in git format now. http://boinc.berkeley.edu/trac/browser/boinc/checkin_notes works for me, if you really want the checkin note file: if you want to know what code the checkin note refers to, try the timeline. |
|
3954)
Message boards :
Projects :
News on Project Outages
(Message 46519)
Posted 30 Nov 2012 by Richard Haselgrove Post: The projected SETI resumption date has been changed to "Tuesday, Dec 4." |
|
3955)
Message boards :
Projects :
News on Project Outages
(Message 46510)
Posted 30 Nov 2012 by Richard Haselgrove Post: They don't say which year. The next time December 6th falls on a Tuesday is 2016... |
|
3956)
Message boards :
Projects :
Matching Projects to Resources
(Message 46486)
Posted 27 Nov 2012 by Richard Haselgrove Post: You might be able to get some idea from WUprop@home - that takes you to a page with a column which gives some idea of data transfer per day, even if the column for data download per WU seems to be broken. |
|
3957)
Message boards :
BOINC client :
- exit code -1073741515 (0xc0000135)
(Message 46332)
Posted 17 Nov 2012 by Richard Haselgrove Post: Hi, The generic description is "The application failed to initialize properly". That most commonly means that one of the application's components or support files - typically a Dynamic Link Library (DLL) - is missing. If you're experienced in peering under the hood in Windows, Dependency Walker can help you identify what's missing. Another possible cause of the 0xc0000135 error is file system security settings not allowing the application to run, but in my experience that's less likely. |
|
3958)
Message boards :
GPUs :
Incorrect GPU detection
(Message 46311)
Posted 16 Nov 2012 by Richard Haselgrove Post: could you tell me, please, which method is using to detect GPU and their number? http://boinc.berkeley.edu/trac/browser/boinc/client/gpu_amd.cpp http://boinc.berkeley.edu/trac/browser/boinc/client/gpu_opencl.cpp |
|
3959)
Message boards :
Web interfaces :
Trac links via BBcode in BOINC message boards
(Message 46284)
Posted 15 Nov 2012 by Richard Haselgrove Post: Just starting to explore and test this, following the transition to GIT. For example, old-style changeset links like [trac]changeset:25601[/trac] produce unfortunate results. I have a preserved url to the same place, which doesn't work either: http://boinc.berkeley.edu/trac/changeset/25601/boinc/trunk/boinc/lib/error_numbers.h The replacement should be [trac]changeset:7f6ab8d[/trac] or http://boinc.berkeley.edu/trac/changeset/7f6ab8d49eeaa6df58bf3bc4f61f2f7540d2c7e7/boinc - let's see how we get on with those. |
|
3960)
Message boards :
Projects :
MilkyWay - N-body application errors
(Message 46265)
Posted 13 Nov 2012 by Richard Haselgrove Post: The next error, once the DLLs are in place, is usually 0xc0000374 for memory heap corruption. They'll need to go back to code, and see where they're writing outside the bounds of allocated memory, for that one. |
|
3961)
Message boards :
Projects :
MilkyWay - N-body application errors
(Message 46262)
Posted 13 Nov 2012 by Richard Haselgrove Post: Done Thanks Byron. Since then, I've found another dependency (on pthreadGC2_64.dll) - with both DLLs manually downloaded and popped into the project folder, the app just about runs: http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=479865&offset=0&show_names=0&state=3&appid=7 But 0.36 credits still doesn't allow me to post. Perhaps you could do the honours again? |
|
3962)
Message boards :
Projects :
MilkyWay - N-body application errors
(Message 46259)
Posted 13 Nov 2012 by Richard Haselgrove Post: Could somebody with posting rights copy the following information to the project's News thread, please? Their message boards are locked up so tight that a new users can't even post an explanation for why their new app isn't going to give them enough credit to post... 0xC0000135 isn't exactly an unknown error - it means 'The application failed to initialize properly'. That's usually a missing DLL - and no, thank you Google, NOT usually the silly 'dot Net' framework. |
|
3963)
Message boards :
Questions and problems :
Odd behavior of POEM@Home?
(Message 46251)
Posted 12 Nov 2012 by Richard Haselgrove Post: POEM seems to have the normal message board structure at http://boinc.fzk.de/poem/forum_index.php. |
|
3964)
Message boards :
Questions and problems :
Cross project stats and links missing from Project Account pages
(Message 46196)
Posted 3 Nov 2012 by Richard Haselgrove Post: The lights are back on at Netsoft. |
|
3965)
Message boards :
Questions and problems :
Cross project stats and links missing from Project Account pages
(Message 46194)
Posted 3 Nov 2012 by Richard Haselgrove Post: No, it comes from netsoft-online. See line 81 of http://boinc.berkeley.edu/trac/browser/boinc/html/inc/user.inc |
|
3966)
Message boards :
Questions and problems :
Cross project stats and links missing from Project Account pages
(Message 46192)
Posted 3 Nov 2012 by Richard Haselgrove Post: That's because the data source - BOINC Combined Statistics - is having problems at the moment. In the left-hand column: BOINC Project Statistics The whole world has gone dark there - is that in sympathy with Hurricane Sandy? |
|
3967)
Message boards :
Questions and problems :
BOINC 7.0.25 High Priority Probelms
(Message 46124)
Posted 25 Oct 2012 by Richard Haselgrove Post: Computing preferences is a Global preference, you only need to set it one project, once Boinc contacts that project it'll update it's global preferences file, That would be 'latest' preferences - the preferences most recently set, no matter where you set them. |
|
3968)
Message boards :
Questions and problems :
BOINC 7.0.25 High Priority Probelms
(Message 46121)
Posted 25 Oct 2012 by Richard Haselgrove Post: You're looking in two different places. Both project account pages have three sets of preferences: * When and how BOINC uses your computer * Message boards and private messages * Preferences for this project At SETI, you're looking at the top set, which includes the 'enough tasks' and 'additional tasks' values - which are transmitted from project to project, so the same for all BOINC projects. At Milkyway, you're looking at the bottom set, which is a different collection of numbers entirely - as your paste shows. |
|
3969)
Message boards :
Questions and problems :
BOINC 7.0.25 High Priority Probelms
(Message 46119)
Posted 25 Oct 2012 by Richard Haselgrove Post: If you look at the top of the screen where you set that SETI '2 days plus 7 days' preference, it says: These apply to all BOINC projects in which you participate. So, you'll have set the same preferences for Milkyway as well. Milkyway seems to have a turnround period - from issue to deadline - of 12 days. With your settings, you're asking Milkyway to keep 9 days' worth of tasks on hand, and finish them no later than 10 days after issue. That's because "maintain enough tasks" is designed to cover the case where you can't connect to the internet for - in your case - two days. There's no point in finishing a task 30 seconds before deadline if you're not going to be able to report it until the next connection, in two days time. So 'maintain enough' doubles up as 'and finish them this early, too'. Reduce the size of the cache, and Milkyway will relax a bit. |
|
3970)
Message boards :
Projects :
D Wave Quantum Computers
(Message 46095)
Posted 23 Oct 2012 by Richard Haselgrove Post: For a while, they ran a BOINC project called AQUA@home, and helped with testing the infrastructure for multi-threaded CPU applications. They also tried GPU computing, both cuda (which ran in production mode for a while) and OpenCL (which never made it off the test bench). They shut down the BOINC side of their operations in about July 2011. |
|
3971)
Message boards :
Questions and problems :
Issue with WCG HCC GPU 6.56 LAIM
(Message 46061)
Posted 20 Oct 2012 by Richard Haselgrove Post: It appears that Leave Applications in Memory is not working with GPU tasks. No - it was deliberately taken out. CPU (main system) memory is routinely swapped out to a pagefile by the operating system if things get tight - so LAIM has minimal effect on system performance. But GPU memory has no swapfile system - so anything left in memory, is in memory. That bit of BOINC was written when 512MB GPUs were common, and many projects (including Einstein, and SETI with cuda 2.3 and above) can only fit one task in that little VRAM. Leaving an app behind in VRAM would prevent any other GPU task running. |
|
3972)
Message boards :
Questions and problems :
Folder data and error 417
(Message 46053)
Posted 19 Oct 2012 by Richard Haselgrove Post: Re-posting a reply I gave to a very similar question on the boinc_projects mailing list this afternoon: IP addresses starting 192 are typically part of the private 192.168.0.0 IP address space which cannot be routed over the internet. See |
|
3973)
Message boards :
Questions and problems :
Nvidia Upgrade - Seti ultra slow down
(Message 46040)
Posted 18 Oct 2012 by Richard Haselgrove Post: This is a SETI project question. Please post it on the SETI message boards ('Number Crunching' is best), because the volunteers there will be able to link through your account to your host details, and - hopefully - find something meaningful to comment on. |
|
3974)
Message boards :
The Lounge :
Parallella and the Epiphany processor
(Message 46026)
Posted 17 Oct 2012 by Richard Haselgrove Post: It's also being discussed on Einstein: http://einstein.phys.uwm.edu/forum_thread.php?id=9686 |
|
3975)
Message boards :
GPUs :
Work failing on GTX 6xx 'Kepler' GPUs.
(Message 46022)
Posted 17 Oct 2012 by Richard Haselgrove Post: Instructions for working round this problem have been consolidated into NVidia driver problems which cause computation errors |
|
3976)
Message boards :
Questions and problems :
BOINC 7.0.27 manager can't connect to client
(Message 46021)
Posted 17 Oct 2012 by Richard Haselgrove Post: Same problem here. Ubuntu 12.04, repository 7.0.27, starting Boinc manager from a gui startup folder. Always, the manager cannot connect to core client. So I stop the manager and restart. Then it works. This tells me all my tcp/ip settings are OK and that the core client starts running. For BOINC Manager to connect to a client, either (a) the client must be running already, or (b) the operating system must be in a suitable state for the client to start up when the Manager instructs it to. If your manager fails to start the client when run automatically from a startup folder, but works after a stop/restart (presumably, after some little time has elapsed), that suggests to me that everything is in the right place and findable. Some years ago, there were regular discussions here about inserting delays into Linux startup scripts: the idea being that BOINC had dependencies on other system components (network drivers? graphics X-drivers?), and couldn't start until the OS had fully initialised those components. I'm not a Linux person, but Jord may be able to find and point you to those earlier discussions. |
|
3977)
Message boards :
GPUs :
Question regarding Radeon HD 5450
(Message 46020)
Posted 17 Oct 2012 by Richard Haselgrove Post: b) Find and use the BOINC software that still supports AGP. Well, it's more that the driver provides the channel of communication between BOINC and the hardware, so that BOINC can use the API (Applications Programming Interface) to ask questions and get answers back. But you're right, it will be the driver which provides the AGP support (or not, as the case may be). BOINC doesn't interact with hardware at that level. AGP cards may require Catalyst AGP Hotfix before the main drivers will install. |
|
3978)
Message boards :
GPUs :
Work failing on GTX 6xx 'Kepler' GPUs.
(Message 46007)
Posted 16 Oct 2012 by Richard Haselgrove Post: The answer has been posted in several SETI@home threads, but the shortest and simplest is probably Nvidia GT640. The problem is generic to all GTX 6xx 'Kepler' cards, so the same answer applies for this questioner too. Jord, could you perhaps edit the typo in the thread title to remove the reference to a non-existent card, and maybe substitute that generic "GTX 6xx 'Kepler'" formulation? That might save us a FAQ... (We could probably use a sticky at SETI to cover this - I'll talk to the mods there) |
|
3979)
Message boards :
Questions and problems :
No usable GPU found
(Message 45960)
Posted 12 Oct 2012 by Richard Haselgrove Post: It was the drivers. There is no APP SDK runtime support for XP in any driver above 11.8 and that is needed to run the tasks. I can only guess that my version 12.3 drivers are working because I just installed them over the previous one. That may also explain why I got BSODs when I tried to update to 12.4. Actually, the last AMD driver with OpenCL support for Windows XP is 12.1 - as you can check by looking at C:\AMD\Support\12-1_xp32_dd_ccc\Packages\Apps\OpenCL after letting the download package unpack itself, as the first stage of driver installation. It is indeed unfortunate that AMD's documentation differs so widely from the distribution packages it is supposed to describe. |
|
3980)
Message boards :
Questions and problems :
Recommended BOINC 7.0.28 x64 can't detect available NV GPU
(Message 45874)
Posted 3 Oct 2012 by Richard Haselgrove Post: Please read message 40316. "Running as a daemon" == "installed as a service" |
|
3981)
Message boards :
Questions and problems :
Windows 7x64 Logon Screen (in Domain) problem with Boinc
(Message 45833)
Posted 28 Sep 2012 by Richard Haselgrove Post: It is the last version of Boinc. Or he may know something we don't know. If all development work on BOINC has stopped, then the latest version will indeed be the last version. |
|
3982)
Message boards :
Questions and problems :
Config file to run three different GPUs
(Message 45801)
Posted 26 Sep 2012 by Richard Haselgrove Post: Ask the guys at Einstein. Double precision isn't a matter of 'optimisation', it's a question of whether their mathematical search algorithm requires the extended precision. From what I understand, it doesn't - so it would just be a waste of their programming time to code for a DP card, especially since that requirement would exclude a lot of lesser cards from the project. |
|
3983)
Message boards :
GPUs :
Question regarding Radeon HD 5450
(Message 45767)
Posted 24 Sep 2012 by Richard Haselgrove Post: Yes, an APP driver is more relevant for end-users than a full-blown developer SDK. |
|
3984)
Message boards :
GPUs :
Question regarding Radeon HD 5450
(Message 45763)
Posted 24 Sep 2012 by Richard Haselgrove Post: Yes, most projects use ATI cards in OpenCL mode, so having a driver which supports OpenCL is likely to be important. But BOINC should still detect the card in native (CAL) mode, shouldn't it? Or has ATI withdrawn that as well? |
|
3985)
Message boards :
Questions and problems :
What is vina_3.0_windows_intelx86.exe and why such a hog?
(Message 45724)
Posted 20 Sep 2012 by Richard Haselgrove Post: Well, since you're posting here, you presumably suspect it might belong to a BOINC project. A few clues, like a list of the projects you're attached to and are active on the computer in question, might have helped - but a short session with a well-known search engine suggested to me that it might belong to FightMalaria@Home. It looks like they upgraded their applications to version 3.00 just three weeks ago. Maybe there are still some issues to overcome? The project website would be a good place to ask. |
|
3986)
Message boards :
Questions and problems :
completed job not uploading
(Message 45699)
Posted 17 Sep 2012 by Richard Haselgrove Post: It's usually best to discuss project-specific problems with the project concerned, but since you ask: Friday 31 Aug: "HTTP/1.1 500 Internal Server Error" - a disk had failed, and a replacement ordered. Monday 03 Sep: replacement disk installed Tuesday 04 Sep: "we are in the congested phase" (uploads working again, slowly) Friday 07 Sep: "Server is out of disk space" Friday 14 Sep: "she is virtually full", currently feels like moving deckchairs on the Titanic. Monday 17 Sep: "Uploader1.atm is now back in action with 750 GB free." Won't take long to fill that up... |
|
3987)
Message boards :
Questions and problems :
Work fetch problem with more than 1 ATI GPU, app_info & exclusions
(Message 45650)
Posted 15 Sep 2012 by Richard Haselgrove Post: Hopefully we'll have a new BOINC build to test soon. "Priority" is not something you can set - it's calculated internally by your BOINC client. Enable the <work_fetch_debug> logging flag to find out what's going on - post one complete cycle here (it recalculates once per minute). We would need from ------- start work fetch state ------- to ------- end work fetch state ------- |
|
3988)
Message boards :
GPUs :
Is there any non-beta research project for ATI\AMD gpu cards?
(Message 45641)
Posted 13 Sep 2012 by Richard Haselgrove Post: GPUGrid has AMD OpenCL apps. I'm afraid not. GPUGrid is nVidia only, and cuda only (not OpenCL), according to their applications page. They are associated with donate@home, which has functioned as their AMD/ATI testbed, but I don't think many people regard it as a "non-beta ... medicine, biology, physics or chemistry" project. Make your own mind up at AMD GPUs are now usable at Donate@Home, a different project from GPUGrid. |
|
3989)
Message boards :
Questions and problems :
BOINC is uploading to which site?
(Message 45616)
Posted 11 Sep 2012 by Richard Haselgrove Post: LHC (the classic v1.0, not the rather specialised Test4Theory) seems to use <url>http://lhcathomeclassic.cern.ch/sixtrack_cgi/file_upload_handler</url> I don't run Rosetta, but the techniques should still work. |
|
3990)
Message boards :
Questions and problems :
BOINC is uploading to which site?
(Message 45614)
Posted 11 Sep 2012 by Richard Haselgrove Post: There are two ways of finding out this information. No, it probably isn't the project's main site - Einstein uploads some results to Germany, although the project servers are in America, and CPDN uploads different files from the same task to different countries, to spread the massive data load. You can: 1) Inspect the contents of client_state.xml (use NotePad, or a simple text viewer like that. Don't use anything which might lock the file and prevent BOINC updating it while you search). Most projects use a file_upload_handler like <url>http://cpdn-upload2.oerc.ox.ac.uk/cgi-bin/file_upload_handler</url> <url>http://cpdn-restarts.oerc.ox.ac.uk/cgi-bin/file_upload_handler</url> <url>http://srv1.cpdn.psu.edu/cpdn_cgi/file_upload_handler</url> (that's three for one project) <url>http://einstein-dl.aei.uni-hannover.de/cgi-bin/file_upload_handler</url> 2) Alternatively, you can enable <http_debug> logging in cc_config.xml, and read what sites BOINC tries to reach in the message/event log, when you re-try an upload. |
|
3991)
Message boards :
BOINC client :
Problem with CPU work request, BOINC 7.0.28
(Message 45600)
Posted 9 Sep 2012 by Richard Haselgrove Post: What were the values of dt and inc shown in [WFD]? It isn't. CPU work request inhibition is caused by Scheduler request completed: got 0 new [CPU] tasks With [Work_Fetch_Debug] you would have seen that GPU fetch was also disabled at that point, for the same reason. 'Computation for task 27mr10ac.13328.12751.12.10.242_0 finished' cleared the request backoff for that resource type (only) - it must have been a GPU task, because the replacement task has version 610 and plan_class ati13ati. So, the next work request is for the not-backed-off resource only. That's all according to design specification. Work fetch backs off when no work is received in response to a request. Where I agree with you is that the duration of the backoff should be related to the reason for the lack of work. My backoff was because MilkyWay haven't finished re-writing the N-Body Simulation application for Windows yet. Another week or two in backoff would have been acceptable. Your backoff was because the BOINC server design can't hold sufficient tasks to fulfill a work request for 2 days GPU crunching in one go, chooses (preferentially) to supply work for the faster resource, and, I suppose, gives no extra weight to the idle resource when choosing which part of a mixed request to give priority to. |
|
3992)
Message boards :
BOINC client :
Problem with CPU work request, BOINC 7.0.28
(Message 45597)
Posted 8 Sep 2012 by Richard Haselgrove Post: What were the values of dt and inc shown in [WFD]? 08/09/2012 22:36:41 | Milkyway@Home | [work_fetch] CPU: fetch share 0.000 rsc backoff (dt 2289.16, inc 4800.00) If dt > 0.00, fetch is inhibited for that resource - no point in hammering a server with requests for work when they haven't finished writing the app yet... |
|
3993)
Message boards :
Questions and problems :
flagged as unrecognisable.
(Message 45587)
Posted 7 Sep 2012 by Richard Haselgrove Post: Good question. <zero debts> has been deprecated, because debt itself has been superceeded as a scheduling mechanism. But now, how do we reset REC and Prio when the need arises (after a testing session has gone haywire)? |
|
3994)
Message boards :
Questions and problems :
PC with GTX285 + GTX560Ti - BOINC is only scheduling jobs for the GTX560Ti.
(Message 45577)
Posted 6 Sep 2012 by Richard Haselgrove Post: This is by design. If you have two dissimilar GPUs in the same computer, by default BOINC will only use the "better" (more modern, faster) card. If you want to override this default, you'll have to use a configuration file with the line <use_all_gpus>1</use_all_gpus> - see client configuration for details. |
|
3995)
Message boards :
Questions and problems :
GPU Missing After Turning Off Display?
(Message 45547)
Posted 2 Sep 2012 by Richard Haselgrove Post: Apologies. I'm on 301.42, which when I go to download drivers from the nVidia website is the most recent version they give me for my card. OK, that rules out the 'monitor sleep' bug present in 295.73 and 296.10 (and predominantly affecting DVI connections). I'm not so familiar with HDMI, but I'm sure somebody here will be. |
|
3996)
Message boards :
Questions and problems :
GPU Missing After Turning Off Display?
(Message 45544)
Posted 2 Sep 2012 by Richard Haselgrove Post: I'm on BOINC 7.0.28 and the latest nVidia drivers. Do you really mean you are running the 306.02 Cuda 5 Beta drivers released six days ago? If not, please give us a real number rather than just saying "the latest"... There are known problems, not dissimilar to this, with the 295.73 and 296.10 WHQL drivers. |
|
3997)
Message boards :
Questions and problems :
BOINC doesn't suspend when computer in use
(Message 45502)
Posted 31 Aug 2012 by Richard Haselgrove Post: On the mac, btw, it's Tools -> Computing Preferences. I just noticed today that the dialog box which opens when you select that has a title bar saying BOINC - Preferences No sign of any computing there ;-) (observed on Windows) |
|
3998)
Message boards :
Questions and problems :
BOINC Manager does'nt run multitask
(Message 45487)
Posted 30 Aug 2012 by Richard Haselgrove Post: All the tasks visible in that screenshot are scheduled to run on the (single) NVidia GPU - you can tell that from the 'Application' column, saying "6.10 (cuda_fermi)". Any tasks assigned to the (8x) CPUs would show as 6.03, with nothing in brackets. You need to scroll the list and look for those. If none are visible, then you are either not requesting, or not being assigned, CPU work. If not requesting, check configuration settings, both in the Manager and on the website - verify using Event log that your changes are successful. If requesting, but not receiving, that's a project issue. Read the (regular) new threads on the SETI project message boards about why this happens. |
|
3999)
Message boards :
Questions and problems :
BOINC Manager does'nt run multitask
(Message 45485)
Posted 30 Aug 2012 by Richard Haselgrove Post: A small correction - the BOINC Manager doesn't process anything at all. It's simply there so that you can interact with, and control, the BOINC 'client' (worker) - which is otherwise invisible to you. You don't have to be running the Manager for the client to be working. But that's a minor quibble. What you do need to establish from your team-mates is: is there work available on their computers for the other resources (I'm guessing CPUs) to process? That is, work assigned, downloaded, and shown as "Ready to start"? If there is no work, then it can't be processed - and you have to find why there is no work. What happens when they click the 'update' button? What is shown in the event log? If it isn't immediately clear from the log, you may have to enable additional debug logging - <sched_op_debug> is a good place to start, followed by <work_fetch_debug> if you still need it. But the only way you can solve an issue like this is to gather, and read carefully, information from the machines in question. Nobody here can give you advice without that. |
|
4000)
Message boards :
BOINC client :
command lines of Boinc client
(Message 45467)
Posted 27 Aug 2012 by Richard Haselgrove Post: Hmmm, missed those. You can't capture it, but if you know it (because you set it up yourself, or you've seen it in response to a ping or tracert to a url), you should be able to use it anywhere a url would normally be used. I haven't tested it in the BOINC environment, but I'm pretty sure it would work: I do regularly use http://192.168.1.1 in a browser address bar to manage my router, or http://192.168.1.5 for my network printer's internal web server. All you're doing with that format is bypassing the DNS/hosts name resolution stage. |
|
4001)
Message boards :
BOINC client :
command lines of Boinc client
(Message 45416)
Posted 23 Aug 2012 by Richard Haselgrove Post: The BOINC command line tool Boinccmd isn't used to attach a computer to a project. You use the BOINC binary for that. Well, up to a point. Yes, the BOINC binary actually does the work, but if BOINC is already running, you can use boinccmd to tell it what to do. Your boinccmd list contains the commands: --create_account URL email passwd name --project_attach URL auth attach to project |
|
4002)
Message boards :
Questions and problems :
2500k + GTX 580 beating out 3930k +GTX690????
(Message 45402)
Posted 22 Aug 2012 by Richard Haselgrove Post: Hey. Milkyway@home project requires a Double prediction support from graphics card. Both graphics cards (580 & 690) support DP calculation, but the gtx 580 is designed for DP calculation better than 690. If you want more out from 690 gpu use it for single prediction projects. Example: gpugrid.net project. gpugrid.net do not need DP, so that project works really well on gtx 690. The word would be double/single precision, but with that correction, the explanation makes sense. To get the fastest DP processing on an NVidia card, you need to purchase from the Tesla range, not GeForce. |
|
4003)
Message boards :
Questions and problems :
Bionic crash
(Message 45382)
Posted 20 Aug 2012 by Richard Haselgrove Post: I am the admin and only account on the computer. There is always a separate, hidden, superuser account called 'Administrator', as described in http://www.sevenforums.com/tutorials/507-built-administrator-account-enable-disable.html |
|
4004)
Message boards :
GPUs :
can dual gpu in one card run 2 tasks?
(Message 45243)
Posted 12 Aug 2012 by Richard Haselgrove Post: Yes, a single physical card with dual GPUs (like that one) will be seen by BOINC as two separate cards. You will be able to run one task (from the same or different projects) on each of the two 'halves'. It won't be able to run a single task at double the speed by using both halves. |
|
4005)
Message boards :
Questions and problems :
boinc 7.0.28x64 runs only 2 tasks on system with 4-core and 75% CPU assigned
(Message 45216)
Posted 10 Aug 2012 by Richard Haselgrove Post: And what is the value you set in BOINC its local preferences? If you open that screen for the first time just to look at the values, do not click the 'OK' button to close the screen - that "locks in" the current settings as a global override for the web settings for evermore (or until you clear them manually). Click 'cancel' instead. |
|
4006)
Message boards :
BOINC client :
Status 'Cancelled by server' changed
(Message 45211)
Posted 9 Aug 2012 by Richard Haselgrove Post: Excellent news - between us, we got it sorted. Just 498 more BOINC projects to fix..... ;-) |
|
4007)
Message boards :
BOINC client :
Status 'Cancelled by server' changed
(Message 45202)
Posted 9 Aug 2012 by Richard Haselgrove Post: Because I found another bug in inconsistent deadlines for resents, SIMAP upgraded meanwhile to software version: 25993 Well, whatever they've updated, it seems not to have been the web display code. That crucial result.inc file is still showing as being rev 24964 when I look at the source HTML for one of my SIMAP result pages this morning. Whether you call it a bug or a feature, it won't go away until they update that file. |
|
4008)
Message boards :
BOINC client :
Status 'Cancelled by server' changed
(Message 45186)
Posted 8 Aug 2012 by Richard Haselgrove Post: I tested it again with BOINC v7.0.33 on the SIMAP project. Unfortunately, the SIMAP web code seems not to have been updated: <!-- SVN VERSIONS --> That's the one which matters, and the later version is working properly, e.g. <!-- $Id: result.inc 25858 2012-07-10 20:21:12Z davea $ --> Result 2524003864: Exit status 197 (0xc5) EXIT_TIME_LIMIT_EXCEEDED I'm afraid it's up to the SIMAP web administrators to update some more of their code - you could try dropping them a line. |
|
4009)
Message boards :
Projects :
News on Project Outages
(Message 45129)
Posted 31 Jul 2012 by Richard Haselgrove Post: Seti's forums seem to be down at the moment. They seemed to go into a normal maintenance shutdown around 16:00 UTC, but other pages haven't been reachable since then. They had problems with the webserver a few weeks ago, I guess they're taking the opportunity for some extra work while the rest of the system is dormant. |
|
4010)
Message boards :
GPUs :
CUDA 5 on OS X 10.8 - only one GPU
(Message 45077)
Posted 26 Jul 2012 by Richard Haselgrove Post: The test on 'available ram' is enough to split them: 62 if (c1.available_ram> 1.4*c2.available_ram) return 1; 63 if (c1.available_ram < .7* c2.available_ram) return -1; Wed Jul 25 20:18:34 2012 | | NVIDIA GPU 0 (not used): GeForce 9600M GT (driver version 5.0.17, CUDA version 5.0, compute capability 1.1, 256MB, 143MB available, 120 GFLOPS peak) Maybe the new OS is using a lot more VRAM on GPU 0 - is that one driving the display? |
|
4011)
Message boards :
Questions and problems :
Different OpenCL apps runs at the same time
(Message 45060)
Posted 25 Jul 2012 by Richard Haselgrove Post: There are many reports, on many message boards, that AMD OpenCL applications are currently running slowly if there isn't sufficient CPU time available. So far as I can tell, it appears to be a deliberate decision by AMD to enforce 'busy wait' synchronisation in the runtime support bundled with recent catalyst suites. I'm guessing that they found this was the most efficient mode for a single OpenCL program running on an otherwise idle computer: but they didn't consider the typical BOINC user, who tends to load the CPU with other tasks at the same time as OpenCL - many BOINC users describe the high CPU requirement of OpenCL applications as a driver bug. The Einstein case is different: their application requires CPU support because somes parts of the algorithm aren't suitable for parallel processing on the GPU, and are computed on CPU instead. During that phase of processing, the CPU will be busy with much more real work than the simple busy-wait of OpenCL runtime support - maybe that is what is holding up POEM, when they run together. |
|
4012)
Message boards :
GPUs :
Separate Resource Shares for GPU?
(Message 45041)
Posted 24 Jul 2012 by Richard Haselgrove Post: This would be especially useful in the new BOINC v7 range, where work done (effectively the fpops count, expressed as REC) on both CPU and GPU is used in the same calculation for scheduling priority. That means that if a project has both CPU and GPU applications, the work done on GPU effectively means the CPU tasks never run until under deadline pressure. That can be a problem if some applications haven't been ported to GPU, like at Einstein where only one of the three applications has a GPU version - the CPU-only app versions get badly delayed. |
|
4013)
Message boards :
Questions and problems :
Problem Updating Boinc software
(Message 45031)
Posted 23 Jul 2012 by Richard Haselgrove Post: Have a read of this FAQ. Funny, that. It was only yesterday that I was saying to David Anderson that this was the most frequently asked of all the frequently asked questions. |
|
4014)
Message boards :
Questions and problems :
Network problems = Unrecoverable error...
(Message 45022)
Posted 22 Jul 2012 by Richard Haselgrove Post: Which DNS Servers do you use? The trouble is, all you're doing is to assume that the ISP's DNS is the cause of the problem, and bypassing it. For that solution to work, the ISP's routers and connectivity (both upstream and downstream) have to be fully present and correct. If the problem is BOINC's use of synchronous DNS resolving (as the very interesting quote from Nicolas in that trac ticket suggests), then a better solution would be the installation of a local caching DNS server. But then you have some tricky management decisions to make regarding caching: SETI's download server url, for example, deliberately has a TTL of 5 seconds, and according to the rules shouldn't be cached. There may be others - it depends which projects are running. In Windows, the command ipconfig /displaydns is a useful tool for getting an idea of what caching is allowed on the sites you visit regularly - I've just discovered that SETI have slowed down their round-robin DNS with a TTL of at least 50 seconds (ipconfig shows the remaining TTL since the last lookup, not the full value). |
|
4015)
Message boards :
Questions and problems :
Network problems = Unrecoverable error...
(Message 45012)
Posted 21 Jul 2012 by Richard Haselgrove Post: It is not a coincidence. During the internet problem times, DNS resolution can be affected. If boinc tries connecting to a project server and can't resolve the DNS quickly, it causes the no heartbeat error. Some science applications error out with signal 11 when receiving the no heartbeat. My Linux (Lubuntu 11.10, ver. 7.0.27) has recently errored on 9 Asteroids tasks when DNS resolution was having problems. I agree that the task errors and the internet problems are linked, and I also agree that DNS name resolution on the flakey internet connection is likely to be implicated in that linkage. My suspicion is that when the BOINC client asks the libcurl sub-component to connect (by name) to a project server, everything is put on hold until, at least, the resolved IP address comes back from DNS. If that involves a wait of more than 30 seconds and a timeout (which, in non-corporate environments, is plausible, because the DNS server is likely to live with your ISP at the other end of the local loop), then the heartbeat mechanism may be stalled and the errors follow. An added complication is that libcurl handles all TCP/IP communications for the client, and - as well as project internet comms - that includes localhost loopback messages between the client and BOINC Manager, and any remote RPC calls that might be issued by a local aggregator like BoincTasks or BoincView. Comms are tricky things, and failures anywhere can cause delays and problems. I recently lost a host which was listed by name in my remote_hosts.cfg file: I noticed the other machines on my LAN stuttering as they generated the "Can't resolve hostname in remote_hosts.cfg: xxx" message and notice, far more often than I would have thought was necessary. Edit - communications between the client and the science applications are handled by files written into a shared memory area - a virtual solid-state disk. They should be exempt from the TCI/IP problems. |
|
4016)
Message boards :
Questions and problems :
GPU tasks cause MT tasks to sit idle
(Message 44958)
Posted 19 Jul 2012 by Richard Haselgrove Post: As far as I know, it's normal. With any MT application, it'll wait until all CPU cores are available. As long as part of any core is in use with another application, the MT app won't start. Not quite. I was running AQUA's MT tasks OK with a GPU task alongside - though of course many BONCs have come and gone since AQUA folded, so that isn't current news. I think the clue is in the linked thread - running two GPU tasks at 0.7 CPUs each, so the total CPU reservation is more than one whole CPU - I think that's the point where the MT get blocked. Running one GPU task at a time would be worth a test: if that works, wrapping the GPU app in an app_info that only claims 0.49 <ncpus> might allow two to run - that doesn't change the way the apps run in any way, just fools the scheduler. |
|
4017)
Message boards :
Web interfaces :
Malfunctioning HTTP links in forums
(Message 44935)
Posted 18 Jul 2012 by Richard Haselgrove Post: There do seem to be some browser dependencies. If you follow a link to a message in the middle of a page, apparently Firefox and IE9 scroll the page display to the correct message, but IE8 (last IE available for Windows XP) just sits with the topmost posts on the page visible. |
|
4018)
Message boards :
Questions and problems :
Got new work after last WU is over 80% finished
(Message 44870)
Posted 13 Jul 2012 by Richard Haselgrove Post: 1) Client configuration <fetch_minimal_work>0|1</fetch_minimal_work> 2) Backup project (v6.10.30) If a project has zero resource share, treat it as a "backup project": |
|
4019)
Message boards :
Projects :
News on Project Outages
(Message 44849)
Posted 12 Jul 2012 by Richard Haselgrove Post: 7/12/2012 8:07:37 AM | Collatz Conjecture | Scheduler request failed: Couldn't resolve host name More likely your local DNS resolver barfed. Mine can reach http://boinc.thesonntags.com/collatz/ OK. |
|
4020)
Message boards :
Questions and problems :
BOINC causes random freezing of OS
(Message 44847)
Posted 12 Jul 2012 by Richard Haselgrove Post: OK, you've checked the basics of the PSU - and that's about as far as my hardware knowledge goes. But I was drawing on these two SETI posts (by a CUDA developer with extensive knowledge of the 560Ti) - you might like to check them out. message 1240004 message 1240409 |
|
4021)
Message boards :
Questions and problems :
BOINC causes random freezing of OS
(Message 44845)
Posted 12 Jul 2012 by Richard Haselgrove Post: The other component which you haven't mentioned, but which can be crucial in cases like this, is the PSU (Power Supply Unit). It's not sufficient just to have one which is rated above the minimum required by your graphics card. It has to be in good working order (capacitors inside PSUs can bulge and fail, just as the ones on mothersboards sometimes fail). And it has to be capable of delivering sufficient power at each of the voltages required by the components. For a GPU, the most important one is the "12 volt rail" (or rails). The GF 560 Ti 448 is specified to require 210 watts (18 amps @ 12v), and a minimum 550W system PSU. Earlier versions of the 560Ti were notoriously susceptible to power supply problems, but yours is a later 'special edition' which should be free of those problems. |
|
4022)
Message boards :
Projects :
News on Project Outages
(Message 44835)
Posted 11 Jul 2012 by Richard Haselgrove Post: NumberFields@Home is temporarily down as of yesterday around 1900UTC. Numberfields@Home is back for the time being, but with front page news: Planned Power Outage |
|
4023)
Message boards :
BOINC client :
Status 'Cancelled by server' changed
(Message 44825)
Posted 10 Jul 2012 by Richard Haselgrove Post: Report made, and David has made a fix. So far, deployed on the SETI Main website - look for the line <!-- $Id: result.inc 25858 2012-07-10 20:21:12Z davea $ --> (or later - there's a small typo) in the page source of the result pages on other projects to see when they've deployed it. |
|
4024)
Message boards :
BOINC client :
Status 'Cancelled by server' changed
(Message 44822)
Posted 10 Jul 2012 by Richard Haselgrove Post: OK, will do. I've been having a look round, and there seem to be other inconsistencies - I have a task with Outcome Computation error Client state Compute error Exit status -5 (0xfffffffffffffffb) Unknown error number <core_client_version>6.12.34</core_client_version> <![CDATA[ <message> - exit code -5 (0xfffffffb) </message> <stderr_txt> SETI@home error -5 Can't open file (work_unit.sah) in read_wu_state() errno=2 File: ..\worker.cpp Line: 131 </stderr_txt> ]]> (Just preserving task 2515612720 so I can find it, until it gets purged) I've got to go out now, but I'll work up a bug report when I get back. |
|
4025)
Message boards :
BOINC client :
Status 'Cancelled by server' changed
(Message 44818)
Posted 10 Jul 2012 by Richard Haselgrove Post: Yes, looking again at [trac]changeset:25601[/trac], and specifically at http://boinc.berkeley.edu/trac/changeset/25601/boinc/trunk/boinc/lib/error_numbers.h#, it does look as if those error/exit numbers were substantially revised on 25 April, after v7.0.25 was compiled on 04 April. It doesn't look as if any change was made to the user web code to accurately reflect both old and new versions of the returned status codes. Do you want to report that to the boinc_alpha mailing list, or should I? |
|
4026)
Message boards :
BOINC client :
Status 'Cancelled by server' changed
(Message 44815)
Posted 10 Jul 2012 by Richard Haselgrove Post: Ah. So it's #define EXIT_ABORTED_BY_PROJECT 202 which isn't being picked up by the SIMAP web server code, and is being mis-reported (confusingly reported) as "Aborted by user". Their code is pretty recent: <!-- $Id: util.inc 24968 2012-01-02 05:53:42Z davea $ --> so it's probably quite a widespread problem - as the observations at SETI would confirm. |
|
4027)
Message boards :
BOINC client :
Status 'Cancelled by server' changed
(Message 44814)
Posted 10 Jul 2012 by Richard Haselgrove Post: I don't think this is a client issue. It's possible that CP has been misled by a server code update happening at the same time as the new client release - it would be helpful to know which project these messages were seen at - but I suspect this is a mis-reading of what the project is reporting. There is a difference between 'Cancelled by server' (and the closely related 'Didn't need', for a task which is cancelled before being sent out), and 'Aborted by user'. Where there is certainly a problem is in the web code which renders the task outcome into text for the user to see. We had a discussion about this at Einstein, some time ago: message 111673, 111705. Ever since BOINC v6.6.12 (March 2009), the client will self-abort a task in the cache which hasn't been started before the deadline is reached. In that case, the client returns #define EXIT_UNSTARTED_LATE 200 rather than the alternatives (from http://boinc.berkeley.edu/trac/browser/boinc/trunk/boinc/lib/error_numbers.h) #define EXIT_ABORTED_BY_CLIENT 194 #define EXIT_ABORTED_VIA_GUI 203 I think it would be a great improvement if exit codes 194 and 200 could be rendered on the website task lists as "Aborted by client", and only exit code 203 rendered as "Aborted by user". The other possible exit codes need appropriate handling too. @ CP, You can see the actual exit code/status returned by the client in the task detail page on the project website - it might be helpful if you could check that for the tasks that you're worried about, and post it back here. Edit - I see (in preview) that CP has posted while I was writing. I'll post the above anyway, even though some of it may have been overtaken by events. |
|
4028)
Message boards :
Questions and problems :
BOINC 7.0.25 / Win 7 64-bit / Manager not able to start
(Message 44804)
Posted 9 Jul 2012 by Richard Haselgrove Post: I'm not familiar with the bug reporting practices, so I have not personally done any additional measures to inform this problem. Maby the development team doesn't even know that we finns are suffering... The project team are aware of the problem, and have made a correction - too late for version 7.0.28, unfortunately, but included in version 7.0.31, currently being tested. What I haven't seen is any attempt to "harden" the translation handling system, so that typos like this don't result in the complete launch failures described in this thread. It would be far better if a badly-formed translation was handled by: * warn (error message in log) * ignore (use untranslated form) * proceed |
|
4029)
Message boards :
Questions and problems :
Functionality suggestion
(Message 44658)
Posted 28 Jun 2012 by Richard Haselgrove Post: Hi, and welcome to the board. Yes, this is a good place to make suggestions like that - though you may find it takes a long time for them to be acted on. You are right to suspect a full hard disk as a possible cause of the problem, though the actual detail was slightly more subtle. The server handling the actual upload was fine, but it was supposed to transfer the completed file to a second server after the upload was complete - and the Climate Prediction project was close (too close) to the limit of what they were allowed to store on that second server. BOINC will detect that the upload server is full, and refuse to accept more uploads until the administrators have dealt with it: but BOINC doesn't (yet) detect the quota problem on the offload file storage, which is why it continues to accept the files, but reports (32 MB later) the error you have quoted. If it's any consolation, one of my files uploaded 10 times overnight, with an error each time, until a successful eleventh upload this afternoon. Fortunately, I'm on a wired connection, with no bandwidth cap or surcharge, but I can see that this problem could be an irritation, and possibly an expense, for some people. CPDN will be working behind the scenes to try and prevent this happening again, but that may take a little while. If it happens again, and you happen to be near the computer at the time, there is one BOINC feature that you could use without shutting down BOINC completely. Go to the BOINC Manager's "Advanced" view, and from the "Activity" menu, choose the very last option: 'Network activity suspended'. That will stop BOINC trying all file transfers - uploads and downloads - until you choose another option. There are two problems with that: firstly, it blocks communications for every BOINC project you may be running, not just the one with the faulty server: and secondly, you have no way of knowing when the problem is solved, except by re-enabling network activity, waiting until the file has uploaded, and looking to see if the error message is shown again. We did, a long time ago, make a request very similar to the one you describe - ticket [trac]#139[/trac]. Unfortunately, the 'powers that be' were unsympathetic then, and they show no sign of softening their attitude now - but no doubt this latest episode will encourage another attempt. |
|
4030)
Message boards :
Questions and problems :
BOINC not getting tasks, despite being below the low-water mark
(Message 44483)
Posted 13 Jun 2012 by Richard Haselgrove Post: Is, by any chance, one of the projects you're attached to SETI@home? SETI is suffering from a configuration problem at the moment (and has been since around 21:00 UTC last night - 12 June 2012). As a result, all SETI tasks will be marked as running 'high priority', and no other projects will be scheduled. Hopefully, this is a glaring enough boo-boo that it will be rectified in short order (we've notified them already). But, in the meantime, you can bypass it by upgrading to BOINC v7.0.28 from http://boinc.berkeley.edu/download_all.php - although this is marked as a 'may be unstable' version, it's actually better than v7.0.25 in several ways, and it's free of the bug triggered by SETI's new configuration. |
|
4031)
Message boards :
Questions and problems :
I can't install 7.0.25
(Message 44421)
Posted 7 Jun 2012 by Richard Haselgrove Post: Try the steps in this BOINC FAQ. |
|
4032)
Message boards :
Projects :
News on Project Outages
(Message 44399)
Posted 4 Jun 2012 by Richard Haselgrove Post: SIMAP - Complete power shutdown of BOINCSIMAP on June 7/8th Due to general maintenance of the electricity network of our building we will have to shutdown all our servers from June 7, 4 p.m. UTC to June 8, 4 p.m. BOINCSIMAP will be completely offline during that 24 hours. We apologize for any inconvenience. We assume that all workunits [for the June monthly run] will have been distributed at that time. |
|
4033)
Message boards :
News :
BOINC server downtime
(Message 44364)
Posted 30 May 2012 by Richard Haselgrove Post: Time Zone Converter |
|
4034)
Message boards :
Questions and problems :
Disk space full but no tasks showing for some projects
(Message 44348)
Posted 29 May 2012 by Richard Haselgrove Post: ClimatePrediction creates a separate folder (and several sub-folders) within the main climateprediction.net project folder, for each live model. Each folder is named the same as the model it is supporting, and they can grow very large - my hadcm3n_o2ai_2020_40_007956542 folder contains 1.36GB. They should be, and in general are, deleted when the model completes successfully. The problems come when models fail with an error part way through their run - that's when the folders get orphaned. Double-check whether you have any live tasks on your system - check both BOINC Manager (paying attention to the show active/all tasks toggle button), and the tasks listed for that computer on your account page at the CPDN website. Once you're sure they're not active, delete any left-over model folders within the climateprediction.net project folder. Don't delete any individual files (even .zip or .gz files, which may be displayed as folders) - they are kept deliberately, as many of them can be re-used by another model. |
|
4035)
Message boards :
Questions and problems :
Ubuntu 12.04 boinc fails SETI@HOME and Cosmology@HOME
(Message 44297)
Posted 26 May 2012 by Richard Haselgrove Post: "Output file absent" isn't an error by itself, it's just an indication that something went wrong and the task didn't complete properly. You need to go to the websites of the projects where the tasks are failing (seti and cosmology), log into your account there, and check the records for the failing computer. Go into the list of tasks, find one which shows that an error was reported, and click on that task's ID number. At the very least, you'll be able to get the 'Exit status' number, and there may be some more diagnostic information in 'Stderr output'. Those will give you (and us) an idea of where to start looking. |
|
4036)
Message boards :
Questions and problems :
BOINC-Manager - Multi-monitor Problem
(Message 44291)
Posted 25 May 2012 by Richard Haselgrove Post: That XPos is huge - almost 3,000 pixels to the right of the origin. Most of the manager window would be on a third monitor, if I had one, beyond my two (large, high-res) 20" LCDs. What are the equivalent X and Y Pos immediately after the Manager starts up on monitor 1? Look at the monitor layout in the display control panel ('Adjust resolution' or 'Change display settings' - they both seem to lead to the same place on my W7/64 box). Does that properly match the monitors you actually have, and the way they're arranged? Sometimes - though more often with older versions, W7 seems better - Windows thinks you have a monitor that isn't really there any more, and tries to display things on it. |
|
4037)
Message boards :
Questions and problems :
BOINC-Manager - Multi-monitor Problem
(Message 44288)
Posted 25 May 2012 by Richard Haselgrove Post: It might be helpful to take this one alongside Boinc manager position second screen. Unfortunately, there's no XPos and YPos stored for DlgAdvPreferences, but if you're comfortable peeking into the registry, it might help the developers if you could post the XPos and YPos values for the main BOINC Manager screen, as shown in HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager |
|
4038)
Message boards :
Questions and problems :
HTTP error 0 with World Community Grid and BOINC 6.12.34
(Message 44224)
Posted 21 May 2012 by Richard Haselgrove Post: In that case, you can try to set it to another test URL or IP address, other than Google. This may be an internal URL, just as long as it can be used to ping and get an answer. Or use the <dont_contact_ref_site>1</dont_contact_ref_site> option to suppress the ping traffic entirely. I've set that flag as default for years now - doesn't seem to cause BOINC any problems. |
|
4039)
Message boards :
Questions and problems :
BOINC network usage
(Message 44181)
Posted 19 May 2012 by Richard Haselgrove Post: What monitoring tool are you using to observe this 'internet' traffic? boincmgr.exe should never access the internet - all external communications are handled by boinc.exe However, boinc.exe and boincmgr.exe talk to to each other all the time, using TCP/IP on localhost (127.0.0.1:31416). Some network analyers/firewall monitors can display this purely internal activity in a way which fails to distinguish it from external traffic. A simple five-second test: pull out the network cable, or turn off the WiFi switch. If the activity continues, it's internal, and nothing to worry about. Having said that, I have an open bug report about BOINC sometimes contacting project schedulers when network activity is suspended, but that doesn't sound like your situation. |
|
4040)
Message boards :
Questions and problems :
CPU recognition problems
(Message 44138)
Posted 14 May 2012 by Richard Haselgrove Post: If you look at the full Intel data sheet, you will see that the same statement is repeated there: # of Cores 2 # of Threads 4 It's a technique known as 'hyperthreading', and it's designed to work that way: although each individual task is completed more slowly when hyperthreading is active, the machine as a whole is more productive because four tasks are being completed in not much more time than two would otherwise take. If you feel that four tasks at a time are too many, or make your laptop too hot, you can ask BOINC not to use all the available threads. On the 'computing preferences' page of any one of your attached projects, set On multiprocessors, use at most: 50% of the processors Enforced by version 6.1+ |
|
4041)
Message boards :
BOINC client :
Why does the Installer for BOINC 7.0.27 require a full computer restart.
(Message 44019)
Posted 6 May 2012 by Richard Haselgrove Post: Are you the same person as user 'B-Man', who just asked a very similar question over at SETI? Even if not, you might be interested in the reply I wrote there. Or you might like to read BOINC's Sandbox design document. |
|
4042)
Message boards :
BOINC Manager :
BOINC requires permission at start-up
(Message 44002)
Posted 5 May 2012 by Richard Haselgrove Post: I run BOINC as a service, and like to move the BOINC Manager shortcut out of the BOINC folder in the Start menu to another folder. However, when I do this I need to grant BOINC Manager permission to start each time I reboot my PC. Two things to add to what Jord said. 1) When you're running as a service, you don't have to grant BOINC Manager permission to start each time - in fact, you don't have to start BOINC Manager at all. BOINC itself - the part that does the processing - should start silently, and you only need to start the Manager (in addition) if/when you want to look at what it's doing. 2) If you want a BOINC Manager shortcut somewhere else - on the desktop, say - the easiest way is to *copy* (Ctrl-drag) the one in the BOINC program group in the start menu. That gives *you* a shortcut with all the same attributes and permissions where you like it, and *anybody else* who uses the machine can find the original where they expect it to be. Doesn't matter much for BOINC, of course, but when I'm called in to fix other people's machines, it's a real *#@$^&* when I find that some silly b***** has moved windows component shortcuts out of their natural place in the order of things.... |
|
4043)
Message boards :
Questions and problems :
7.0.25 Doesn't fetch work when cc_config excludes GPU's
(Message 43959)
Posted 4 May 2012 by Richard Haselgrove Post: Sorry not to reply sooner but sleep got in the way. I see you have the link. Yes, the Windows installer does allow that - I take it you're talking Windows, from your avatar. You'll find that a downgrade installation ends with an error. Just re-run the installer (so dowdload it to your local disk, and run from there), and choose the 'repair' option that it'll offer. It's always worked at the second attempt, for me. |
|
4044)
Message boards :
BOINC client :
BOINC 7.0.26
(Message 43929)
Posted 2 May 2012 by Richard Haselgrove Post: A 7.0.27 tag has been created in the tree so an official release is getting nearer ;) The Windows downloads are already accessible, but not yet Linux or macOSX. |
|
4045)
Message boards :
BOINC client :
BOINC 7.0.26
(Message 43879)
Posted 29 Apr 2012 by Richard Haselgrove Post: Since there is nowhere feedback to be found for this build, I give some here. Interesting philosophical debate. In a situation where the "receommended" version has bugs which are complete show-stoppers for some projects, is it better to wait and promote the bug-free version, or to promote an interim "not quite so bad" version. Hint: if we waited for bugs like [trac]#336[/trac] - described as 'critical' by the lead developer, when he reported it five years ago - to be fixed, we would still be in a pre-beta testing phase. So why wait now? |
|
4046)
Message boards :
Questions and problems :
Latest Boinc Version Seems To Have Stopped Polling For Work
(Message 43847)
Posted 28 Apr 2012 by Richard Haselgrove Post: Thank you. It installed and once up and running immediately started polling for work. Will monitor progress. And 7.0.26 still has problems, though not that particular one. We're still waiting for v7.0.27 - have been for a week, now - to see how many of them have been fixed. Treat v7.0.26 as a usable stopgap for now, but keep an eye open for a proper new release in due course. |
|
4047)
Message boards :
Questions and problems :
7.0.25 Doesn't fetch work when cc_config excludes GPU's
(Message 43797)
Posted 26 Apr 2012 by Richard Haselgrove Post: PS: Still doesn't explain why .25 and .26 always run Milkyway@home GPU tasks in high priority... That one was fixed by a volunteer here - [bug report, patch] Always running high priority - David adopted it (without citation) as [trac]changeset:25593[/trac]. Also should be in v7.0.27 |
|
4048)
Message boards :
Projects :
News on Project Outages
(Message 43764)
Posted 25 Apr 2012 by Richard Haselgrove Post: 4/23/2012 10:21:41 AM | climateprediction.net | [error] Error reported by file upload server: Server is out of disk space Cimate Prediction currently has multiple upload server problems. It's the Oregon State server handling PNW tasks which is out of disk space: two Oxford servers handling the other regional models have serious HDD problems which are being worked on. About the only model type unaffected is the "Coupled Model Full Resolution Ocean". |
|
4049)
Message boards :
Questions and problems :
BOINC 7.0.25 and Proxies
(Message 43626)
Posted 20 Apr 2012 by Richard Haselgrove Post: Any advice? |
|
4050)
Message boards :
Questions and problems :
When is version 7.0.26 going to be released
(Message 43583)
Posted 18 Apr 2012 by Richard Haselgrove Post: A 7.0.26 tag has been created so the berkeley version shouldn't be too far away. Once it's released if you don't want to/can't use the berkeley version you will have to bug your distro's devs for a version bump, or you can try your hand at compiling it yourself following these instructions. Windows and Mac versions are available now in the download directory, and the Windows ones (tested both 32-bit and 64-bit) seem to be running OK. Hopefully the Linux version will appear in the same place soon. |
|
4051)
Message boards :
Questions and problems :
HTTP error 0 with World Community Grid and BOINC 6.12.34
(Message 43541)
Posted 16 Apr 2012 by Richard Haselgrove Post: ive just upgraded to 7.0.25, after having to stay on 6.12.33 to get comms, this version seems to have the same problem any fixes? An apparently successful workround is posted at https://secure.worldcommunitygrid.org/forums/wcg/printpost_post,372940 |
|
4052)
Message boards :
Questions and problems :
Cross Project ID Split
(Message 43511)
Posted 14 Apr 2012 by Richard Haselgrove Post: ... it look like the id is tied to your address.. Indeed it is. |
|
4053)
Message boards :
Questions and problems :
Ver 7.0.25 and CUDA loss
(Message 43496)
Posted 14 Apr 2012 by Richard Haselgrove Post: Might you be running as service mode/Protected Application Execution? If so, this exchange may be of assistance. |
|
4054)
Message boards :
Questions and problems :
Ubuntu Update Manager update to 7.0.24 today caused many errors
(Message 43425)
Posted 12 Apr 2012 by Richard Haselgrove Post: I am getting computation errors on: Add a report from NumberFields@home Project administrator says their Linux app is statically linked, so feels it's a client problem. |
|
4055)
Message boards :
BOINC client :
7.0.25 install / uninstall nightmare:- discuss
(Message 43420)
Posted 12 Apr 2012 by Richard Haselgrove Post: I wonder if the developers could have given slightly more prominence to the change in Service Mode behaviour under Windows XP? v6.12.34: GPUs can be used v7.0.25: GPUs can't be used Just guessing that this is the issue here..... |
|
4056)
Message boards :
Questions and problems :
Select A Computer?
(Message 43419)
Posted 12 Apr 2012 by Richard Haselgrove Post: Under BOINC if you select Advance and Select a Computer, you get no compter option when you press the down arrow key of the combo box. That combo box functions as a MRU (Most Recently Used) list - like the lists of recently opened documents in word processors and spreadsheets. The first time you use it, you have to type in the name of the remote computer you wish to control. But once you have made the connection, the computer name will be remembered so you don't have to keep typing it for future sessions. |
|
4057)
Message boards :
Questions and problems :
Downloading Tasks Seti@home
(Message 43201)
Posted 29 Mar 2012 by Richard Haselgrove Post: I downloaded the program last night and set it up with no problems, but it has been downloading work from the server for several hours. Is that normal? Unfortunately, yes. SETI has many, many users already, and they are all, always, trying to download work over an inadequately-sized internet connection. If you have 'just joined' SETI, you are probably also using a version of BOINC which actually spends much more of its time waiting politely in line, to let other users download work - a strategy designed for a different sort of connection problem completely. My suggestion would be to join - in addition - one of the many other BOINC projects, so that you can see BOINC working properly while you are waiting for your SETI tasks to download. Einstein is popular among SETI users, because it is well enough resourced to maintain an adequate supply of work. But many other projects, from a variety of scientific disciplines, are listed here - I'm sure you will find one to interest you, even if SETI is your primary choice. |
|
4058)
Message boards :
Questions and problems :
Tasks are not started with shortest deadline first
(Message 43105)
Posted 17 Mar 2012 by Richard Haselgrove Post: 24/7/52.14285714 And a day off in leap years... |
|
4059)
Message boards :
GPUs :
GPU Setup Question
(Message 43074)
Posted 17 Mar 2012 by Richard Haselgrove Post: I was using the correct driver for my GPU - Thank you again. "driver version 29573" may be correct according to the documentation, but you will find that one of your chosen projects - Einstein@home - will not allow you to process any of their work until you downgrade it - it's one of the buggy ones. Please read Einstein's Technical News. |
|
4060)
Message boards :
GPUs :
GPU Setup Question
(Message 43062)
Posted 16 Mar 2012 by Richard Haselgrove Post: By the way, you will also need to install videocard drivers from the GPU manufacturer's web site. For your NVidia GTX 260, use something like GeForce 285.62 Driver from the beta/archive download search. NVidia have acknowledged a major bug in the more recent drivers - they have promised the Einstein project that it will be fixed before the next release. |
|
4061)
Message boards :
Questions and problems :
Tasks are not started with shortest deadline first
(Message 43047)
Posted 15 Mar 2012 by Richard Haselgrove Post: And it matters not in the slightest. If there is any danger of a task missing deadline, it will be run in high priority mode, with the earliest deadline first. If there is no danger of missing deadline, then they will be done in the natural order of arrival, like any normal queue. What's so problematic about that? |
|
4062)
Message boards :
Questions and problems :
Manager says "Task suspended" but is computing
(Message 43042)
Posted 15 Mar 2012 by Richard Haselgrove Post: Great. In the meantime, I investigated a little bit more about the gpu-non-suspending problem. It appears that the problem is known since 2009: Unfortunately, although that sounds like a simple solution on the surface, in practice it'll cause more problems than it solves. Especially with modern NVidia drivers, abruptly terminating a CUDA application like that can trigger a protective reaction from the video driver, slowing the card down to a minimal clock rate - from which it can't recover without rebooting the computer. We'll have to be a bit more subtle than that. |
|
4063)
Message boards :
Questions and problems :
Bug reporting
(Message 42983)
Posted 11 Mar 2012 by Richard Haselgrove Post: I am well aware of the current procedure for reporting bugs. Agreed. I would add 3. Having the best possible chance of the developers fixing said bug. I am aware of four distinct methods of reporting bugs: A: Posting on this board achieves (1) and (2), fails (3) B: Filing a trac ticket achieves (1), poor results for (2) and (3) C: Posting to boinc_alpha mailing list fails (1), moderate results for (2) and (3) D: Having a quiet word in Jord's ear Best results for (3), fails (1) and (2) abysmally :-) |
|
4064)
Message boards :
Projects :
Cant download milkyway_nbody_0.84_windows_x86_64__mt.exe
(Message 42929)
Posted 7 Mar 2012 by Richard Haselgrove Post: It's possible that the executable files - .exe and .dll - are being blocked by your anti-virus program, particularly if you have one that implements "heuristic" scanning. Norton Sonar and Comodo Firewall have both caused problems like that in the past - they've always turned out to be false positives. Of course, you'll want to take all precautions, like pasting the download url (which can be found in client_state.xml) into https://www.virustotal.com/ - but in general the applications written and distributed by reputable BOINC projects are trustworthy (they get very close scrutiny). I'd certainly include Milkyway in the 'reputable' group. It way even be that the 7K you're seeing being downloaded is an AV report, instead of part of the file itself. Try looking at your AV/firewall console, to see if it's blocking anything. |
|
4065)
Message boards :
Questions and problems :
Communication Deferred errors
(Message 42911)
Posted 6 Mar 2012 by Richard Haselgrove Post: There was a particular problem to do with the configuration of a new server at SETI, which has been causing problems for the last 24 hours or so. It was found and fixed just before the project went into routine maintenance today, so hopefully it will run more smoothly once maintenance is complete. |
|
4066)
Message boards :
Web interfaces :
Boinc + Optimus + Ubuntu + GPU
(Message 42851)
Posted 4 Mar 2012 by Richard Haselgrove Post: I have a similar Dell XPS L501X laptop. Under Windows, BOINC has no problem finding and using the GT 420M GPU, even when the active display is set to use the Intel HD graphics chipset. There is a minor problem with the 'optimus' technology in current Windows versions of BOINC - the client returns "(driver version unknown)". This has been fixed for BOINC v7 (in testing), but didn't prevent CUDA applications from running. |
|
4067)
Message boards :
BOINC client :
multiple NVIDIA GPUs
(Message 42826)
Posted 2 Mar 2012 by Richard Haselgrove Post: Afterthought - has anybody discussed this potential cross-project "play nice" issue at Moo! ? It's similar to issues we saw with BOINC in the early days of multi-threaded CPU applications. I think the solution there was to unceremoniously dump all single-threaded tasks when a MT app was scheduled to run (whether or not they should have continued running under 'Task Switch' rules). That wasn't very nice, either. |
|
4068)
Message boards :
BOINC client :
multiple NVIDIA GPUs
(Message 42825)
Posted 2 Mar 2012 by Richard Haselgrove Post: Are you attached to other projects? Was anything else active at the time? I think that's the message you would see if, for example: Two count=1.0 applications are running, separately, one on each GPU. One of the two tasks finishes, leaving one GPU free and one GPU occupied. The client scheduler considers scheduling a Moo! app next - it finds one available GPU, needs two, and backs out with the 'insufficent' message. In general, BOINC doesn't like to pre-empt CUDA applications - it can be inefficient. So, Moo! probably won't run until both GPUs are freed by single tasks finishing at exactly the same time (unlikely), the other projects run out of work, or Moo! is forced to run in High Priority by deadline pressure. One possible work round would be to set a really low Task Switch Interval, so that the other tasks become pre-emptible quickly. |
|
4069)
Message boards :
Projects :
News on Project Outages
(Message 42814)
Posted 1 Mar 2012 by Richard Haselgrove Post: SETI recovered from the last one, but has another coming up. Monday Morning Outage (come to think of it, this probably affects BOINC as well) |
|
4070)
Message boards :
Questions and problems :
Problem with the setting "Maintain additional tasks to keep busy for up to"
(Message 42668)
Posted 22 Feb 2012 by Richard Haselgrove Post: And which project typically misses deadlines? Edit - if you're going to get to the bottom of this particular problem (and the somewhat related one you're discussing in an adjacent thread), you're going to need to do a bit of research. And without that research, you won't be sure if BOINC is working the way the developers intended. In particular, pay attention to how much work BOINC (on your computer) asks for, and to how much work BOINC (on the project server) sends you in response. If there is a problem, it could be in either of those places. |
|
4071)
Message boards :
GPUs :
Need help with GPUGrid app_info.xml
(Message 42645)
Posted 20 Feb 2012 by Richard Haselgrove Post: But still: <name>acemdlong</name> doesn't match <app_name>acemdlong_6.15_windows_intelx86__cuda31</app_name> I don't know what the correct value is - you'll have to get that from client_state - but they have to be the same (and right, of course :P) |
|
4072)
Message boards :
GPUs :
Need help with GPUGrid app_info.xml
(Message 42641)
Posted 20 Feb 2012 by Richard Haselgrove Post: Another place that's always worth checking is the apps.php page for the project - which should exist, even if it's hard to find an 'Applications' link on their website. In this case, it's not easy to see what's going on. But I checked my own rig running short tasks, and found <app> It's the <name> which matches the <app_name> lower down, as I suspected. Those two are the important ones. |
|
4073)
Message boards :
GPUs :
Need help with GPUGrid app_info.xml
(Message 42639)
Posted 20 Feb 2012 by Richard Haselgrove Post: Not every project has a thriving tradition of optimisation. In the specific case of GPUGrid, it may even be counterproductive. The apps themselves are pretty tightly written, and you're not aiming to improve them by re-coding the program: you're trying to increase your total throughput by running two at a time. In every case I've seen, that results in each individual task running more slowly than it would when running on its own. At GPUGrid, each scientific task is broken down into separate sections, each of which is derived from the results returned by the previous cruncher, and in turn each supplies the input for the following task. So the overall running time from issue to return is important to the project, and what is optimal for you (GPU usage) may not be optimal for the project (they'd prefer fast turnround). Having said that, I've run other projects under an app_info.xml written from first principles without help, so I can suggest what to look for. It's easiest to start by attaching to the project normally, and running at least one task. That downloads the applications, DLLs and whatever you're going to need: but it also gives you a working set of <app>, <file> and <app_version> entries in client_state.xml - you can use those to get the right values to put into your app_info framework. I haven't checked the whole file you posted, but two lines leap out at the eye: <app> <app_name>Long runs (8-12 hours on fastest card) (cuda31)</app_name> I think you'll find those two have to match, so that BOINC knows which application is associated with which app_version. |
|
4074)
Message boards :
BOINC client :
CRAZY? Temperature sensor
(Message 42584)
Posted 16 Feb 2012 by Richard Haselgrove Post: And a little like TThrottle, available for free by following the links in Automatic Temperature regulation on this board. |
|
4075)
Message boards :
Questions and problems :
Disabling BOINC messages?
(Message 42583)
Posted 16 Feb 2012 by Richard Haselgrove Post: David Anderson (private email) says that under these circumstances - service install, non-admistrative user logged in - the Manager shouldn't run at all (without going through all the hassle with DisableAutoStart or local Manager options), so your workers shouldn't see any notifications from it. I'd check that out, before going in to the Group Policy arena. |
|
4076)
Message boards :
Projects :
News on Project Outages
(Message 42580)
Posted 15 Feb 2012 by Richard Haselgrove Post: Seti: SETI is back in working order, but competition for WUs will be intense. |
|
4077)
Message boards :
Questions and problems :
Disabling BOINC messages?
(Message 42565)
Posted 14 Feb 2012 by Richard Haselgrove Post: Yes, a "service" installation would be the way to go - that way, BOINC could be doing something useful even when no user was logged in to the workstation. It's still available, though known as "Protected Application Execution" in recent versions. If you're installing manually, you'll need to click the 'Advanced' button on the third installer screen to give yourself the ability to select/deselect the option. Disabling the messages automatically for all users is trickier. They come from BOINC Manager - which you don't need to run in service mode - but by default the Manager will run at logon for every user, until each user on each machine disables the "Run Manager at login?" option from their own account. If the machines are part of a Domain, you might be able to turn off the autorun feature globally by deploying this registry key via group policy: [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager] "DisableAutoStart"=dword:00000001 Also, for a setup like this, I'd recommend setting up the tools for remote control as you do the installations, so that you can do minor maintenance centrally using tools like BoincTasks or even the venerable BoincView. |
|
4078)
Message boards :
BOINC client :
7.0.14 ignores setting of change project every x minutes
(Message 42525)
Posted 11 Feb 2012 by Richard Haselgrove Post: Remember that with BOINC v7.x.xx, we can't use the word 'debt' any more - at least, not in the old time-based units that we're familiar with. Even the <debt_debug> logging flag doesn't work any more, having been changed to <priority_debug> ([trac]changeset:24429[/trac]). Scheduling priority is now determined by credit awarded by projects, rather than by time. It's written up in http://boinc.berkeley.edu/trac/wiki/ClientSchedOctTen and http://boinc.berkeley.edu/trac/wiki/ClientSched, but no - I don't understand it either. But I'm seeing something similar. I have a machine with CPU tasks waiting to run, but which are never getting scheduled. They've been sitting there, waiting, for five days now. Their original deadline was 14 days, so they're now down to 9 days. I'll wait and see if they're held back until high priority running becomes necessary, and try to remember to watch the priority logs in the meantime. |
|
4079)
Message boards :
GPUs :
what is the cuda "compute capability" figure
(Message 42503)
Posted 9 Feb 2012 by Richard Haselgrove Post: It's a sort of hardware version number. Have a look at http://en.wikipedia.org/wiki/CUDA#Limitations and the table below. |
|
4080)
Message boards :
Questions and problems :
running 2 BOINCs ?
(Message 42478)
Posted 8 Feb 2012 by Richard Haselgrove Post: 1) Normally, only one copy of BOINCMGR is allowed to run at once. But there is a command (Advanced view, Advanced menu, next to last item): "Launch another BOINC Manager". That bypasses the 'only one copy' restriction, and allows you to fill in the 'localhost:31417' that Claggy was suggesting. 2) 'GPUs have become unusable' isn't a function of the Manager: the Manager is merely reporting the status of the underlying client. With Windows 7 (or Vista), the core client can only access the GPU drivers if it is running under the same user account as the currently logged-in active user: so if you switch users, the core client loses contact with the drivers and hence with the GPU. You can get round that by shutting down the core client and restarting it in the new user context. |
|
4081)
Message boards :
Questions and problems :
Stuck on no work units and 2 resluts quota...
(Message 42426)
Posted 3 Feb 2012 by Richard Haselgrove Post: I'm not a member of worldcommunitygrid, but 'Computer ID 1580833' should enable somebody who is to find your records and read whatever error messages are available. |
|
4082)
Message boards :
Questions and problems :
Stuck on no work units and 2 resluts quota...
(Message 42416)
Posted 3 Feb 2012 by Richard Haselgrove Post: Not unless you tell us which project, and give us the ID or other link to the host which is suffering problems. But, on the exceedingly sparse information you've given us, I'd guess you're missing required libraries. "Daily quota of 2 WUs" suggests that you are erroring every task. |
|
4083)
Message boards :
GPUs :
both gpus usable in SLI or Crossfire?
(Message 42378)
Posted 31 Jan 2012 by Richard Haselgrove Post: so boinc will use 2 GPU's but only if I put that file in ? I was reading in another topic something about dummy loads not being needed anymore - have there been updates to boinc ? BOINC will use two (or more) GPUs without configuration if they are of similar specification. If the GPUs are different, it will only use the 'best' device (or devices), unless the 'lesser' devices are manually enabled. How easily multiple devices can be seen by Windows seems to depend on both the Windows version and the device driver in use. Be prepared to experiment. |
|
4084)
Message boards :
Questions and problems :
Help, I can no longer install boinc
(Message 42285)
Posted 27 Jan 2012 by Richard Haselgrove Post: There can be a number of reasons for this. Most commonly, you can get round it by re-installing the old version of BOINC you were running before the trouble started. There are more explanations and workrounds in this BOINC FAQ. |
|
4085)
Message boards :
GPUs :
Wrong GPU being declared Not Used
(Message 42194)
Posted 24 Jan 2012 by Richard Haselgrove Post: cc_config.xml is an optional configuration file: it will only be present if you created it yourself. If you haven't got one, we can cross it off the list of suspects. BOINC tries to use the "best" GPU in a system. BOINC's assessment of "best" is done in the '_compare' functions in coproc_detect.cpp. If I'm reading it right, the priority order for NVidia GPUs is: 1) compute capability 2) CUDA version 3) Available memory 4) Speed In your case, the CC 2.1 GT520 is automatically declared the 'best' above the CC 2.0 GTX570 in the first round of voting: everything else is ignored. That suggests that the code is working as designed - which isn't the same thing as saying that the design is intuitive, or even the best possible. |
|
4086)
Message boards :
Questions and problems :
Wrong Signature
(Message 42114)
Posted 16 Jan 2012 by Richard Haselgrove Post: For some reason Boinc is displaying someone else's stats. Where do I find my ID number? Thank you, in advance. Signatures (like mine below) usually come from an external statistics site, according to the details you enter on the 'Community Preferences' page of an individual project web site. I don't see any signature below your post here, so I don't know what sort of signature you intended to use. However, I do know that one of the popular ones - BOINCstats - had a database problem sometime last year and re-issued user ID numbers: that had an effect similar to the one you're describing. Go to the project where the faulty signature is displaying: log in to your account: follow the link on your account page to BOINCstats (or whichever site you're using): collect the new signature image string for your account: and put that back into your community preference page to replace the corrupted one. |
|
4087)
Message boards :
BOINC Manager :
Boinc manager position second screen
(Message 42098)
Posted 15 Jan 2012 by Richard Haselgrove Post: I've used multiple monitors in Windows ever since they were first supported by Windows 98, and I've even tried to program for them. It ain't easy, but it can be done. My current main machine has two 1600 x 1200 screens side by side. I have them in the opposite arrangement to Renato - the primary screen with task bar and icons to the left, the secondary screen to the right. That makes the co-ordinate origin of the secondary screen (top left corner) to be (1600,0) Using BOINC Manager 7.0.3 on Windows 7, I can maximise the manager on the secondary display. The relevant data written to the registry is: Height 1216 Width 1616 XPos 1592 YPos 4294967288 That YPos looks better in hex: 0xfffffff8, or -8 :-) So it seems as if the BOINC Manager display is slightly over-sized, hanging 8 pixels over the edge on all four sides. But apart from that, the registry values look right for my geometry. If I close BOINC Manager, the registry values are unchanged. But when I open it again, the values change to Height 1176 XPos -8 - in other words, the Manager has moved to the primary screen: become slightly shorter to accommodate the task bar, and moved 1600 pixels to the left. This would appear to be something specific to BOINC's start-up logic. Most other applications that I use on this machine (the ones that have a windowsize memory - from Notepad upwards) do not adjust XPos on startup, and stay on the monitor I've most recently parked them on. |
|
4088)
Message boards :
Questions and problems :
Boinc 7.0.3 and Laptops.
(Message 42014)
Posted 11 Jan 2012 by Richard Haselgrove Post: Tested with Windows BOINC v7.0.3 (64-bit version). Using local settings throughout, BOINC responded as expected: With 'run according to preferences', and mains power disconnected: computation suspended when I unchecked "While computer is on batteries", and resumed when I checked the box. With "While computer is on batteries" unchecked: computation suspended when I removed the mains power cord, and resumed when I reconnected it. Laptop is a Dell XPS L501X. |
|
4089)
Message boards :
GPUs :
CUDA Work Units do not Suspend when Computer in use
(Message 41636)
Posted 14 Dec 2011 by Richard Haselgrove Post: Which version of BOINC are you running? 'Use GPU based on preferences' is a global setting, not per project. From what he says about SETI suspending, sounds like it's in operation already. Maybe there's something odd about the way Einstein's app does, or doesn't, respond to BOINC's "suspend" directive - but I haven't heard anything about that being a common Einstein problem. The Einstein message board would be the place to ask. |
|
4090)
Message boards :
Questions and problems :
DCF issue an back offs
(Message 41617)
Posted 13 Dec 2011 by Richard Haselgrove Post: I've been told there's a problem with the DCF swinging back and forth, any idea when this will be fixed Well, not quite. I think the questioner is referring to a specific set of circumstances, which you may not be following. Under the "runtime estimates" code changes associated with CreditNew, DCF is effectively redundant. Instead, averages are maintained on the server - not only per application, but even per application_version. The net figure accessible to users is the APR or Average Processing Rate - when this is being properly managed by project servers, client DCF remains close to 1.0000, even for a mix of applications and devices (CPUs and GPUs). It actually works, and has worked at SETI for about 15 months. Unfortunately, a well-meaning but flawed attempt to cope with problems caused by anomalous results (which confuse the averaging system) led to DA himself applying a code patch which broke the APR mechanism at SETI. That was in early September: the project is still suffering the after-effects. It's well known what needs to be done to rectify matters, but unfortunately a straight reversal of the code patch will cause as many problems as the patch did in the first place. It has to be a carefully graduated sequence of changes: neither David, nor the over-stretched SETI staff, have found time amongst their other fire-fighting duties to put them into effect. |
|
4091)
Message boards :
GPUs :
Why The Need For Double Precision?
(Message 41562)
Posted 10 Dec 2011 by Richard Haselgrove Post: For most of the science projects which run on the BOINC platform, they don't. There may be one or two projects (I believe MilkyWay is one) which have chosen to demand full 64-bit hardware, and haven't programmed their science application to use the well-documented mathematical techniques for emulating 64-bit accuracy on 32-bit hardware, but you'd perhaps be better asking at whichever project is causing you to ask the question. Alternative answer: these days, even a 32-bit CPU has an internal IEEE-compliant floating point maths coprocessor. GPUs, in general, don't. |
|
4092)
Message boards :
Questions and problems :
When will someone fix the Boinc installer to deal with installation on a Domain Controller?
(Message 41548)
Posted 9 Dec 2011 by Richard Haselgrove Post: The BOINC developers are stuck between two choices. On the one hand, they want to move to a new installer technology and resolve all of these long-standing issues with the new software. On the other hand, they know and understand the old installer, and it's comfortable to go on using it - bugs and all. And it's not worth spending any development effort on old technology. At least that's my interpretation of the answer I received when I asked that specific question (along with several others in the same vein) on the mailing list earlier this week. Unfortunately, their comfortable niche between the two forks in the road has satisfied them for several years already, and shows no sign of being vacated. |
|
4093)
Message boards :
Projects :
News on Project Outages
(Message 41504)
Posted 6 Dec 2011 by Richard Haselgrove Post: Their home pages are nowhere to be seen, is what I meant. No home pages, no forums. I didn't check in BOINC, since mine isn't running. All (Einstein, didn't check Albert) visible online here. May have been a brief routing problem? |
|
4094)
Message boards :
Questions and problems :
Edit Prefences screen won't Scroll. Screen clipped. Linux. Gnome.
(Message 41497)
Posted 6 Dec 2011 by Richard Haselgrove Post: Charlie was working in this area almost two years ago: http://boinc.berkeley.edu/trac/changeset/20727/boinc |
|
4095)
Message boards :
Questions and problems :
BoincTasks alternative BOINC manager
(Message 41465)
Posted 3 Dec 2011 by Richard Haselgrove Post: Here's a bug I've been able to replicate with BT 1.27 in combination with 6.13.12 BC development: I'm pretty sure this is a feature of the Core Client itself, nothing to do with BoincTasks. I've just replicated the first part of this report (request - and receive - new tasks for a suspended project) on a simple, stand-alone v7.0.2 installation, using the native BOINC Manager alone. Reported to boinc_alpha. |
|
4096)
Message boards :
BOINC Manager :
cannot connect to my BOINC project
(Message 41434)
Posted 1 Dec 2011 by Richard Haselgrove Post: http://192.168.2.5/test/ I think a local network address of that form is allowable (in general networking terms - not saying anything specific about BOINC here) - I often see addresses like that when using a browser to configure routers. But seyfullah - are you sure you're using the IP address of the virtual network adapter in the VM, not of the server itself? |
|
4097)
Message boards :
Questions and problems :
GPU CUDA work doesn't auto suspend when computer is in use
(Message 41288)
Posted 23 Nov 2011 by Richard Haselgrove Post: I also have serious problems using Cuda tech my PC Win7 x64 reboots automatically when Boinc attached to seti@home is trying to execute fermi packages! Any ideas? Does anyone else have this behaviour? TIA First guess - inadequate power supply. But ask this sort of question in the SETI@home Number Crunching forum - there are hardware guys there who can follow up (and the board is up again now - there were problems earlier). |
|
4098)
Message boards :
Questions and problems :
is BOINC damaging my PC?
(Message 41282)
Posted 23 Nov 2011 by Richard Haselgrove Post: You might also want to re-consider running CPDN. I thought that was geared mainly for systems equipped with GPU's. No, CPDN is very firmly a CPU-only project, and will remain so for the foreseeable future. |
|
4099)
Message boards :
Questions and problems :
Boinc 6.13.12 client stops working and disappears.
(Message 41226)
Posted 19 Nov 2011 by Richard Haselgrove Post: Did you really mean this version number, which dates back to early February and is long superseded. Could you be referring to the current test build v6.13.12? |
|
4100)
Message boards :
Questions and problems :
Does it exist a Boinc for SPARC cpu's?
(Message 41213)
Posted 18 Nov 2011 by Richard Haselgrove Post: There are a number of download sites for other operating systems and processors (including Solaris/SPARC) at http://boinc.berkeley.edu/trac/wiki/DownloadOther |
|
4101)
Message boards :
Questions and problems :
BOINC on Winders Server 2003
(Message 41155)
Posted 14 Nov 2011 by Richard Haselgrove Post: If you were to 'install' with the administrative switch "boinc_installer_name.exe /a", it would unpack without problems. Be careful when doing that on a host where BOINC is already running. Even in 'admin' mode, the installer sets up INSTALLDIR and DATADIR paths pointing to the network unpack directory, as I reported in trac ticket [trac]#847[/trac] - I've just tested v6.13.12, and it's still doing it. If you need to unpack, it's best to do it using a BOINC-free workstation on the network. |
|
4102)
Message boards :
Questions and problems :
Running Headless - Problem Connecting to the Projects
(Message 41152)
Posted 13 Nov 2011 by Richard Haselgrove Post: And back to the subject of BOINC. BOINC can, and does, try to distinguish between "can't connect to server" and "can't connect to internet". The former is likely to be a server problem, and requires anti-DDOS measures when service is resumed. The latter is likely to be a local problem, so doesn't. |
|
4103)
Message boards :
Questions and problems :
Set affinity for one project?
(Message 41100)
Posted 9 Nov 2011 by Richard Haselgrove Post: Is there any way to set affinity (which processors it runs on) for one project? Use Process Lasso. But why? |
|
4104)
Message boards :
Questions and problems :
Boinc server: How to find the current version
(Message 41091)
Posted 9 Nov 2011 by Richard Haselgrove Post: Some, but not all, projects display the server code revision number on their Server Status Page. Thus, NumberFields is running software including revision 24527. |
|
4105)
Message boards :
BOINC Manager :
Feature request - task control
(Message 41048)
Posted 7 Nov 2011 by Richard Haselgrove Post: And I got caught the same way when suspending tasks in BOINC v6.13.10 In my case, I got a dollop of NumberFields, which - to add to the stress - the server had allocated without reference to the client's current DCF of something over 7.5: two and three quarter million unexpected seconds of work.... See BOINC Dev mailing list. |
|
4106)
Message boards :
BOINC Manager :
Feature request - task control
(Message 41044)
Posted 7 Nov 2011 by Richard Haselgrove Post: Hard to believe, but maybe we're talking different things. There was a time that suspending tasks were not counted for the cache and that's been gone a very long time. If suspending tasks, when technically the cache is full, then fetching work from other projects to exceed the total value of cache, I'd be surprised. I've had a task for longer in suspend [the last task], not the project, and BT tells me the sum of all work is still what I've got set, and never requesting more unless that projected total [of all projects] drops below the set value. Same observation for 6.10 and 6.12. And I got caught the same way when suspending tasks in BOINC v6.13.10 |
|
4107)
Message boards :
Questions and problems :
Restriction of multiple CPU requests
(Message 41037)
Posted 7 Nov 2011 by Richard Haselgrove Post: Personally the *pigs* projects might want to consider an option on their crunching profile where you can set the number of allowed threads, but it will likely require the client to listen for that parm, though Richard Haselgrove's mod suggest that the the project could send a new app_info file to facilitate i.e. no BOINC client change requirement. Well, it wouldn't be an app_info file, because that's something that an individual user would use to run under non-standard conditions ("anonymous platform"). But a project could certainly fine-tune their application setting through the <app_version> block describing the application in question. [You can find that in client_state.xml]. This is how AQUA did it: <app_version> The ncpus lines are a hint to BOINC about how to schedule the app, and the cmdline was AQUA's way of controlling their own application. Changing both to three would effect the sort of change Kirk is suggesting. Since Kirk is seeing 8 cores in use, and Sekerob just 4, I would assume that MilkyWay has similar controls available for their app - it would just be a matter of persuading them to allow user control of them via preferences. |
|
4108)
Message boards :
Questions and problems :
Restriction of multiple CPU requests
(Message 41033)
Posted 7 Nov 2011 by Richard Haselgrove Post: There is no way to do what you want. Just let the task run. It will run for a while and then your other projects will get their share of the CPU. Without knowing the project in question, I'd hesitate to say "no way". On the previous multi-threaded project, AQUA, I ran one of their apps successfully on three out of four cores (leaving one core running other projects) by writing a custom app_info.xml file. |
|
4109)
Message boards :
BOINC Manager :
Password invalid on 6.13.6
(Message 40874)
Posted 28 Oct 2011 by Richard Haselgrove Post: Now testing v6.13.10 |
|
4110)
Message boards :
Questions and problems :
Application uses too much bandwith.
(Message 40774)
Posted 22 Oct 2011 by Richard Haselgrove Post: And tell them that BOINC itself requires very little bandwidth. By far the greatest part of the bandwidth requirement is for the science applications and data that BOINC is simply managing. Do some research on the various projects which can be run on the BOINC framework. Try to find one which both (a) has relatively low data requirements, and (b) is in a field of scientific research that has some sort of relationship with the classes being taught in the lab in question. Then go and talk to the educationalists (not the administrators) about how beneficial it would be for the students to see some real scientific research happening on their computers. Administration should exist to facilitate primary purpose, not the other way round. |
|
4111)
Message boards :
Questions and problems :
Using BOINC with more than one user logged in.
(Message 40363)
Posted 25 Sep 2011 by Richard Haselgrove Post: When I tried this on Windows 7, I found I could crunch with the GPU after fast user switching, simply by shutting down the BOINC core client and restarting it. The key seems to be the line 16-Sep-2011 21:03:53 [---] Running under account Richard Haselgrove in the startup messages: if that line matches the account of the foreground user, you get a GPU: but if you switch, so the BOINC user is in the background, you don't. Whether there's any way of detecting a user switch, and automatically stopping/restarting BOINC when it happens, I wouldn't begin to guess. |
|
4112)
Message boards :
Questions and problems :
Boinc no longer connects in protected (service) mode
(Message 40350)
Posted 24 Sep 2011 by Richard Haselgrove Post: I expect W8 Preview is a Pro version. That changeset is the source code for the client, determining how it behaves after installation. skgiven's problem is with the installer, which will use different code entirely. Agreed, they should both have been updated at the same time, and should most likely both have been updated from the same list - but it ain't necessarily so. While I'm on, a couple of messages back Jord wrote ... the temporary workaround that disables ATI GPU detection at all when installed as a service. The actual checkins say "Disable GPU detection of Windows when running as a service" - no mention of limiting it to ATI only, and the code change appears to be in a section that applies to both ATI and CUDA. I'd prefer it the way Jord wrote it - I don't see why my NVidia/XP hosts should need to be reconfigured because of an ATI coding choice. |
|
4113)
Message boards :
Questions and problems :
Running BOINC on GC (PDC)
(Message 40331)
Posted 23 Sep 2011 by Richard Haselgrove Post: Does 5.10.45 recognise GPU's? No, the first one that does - and probably only NVidia GPUs, in case you wondered - is v6.4.5 As you've discovered, the problem with domain controllers is installation. Later versions apparently run OK, if you position the files manually, though obviously there's not much experience with Windows 8 yet. |
|
4114)
Message boards :
Projects :
News on Project Outages
(Message 40282)
Posted 21 Sep 2011 by Richard Haselgrove Post: I can reach the Seti site and forums without a problem. If you still can't, it's probably something local. The site and forums run off a different internet connection than the work connection does. There was a complete communications blackout for both data and web services between about 04:00 and 06:00 local (Berkeley) time today - I usually take that to mean some planned network maintenance by comms staff in the Bay area, not necessarily either in the SETI lab or the Berkeley campus. All services are back up and running now, albeit at their usual congested (glacial) speed. |
|
4115)
Message boards :
Projects :
LHC@Home a.k.a Test4Theory..now SIXTRACK a.k.a LHC@Home 1.0??
(Message 39975)
Posted 2 Sep 2011 by Richard Haselgrove Post: The last I heard was that the old LHC/sixtrack server was being 'repatriated' back from QMC London (I think it was) to CERN in Geneva. It was in much need of an overhaul and updating, anyway. I believe the plan was to bring it back into use, but I'm guessing it's taken longer, and more resources, to get the VM version to this stage than expected. |
|
4116)
Message boards :
Projects :
Aqua@Home
(Message 39819)
Posted 23 Aug 2011 by Richard Haselgrove Post: Home-page announcement at AQUA@home Thank-you all |
|
4117)
Message boards :
BOINC Manager :
unlimited wu download?
(Message 39809)
Posted 23 Aug 2011 by Richard Haselgrove Post: That depends on the server administrators on each project. They have to make choices depending on the number of transactions they can process, the size they are prepared to let their database grow to, the storage space available for data and result files, backup facilities available, the time it takes to backup and manage their database, etc. etc. etc. |
|
4118)
Message boards :
Questions and problems :
BONIC Runtime Error Message keep being received
(Message 39802)
Posted 22 Aug 2011 by Richard Haselgrove Post: Thanks, On the off chance it saves time, why not look at the task lists for the host in question, at each of those project web sites? Most, if not all, will allow you to apply an 'error' filter to the list - any project with a non-zero recent error count stays on your short-list. |
|
4119)
Message boards :
Questions and problems :
can't upload when using dsl 512/64
(Message 39774)
Posted 21 Aug 2011 by Richard Haselgrove Post: I see 21/08/2011 13:23:41 [http_debug] [ID#1] Received header from server: Via: 1.1 greencache3:3128 (squid) Are you correctly setting/resetting BOINC's proxy settings when switching between the 1mb/256k and 512/64 dsl accounts? |
|
4120)
Message boards :
Projects :
News on Project Outages
(Message 39694)
Posted 16 Aug 2011 by Richard Haselgrove Post: And Einstein is fully back. No explanation so far. |
|
4121)
Message boards :
Projects :
News on Project Outages
(Message 39693)
Posted 16 Aug 2011 by Richard Haselgrove Post: Einstein seems to be down - no website, no (gravity wave) uploads, no reporting completed tasks. I think all of those are handled at the American server-site, not in Germany. |
|
4122)
Message boards :
Questions and problems :
BOINC MANAGER
(Message 39636)
Posted 12 Aug 2011 by Richard Haselgrove Post: 2. There is no security tab under properties there is a general, sharing and customize. With XP Pro (as stated in Message 39591), full security settings are available to you. It sounds like you have 'simple file sharing' active - follow the instructions in Microsoft Knowledge Base article 307874 to turn it off (temporarily, at least). 5. this system didn't come with a windows install disk. If you have a pre-installed copy of Windows, the files needed by SFC should already be on your hard disk, and SFC will know where to find them. You can run it anyway. |
|
4123)
Message boards :
Questions and problems :
BOINC MANAGER
(Message 39623)
Posted 12 Aug 2011 by Richard Haselgrove Post: Note that the original location that you asked about, c:\program files\boinc\, has different uses in different versions of BOINC. In BOINC version 5 (very old, now), absolutely everything got shovelled in there - programs, data and all. In newer versions of BOINC, that folder is reserved for BOINC's programs (executable files) only. It is, on some versions of Windows, locked against modification by anyone except BOINC's own user group, as a secutity measure against malware. All the files you would normally be interested in modifying should be in a separate data directory. |
|
4124)
Message boards :
Projects :
News on Project Outages
(Message 39560)
Posted 10 Aug 2011 by Richard Haselgrove Post: Aqua remains in a curious 'Predictor-like' status -- the servers are up and running, the message boards are up and running, but there has been no work and no comment from Dwave (who used to run the project) since the project admin left a couple of weeks ago. There's a distinction to be drawn between DWave the company (who sponsor the project, own the server, and commission/use the research - and are, so far as I know, fully functioning as normal) ... ... and the small unit or department which used to interact with us through BOINC - that's the bit which is going through staff changes. |
|
4125)
Message boards :
BOINC client :
Feature Request: Hierarchical project ordering
(Message 39548)
Posted 9 Aug 2011 by Richard Haselgrove Post: If you can enter zero for resource share in the preferences box on a project website, and if you see a zero resource share for the project in BOINC Manager after updating, then that's "support" enough. Al that controls is whether the BOINC client will request new work from the project - it'll only request work if no other project is available and willing to supply work. Once the work is downloaded, it doesn't run with a resource share of zero - that would imply that any task in progress when the main project came back on line would never get completed and returned - which isn't the idea at all. In contrast, it actually behaves the opposite way, running continuously until completion (as it it were running high priority), so it can be cleared out of the way as quickly as possible and return you to your primary project. From that perspective, Climate Prediction (which only has multi-day tasks on offer) might be seen as an odd choice for a backup project. |
|
4126)
Message boards :
BOINC Manager :
How to stop Boinc from auto starting on system boot?
(Message 39542)
Posted 9 Aug 2011 by Richard Haselgrove Post: Thanks, that seems to be what I missed. I actually didn't think about looking for an auto-start option within a "display and network" option panel. Boinctray is a small utility which monitors user activity (keyboard and mouse) to assist BOINC to implement preferences like "don't run while user is active". It's part of the suite, sure, but not one which will cause you any problems. |
|
4127)
Message boards :
BOINC Manager :
How to stop Boinc from auto starting on system boot?
(Message 39521)
Posted 8 Aug 2011 by Richard Haselgrove Post: It is in the registry: I think Rom explained it once on boinc_alpha. The HKEY_LOCAL_MACHINE...Run is a global setting which does start the Manager - every time. But the first thing it does is test [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager] "DisableAutoStart"=dword:00000001 to see whether BOINC should be running for the current user - and if that is set, it closes itself down again before you notice anything. That's why you have to set the option checkbox individually for each user of the machine (it defaults to enabled). I'm not sure exactly what ENABLELAUNCHATLOGON does - it doesn't seem to have a value on my systems, even the ones which do autostart. But because it's in a "Space Sciences Laboratory" key, it can only be read by some BOINC program or other, which means that BOINC program would have to be running already for it to have any effect. Because it's in the 'Setup' subkey, I suspect it may be intended to control the default setting of the 'Run Manager' checkbox on the final installation screen, the next time the installer is run. |
|
4128)
Message boards :
BOINC Manager :
How to stop Boinc from auto starting on system boot?
(Message 39516)
Posted 8 Aug 2011 by Richard Haselgrove Post: Thanks, that seems to be what I missed. I actually didn't think about looking for an auto-start option within a "display and network" option panel. It is in the registry: [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Run] "boincmgr"="\"C:\\Program Files\\BOINC\\boincmgr.exe\" /a /s" I agree about the misleading menu location. |
|
4129)
Message boards :
BOINC Manager :
How to stop Boinc from auto starting on system boot?
(Message 39514)
Posted 8 Aug 2011 by Richard Haselgrove Post: There is, in fact, a control for this. Find the BOINC Manager program - either through the start menu, or as a hidden icon which will be running in the Notification Area (system tray for older versions of Windows). Switch to 'Advanced View', if it isn't already. From the 'Tools' menu, select 'Display and network options...'. Uncheck the option for 'Run Manager at login?'. Close BOINC Manager fully (File|Exit) to ensure that the new setting is flushed to the registry. And you should be good to go. Note that if more then one user has access to the computer, you may have to repeat this exercise for each user login account. |
|
4130)
Message boards :
Questions and problems :
Unintentional "Return Results Immediately"
(Message 39422)
Posted 3 Aug 2011 by Richard Haselgrove Post: I've never been quite sure how much of a deliberate BOINC development policy that was. There's a one minute backoff after a task ends in a computation error, but I agree with you - I'd have thought it would be a good idea to do something similar after a 'success' outcome too. There is now - again, new in v6.12.xx - a pause while the result file with the outcome of the task is uploaded. There's no sign of that in your log - are you sure that this scheduler request wasn't reporting a previous task? |
|
4131)
Message boards :
Questions and problems :
zero disk space after upgrade of BOINC
(Message 39409)
Posted 2 Aug 2011 by Richard Haselgrove Post: 500GB drive with 340GB free That doesn't quite match 7/29/2011 8:50:42 AM | | Data directory: C:\ProgramData\BOINC Are you sure that BOINC ended up on the drive that you intended to install it on? |
|
4132)
Message boards :
Questions and problems :
zero disk space after upgrade of BOINC
(Message 39408)
Posted 2 Aug 2011 by Richard Haselgrove Post: And because you have 7/29/2011 8:50:42 AM | | Reading preferences override file it's the local preferences set through BOINC Manager that you have to check, not any website settings - or you could clear the local settings, and return to using web-managed ones. |
|
4133)
Message boards :
Projects :
Test4Theory (AKA LHC@home 2.0) NEWS
(Message 39397)
Posted 1 Aug 2011 by Richard Haselgrove Post: Wow, it's not for the faint hearted, I see. 220MB tasks, 8GB drive space needed? Anything else I haven't seen yet? 4GB of RAM needed? ;-) From the front page - Before Attaching to us you must first have an up to date installed version of the VIRTUAL BOX software package. |
|
4134)
Message boards :
Questions and problems :
Regularly have trouble getting BOINC to request new WUs in projects
(Message 39374)
Posted 30 Jul 2011 by Richard Haselgrove Post: Because when BOINC tries, but fails, to obtain work from a project, it understands that the project server maybe under stress of some kind. Rather than adding to the stress by repeated requests for work, it backs off the work request rate to the project(s) having difficulty keeping up, and asks another project to top up your cache instead. The longer the project problems which result in "can't supply work" go on, the less your machine will pester it. But your machine will still ask roughly once a day: in effect "have you sorted out your problems yet?". If that request receives work, then you're back in business: if it doesn't, your computer keeps quiet for another day until trying again. Aithough BOINC tracks your relative work across projects, and tries to obey your preferences in the long term, that isn't the most important factor in its decision-making process. Making allowance for projects with unreliable or overstretched servers (of which there seem to be an increasing number these days) is also important. |
|
4135)
Message boards :
Questions and problems :
Excessive IPv4 loopback
(Message 39334)
Posted 28 Jul 2011 by Richard Haselgrove Post: That's worth thinking about. The way I understand it, every second the Manager (the bit we can see) asks the Client (the worker in the background) to describe itself, and say what it's doing. The request is trivial - it'll be the data (text) in the reply which comprises the vast bulk of the loopback traffic. The nature of the traffic is described in GuiRpc - and in that, I only see one get_results(RESULTS&). That leads me to question Claggy's suggestion. It seems to me that the Manager must request (and receive) data for all results, whether or not 'active tasks only' is selected. Perhaps it's the manager which simply discards data relating to inactive tasks: the time-saving allowing screen refresh at 1 Hz might only be in the text-to-graphics rendering in the manager (which is plausible - GUIs always place heavy computational demands on any system. That's why GPUs were invented). The complete 1-second cycle comprises: Render binary data in memory into text (XML) format Exchange text data between two applications by loopback Parse text data, and render it graphically If I'm right, and only one of those steps (the final one) is controlled by the 'active tasks' button, then a further optimisation is obvious. A further GUI_RPC command, get_active_results(RESULTS&) - which should be trivial to derive from the existing code - would reduce the time taken for both the initial memory-to-text conversion, and the communications (in both programs), in addition to the existing GUI savings. Or maybe that's been done already, and it's only the documentation which is out-of-date - wouldn't be the first time that has happened in the history of software development :-) |
|
4136)
Message boards :
Questions and problems :
Unintentional "Return Results Immediately"
(Message 39314)
Posted 27 Jul 2011 by Richard Haselgrove Post: There is a new feature in v6.12.33, that when a project is set to 'No new tasks' it is also, and automatically, set to report results immediately. The idea is that when BOINC is running normally, it is allowed to fetch work as and when needed: but people sometimes know better than BOINC can that, say, a vacation is imminent. If they set NNT in anticipation, without work fetch completed work wouldn't be reported - and sometimes it isn't reported until after the vacation, which can be too late. Hence, the clearout of completed work as it happens. Could you be, inadvertently, invoking that rule? |
|
4137)
Message boards :
Questions and problems :
Excessive IPv4 loopback
(Message 39308)
Posted 27 Jul 2011 by Richard Haselgrove Post: And reverting to an older version won't help - they all work the same way. If you haven't noticed it before, that's because you haven't been looking - it isn't new to v6.12.33 |
|
4138)
Message boards :
Projects :
News on Project Outages
(Message 39218)
Posted 23 Jul 2011 by Richard Haselgrove Post: It seems Surveill@Home bit more off than they could chew. They're down. Ironic, really. Who will provide statistics on their downage? |
|
4139)
Message boards :
Questions and problems :
Terrifying update to boinc_6.12.33_windows_x86_64
(Message 39208)
Posted 23 Jul 2011 by Richard Haselgrove Post: Neither of your solutions rings true. I did download a second copy of the installer (that was the first thing I did before I came to this website), same size, same error. The previous boinc.msi was valid, because BOINC is no longer installed on the machine. Other suggestions? That's odd. The copy of boinc_6.12.33_windows_x86_64.exe that I downloaded a while back (8,417,328 bytes, MD5 hash 4FFC4EE49EFA55AE2A1188909ECD702F) installed just fine and has been running ever since - for 64-bit, I can only offer one installation on Windows 7 Home Premium as evidence. The 32-bit versions were successful on both WinXP and Win7. I'm afraid the lack of 'me too' complaints all over these (and other project) message boards still suggests that your problem is closer to home. |
|
4140)
Message boards :
Questions and problems :
Terrifying update to boinc_6.12.33_windows_x86_64
(Message 39205)
Posted 23 Jul 2011 by Richard Haselgrove Post: 6.12.33 for Windows will not install. It pops the dialog box looking for source "1". When you locate boinc.msi on the hard drive it says it's invalid. I see a lot of other threads where this version has failed. How about posting a link to a previous version that does install correctly while you figure out why this install package is whacked? Because it isn't whacked. There are two possibilities for an invalid boinc.msi: Your copy of the new v6.12.33 installer may have been whacked by a bad download. Delete it, download a fresh copy to your hard disk (save it locally, don't try to run it directly from the download link). Check that the file size is right (i.e. that you've got a complete download) before trying to install again. Or: your previous boinc.msi - needed to uninstall the previous version - may have got damaged since you installed the last one. The other threads, and the linked FAQ, have advice on how to cope with that scenario. |
|
4141)
Message boards :
News :
New project: Surveill@Home
(Message 39196)
Posted 22 Jul 2011 by Richard Haselgrove Post: Does that include BOINC project web sites? :-) |
|
4142)
Message boards :
Questions and problems :
Visual C++ Runtime
(Message 39150)
Posted 20 Jul 2011 by Richard Haselgrove Post: Yes, that's a CPDN file - and since it's called 'graphics', and the problem happens during idle time, I think the finger points to a screensaver. Might be best to disable the BOINC screensaver function until you're able to ask/read at CPDN. |
|
4143)
Message boards :
Questions and problems :
Since Upgrade To Latest - Not Starting With Windows
(Message 39134)
Posted 20 Jul 2011 by Richard Haselgrove Post: If you didn't know, in v6.12.33 you can go (in Advanced view) to the Tools menu, 'Display and network options...' (not intuitive, I know), and find a checkbox for 'Run Manager at login?'. That would be easier than making your own shortcut. Note that when started that way, BOINC runs silently in the system tray/notification area - under Windows7, the icon is invisible by default, and the same may be true under Vista. Before you change anything, make sure BM isn't running already in the background. |
|
4144)
Message boards :
Questions and problems :
Since Upgrade To Latest - Not Starting With Windows
(Message 39125)
Posted 20 Jul 2011 by Richard Haselgrove Post: Which are your 'older' and 'latest' versions - hard numbers, please. Are you following the 'recommended' releases v6.10.58 --> v6.12.26 -->v6.12.33, or are you talking about yesterday's alpha release v6.13.1? |
|
4145)
Message boards :
Projects :
News on Project Outages
(Message 39099)
Posted 19 Jul 2011 by Richard Haselgrove Post: GPUGrid's hardware upgrade is scheduled for today, Tuesday 19 July - time not specified. They are anticipating "a couple of hours maximum during the day". |
|
4146)
Message boards :
Projects :
News on Project Outages
(Message 39054)
Posted 18 Jul 2011 by Richard Haselgrove Post: GPUGrid Server may be down for a couple of hours for hardware upgrade. News, 18 Jul 2011 7:23:29 UTC |
|
4147)
Message boards :
BOINC Manager :
Cannot Install 6.12.33
(Message 39023)
Posted 17 Jul 2011 by Richard Haselgrove Post: Running Vista 64bit, trying to install 6.12.33 Explanations and a variety of workrounds are in this BOINC FAQ |
|
4148)
Message boards :
Projects :
News on Project Outages
(Message 38879)
Posted 11 Jul 2011 by Richard Haselgrove Post: Which is weird, as it's two different connections and systems. I mean, the 100Mbit pipe is only for data, their forums run off a separate 1Gbit pipe through campus. It's also different servers that it all runs on, so it sounds like a collective power-outage, but then how do we type on these boards? ;-) It can't be either a connection, or a power, issue, for the reasons you state. SETI has a number of different servers, and tends to have a separate server for each major BOINC function. At the moment, the file upload server and the scheduling server are both intact, and so is the BOINC database - reported results are being validated. But the download server and the web server are inaccessible. |
|
4149)
Message boards :
Questions and problems :
Cobblestone definition
(Message 38644)
Posted 23 Jun 2011 by Richard Haselgrove Post: Agreed. Done. |
|
4150)
Message boards :
Projects :
News on Project Outages
(Message 38540)
Posted 18 Jun 2011 by Richard Haselgrove Post: Climate has been down for a while... The project development/administrative team are aware of this, and are trying to fine-tune the servers to support the very heavy data traffic that the project requires. From a report this morning (Saturday): Most of the CPDN servers are set to work their little socks off, and they just can't cope. Realistically, it'll probably be Monday before the servers are re-enabled. Additionally, the same (very small) team are trying urgently to eradicate the last few bugs from a new application that the scientific researchers are anxious to get into full production. Other issues, such as credit exports, are taking a bit of a back seat to that. |
|
4151)
Message boards :
Questions and problems :
Multiple Managers getting confused with WU progress for multiple clients
(Message 38497)
Posted 16 Jun 2011 by Richard Haselgrove Post: Just check that the --detach command (as shown in the documentation) works properly for you. I recently hit a problem with it, and found that the actual source code uses --detach_phase_two, which worked better for my purposes. Also, note that the new format --detach_console (shown in the revised documentation) will only start working from v6.12.34 at the earliest, and possibly not until v6.14 |
|
4152)
Message boards :
GPUs :
Can I run BOINC programs on PCI (Not PCI-E) GPUs?
(Message 38416)
Posted 14 Jun 2011 by Richard Haselgrove Post: I don't expect spectacular performance, but I want to maximize my contributions to BOINC projects that use GPUs, like Collatz conjecture or MilkyWay@home. Is there any way to get these projects to run on a PCI GPU? Yes, the bus doesn't matter to BOINC. You'd need to keep an eye on other limitations and specifications: Check the minimum driver level at the projects. Likewise, check the minimum video RAM requirement (cards with only 256 MB RAM are pretty much restricted to Collatz and SETI). I'd be slightly surprised to see a double-precision card (required by MilkyWay) on a PCI bus, but I haven't researched that. Projects which shift a lot of data to and from the GPU, like Einstein, might be slowed down by the older bus. Edit - forgot to state the obvious: it has to be a supported card - 'CUDA' for nVidia (check on http://www.nvidia.co.uk/object/cuda_gpus_uk.html). People have recently found 8400GS and 9500GT cards in PCI format, though they're not sold in all countries. I'm afraid I don't know the equivalent data for ATI. |
|
4153)
Message boards :
Questions and problems :
system getting hacked after install of boinc
(Message 38382)
Posted 13 Jun 2011 by Richard Haselgrove Post: 2. Windows users should install BOINC in Protected Application Execution (PAE) mode which I believe is the default installation mode in the BOINC installer for Windows. Installing in PAE mode severely restricts project executables in what they can do on your computer. (I hope this info about Windows is correct, someone please correct me if I'm wrong, it's been a while since I've run Windows) That's correct. Unfortunately, the protection against malicious behaviour is so strong that project executables can't interact with graphics card drivers under PAE, and that rules out GPU computation. Participants who want to offer the use of GPUs to projects (increasingly common) are forced to opt for the insecure 'user' mode of operation. |
|
4154)
Message boards :
BOINC client :
Need help about client_state.xml
(Message 38338)
Posted 11 Jun 2011 by Richard Haselgrove Post: Pepo can you help me with RPC way? My programming experience is in a different area, but I managed to get RPC running in Visual Basic 6, so I can help with the logic - the two documentation pages I linked in the other thread aren't exactly transparent. As Pepo says, the first stage is working out how to open a network socket in your development language. @ both - we probably ought to move this conversation away from the News area. |
|
4155)
Message boards :
News :
BOINC Monitor
(Message 38331)
Posted 10 Jun 2011 by Richard Haselgrove Post: Thanks for the name-check, but I still think file-scraping is the wrong way to go for the long term. Biting the bullet and switching to RPC will give you greater flexibility and more possibilities, as I suggested in your original thread. |
|
4156)
Message boards :
BOINC Manager :
Aborted work units...?
(Message 38317)
Posted 10 Jun 2011 by Richard Haselgrove Post: Another bit of missing information - your operating system. With a brand new Asus laptop - let me guess that you're running Windows 7. When I bought a new CUDA-enabled laptop earlier this year, the fan didn't run at anything like the speed needed for GPU crunching until I found and enabled the 'Active' system cooling policy under processor power management. |
|
4157)
Message boards :
News :
BOINC Monitor
(Message 38295)
Posted 9 Jun 2011 by Richard Haselgrove Post: Going the GUI RPC route will get up-to-date information, all gathered in one place. BOINC hasn't re-written client_state.xml agressively for a while now, because of the heavy workload it involves with multi-core CPUs and now GPUs. Progress updates have been transferred to individual boinc_task_state.xml files in each slot directory since sometime in 2009. |
|
4158)
Message boards :
Questions and problems :
Boinc failing to close apps
(Message 38219)
Posted 6 Jun 2011 by Richard Haselgrove Post: And version numbers? I refer the honourable gentleman to the answer I gave some moments ago.... OK, I posted in boinc_alpha on 17 April in relation to v6.12.22, but quoted an earlier report from Carola Tralau dated 18 November 2010, referencing v6.12.6 |
|
4159)
Message boards :
Questions and problems :
Boinc failing to close apps
(Message 38210)
Posted 6 Jun 2011 by Richard Haselgrove Post: In my case: For my two Windows 7 machines, I use - - normal mode: automatic "run manager at login" from registry. - maintenance mode: no autorun, start when needed from Manager shortcut on desktop/start menu. For XP machines - - installed as service, start and stop services using services control panel, run Manager from shortcut only when needed. But I don't think the issue is so prevalent on XP. |
|
4160)
Message boards :
Questions and problems :
Boinc failing to close apps
(Message 38204)
Posted 6 Jun 2011 by Richard Haselgrove Post: My impression matches that: "No see exit dialog, exit dialog choices not actioned". I'm wondering if this can be linked to the "graphics button greyed out" that others are reporting, and Charlie is looking into. Seems to happen under similar circumstances. |
|
4161)
Message boards :
BOINC Manager :
Recent modifications to Boinc Mgr look - not liking it [have included suggestions]
(Message 38201)
Posted 6 Jun 2011 by Richard Haselgrove Post: Now do tell, what lines? I do find that on a widescreen monitor there are (fine, unobtrusive) lines between the data rows, which extend into the empty space to the right of the last data column. And that if you have the manager open on a dynamic tab where the number of data rows varies frequently - transfers, for example - the lines remain visible after the data row has been cleared. A slightly untidy, cosmetic bug: I wouldn't call it a show-stopper. Platform: Windows. |
|
4162)
Message boards :
Projects :
News on Project Outages
(Message 38175)
Posted 5 Jun 2011 by Richard Haselgrove Post: I noticed both the CPDN forums are no longer on line. Anyone know about that? Both are online as tested from the UK a few seconds ago. |
|
4163)
Message boards :
Questions and problems :
Boinc-6.12.26 in Gentoo, buffer overflow
(Message 38128)
Posted 2 Jun 2011 by Richard Haselgrove Post: So far I can only confirm that WCG causes the above error, as soon as I attach it. Or, perhaps, it's the encryption required by HTTPS which invokes the call to SSL, and hence triggers the (failing) library access? |
|
4164)
Message boards :
BOINC client :
6.12.26: insufficient nVidia tasks
(Message 38112)
Posted 1 Jun 2011 by Richard Haselgrove Post: You would need to look in detail at the debt values - "Work Fetch Priority" x3, on each project properties page in BOINC Manager - and track it over time. I did that for months on end while we were trying to get GPU scheduling right - I think it's pretty good now, and respects requested resource share over the long term. If you think you're seeing something different, best to set a well-chosen set of those logging flags Jord suggested, and present the evidence. |
|
4165)
Message boards :
Projects :
News on Project Outages
(Message 38109)
Posted 1 Jun 2011 by Richard Haselgrove Post: Looks like SETI just had another crash. It was only down for a few minutes, mostly back up now. Jord requested no sigs in this thread, I think. |
|
4166)
Message boards :
GPUs :
Missing ATI GPU after update to 6.12.26 (x64)
(Message 37993)
Posted 25 May 2011 by Richard Haselgrove Post: Thanks! :) That fixed it, however I am somewhat confused. I have that same setting selected on my desktop and its graphics card is found just fine :/ You can run in 'protected application execution' mode in Windows XP, but not in Vista or Windows 7. |
|
4167)
Message boards :
Questions and problems :
Ubuntu 11.04: BOINC can't connect to localhost
(Message 37991)
Posted 25 May 2011 by Richard Haselgrove Post: Rumours have it [read it I think in the alpha mail list, though can't remember if this is the 64 bits], that change in Ubuntu libraries broke the BOINC .sh builds. I'm running with the Synaptic RPM v 6.10.59 dbg now on 11.04. The download list does show 6.12.27 for 32+64 bits, but that's alpha. Don't use v6.12.27. Bug reporting in progress. |
|
4168)
Message boards :
Questions and problems :
Cannot upgrade or run boinc
(Message 37921)
Posted 21 May 2011 by Richard Haselgrove Post: A search for "error 1714" will lead you to the thread Unable to upgrade. |
|
4169)
Message boards :
Questions and problems :
Installation on Windows 2008 R2 Domain Controller
(Message 37836)
Posted 16 May 2011 by Richard Haselgrove Post: Hi All, This is a known, and long-delayed, problem - see trac ticket [trac]#652[/trac]. The simplest workround is install version 5.10.45 (or something else from the v5 sequence), unless you require new hardware support - I would imagine GPU hardware would be rare on domain controllers - or intend to attach to a project like AQUA wich uses other advanced features of the later BOINC versions, like multi-threading. In that case, there are more complicated workrounds available - ask again - but let's keep it simple for now. |
|
4170)
Message boards :
BOINC client :
Need help about client_state.xml
(Message 37835)
Posted 16 May 2011 by Richard Haselgrove Post: In any event, wouldn't it be better practice for your gadget to obtain BOINC's state by GUI RPC (protocol)? You are currently reading a file which is continually being re-written by BOINC. There is a danger that the file may be inaccessible to you during re-writes, or worse that BOINC's operations may fail because you are holding the file open at the critical moment. (And if you use GUI RPC, it is easier to extend the gadget to be network aware, and hence able to monitor remote computers. That would be welcomed by many in the BOINC community.) |
|
4171)
Message boards :
BOINC client :
Need help about client_state.xml
(Message 37827)
Posted 16 May 2011 by Richard Haselgrove Post: I've just looked through a couple of client_state.xml files for multiple-project computers. The only semi colon characters I found were separating the input file names, in the <command_line> of Einstein workunits, and the <stderr_txt> of Einstein results. Where, exactly, is your gadget expecting to find a semi-colon? |
|
4172)
Message boards :
BOINC client :
My Wish List
(Message 37779)
Posted 12 May 2011 by Richard Haselgrove Post: From client configuration, there's a command-line option --exit_when_idle which is close to what you want, though not as 'user friendly' (!) as editing cc_config.xml |
|
4173)
Message boards :
Questions and problems :
BOINC 6.12.26 and CUDA Task Problem
(Message 37762)
Posted 11 May 2011 by Richard Haselgrove Post: There's also 'abort because not started before deadline', which has happened at Einstein with some over-filled caches recently. That's an automatic BOINC (core client) action, although it shows as 'aborted by user' on the website. |
|
4174)
Message boards :
Questions and problems :
Multiple Instances of BOINC Manager
(Message 37731)
Posted 9 May 2011 by Richard Haselgrove Post: The newer BOINC installers put the "run" instruction into the registry. The old BOINC installers used to put a shortcut link in the start menu startup group. [...] Rom explained it once. HKLM gives BOINC a chance to sniff around. If it finds HKCU saying 'no', it goes away again. The downside is: each new user starts with a default presumption of 'yes', and users who don't share the enthusiasm of the person who first installed BOINC have to manually switch the option to 'no'. I suppose they get more cycles that way, but it's debatable.... |
|
4175)
Message boards :
Questions and problems :
Multiple Instances of BOINC Manager
(Message 37728)
Posted 9 May 2011 by Richard Haselgrove Post: Is it possible, that your (or All users') "Start Menu\Programs\Startup" contains a link to BOINC Manager? (BOINC installer likes to create an entry in "HKLM\Software\MS\Windows\CurrVersion\Run", even if the "Allow all users to control BOINC" checkbox switched off.)-: The newer BOINC installers put the "run" instruction into the registry. The old BOINC installers used to put a shortcut link in the start menu startup group (sorry, I can't remember the transition point between 'new' and 'old' in that sentence). But I think all of us who were active upgraders at that point went through the transition with double icons, and had to remove one of them. Since the current (and likely future) standard is the registry startup, it would be best to declare the program menu startup entry redundant, and delete it - that's easier (with a right-click/delete) anyway. |
|
4176)
Message boards :
Questions and problems :
random bsods
(Message 37684)
Posted 3 May 2011 by Richard Haselgrove Post: Does Boinc use .NET? No, it doesn't. |
|
4177)
Message boards :
Questions and problems :
random bsods
(Message 37682)
Posted 3 May 2011 by Richard Haselgrove Post: NDIS.SYS (from your first screenshot) is usually part of a network card driver. The other two are from the (Windows) NT-based O(perating) S(ystem) Kernel. The BOINC program itself doesn't touch any areas like that, and it's highly unlikely that BOINC itself is directly implicated in the crashes - Windows 7 has lots of security features to keep applications like BOINC away from the heart of the operating system and the hardware drivers. More likely, one or other of the science applications supplied by the projects you're attached to is making your computer work harder than it has done before, and the extra stress (and possibly heat) is revealing weaknesses or mis-configurations which didn't matter before. Googling your error messages led me to http://windows7forums.com/blue-screen-death-bsod/55435-occasional-ntoskrnl-bsod.html, which is a good example of the kind of things which can go wrong, and the steps you may need to go through to get things working right again. |
|
4178)
Message boards :
Questions and problems :
random bsods
(Message 37678)
Posted 2 May 2011 by Richard Haselgrove Post: BSODs can also be re-created from the minidump files, using free tools like BlueScreenView. Edit - removed default signature. Could we have [trac]changeset:23466[/trac] on this board, please? |
|
4179)
Message boards :
Projects :
News on Project Outages
(Message 37533)
Posted 19 Apr 2011 by Richard Haselgrove Post: Or perhaps a bit longer.... As posted in the independent forum previously linked (that's an official CPDN website, just independent of the BOINC software that's being upgraded at the moment): The database maintenance that has been ongoing for the last few days is likely to continue to over run until Thursday evening (21 April 2011). |
|
4180)
Message boards :
Questions and problems :
One account and 2 ID's
(Message 37525)
Posted 19 Apr 2011 by Richard Haselgrove Post: Hello, Ask at BOINCstats, or read their forums relating to 'Split CPID'. It's an issue to do with BOINCstats, more than BOINC itself. |
|
4181)
Message boards :
Projects :
News on Project Outages
(Message 37489)
Posted 14 Apr 2011 by Richard Haselgrove Post: SETI@Home Doesn't seem to be as serious as they feared. Work is flowing again, but as usual slowly while the userbase catches up from the outage. Storage Server Issues |
|
4182)
Message boards :
Projects :
News on Project Outages
(Message 37486)
Posted 14 Apr 2011 by Richard Haselgrove Post: SETI@Home SETI seems to have gone offline. I can't load any pages and the Cricket graphs are showing no traffic. I hope something really bad has not happened like last years AC failure. Their workunit storage server has broken down, again: Gowron failed a drive and is hung. It will probably be 3-4 days to re-sync the RAID array, so no work until then. But the rest of the project, including the message boards, is up at the moment, and it still seems to be possible to upload and report completed results. |
|
4183)
Message boards :
GPUs :
Dual gpu problem: 99% utilization hangs two gtx-280s
(Message 37452)
Posted 9 Apr 2011 by Richard Haselgrove Post: lol... excuse me. I was half and half busy with my electricity bill. That's inflation for you. It does things like that - especially to the bill ;-) |
|
4184)
Message boards :
Questions and problems :
2 icons in system tray
(Message 37436)
Posted 8 Apr 2011 by Richard Haselgrove Post: I've come back to Boinc after several years away. Boinc 6.10.60 is installed on 2 of my W7 32bit computers, one a quad core, the other a dual core. On both computers Boinc is running twice in the system tray displaying duplicate information on hover. I can exit one, but both come back after a reboot. Quite likely. The v6.10.xx series of BOINC are, normally, started via an entry in the Windows Registry. Earlier versions of BOINC were started via a shortcut in the 'startup' group in the Windows Program menu. That shortcut is now redundant, but there may be a leftover lying about. For Windows 7, specifically: Click the start button, or press the Window key. Click on 'All Programs' Scroll past the individual program names, and through the folder list to 'S' for Startup. Click on Startup. If you see a BOINC entry there, right-click and delete it. That should be all. |
|
4185)
Message boards :
BOINC Manager :
BoincManager Accessibility
(Message 37420)
Posted 7 Apr 2011 by Richard Haselgrove Post: Hi It might be helpful if you and your friend could review trac ticket [trac]#147[/trac] and see if anything has been missed. Your comment about not being able to read the names of tabs sounds familiar - that must be in Advanced View, right? Reading the latest comments on the ticket, BOINC should already switch automatically to Advanced View when it detects a screen reader. Does that happen with NVDA, or just with JAWS? |
|
4186)
Message boards :
Questions and problems :
A project involving BOINC and multi-agent systems
(Message 37359)
Posted 1 Apr 2011 by Richard Haselgrove Post: It looks to me like a formal, technical, term. Google found |
|
4187)
Message boards :
Questions and problems :
processing in different computer
(Message 37277)
Posted 24 Mar 2011 by Richard Haselgrove Post: i cant locate the Boinc data files, where are they? im on MacOSX See The Big BOINC 6 Answer Thread, from the FAQs linked in Jord's signature. The data should be in /Library/Applications Support/BOINC/ |
|
4188)
Message boards :
Questions and problems :
Pause/Resume Question
(Message 37215)
Posted 19 Mar 2011 by Richard Haselgrove Post: ... Surely the 'current directory' for boinccmd would be the program directory? (boinccmd - though I agree 'same directory as the client' could be worded less ambiguously). In any event, it didn't seem to be necessary when I tried it recently. |
|
4189)
Message boards :
GPUs :
Project GPU requirements
(Message 37194)
Posted 16 Mar 2011 by Richard Haselgrove Post: AQUA has an NVidia/CUDA application - Windows and Linux - but never issues any work for it. It's a ghost application. |
|
4190)
Message boards :
Questions and problems :
how about mobile
(Message 37175)
Posted 14 Mar 2011 by Richard Haselgrove Post: ok. Es gibt zich Betriebssysteme für smartphones, aber warum kein "mobile boinc"? Battery life is the usual answer. |
|
4191)
Message boards :
Questions and problems :
BOINC hanging mid 2011 MacBook Pro 10.6.6
(Message 37130)
Posted 9 Mar 2011 by Richard Haselgrove Post: It looks as if Bernd Machenschalk of Einstein@Home is actively working with NVidia on the problem: http://einstein.phys.uwm.edu/forum_thread.php?id=8761&nowrap=true#110999 |
|
4192)
Message boards :
Questions and problems :
disabling running high priority
(Message 37117)
Posted 8 Mar 2011 by Richard Haselgrove Post: anyway to disable is as its showing up on one of my wu's which seem to have made it work on the wu with the longest deadline and its been working on it for 30mins and 50% complete not exactly going to miss a 8 day deadline Yes - don't choose a large cache setting. Usual advice is "Shortest typical deadline of all attached projects" (so probably 8 days in your case) "divided by number of attached projects in total". |
|
4193)
Message boards :
BOINC Manager :
Disable Notices Popup?
(Message 37085)
Posted 4 Mar 2011 by Richard Haselgrove Post: And be aware that v6.12.16 has a significant change to that reminder dialog - the reminder interval has been re-scaled to hours, and new min/max limits set. From what I understand of the code, the first arrival of a new notice will always trigger as alert: the dialog concerns the frequency of reminders (ie after the first) of unread notices. I'd certainly agree with testing the newest version, rather than one which dates back to late January (and was probably changed for a reason). |
|
4194)
Message boards :
BOINC client :
Use at most % processors when computer is not in use
(Message 36990)
Posted 23 Feb 2011 by Richard Haselgrove Post: Genuinely thought this could be scripted using boinccmd or cc_config.xml by building a launch sequence to apply the changes to number of cores or % of cores permitted, and then on exiting for instance WOW to put it back to 100%, but there is not. It is possible to make multiple global_prefs_override.xmls, copy over and force BOINC to read it in with boinccmd No, no benchmark yesterday when I used 75% of cores for a while with v6.12.15 Just one task preempted/resumed, the others continued unperturbed. |
|
4195)
Message boards :
Questions and problems :
Error: Free space problem: 100Gb free, need 2Gb
(Message 36974)
Posted 22 Feb 2011 by Richard Haselgrove Post: AQUA's scheduler has resumed sending out the smaller Fokker-Planck tasks - they also failed last night, with a message about the application not being permitted by preferences ("No work available for the applications you have chosen"). Edit - and now the message boards are up as well. |
|
4196)
Message boards :
Questions and problems :
Error: Free space problem: 100Gb free, need 2Gb
(Message 36972)
Posted 22 Feb 2011 by Richard Haselgrove Post: Looks as if AQUA is trying to send multiple WUs, sometimes even multiple tasks from the same WU. My sched_reply file has 14 </workunit> tags. Since the ROQS tasks are huge, and the limit should be 2 per host (and I had one already, so should have been limited to 1 this time), that might explain it. The sched_reply also contains <message priority="notice">D-Wave's Redfield Open Quantum Simulation: Multi-Threaded needs 1414.81MB more disk space. You currently have 969.37 MB available and it needs 2384.19 MB.</message> I'll try and email them through, but my ISP can't find Berkeley at the moment. |
|
4197)
Message boards :
Questions and problems :
Error: Free space problem: 100Gb free, need 2Gb
(Message 36970)
Posted 22 Feb 2011 by Richard Haselgrove Post: I went through all the entries from 23007 to 23079, unless one of the preferences changes did it, there hasn't been anything else. Yet I agree with Peter that 23079 is for the client only and so you will see this change in 6.12.16. It shouldn't affect any server. Front page is live, but if I try to navigate from there to the message boards I get an HTTP 500 error. Sounds like the server is well scrambled. |
|
4198)
Message boards :
Questions and problems :
Error: Free space problem: 100Gb free, need 2Gb
(Message 36968)
Posted 22 Feb 2011 by Richard Haselgrove Post: It seems to affect exclusively ROQS tasks... Not many to choose from. They were at least at [23007] beforehand, because that one was written to fix a bug which AQUA themselves encountered and reported - and the fix worked for two weeks, until last night's upgrade. I've got those logs and request/reply now - trying to make sense of them before I try posting to boinc_alpha. |
|
4199)
Message boards :
Questions and problems :
Error: Free space problem: 100Gb free, need 2Gb
(Message 36963)
Posted 22 Feb 2011 by Richard Haselgrove Post: In updating, they also happened to catch changeset [trac]changeset:23079[/trac], which I suspect is implicated in this problem. The reported errors are all of the "message from server" type. I think it's highly unlikely that so many people would both compile their own v6.12.16 clients, and yet not mention it when posting. It seems to affect exclusively ROQS tasks, which are huge - in both memory and time. I should have one finishing in a couple of hours (after 24 CPU-days!), so I'll try and fetch a replacement (AQUA limits to two tasks per host, and I have my spare already). That should give me both some logs, and a matched pair of sched_request and sched_reply files to pore over. |
|
4200)
Message boards :
Questions and problems :
Error: Free space problem: 100Gb free, need 2Gb
(Message 36960)
Posted 22 Feb 2011 by Richard Haselgrove Post: This is a problem which appeared at AQUA immediately after they upgraded their server to correct some user web page errors. In updating, they also happened to catch changeset [trac]changeset:23079[/trac], which I suspect is implicated in this problem. Last I heard (http://aqua.dwavesys.com/forum_thread.php?id=747&nowrap=true#10502), they were trying to revert to an earlier version, but they don't appear to have succeeded. Also, last night I couldn't get any mail through to Berkeley (my ISP didn't have a DNS MX record for them), so the developers may not be aware yet. It may be several more hours before anybody starts working on this problem. |
|
4201)
Message boards :
Questions and problems :
Problem attaching to malariacontrol.net
(Message 36912)
Posted 18 Feb 2011 by Richard Haselgrove Post: ... This might be included in a handy list of "Common Gotchas for new project admins". In the past, I've received emails like You have received a new private message at AQUA@home. where the sender has appeared as Apache <apache@localhost.localdomain> If I were an ISP, I think my spam filter would block that one too! |
|
4202)
Message boards :
Questions and problems :
BOINC requests not needed CPU WUs
(Message 36896)
Posted 17 Feb 2011 by Richard Haselgrove Post: This sounds like a problem which has been niggling Einstein for some little time now - see the thread Scheduler went nuts on the Einstein message board. It would be most helpful to compare detailed client and server logs for your machine, to try and track the problem down. But I can't find a user called "Zarbis" at Eintstein, so I have no way of tracking down your computer, either. Perhaps you could identify yourself in that Einstein discussion, and we can take it from there? |
|
4203)
Message boards :
BOINC client :
Feature request: Make EDF mode Hyper-threading aware
(Message 36884)
Posted 16 Feb 2011 by Richard Haselgrove Post: Oh, detecting whether the CPU can do hyper-threading isn't the problem. The problem is to disable the virtual CPUs. How will any program know which are the real and which are the virtual processors? The OS doesn't even differentiate between the two, it just shows how many CPUs you have. There a problem all right, but this isn't it. The real problem is: how do you distinguish between a four-core CPU with hyperthreading disabled (four logical cores) and a dual-core CPU using hyperthreading (four logical cores). Maybe Jord's EAX=1 test on bit 28 answers that one, making it possible to branch into different execution paths depending on whether HT is active or not. And then what should BOINC do? Disable hyper-threading, a function that is only done in the BIOS? Or at a guess run with a max of 50% of the CPUs? As when it does that, it could well be that in your case of an i7/4/8, there's still 2 real/2 virtual CPUs running, or 1 real/3 virtual, or 3 real/1 virtual, or 4 real/0 virtual or 0 real/4 virtual. Oh, come off it - this is no problem at all. BOINC doesn't employ CPU affinity, and nor - despite what Crunch3r and others keep saying - should it. BOINC just launches 'n' threads of work, and leaves it up to the underlying OS to manage the hardware. It would be perfectly possible, provided the HT detection works as advertised, for BOINC to launch (or preempt) tasks until the number of active threads was 'n/2', in the circumstances described by the OP. That may not get exactly the same benefit as disabling HT entirely in BIOS for the duration, but I bet you it would show some benefit with any OS sensibly paired with an i7. archae86 could probably help us with that one. |
|
4204)
Message boards :
Questions and problems :
Stolen laptop. Possible to track IP-number via Boinc?
(Message 36873)
Posted 15 Feb 2011 by Richard Haselgrove Post: Yes, it can be done, and it has been done using BOINC, more than once - see this news item: BOINC helps recover stolen laptop. Log into your AQUA account, using the machine you're posting from now. Go and look at the list of your computers - see if the laptop has contacted the servers since it was stolen. If it has, click on the 'Details' link in the 'Computer ID' column, then click on 'Show IP address'. You'll probably see two IP addresses: record them both, but it's the "External IP Address" which is most important. Start keeping detailed records. Only you (and the project admins) can see that external IP address. Record it, and alongside it, the "Last time contacted server" - to the second. Be very, very careful to record the values accurately, and to keep the records - you will probably need to pass them on to the police, and the police (depending on the legislation in your local jurisdiction) will probably need to take the list to court as evidence - if you can get somebody independent to witness your recording, and check the records as you make them, so much the better. Then, send a personal message to AQUA users 'Boinc Admin' and 'Neo' - they're the project administrators, very friendly and helpful. Explain your situation, and pass on the advice I've given here. Ask them to contact David Anderson on your behalf, and get some more details of how it was done last time. I'm sure he'll be willing to help. Best of luck - let us know how you get on. |
|
4205)
Message boards :
GPUs :
Overclocking a GPU
(Message 36835)
Posted 12 Feb 2011 by Richard Haselgrove Post: Would BOINC request work from a project for a GPU even if the project was not designed to take advantage of GPUs? Yes, it would request work - at random times, but roughly once per day. It won't ever get any, though. |
|
4206)
Message boards :
Questions and problems :
Problems with BOINC
(Message 36803)
Posted 10 Feb 2011 by Richard Haselgrove Post: This is a known problem - you can read the various reports in trac ticket [trac]#815[/trac], with several possible workrounds. My workround was to use the a registry editor to change all references to the failed drive (F: in your case) in HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup and replace them with references to E:, or any known working drive (don't use your DVD drive letter!) You'll probably find that your experiment with BOINC v5 has fixed INSTALLDIR, but that you still need to edit DATADIR, and possibly MIGRATIONDIR, manually. |
|
4207)
Message boards :
Questions and problems :
Requesting tasks for GPU?
(Message 36749)
Posted 5 Feb 2011 by Richard Haselgrove Post: Requesting GPU tasks is not "instead of" requesting CPU tasks - it is "in addition to" CPU tasks. Separate queues, and separate long term debt values, are maintained for the different processors. If your computer doesn't request CPU tasks, it's because your CPU doesn't need CPU tasks, or because the CPU queue is stalled for some CPU-related reason, not because it is trying to get GPU tasks. Having said that, I don't like the redundant GPU requests to non-GPU projects either, but the developers haven't responded to my representations yet. |
|
4208)
Message boards :
Projects :
News on Project Outages
(Message 36735)
Posted 4 Feb 2011 by Richard Haselgrove Post: SETI web pages and message boards are back up, but it's unlikely any new work can be downloaded until Monday. |
|
4209)
Message boards :
Questions and problems :
Boinc does now not work with Smoothwall Gaurdian Installed......
(Message 36711)
Posted 3 Feb 2011 by Richard Haselgrove Post: SETI hasn't had any communication problems that I've seen since the end of this week's maintenance window, around 16:00 Pacific time Tuesday 01 Feb, 00:00 UTC Weds 02 Feb. Even the proposed lab-wide outage on Wednesday didn't cause any comms loss, so far as I could see. But as always, congestion after maintenance may have meant that individual works requests failed for a while. Jord, surely the 'message from server' and the subsequent reason should arrive in the same sched_reply_...xml from the server: and that must have arrived in its entirety, otherwise the message log would contain 'parse error' or 'incomplete' remarks. And surely, BOINC scheduler RPC generators - both client and server - should automatically put "no cache" directives, or whatever the proxy language equivalent is, in front of scheduler RPCs - otherwise the sequence numbers might get scrambled, and we'd be seeing random CPID changes.... ....oh, sorry, I forgot for a moment. We are seeing random CPID changes, already, aren't we? |
|
4210)
Message boards :
BOINC client :
Seti@Home Beta keeps on getting new GPU tasks
(Message 36704)
Posted 2 Feb 2011 by Richard Haselgrove Post: I would imagine the info is not from the Event Log? I had a look the Job Log in the C:\ProgramData\Boinc folder but that had strange info that made no sense. If you modify cc_config.xml as described, the extra information will appear in Event Log (once you have used 'Read Config File', Advanced view, Advanced menu). Yes, add that line. See client configuration for details. |
|
4211)
Message boards :
Projects :
News on Project Outages
(Message 36579)
Posted 25 Jan 2011 by Richard Haselgrove Post: And they're nearly there. From the forum (Yay! We can read that): AQUA@home : BOINC daemons will be restarted shortly The DB is restored and things seem OK. I'll be starting BOINC's processes shortly. |
|
4212)
Message boards :
Projects :
News on Project Outages
(Message 36565)
Posted 25 Jan 2011 by Richard Haselgrove Post: Aqua remains offline - perhaps a full up drive replacement plus RAID rebuild or restore is going on there. They're on it. Front page now says "The AQUA server is being reconfigured. Please wait." |
|
4213)
Message boards :
Projects :
News on Project Outages
(Message 36538)
Posted 22 Jan 2011 by Richard Haselgrove Post: Advance notice Einstein@Home downtime The project will be taken offline for important database maintenance late on Tuesday Jan 25 (UTC). Estimated downtime is 24h, we'll try to keep it shorter if possible. There is some ambiguity over the timing - it appears that the outage will extend into Wednesday Jan 26. GPUGrid.net The server will be down for some minutes due to power maintenace operations in the building, next Thursday January 27th at 10am (UTC+1). |
|
4214)
Message boards :
Projects :
News on Project Outages
(Message 36536)
Posted 22 Jan 2011 by Richard Haselgrove Post: Aqua went AWOL this morning -- no access to their home page or anything else (about 4 hours ago - 7AM PST). They also posted worries about a possibly degraded hard disk yesterday. |
|
4215)
Message boards :
Projects :
News on Project Outages
(Message 36483)
Posted 19 Jan 2011 by Richard Haselgrove Post: The IP addresses of some of the servers have been changed. We do not need to do anything ourselves... The libcurl patches were applied in changeset [trac]changeset:19774[/trac], 03 December 2009 - so BOINC versions 6.10.22 and above should be immune. Clients earlier than that may need a simple BOINC restart before uploads resume. |
|
4216)
Message boards :
Questions and problems :
Memory
(Message 36437)
Posted 14 Jan 2011 by Richard Haselgrove Post: Which projects are waiting for memory? Different applications have different requirements. |
|
4217)
Message boards :
Questions and problems :
CUDA question.
(Message 36360)
Posted 7 Jan 2011 by Richard Haselgrove Post: I have 256MB of VRAM, but BOINC only reports that I have 244MB. Einstein@home requires 250MB. I know that one way to solve this is to rollback my drivers, but as the drivers that would work are significantly older I would rather not do that. My first question is,is there ANY other way around this without rolling back the drivers, and, why the is there a discrepancy between the VRAM I actually have and what BOINC detects? Thanks. I suspect that BOINC doesn't detect the VRAM size directly, but instead queries the driver (via the Applications Programming Interface), and reports the answers it gets. As to why different drivers report a different RAM size, you'll have to ask NVidia. I can only presume that newer drivers reserve a bit of space for their own private use. So, even if BOINC was able in some way to subvert NVidia's programming, and report the raw underlying hardware memory size, it still wouldn't do any good. Either the memory would still be reserved for other uses, and unavailable for Einstein: or you could try to use it anyway, trample all over somebody else's code/data, and probably end up with a crash, bluescreen, or some other nasty. Also, check reports at Einstein that their app actually uses (maybe only at certain stages of the calculation) significantly more than the 250MB they 'declare' as being required. But AFAIK, those tests haven't yet been rerun for the new version of the application deployed today. You may be in luck with that, but I wouldn't hold my breath. |
|
4218)
Message boards :
Questions and problems :
exclusive_gpu_app perhaps not working?
(Message 36300)
Posted 31 Dec 2010 by Richard Haselgrove Post: If BOINC had stopped the CPU apps (actually issued a 'suspend' instruction), you would see restart/resume messages in the log, same as you're seeing for the GPU apps. I can only assume that - as far as BOINC is concerned - the CPU apps are running normally, and clocking up time. Remember that current BOINC managers display elapsed (wallclock) time, not CPU time. And I guess the apps themselves were fighting for whatever resources (CPU, memory) were left over from from the video editing. Have you checked how heavy the CPU use of that app is? |
|
4219)
Message boards :
The Lounge :
DAMN Cool Science (YouTube comedy series)
(Message 36284)
Posted 30 Dec 2010 by Richard Haselgrove Post: Shamelessly nicked from a BOINC project far, far away. http://www.youtube.com/watch?v=5t7BA9HQJWg This guy is for real. He's a BOINC project scientist, application developer, and many things besides. Really. See if you can guess where. |
|
4220)
Message boards :
Questions and problems :
BOINC cant connect to localhost after Einstien@home added
(Message 36279)
Posted 30 Dec 2010 by Richard Haselgrove Post: It is also to be noted that Einstein tasks have a very long and memory-intensive startup (or restart) phase, during which the BOINC core client is unresponsive. I often see both the 'cannot connect to localhost' warning, and a gap in the time sequence of message log entries, if I try to start the manager too soon after a client restart, and there are Einstein tasks in the current mix. In short, if the problem is only observed at startup, it's nothing to worry about - go and have a cup of tea, and things will be fine when you get back. |
|
4221)
Message boards :
Questions and problems :
Trojan found in Boinc Data
(Message 36185)
Posted 23 Dec 2010 by Richard Haselgrove Post: Since nobody has reported back on this, I've done my own test. VirusTotal test results have been posted in the SETI@home News message board area, where the first warning was posted. Note my comment on methodology - I pasted the download url for the SETI file directly into VirusTotal, so the file was downloaded directly from Berkeley to VirusTotal with no risk of contamination or modification on my machine. |
|
4222)
Message boards :
BOINC Manager :
Please Remember Window Information Between Sessions
(Message 36150)
Posted 19 Dec 2010 by Richard Haselgrove Post: With 6.12.8 it will remember screen position and approximate size it's in, just not if it was maximized or not. BOINC Manager never starts up maximized as then it could take up focus, placing it over other programs you want to track (upon Windows start up, for instance). Just tested on Windows 7, single video card, dual monitors. (It's a dual-DVI NVidia cuda card, with 2 x 1600x1200 monitors - primary to the left, secondary to the right). When I drag both BOINC Manager windows (main tabbed display, and subsidiary Event Log window) to the right-hand (secondary) monitor, and then exit the Manager: On reopen, the windows both open with their previous size, on the primary monitor, but as far to the right (as close to the secondary monitor) as possible. I remember wrestling with this problem soon after the release of Windows 98, the first to support multiple monitors. It's extremely difficult to program to cover all eventualities (for example, on my current system, the driver drops the resolution of both monitors to 1024x768 whenever I use the DVI-KVM to switch the primary monitor to a different host: should BOINC know about that?) The present balance (I'm testing with v6.12.8) is probably as reasonable as we're likely to get. Unless any programmers out there can contribute better (cross-platform) code? |
|
4223)
Message boards :
Questions and problems :
Boinc doesn't switch between tasks
(Message 36128)
Posted 18 Dec 2010 by Richard Haselgrove Post: It would help if the OP (and all others asking for help with similar problems) could name the specific projects they are having problems with. For example, as well as the checkpointing problem, task switching is known to suffer from problems when the AQUA multi-threaded applications are included in the mixture. If AQUA is included, v6.10.56 might be a better choice than v6.10.58 |
|
4224)
Message boards :
Questions and problems :
Boinc doesn't respond to mouse movement
(Message 36059)
Posted 13 Dec 2010 by Richard Haselgrove Post: Well, whatever host 2675774 is doing, it isn't Einstein - no tasks listed. NB, if you look at that page while logged in under your "smartcooky" account, you'll see a lot more detail - like IP addresses - than we can. But at first glance, it doesn't sound like a networking, or even a BOINC-related, issue. I'd do a thorough check on that machine, and make sure it isn't running any malware. Even a single-core Celeron shouldn't thrash like that when you reactivate it. |
|
4225)
Message boards :
Questions and problems :
Boinc doesn't respond to mouse movement
(Message 36053)
Posted 12 Dec 2010 by Richard Haselgrove Post: Won't be SETI. I don't know Rosetta's characteristics, someone else will have to chip in. Einstein is usually well-behaved while running, but has a very heavy memory footprint when starting a task (especially the Global Correlations S5 search #1 v3.02). There's only one Ian Cook active at Einstein, and he only has one computer, running Windows XP: is this yours? If so, you haven't returned a task in over a month. If the tasks are starting, but failing to get past the initial 'high memory' phase, then you have a bigger problem than just a missing pop-up. Check whether an anti-virus program might be blocking the application, or whether it's having difficulty reading files off the disk (run chkdsk or similar). |
|
4226)
Message boards :
Questions and problems :
Boinc doesn't respond to mouse movement
(Message 36050)
Posted 12 Dec 2010 by Richard Haselgrove Post: That rather sounds as if the keyword in your opening post is "immediately": BOINC is shutting down the running science application(s), but just not as quickly as you would like. In turn, that sounds as if you're attached to a project which uses a very large amount of RAM, relative to the amount of free RAM available in that particular machine. Which project, which application (some projects have multiple applications for different work types), and how much RAM do you have in the machine? I wouldn't even attempt to run WinXP in under 512MB these days, preferably 1GB minimum with BOINC on top - and more for heavy-duty projects. |
|
4227)
Message boards :
Questions and problems :
Boinc doesn't respond to mouse movement
(Message 36044)
Posted 12 Dec 2010 by Richard Haselgrove Post: Owing to an unfortunate naming choice by the BOINC developers, you have fallen into a common misconception. The icon you see in the system tray on Windows machines actually belongs to BOINC Manager. As the name Manager implies, this is a command-and-control module, not a program which does any actual work. It is also the module responsible for handling communications with you, the user, which is why you see bubble notifications coming from it. boinctray.exe, despite the name, has no connection with the system tray. It is a small helper application, which runs entirely silently and without an icon - indeed, it has no user interface at all. The main work is done by boinc.exe, and by the science applications it controls. It is entirely possible to run boinc.exe, and to be doing useful science, without ever seeing BOINC Manager or its icon after the initial setup. So, having got that confusion out of the way, what exactly is the problem? Are you sure that BOINC's science applications continue to run, using 100% (or whatever) CPU time, even after you touch the keyboard/mouse/screen? Open Windows Task Manager, and watch what CPU %age is shown against the System Idle Process. Or could it be that the major concern is the lack of reassurance you're getting, because of the absence of the popup window? |
|
4228)
Message boards :
Questions and problems :
domain controllers allowed ??
(Message 36042)
Posted 12 Dec 2010 by Richard Haselgrove Post: It was not lifted. See also the wiki release notes. Also see the revision notes on trac ticket [trac]#652[/trac]. |
|
4229)
Message boards :
Projects :
Seti Server(s) Unavailable?
(Message 36001)
Posted 8 Dec 2010 by Richard Haselgrove Post: Front page news at SETI@home: We've been offline for over a month to remodel our database servers. The bulk of this work is complete, and work is being distributed. Please note we may continue to bring servers up/down over the next few weeks (outside of regularly planned outages) as we wrap up the final stages of this major endeavor. |
|
4230)
Message boards :
Questions and problems :
Win 7 x64 Installation Problems
(Message 36000)
Posted 8 Dec 2010 by Richard Haselgrove Post: And are you certain about the operating system? That's the same error message you'd get if installing on a Domain Controller, such as Windows Small Business Server 2008. |
|
4231)
Message boards :
Questions and problems :
Bit of a problem
(Message 35970)
Posted 4 Dec 2010 by Richard Haselgrove Post: This has probably been asked before, but the search function was unable to find it. I have moved my 'Documents' folder to another hard drive as I had run out of space, now Boinc cannot connect to my projects. How do I point Boinc at the new location? Thanks in advance to anyone who can help. Not enough information - need to know platform and version. Making the (big) assumption that you are running some version of Windows, and something from the BOINC version 6 line, the easiest thing is to edit the registry. Find [HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup] and edit the value for DATADIR to point to the new location of your folder. You may find it helpful to refer to The Big BOINC 6 Answer Thread for the common default locations under different operating system versions. |
|
4232)
Message boards :
BOINC client :
AQUA won't run with v6.12.5
(Message 35968)
Posted 4 Dec 2010 by Richard Haselgrove Post: LOL :-) |
|
4233)
Message boards :
BOINC client :
AQUA won't run with v6.12.5
(Message 35964)
Posted 4 Dec 2010 by Richard Haselgrove Post: Are you following me ... :P No, I've been here for years, and at AQUA almost as long ;-) But posting the same question on different boards doesn't always get you a more palatable response each time you try it :P |
|
4234)
Message boards :
BOINC client :
AQUA won't run with v6.12.5
(Message 35961)
Posted 4 Dec 2010 by Richard Haselgrove Post: Don't know if this has been reported or not but AQUA won't run on my Win XP Pro 64-Bit Box's with v6.12.5 installed, I went back to v6.10.58 and the Wu's ran then ... Yes, there was a fix in v6.12.6 (12 November 2010), and a further fix in v6.12.8 (yesterday). If you're going to run test/development versions of BOINC, it pays to read the messages here and on the various mailing lists. Oh, and to report any bugs that you find. |
|
4235)
Message boards :
Questions and problems :
Help in creating custom installer
(Message 35943)
Posted 2 Dec 2010 by Richard Haselgrove Post: When I reminded myself of the DC problem at Einstein recently (a side-effect is that the only recognised way to run BOINC on a small business server is to stick at v5.10.xx, which doesn't recognise the plan_class approach to optimised apps), I was referred to http://boinc.thesonntags.com/collatz/forum_thread.php?id=455&nowrap=true#7137. That seems to suggest that there is no actual problem with running BOINC v6 on a DC (though I don't recommend that anyone else adopt Crunch3r's cavalier approach to security), but that all the problems reside in the Berkeley BOINC installer. |
|
4236)
Message boards :
Questions and problems :
GPU found under cmdline, but not when runned as service.
(Message 35883)
Posted 26 Nov 2010 by Richard Haselgrove Post: 26-11-2010 19:53:52 NVIDIA GPU 0: GeForce GTX 460 (driver version 26099, CUDA version 3020, compute capability 2.1, 993MB, 363 GFLOPS peak) Seems to have been, though that's not how I would have read the documentation. Either or both of us ought to check how that option works on single-GPU machines with the latest (v6.12.6) beta client. someone said to me to use this: The question is, which "someone" told you to do such a thing, and what were they trying to achieve when they gave you that "advice"? "Use all..." is only indicated when you have more than one GPU and they have dissimilar capabilities. |
|
4237)
Message boards :
Questions and problems :
GPU found under cmdline, but not when runned as service.
(Message 35881)
Posted 26 Nov 2010 by Richard Haselgrove Post: Uninstall BOINC using "add/remove programs" (your account data, work in progress, stats data etc. will be left in place), and then re-install it but de-select 'Protected Application Execution' on the advanced options page. That should cure it. Odd - should be OK. You have non-standard files for both cc_config.xml and global_prefs_override.xml - just to be on the safe side, I'd review the contents of both of them, or even temporarily disable them by renaming or moving them. To re-test a changed cc_config.xml for GPU detection, you'll have to restart the BOINC core client, not just re-read the config file - this is one change you can't do on the fly. |
|
4238)
Message boards :
Questions and problems :
GPU found under cmdline, but not when runned as service.
(Message 35879)
Posted 26 Nov 2010 by Richard Haselgrove Post: Started Boinc Manager on the normal way through the Desktop Icon Those three lines, taken together, are your problem. When "running as a daemon" (aka service mode, Protected Application Execution), BOINC runs under its own account, which doesn't have (and can't be given, under Vista or Windows 7) the necessary rights to access the NVidia video driver. In your second log, BOINC is running under your login - you have the right to access the driver, and so does BOINC. Uninstall BOINC using "add/remove programs" (your account data, work in progress, stats data etc. will be left in place), and then re-install it but de-select 'Protected Application Execution' on the advanced options page. That should cure it. |
|
4239)
Message boards :
Questions and problems :
Remove the client without losing any work.
(Message 35849)
Posted 23 Nov 2010 by Richard Haselgrove Post: Wish there was an option to say "report all work upon client exit" Think many part time crunchers would be happy with that too... particular those that move around with a laptop. There's a lot of relevant stuff in Client configuration. Option: <report_results_immediately> Command line: --exit_when_idle |
|
4240)
Message boards :
Questions and problems :
Under this scenario, BOINC won't get work
(Message 35793)
Posted 18 Nov 2010 by Richard Haselgrove Post: So I think you are saying that there is some sort of limit applied to all tasks from a given project. If that is correct, how is that limit computed? That isn't what I was saying at all. True, there are mechanisms available to BOINC project administrators to limit the number of jobs you receive at any one time, or the number of tasks you can hold in reserve or in progress on your machine. But those would all imply that your machine was actively asking the project server for more work, and being refused because of one or other of those limits. Those refusals would show up in your message log, and I see no evidence from what you've written so far that that is the case. Instead, you've told us that you suspended two primegrid tasks. While those two tasks are suspended, you will find that BOINC doesn't request any new work from the primegrid project. And if you don't request work, you'll never receive work - work is never forced on you, by any project. Perhaps your question would be better directed to the Primegrid administrator, and the way he designs his strategy to get the short tasks returned quickly when there is long-running work already in progress. He or she will have a better idea of the relative need the project has for work of each type. |
|
4241)
Message boards :
Questions and problems :
Under this scenario, BOINC won't get work
(Message 35789)
Posted 18 Nov 2010 by Richard Haselgrove Post: The distinction is that Jim is considering the two sets of tasks, short and long, as being distinct: and feels that by suspending one, he should be able to get more of the other. But BOINC sees just one set of tasks, Primegrid, and makes no distinction. By suspending one, BOINC, in its search for another, is forced to look further afield, in this case to Yoyo. The debate between "multiple projects" and "multiple applications within one project" takes us deep into the design fundamentals of BOINC. |
|
4242)
Message boards :
Projects :
News on Project Outages
(Message 35738)
Posted 15 Nov 2010 by Richard Haselgrove Post: Has SETI been having problems I havent been able to download any data since Nov 6, my system has been checked and is working fine just keeps saying Message from Server: Project has no tasks available" Information is posted on the project home page, and the dedicated thread in the 'questions and problems' area. |
|
4243)
Message boards :
BOINC client :
6.12.5 (Alpha) - Breaks GPU Pause settings?
(Message 35675)
Posted 10 Nov 2010 by Richard Haselgrove Post: There's also a report at GPUGrid of a breakage: It's a disaster here. All 5 machines that I installed 6.12.5 on quit running a full complement of CPU programs and refused to run AQUA at all. Reverted to 6.12.4 and they all started behaving properly again except of course for the GPU cache shrinkage bug. User later reports running FreeHAL [nci], which I can't reproduce at the moment, but might be a clue. Too late for me to work on tonight, but I'll have another look in the morning. Reference: http://www.gpugrid.net/forum_thread.php?id=2231#19393 |
|
4244)
Message boards :
BOINC Manager :
installing and connecting problems
(Message 35653)
Posted 8 Nov 2010 by Richard Haselgrove Post: Message 35647 - Posted 8 Nov 2010 21:46:10 UTC I did wait until after you had posted ;-) Somebody has to run the first end-to-end check that the downloaded file is working - and it may as well be me. At least one release failed that test. |
|
4245)
Message boards :
BOINC Manager :
installing and connecting problems
(Message 35648)
Posted 8 Nov 2010 by Richard Haselgrove Post: I'm in the process of Updating them all to v6.12.4, that should do me until the 6.15.xx comes out ... lol v6.12.5 is out, working, and an improvement on v6.12.4 - until I can find the next bug ;-) |
|
4246)
Message boards :
Projects :
News on Project Outages
(Message 35589)
Posted 2 Nov 2010 by Richard Haselgrove Post: Seti seems to be having difficulties again. Servers are disabled/not running again. Nothing posted to SETI's front page yet. looks like it happened around one half hour ago from the router graph page. It's Tuesday. Weekly maintenance day. |
|
4247)
Message boards :
Questions and problems :
Static remote gui connection
(Message 35578)
Posted 1 Nov 2010 by Richard Haselgrove Post: Is there some XML file The next version of BOINC (currently testing as v6.12.4) will allow you to start the Manager with the hostname, username, and password as parameters on the command line to achieve what you want. In Windows, I would do that by fiddling with the properties of a shortcut file: you want to use the Mac as the viewer, if I'm reading you right. I'm not sure how, or even whether, it's possible to fine-tune whatever your equivalent or a 'shortcut' is, but it might be worth testing - if you don't mind testing the rest of the manager while you're at it! You wouldn't need to upgrade to 'worker' machine to v6.12.4 - just the 'viewer', to get access to the new manager. |
|
4248)
Message boards :
BOINC Manager :
installing and connecting problems
(Message 35556)
Posted 30 Oct 2010 by Richard Haselgrove Post: I had already said there was nothing there & that I had reinstalled BOINC several times & that I had done a Repair install of the OS already so it stands to reason that there would be either no stderrdae.txt or there would be nothing in it if it can't connect at all. Thanks for the help though anyway ... False inference, I'm sorry to say. stderrdae.txt, and its close cousin stdoutdae.txt, are written by the core client, boinc.exe The client can be, and frequently is, running and logging messages into those files, irrespective of whether a BOINC manager or other monitoring tool is connected to display the same messages on screen. As has been said already in this thread, you need to distinguish between the case where the core client fails to run, or starts and immediately crashes (in which case the log files are your friend), versus the case where the client is running properly, but the manager fails to connect to it (which is where the discussion about firewalls should start). |
|
4249)
Message boards :
Questions and problems :
BM downloads new task, but doesn't show it under "Tasks" :(
(Message 35522)
Posted 29 Oct 2010 by Richard Haselgrove Post: Try clicking the new 'Show All Tasks' button. |
|
4250)
Message boards :
Questions and problems :
[Gentoo] Boinc stop the boot
(Message 35501)
Posted 29 Oct 2010 by Richard Haselgrove Post: Jord, according to the Wiki the <start_delay> means Specify a number of seconds to delay running applications after client startup I don't think it'll have any effect on the startup for BOINC itself - in fact, thinking about it, BOINC will have to be running in order to read cc_config.xml |
|
4251)
Message boards :
Projects :
News on Project Outages
(Message 35482)
Posted 28 Oct 2010 by Richard Haselgrove Post: Quantum Fire Alpha is down for a day or two due to a power cut at Cavendish Lab. QuantumFIRE Alpha have indicated that they are planning an extended outage (several weeks), while they plan and re-equip for an enlarged QuantumFIRE Beta. The plan was to keep the existing server running for as long as possible to allow work in progress to be returned. QFA BOINC services are down again, and the website is inaccessible. Does anyone know whether this is the final Alpha shutdown, or just another random glitch? I've still got a few tasks that they're welcome to if they'd like them. |
|
4252)
Message boards :
Questions and problems :
1 Project at a time.
(Message 35479)
Posted 28 Oct 2010 by Richard Haselgrove Post: Sandman, I'm only going to say this once, even though you seem to be intent on repeating your assertion in every thread here. BOINC v6.10.58 runs on all available cores just fine. My two remaining installations (4 cores and 8 cores) are running 4 and 8 tasks, respectively, as I type. It's always possible for you to throttle it back to a single task through unplanned, or incompletely understood, preference settings, but that's something under your, the user's, control. It can be reversed, if you can be humble enough to ask for, and accept, help. But I'm afraid that if you continue to post untrue statements on these boards, you may find that the willingness of your fellow volunteers to help you resolve whatever problem you have starts to diminish. |
|
4253)
Message boards :
Questions and problems :
BOINCstats
(Message 35405)
Posted 24 Oct 2010 by Richard Haselgrove Post: I can follow both links with IE8, both the Windows XP and the Windows 7 versions (and I've never had any problems reaching my own stats witheir either version). I can only surmise that the OP's Windows 7 IE installation has something unusual, perhaps a security setting or an ad blocker, getting in the way. I've also occasionally seen third-party toolbars with pop-up blockers cause unexpected problems. |
|
4254)
Message boards :
Projects :
News on Project Outages
(Message 35400)
Posted 24 Oct 2010 by Richard Haselgrove Post: SETI@home _ we are back. Well, the message boards are back, but SETI isn't dishing out or accepting any science yet. |
|
4255)
Message boards :
Questions and problems :
BOINCstats
(Message 35342)
Posted 21 Oct 2010 by Richard Haselgrove Post: I always use http://BOINCstats ... That needs to be http://boincstats.com/ |
|
4256)
Message boards :
Projects :
News on Project Outages
(Message 35294)
Posted 19 Oct 2010 by Richard Haselgrove Post: GPUGrid.net It started with Storage server problem Now visible again on their front page. Some functionality is back, but their database server only seems to be partially working (can see account details, but not message boards). |
|
4257)
Message boards :
Questions and problems :
How to Transfer BOINC to Another Mac
(Message 35286)
Posted 18 Oct 2010 by Richard Haselgrove Post: ...once you attach your new Mac to all the projects you were busy with on the old one, you'll get your credits back. Provided you attach to your existing account, and don't create a new one. Make sure you use exactly the same email address as you used last time when you attach. |
|
4258)
Message boards :
Questions and problems :
Unable to intsall BOINC
(Message 35235)
Posted 15 Oct 2010 by Richard Haselgrove Post: Error 1327 is slightly different from the common .msi woes. There are a couple of workrounds in the (open) bug report: trac ticket [trac]#815[/trac], but unfortunately both involve editing the registry - either HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\User Shell Folders or HKEY_LOCAL_MACHINE\SOFTWARE\Space Sciences Laboratory, U.C. Berkeley\BOINC Setup The second one worked for me: change any references to M: in INSTALLDIR and DATADIR to C:, or some other valid local drive - that should give you control over the installer again, and you can make your own choices on the 'advanced' screen as normal. |
|
4259)
Message boards :
Questions and problems :
GPU trouble (8800 GT)
(Message 34688)
Posted 15 Sep 2010 by Richard Haselgrove Post: Currently, I am testing BOINC on just one computer of each configuration. BOINC is not recognizing the 8800GT (I am running AQUA as my project). Also, be aware that at present, AQUA is not making any work available for GPU processing - their multi-threaded CPU application handles the kind of work they need processed far more efficiently than GPUs do. (See the posts by the project administrators on the AQUA message boards for fuller details). The only reason the CUDA app is loaded on their servers (and hence AQUA is listed as a CUDA project by BOINC) is to make the "Use NVIDIA GPU" preference visible on their website, so users can switch off the futile work requests..... |
|
4260)
Message boards :
BOINC client :
Fermis shader count - how to
(Message 33782)
Posted 15 Jul 2010 by Richard Haselgrove Post: The number of shaders are being reported incorrectly for Fermi cards by Boinc. We got that one sorted out back in March: message 31859. So everything from BOINC v6.10.45 onwards should be reporting correctly. The snip you posted looks like a GPUGrid stderr_out snip: and it was GPUGrid that first alerted BOINC to the mis-reporting problem (as you can see from the thread I've linked). You can also see that you were an active participant in that earlier thread. So I'm not sure what the new problem is, exactly. |
|
4261)
Message boards :
BOINC Manager :
My Wish List - part 3.
(Message 33770)
Posted 14 Jul 2010 by Richard Haselgrove Post: My wish (list): You could also look at Automatic Temperature regulation on this very board. |
|
4262)
Message boards :
BOINC Manager :
My Wish List - part 3.
(Message 33731)
Posted 10 Jul 2010 by Richard Haselgrove Post: The project wide back-off on uploads and downloads will be increased though. At this time it has a maximum of 4 hours, while (each backoff is multiplied by a random number 0..1, so the max average of 2 hours. Which means that at the end of a planned outage, when work-supply issues are being managed at the server (such as is currently implemented at SETI), free-running clients will be idle for up to an extra 12 hours because there's no way to give them a "you can come back in now" signal. They should talk to Ned Ludd about his idea of using DNS as an out-of-band signalling mechanism. |
|
4263)
Message boards :
Questions and problems :
Requesting new tasks
(Message 33730)
Posted 10 Jul 2010 by Richard Haselgrove Post: My PC does not have an NVIDIA card and I set the option "never use GPU". However, the message log tells me that the BOINC manager continues to request work for the GPU and - of course - always gets the answers "scheduler request completed: got 0 new tasks" and "message from server: no work sent". Well, it's plainly silly and wasteful, but it's not actually a BOINC Manager bug. If you are attached to a project that has new enough server code, and which has a ATI application available to send out, you'll see an option in your account's project preferences web page: Use ATI GPU Enforced by version 6.10+ Setting that to 'no' does work - the GPU work requests from BOINC will be inhibited. But if the project hasn't got a ATI application, there's no way you can stop BOINC asking for work for the non-existing application for an ATI card - crazy. It got so bad at AQUA that they re-installed a dummy CUDA application (but made no work for it) just so people could turn off the work requests. It would be simple to make the server send out the 'don't ask' signal if there was no application installed, but they won't do it - I've tried suggesting it a couple of times now. |
|
4264)
Message boards :
BOINC Manager :
My Wish List - part 3.
(Message 33729)
Posted 10 Jul 2010 by Richard Haselgrove Post: I was rather thinking of a network/no network button, like NNT. Hello Carola - fancy meeting you here! As it happens, this has already been worked on - see trac [trac]#139[/trac]. Milo (CPDN admin) was running out of disk space, and needed to specify / finance / order / take delivery / install / configure a new server (and you know how long that takes!) before he could accept another upload. At the time, the upload delay limit was just 14 days, after which the science was just thrown down the tubes by BOINC. So Thyme Lawn wrote all the code needed, and I tested it for him prior to submission - on the 6.6 codebase, so it works well enough with CUDA - not fully updated for ATI, but IIRC that doesn't matter for you. I'll still have the test downloads available somewhere, if you're interested - I'm sure TL wouldn't mind. But, as usual, even with fully developed code patches, it still got a "not invented here" response from David - he just upped the time limit, as he has again this time. |
|
4265)
Message boards :
Questions and problems :
BOINC CPU Allocation Big Problem
(Message 33579)
Posted 28 Jun 2010 by Richard Haselgrove Post: SETI (and of course SETI Beta) was the high-demand project I was thinking of. In the 'status' column of BOINC Manager (Advanced view, Tasks tab), for a GPU task it should show "Running - x.xx CPUs and 1.00 GPU", or words to that effect. What's the x.xx? Sorry, inadequately phrased question. I know what it's for: I was asking what value the BOINC server had calculated as appropriate on your particular system (it's not a fixed value, but calculated dynamically from, e.g., the relative speeds reported by your host for its CPU and GPU). Edit - and before we get into another misunderstanding, could you confirm whether you run the stock application, as issued by the SETI project from Berkeley: or do you install applications yourself, using the 'anonymous platform' mechanism and an app_info.xml file? In the first case, the x.xx I'm after will be calculated and issued by the server: in the second, BOINC will simply read back the value you yourself have put into the file. Which is less meaningful, but easier to cope with. |
|
4266)
Message boards :
Questions and problems :
BOINC CPU Allocation Big Problem
(Message 33577)
Posted 28 Jun 2010 by Richard Haselgrove Post: OK. I give the upgrade a try first and see if that solves it. For the record, the Fermi apps I'm running are Seti@home, Seti@home beta, Collatz conjecture, and GPU Grid. Although, GPU Grid is currently dry so you can count that one out of this scenario. SETI (and of course SETI Beta) was the high-demand project I was thinking of. In the 'status' column of BOINC Manager (Advanced view, Tasks tab), for a GPU task it should show "Running - x.xx CPUs and 1.00 GPU", or words to that effect. What's the x.xx? (may have to scroll / expand column to read it) |
|
4267)
Message boards :
Questions and problems :
BOINC CPU Allocation Big Problem
(Message 33574)
Posted 28 Jun 2010 by Richard Haselgrove Post: BOINC v6.10.18 is a bit elderly for AQUA - there was an AQUA-specific bug-fix in v6.10.19 - and another in v6.10.57, one beyond the currently-recommended v6.10.56 Not worth hunting out .57 - different problem fixed - but .56 would be worth installing. There may be an issue here with the somewhat extreme combination of 2 x GTX 480, AQUA, and Vista - I doubt that the developers can afford such a rig to test on. Depending what project you run on the Fermis (you didn't answer that question): I'm seeing that some projects declare a high CPU requirement for Fermi support, especially with Vista/Win7. If the project application configuration ends up deciding that a GPU job requires more than 0.5 CPUs, then two of them are going to take you above 1 CPU reserved. Under those conditions, BOINC would run 7 single-threaded CPU jobs: I don't know how it would cope with an 8-CPU call from AQUA. I think the best way to proceed would be to start with an upgrade to v6.10.56, and observe how that behaves. If there's still a problem, I think you've just volunteered yourself as guinea-pig in chief: we may have to call on you to enable some debug logging, and post the resulting messages, to aid diagnosis. Is that OK? But do the simple upgrade before anything else. |
|
4268)
Message boards :
Projects :
News on project outages
(Message 33548)
Posted 27 Jun 2010 by Richard Haselgrove Post: GPUGrid fell off the web last night, and hasn't reappeared today - on previous form, I doubt it'll restart until tomorrow morning. Refugees have started to appear at other projects that BOINC flags as NVidia-capable. |
|
4269)
Message boards :
Questions and problems :
Cores vs Threads ? (hyperthread matter?)
(Message 33481)
Posted 21 Jun 2010 by Richard Haselgrove Post: I think Mo meant to draw a distinction with 'multithreading': the distinction is probably between 8 separate applications, running on 8 different tasks with eight different datasets: or a single application, spawning eight threads, all working on different aspects of the same task, and only accessing one dataset. Hopefully the latter case would suffer from far less memory bus contention. Those CPDN HadAM3P jobs have a heavy memory demand at the best of times: I've got one at the moment which is holding almost 220 MB in RAM even while 'waiting to run'. |
|
4270)
Message boards :
Questions and problems :
Cores vs Threads ? (hyperthread matter?)
(Message 33447)
Posted 18 Jun 2010 by Richard Haselgrove Post: Thanks. IIRC, some of the earlier NetBurst experiments showed much better results with dis-similar tasks - one SETI with one Einstein was a favourite pairing. I know Mark is an avowedly one-project cruncher (SETI first, the rest nowhere) - and shortly to become a no-project cruncher, such is his disgust at the cack-handed way the latest BOINC server updates have been rolled out - so presumably mixed project pairing wasn't part of your tests with him. We've also lost Tony (mmciastro), who used to do similar testing with mainly AMD processors (if I may be permitted to use those letters in this company!) So I wonder if the opening question has yet been answered with iN technology and diverse projects? (posting from a Williamette, as it happens) |
|
4271)
Message boards :
Questions and problems :
Cores vs Threads ? (hyperthread matter?)
(Message 33413)
Posted 16 Jun 2010 by Richard Haselgrove Post: The answer is highly application dependent, and also will vary with the particular CPU implementation. Peter, Would it be fair to say that those earlier comparisons were done on NetBurst-era HT processors? Have you had any chance to repeat them on the Core iN range, or do you know anyone else who has? |
|
4272)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 33391)
Posted 16 Jun 2010 by Richard Haselgrove Post: Happened again at the beginning of this month. Since the last show, I've upgraded the BOINC version (to v6.10.34), which means I can't (so easily) re-run the "runs in user mode / gets stuck in service mode" test I did with BOINC v5. Still gets stuck in advapi32.dll: 
|
|
4273)
Message boards :
Questions and problems :
Not reporting or requesting tasks
(Message 33364)
Posted 14 Jun 2010 by Richard Haselgrove Post: Why not ask this on the Climate Prediction message board? |
|
4274)
Message boards :
BOINC Manager :
Trouble accessing remote computers
(Message 33362)
Posted 14 Jun 2010 by Richard Haselgrove Post: No. Every computer will certainly have a password (and as Gundolf says, if you can't see a gui_rpc_auth.cfg file, you're looking in the wrong folder). But they don't have to be the same. All that's required is that you (the user) knows the password for the remote computer (the one you're trying to control), and you can type, paste or otherwise enter it into the manager dialog on the computer you're using to do the controlling. If you failed to find the right gui_rpc_auth.cfg file (and the fact that you got a bad password messge suggests that is the case), then you probably put the remote_hosts.cfg file in the wrong folder too. |
|
4275)
Message boards :
API :
determining the number of cpus available to bionc client & scheduling question
(Message 33346)
Posted 13 Jun 2010 by Richard Haselgrove Post: thanks! would either of you happen to have an example for the mt plan class, namely what needs to be done for xadd and update_versions to recognize the mt plan class mentioned on the wiki? an example of the file structure would also help for the apps directory The best idea would be to ask the devs (Neo and Boinc_Admin) at the AQUA project - they've been using it live for about a year now, and have helped to get a lot of the bugs out of the BOINC [mt] system. |
|
4276)
Message boards :
Questions and problems :
BOINC Linux uses wrong browser
(Message 33335)
Posted 11 Jun 2010 by Richard Haselgrove Post: Sorry, my mistake. Bad copy'n'paste. |
|
4277)
Message boards :
BOINC client :
Boinc Port Number
(Message 33326)
Posted 10 Jun 2010 by Richard Haselgrove Post: Yes, 31416 is the default port for manager/core client communications, though it can be overridden. |
|
4278)
Message boards :
Questions and problems :
BOINC Linux uses wrong browser
(Message 33322)
Posted 9 Jun 2010 by Richard Haselgrove Post: FAQ: How to set up your default browser in Linux? |
|
4279)
Message boards :
Questions and problems :
flops estimate for different app plans
(Message 33234)
Posted 2 Jun 2010 by Richard Haselgrove Post: Question is: do separate time estimations for different app_versions implemented already and in what BOINC version if yes ? They always were. Re-read the very first post ever written about GPU flop entries: http://setiweb.ssl.berkeley.edu/beta/forum_thread.php?id=1536&nowrap=true#36887 Almost 15 months ago. It says your machine's floating point benchmark, multiplied .... your graphics card's CUDA FLOP count, multiplied .... Edit - in case this is the point of confusion: the values to put in app_info.xml are speed estimates, which relate to your hardware: not counter values relating to the task size. |
|
4280)
Message boards :
Questions and problems :
flops estimate for different app plans
(Message 33231)
Posted 2 Jun 2010 by Richard Haselgrove Post: Question is: do separate time estimations for different app_versions implemented already and in what BOINC version if yes ? They're implemented on the server only - do keep up! And only on SETI Beta so far. |
|
4281)
Message boards :
Questions and problems :
Why not asking for GPU work (look posted log)
(Message 33222)
Posted 2 Jun 2010 by Richard Haselgrove Post: Enable logging!!!! |
|
4282)
Message boards :
Questions and problems :
Why not asking for GPU work (look posted log)
(Message 33203)
Posted 2 Jun 2010 by Richard Haselgrove Post: Probably overworked (excess cuda_long_term_debt). But unless you enable [work_fetch_debug] logging, or at least look at the properties page for the project, you'll never know. |
|
4283)
Message boards :
Projects :
News on project outages
(Message 33181)
Posted 1 Jun 2010 by Richard Haselgrove Post: http://boinc.netsoft-online.com seems to be down. That's the site which provides the cross-project stats listing of other projects on your account page, on every BOINC project. It's causing account pages to load extremely slowly, even on projects which aren't having any problems of their own. |
|
4284)
Message boards :
Questions and problems :
BOINC 6.10.58 released to public
(Message 33117)
Posted 29 May 2010 by Richard Haselgrove Post: Full 5 days worth of work. Is five days your CI or AW setting? For a backup project, shouldn't we ask for it to be CI only? In fact, I thought that was what David said when he checked in the change. I'll have to go back through the notes. Edit - [trac]changeset:20286[/trac]: no, he didn't. Memory playing tricks again. |
|
4285)
Message boards :
Questions and problems :
Manually creating a boinc service
(Message 33087)
Posted 27 May 2010 by Richard Haselgrove Post: On my service installation, created by the installer under Windows XP on a non-domain home computer, the service executable line is "C:\Program Files\BOINC\boinc.exe" -daemon It also has a dependency on the Remote Procedure Call (RPC) service, it's running under account .\boinc_master, and I know it has to find necessary data files from registry settings. Your best bet would be to run the installer on your workstation so you can examine all of those, and then run setup with the traditional /a administrator's switch (it's implemented) to get a clean folder layout on a network share for your xcopy. But before you do, I'm worried that you appear to want to hide the fact that you're running BOINC from your work colleagues, and your work colleagues don't appear to want to run it. Doesn't sound like a happy ship to me. Are you sure you've cleared the "owner's permission" hurdle before starting out? |
|
4286)
Message boards :
Questions and problems :
Any upgrade guide from version 5 to 6?
(Message 33083)
Posted 27 May 2010 by Richard Haselgrove Post: This point about running High Priority keeps being mentioned, but usually by crunchers without your powers of observation and understanding of logic. I don't see why it should happen, but all my reasonably high-powered machines have been running v6 for so long that their caches and system metrics have settled down to non-priority running years ago. Do you have any idea (or could you try and look, to find out) why BOINC is so panic-struck? High Priority shouldn't happen any more often than it did under v5. Too much work? Bad runtime estimates? Bad speed estimates (have the benchmark results changed?) Bad availability estimate? Something must be causing it, and it's not clear what. |
|
4287)
Message boards :
Server programs :
GPU scheduling
(Message 33073)
Posted 27 May 2010 by Richard Haselgrove Post: The job is associated with an application though, which has strict requirements upon the client's capabilities as defined by the application plan. Reading the application planning page, I can see that an application, and all subsequent jobs associated with it will be tied to its definition: NAME_VERSION_PLATFORM[__PLAN-CLASS]. So how can an application, and a job created for this application, have: Usually yes, a particular piece of hardware and an operating system will between them need a unique application. But have a look at this list of applications (from SETI Beta): Platform Version Installation time Linux/x86 6.03 20 Aug 2008 18:30:38 UTC Windows/x86 6.03 31 Jul 2008 0:24:04 UTC Windows/x86 6.08 (cuda) 15 Jan 2009 18:55:47 UTC Windows/x86 6.09 (cuda23) 13 Aug 2009 22:38:38 UTC Windows/x86 6.10 (cuda_fermi) 19 May 2010 22:15:39 UTC Mac OS X 6.03 20 Aug 2008 23:50:39 UTC SPARC/Solaris 5.17 3 Aug 2006 21:20:53 UTC Linux/x86_64 5.28 26 Sep 2007 16:14:58 UTC SPARC/Solaris 5.17 3 Aug 2006 21:20:53 UTC Mac OS X/Intel 6.03 20 Aug 2008 23:50:39 UTC Linux/x86_64 6.03 20 Aug 2008 18:59:32 UTC Any one of those eleven application/OS/hardware combinations would be capable of doing the same job. Each job has eleven compatibility labels on it while it's in the queue: it'll only be allocated to one of the combinations that you have available and are requesting work for. |
|
4288)
Message boards :
Server programs :
GPU scheduling
(Message 33071)
Posted 27 May 2010 by Richard Haselgrove Post: Not quite. A 'job' is just a description of something that needs doing. It may be data that needs searching or manipulating, or parameters that need testing. The description of that job that needs doing doesn't say anything about how the work is going to be done, or what is going to do it. Windows or Linux? Intel, PowerPC or Sparc? It'll be a computer program of some sort - supplied by the project or by the user (optimised application)? There's a lot of pattern-matching going on: the objective is to assemble, out of the available jigsaw pieces, both a tool and a task which together can occupy the idle resource productively. |
|
4289)
Message boards :
Server programs :
GPU scheduling
(Message 33068)
Posted 27 May 2010 by Richard Haselgrove Post: Addendum: cross-posted. Your computer doesn't ask for work for a particular application, but it does ask for work for a particular 'resource' - CPU, NVidia GPU, or ATI GPU. Usually just one type of resource at a time, although it's capable of asking for all three at once. It's the server (scheduler) which is reponsible for asking "Who is this guy, anyway? What sort of work does he like doing?", and extracting your application preferences from your preference set on the server (it may also be influenced by any applications you installed manually via Anonymous Platform, or data files you already hold for Locality Scheduling). |
|
4290)
Message boards :
Server programs :
GPU scheduling
(Message 33067)
Posted 27 May 2010 by Richard Haselgrove Post: A job is a job is a job. The purpose of the feeder is always to keep a few jobs close at hand, just in case. Your computer comes knocking on the door, and says "Can I have a job for my GPU, please?" The feeder will look at the jobs lying around the place, pick one up, ask itself "Does the label on this one say it can be done by (a) (that sort of) GPU?", and if it can, the feeder will give the job to your computer. A millisecond later, my computer might come knocking, and say "Can I have a job for my CPU, please?" The feeder might pick up an *identical* job, say "Can this be done on a CPU?", find that it can, and pass it out. Jobs can have more than one compatibility label on them. They only become "a job for a GPU" or "a job for a CPU" when they've been assigned to a particular GPU or CPU, respectively. |
|
4291)
Message boards :
Questions and problems :
Any upgrade guide from version 5 to 6?
(Message 33040)
Posted 26 May 2010 by Richard Haselgrove Post: boinctray.exe is the program that looks for mouse / keyboard activity. And boinctray.exe, despite it's ridiculous name, does not have an icon or other visual presence in the system tray. It's completely silent when running, from the user's point of view. So your two icons must relate to the two running copies of BOINC Manager, one started by the registry (keep it), and one by your roaming profile (not sure what you'll have to do to prevent that one resurrecting itself). Edit - you said there was one running copy of BOINC Manager? Are you viewing 'tasks from all users'? The roaming profile will start an instance belonging to user 'Peter', but the registry one may be a different user (I'm on XP, running as a service, on this machine, so I can't easily check). |
|
4292)
Message boards :
Questions and problems :
beginner question
(Message 33036)
Posted 26 May 2010 by Richard Haselgrove Post: I don't care at all how slow my computer runs. It is left on and idle all day and some nights, is there any good reason not to keep the % CPU time and memory usage at 100% ? Provided there is adequate ventilation for cooling, you keep the fans and vents clean, and you can afford the electricity bill - none at all. |
|
4293)
Message boards :
Questions and problems :
Any upgrade guide from version 5 to 6?
(Message 33034)
Posted 26 May 2010 by Richard Haselgrove Post: I have also recently upgraded a host from v5 to v6 - in my case, also upgrading from Vista to Windows 7, and installing new hardware. My running tasks remained intact. But I'm running 32-bit Windows: I believe that there are greater difficulties with 64-bit platforms, especially with app_info and perhaps 32-bit applications. Running two copies of BOINC Manager - v6 is started from the Registry. v5 was usually started from a shortcut in the Programs menu (startup group) - I can't even remember whether BOINC created that automatically as part of the installation, or we did it outselves. Either way, it can simply be deleted now. |
|
4294)
Message boards :
Questions and problems :
BOINC 6.10.56 does not install under Windows 7 (32/64bit)
(Message 33010)
Posted 25 May 2010 by Richard Haselgrove Post: Assuming that you got a complete / uncorrupted download (check that the file size is correct), the problem isn't to do with installing the new version of BOINC, but with removing the old version - which is done automatically at the same time. Various solutions are described in the BOINC FAQ Service. |
|
4295)
Message boards :
Projects :
News on project outages
(Message 33006)
Posted 25 May 2010 by Richard Haselgrove Post: The Einstein project seems to be missing all of a sudden. We're trying to find out what happened. Einstein seems to have returned from its own private black hole. |
|
4296)
Message boards :
Questions and problems :
Windows 7 Issues
(Message 33002)
Posted 25 May 2010 by Richard Haselgrove Post: Check your power settings, My Windows 7 desktop was going to sleep after an hour or so. Yes, my recent Windows 7 upgrade on a long-time Vista cruncher changed that setting too. |
|
4297)
Message boards :
Questions and problems :
Don't start Boinc if another program is running. Possible?
(Message 32970)
Posted 23 May 2010 by Richard Haselgrove Post: You're thinking of the <exclusive_app> option directive in cc_config.xml See client configuration. |
|
4298)
Message boards :
News :
AQUA@home deploys first OpenCL application on BOINC
(Message 32967)
Posted 23 May 2010 by Richard Haselgrove Post: I have also run AQUA's test offline on a Fermi (GTX 470). In my particular case (possibly slightly atypical hardware), the OpenCL version was approximately 10% slower than the CUDA version, and in addition used much more CPU resources. I think that we are still at the 'proof of concept' stage - we await more mature OpenCL programming and runtime support. But it's good to know that the possibility exists. |
|
4299)
Message boards :
BOINC client :
BOINC Allows WU's to go Passed Deadline.
(Message 32863)
Posted 18 May 2010 by Richard Haselgrove Post: You mean all those megabytes of 18-May-2010 19:52:58 [QuantumFIRE alpha] [rr_sim] casino_p2-hno_04_parasweep.1000084_0 dur: 26513.53 = 0.335*26079.79 + 0.665*26732.13 are never going to be read? ;-) Just for the record, there's a "misses deadline by 21829.71" in there, and it has fetched new work for AQUA whilst in deadline trouble. I'm comfortable with leaving this in trunk and not holding up v6.10.56 - though if it gets called back yet again, and we have to go through another round of v6.10 testing, I'd suggest including this in the next re-release. |
|
4300)
Message boards :
BOINC client :
BOINC Allows WU's to go Passed Deadline.
(Message 32861)
Posted 18 May 2010 by Richard Haselgrove Post: OK, I've updated my Quantum computer to v6.10.56, brought forward the QF deadline to 22:30 this evening (about 30 minutes longer than BOINC estimates it would need), and set the log flags. I have just three tasks on the machine at the moment - one QF and two AQUA. I suspended all three tasks while I did the fiddling around: then I first resumed the QF. BOINC reported it running High Priority. Then I resumed an AQUA: BOINC preempted the QF, and ran AQUA instead (task duration ~70 minutes, deadline 10 days away). I'll send John (and David?) edited highlights of the log as it approaches and passes the artificial deadline. Fiddling around with Unix time converters, I find that my phone number converts to nest Saturday evening. I don't think that proves very much, but it was diverting.... |
|
4301)
Message boards :
BOINC client :
BOINC Allows WU's to go Passed Deadline.
(Message 32858)
Posted 18 May 2010 by Richard Haselgrove Post: I think TGG is right to finger the handling of multi-threaded (i.e. AQUA) tasks for this one. We went through several stages of MT scheduling in development testing, including the currently recommended v6.10.18 which has a tendency to leave single-threaded tasks unstarted (or waiting to run) if there is an AQUA task in the mix but not active. I have a dual-core attached to AQUA and QuantumFIRE (you could call it my quantum computer...). At the moment, the AQUA admins are pressing harder for results, so I have NNT set for QF. Each time I do that, I have one orphaned QF task left over, which is never scheduled to run because there's nothing to pair it with - even though debt is on the limit, with +/- 86,400 seconds for the two projects. Up to now, I haven't let anything reach deadline (I've manually allowed the orphan a playmate when deadlines approach, and then they get scheduled together until one completes). But if we still need logs when the time comes (early hours of 24 May for the current one), I can supply them. On the other hand, I think David will recognise that this is a consequence of the current design. |
|
4302)
Message boards :
BOINC Manager :
Poll: BOINC Manager remote control
(Message 32834)
Posted 16 May 2010 by Richard Haselgrove Post: There is also a significant but hard-to-fix problem: when you connect, the client resolves all hostnames in remote_hosts.cfg before deciding if it will let you in. If DNS is down, the client may get stuck in that process for up to 30 seconds per hostname. While it's doing that, it doesn't answer science app heartbeats, so apps may quit. But it can't be changed to only resolve hostnames on startup, because if the corresponding IPs change later, it wouldn't let you connect... In my case, it's a small home Windows-only network, with a mixture of most versions of Windows from 98 to Vista. There's no Active Directory involved, so the question of name resolution is separate and distinct from any possible DNS issues. I think my issue started with the increased security on NETBIOS these days. At this domestic level, there's always a tension between plug'n'play simplicity and educating users into why and when proper security is necessary. For example, I recently changed ISP at another location. The new ISP supplied a free router: I unboxed it, connected power, DSL and ethernet, and I was online. No fuss, no configuration, no loading software, no username/password/IP (they all exist, of course, and I've been notified of them - but the autoconfig 'just worked'). That's the way the future is moving. Given that remote_hosts.cfg only gives access to BOINC RPC calls, and (until the first virus@home project is launched) doesn't compromise the rest of the host, I think most home users would be satisfied with "private" or "shared", without knowing the arcana of precisely how the sharing is achieved or controlled. Would it be possible to define a keyword for remote_hosts, or a switch for cc_config, which permitted access from "this subnet only" - without forcing people to even look up their IP address/subnet mask? |
|
4303)
Message boards :
BOINC Manager :
Poll: BOINC Manager remote control
(Message 32815)
Posted 15 May 2010 by Richard Haselgrove Post: 1) Yes, I've used BoincView for years. I've also written my own homebrew application which displays, logs and optionally resets individual debt values (I don't think the new v6.10 'top limited' debt system is working properly yet). The RPC for setting GPU_STD doesn't exist yet - if you happen to be passing through......? 2) remote_hosts.cfg 3) Four 4) Started with two names only, later added matching IP addresses as a fallback (had some resolving issues on my network) 5) I've been pleasantly surprised that my DHCP addresses have never 'wandered', so no, I've not needed lots of entries. DHCP is by a Draytek Vigor 2600 small-office class router. 6) Sounds logical, but perhaps even more of a minority interest than the present system. From my personal perspective, not a high priority, but nice. |
|
4304)
Message boards :
Questions and problems :
Multiple WU progress toward completion
(Message 32814)
Posted 15 May 2010 by Richard Haselgrove Post: BTW, is Report deadline time UTC? A task deadline is a fixed, universal point in time, and it's displayed in UTC on project websites because somebody might be looking at it from anywhere in the world. However, the BOINC Manager on your own machine will display the deadline in whatever local time setting you use. It's really quite clever about that: if you're looking at a task in winter, and the deadline isn't until the summer, BOINC and your computer's operating system should co-operate to make the daylight saving correction to the deadline. |
|
4305)
Message boards :
Questions and problems :
One processor for one project
(Message 32801)
Posted 14 May 2010 by Richard Haselgrove Post: Are you sure that value is accepted? Illegal values tend to be interpreted as default values (60 seconds in this case). Minutes? |
|
4306)
Message boards :
Questions and problems :
Help! I want to add a project, but had problems with BAM
(Message 32755)
Posted 12 May 2010 by Richard Haselgrove Post: GigaMega's I think. But aren't those Terabytes? :-) No, Giga (10^9) Mega (10^6) comes to 10^15. Them's Petabytes. |
|
4307)
Message boards :
Questions and problems :
Do all lines Read In from cc_config.xml immediately take effect?
(Message 32668)
Posted 7 May 2010 by Richard Haselgrove Post: Sorry, I was posting in a hurry while trying to load up for a weekend away. I just wanted to alert readers that ncpus was an instant-acting, rather than deferred, option - Gundolf wasn't sure, and had pointed in the opposite direction. Just had time to run the test and post a single copy of the full results :-( |
|
4308)
Message boards :
Questions and problems :
Do all lines Read In from cc_config.xml immediately take effect?
(Message 32660)
Posted 7 May 2010 by Richard Haselgrove Post: Jord, the <ncpus> setting takes effect immediately, not on restart. See the HT thread at SETI. |
|
4309)
Message boards :
Questions and problems :
Varying success rates of different machines
(Message 32659)
Posted 7 May 2010 by Richard Haselgrove Post: Regarding the amount of work stored, there is no difference betweenComputer is connected to the Internet about every 2.5 days (Leave blank or 0 if always connected. BOINC will try to maintain at least this much work.)andMaintain enough work for an additional 0.25 days Yes there is - there's a factor of 10 diffenence. Isn't there? |
|
4310)
Message boards :
Questions and problems :
Dual core behavior
(Message 32653)
Posted 7 May 2010 by Richard Haselgrove Post: Ah. Which version of BOINC are you using? We keep going back to v6.10.18 as the 'recommended' version, but that goes back a long way: we were actively working our way through some multi-threaded scheduling bugs at the time it was first promoted. Read BOINC v6.10.19 available: corrects 'idle CPU' bug on the AQUA message board. |
|
4311)
Message boards :
Questions and problems :
Do all lines Read In from cc_config.xml immediately take effect?
(Message 32650)
Posted 7 May 2010 by Richard Haselgrove Post: There are certainly lines which only take effect on client restart - <zero_debts> is another one. I agree it would be helpful to have a definitive list. |
|
4312)
Message boards :
BOINC Manager :
10.6.43, how to recover
(Message 32648)
Posted 7 May 2010 by Richard Haselgrove Post: Try <coproc_debug> |
|
4313)
Message boards :
Questions and problems :
Dual core behavior
(Message 32643)
Posted 6 May 2010 by Richard Haselgrove Post: AQUA is a pioneer of "multi-threaded" applications - one task is assigned to all available cores. Most (all?) other projects are "single-threaded" - you would see a separate task for exach available resource: two tasks on your dual core, four on quads, more if you have compatible GPUs as well. Check with Task Manager (assuming Windows) to see what your actual CPU usage is with 1 x AQUA or 2 x anything else. I think you'll find it's the same. |
|
4314)
Message boards :
Questions and problems :
Install failure of ver 6.10.18
(Message 32637)
Posted 6 May 2010 by Richard Haselgrove Post: I don't know what the ideal solution is Yes, I think that's better (in the context). A little education, assuming even less prior knowledge.... |
|
4315)
Message boards :
Questions and problems :
Install failure of ver 6.10.18
(Message 32635)
Posted 6 May 2010 by Richard Haselgrove Post: Since this is a much asked thing, I made a FAQ with screen shots for it: BOINC and science applications continue running when I exit BOINC Manager. Useful. One small comment - the term 'GUI' (for 'Graphical User Interface') is probably unknown to the sort of user this FAQ is aimed at - too many TLAs. I don't know what the ideal solution is, but it seems an increasing number of posters don't demonstrate that they have a clear understanding that there are two quite separate programs - BOINC, and then BOINC Manager to, well, manage it. See, for example, the fairly recent phenomenon of people posting that they have "installed BOINC Manager version such-and-such". |
|
4316)
Message boards :
Server programs :
How to limit number of jobs per host?
(Message 32615)
Posted 5 May 2010 by Richard Haselgrove Post: Just found much the same thing. David accepted the concept of moving 'max_wus_in_progress' from a project scope to an application scope (not app_version, as I originally had it) on 19 March (boinc_dev mailing list, thread "Preemption of very short tasks" (sic!)). But the implementation seems to be bound up with the "new credit" concept, which went to live test at SETI Beta on or about (appropriately) 1 April, but seems to have stalled since. I do hope the useful scoping issues aren't going to get dragged down by credit wars.... |
|
4317)
Message boards :
Server programs :
How to limit number of jobs per host?
(Message 32612)
Posted 5 May 2010 by Richard Haselgrove Post: You didn't actually mention the two science applications in the OP! :-( So many of these legacy options in BOINC still have an implicit assumption that one project == one science application (give or take a platform or three). We're slowly leaning on David and the dev team to separate the controls out per <app_version>: this looks like another one that needs the treatment. |
|
4318)
Message boards :
Server programs :
How to limit number of jobs per host?
(Message 32610)
Posted 5 May 2010 by Richard Haselgrove Post: Doesn't <max_wus_in_progress> N </max_wus_in_progress> from Project configuration do that? |
|
4319)
Message boards :
Questions and problems :
BOINC mgr. GUI
(Message 32554)
Posted 3 May 2010 by Richard Haselgrove Post: @ Jord, Had an updating session on one of my Quads this morning. Installed
I was watching your manager to check everything was working when the 'Communicating with Client' dialog popped up. This is incredibly rare on my systems (I run short caches - only about 20 tasks on the machine in total). |
|
4320)
Message boards :
Questions and problems :
Version question
(Message 32522)
Posted 1 May 2010 by Richard Haselgrove Post: Do any projects you know of require a newer version than 5.10.45 for CPU only crunching? Jord, you just beat me to it. 2010-04-19 13:27:35 [AQUA@home] Master file download succeeded 2010-04-19 13:27:40 [AQUA@home] Sending scheduler request: Project initialization 2010-04-19 13:27:40 [AQUA@home] Requesting 1 seconds of new work 2010-04-19 13:27:45 [AQUA@home] Scheduler RPC succeeded [server version 611] 2010-04-19 13:27:45 [AQUA@home] Message from server: Need version 6.2.0 or higher of the BOINC client. You have 5.10.13. 2010-04-19 13:27:45 [AQUA@home] Deferring communication for 1 days 0 hr 0 min 0 sec |
|
4321)
Message boards :
Questions and problems :
Question about preventing a project from returning WUs
(Message 32497)
Posted 1 May 2010 by Richard Haselgrove Post: Explain "important" ? |
|
4322)
Message boards :
Questions and problems :
Compution error for LHC@home after click 'show graphics'
(Message 32485)
Posted 30 Apr 2010 by Richard Haselgrove Post: happen 3 times already. Been reported on the LHC@home message boards many, many more than three times. And acknowledged by the Project Developer as an item on his 'to do' list. |
|
4323)
Message boards :
News :
BOINC 6.10.43/44 withdrawn
(Message 32473)
Posted 29 Apr 2010 by Richard Haselgrove Post: This is very odd. I saw and reported such a bug in v6.10.46: reverted to v6.10.45, which allowed the task to run: tried v6.10.47: that also prevented the task running: reverted to v6.10.45, and successfully completed the task. So, for that limited experience, v6.10.46 and later were less reliable than v6.10.45 For the record: I had problems with a GPUGrid TONI_CAPBIND task (specifically: f237-TONI_CAPBIND99SB-51-100-RND2375_1) on an NVidia 9800GTX+ 512 MB GPU. I haven't been able to reproduce the problem since, so it may just have been a weird transitional experience with a WU in progress: but I haven't seen any explanation to that effect on the alpha mailing list. |
|
4324)
Message boards :
Questions and problems :
Hard drive spinning up and down frequently...?
(Message 32459)
Posted 29 Apr 2010 by Richard Haselgrove Post: Are you sure it's the hard drive spinning up? I find it hard to tell if disks are even powered up, the modern ones are so quiet: and it would be very unusual for disks to spin up and down so quickly - the power-saving timeouts are usually measured in minutes rather than seconds, even for a laptop. The noisy bits of the computer - the ones you can hear - are usually fans. |
|
4325)
Message boards :
Questions and problems :
BOINC 6.10.43/6.10.44 no longer released for public
(Message 32447)
Posted 29 Apr 2010 by Richard Haselgrove Post: It won't be anything to do with the installation of the new version: it's the uninstallation of the previous version - which happens automatically as part of the process - which is causing the problem. And before you jump to conclusions, it's a glitch with the Microsoft Installer technology, not with BOINC: the 'feature' that is unavailable is the previous installation data. The easiest way of sorting out a problem like this is to re-run the installer for the exact same version you were running before (I always keeps the download files, for precisely this reason). That should work, and offer to "repair" your current BOINC: once that's complete, the new version installer should work as well. |
|
4326)
Message boards :
Questions and problems :
Message received - 'Suspending Computation-CPU usage is too high
(Message 32425)
Posted 27 Apr 2010 by Richard Haselgrove Post: It's 25% of the total power of the machine. If you have four cores, that would be the equivalent of 1 core full time, and nothing at all on the other cores - except no self-respecting operating system would ever run the hardware that way. It's far more likely to be 25% of each of however many cores you have, +/- 25% in every case. And please do check that you're looking at this from the right end of the telescope. BOINC excuses itself and gets out of the way if other tasks are using 25% of the CPU(s). Or more. So 26%, 27%, 30%, 50%...... stop BOINC. Conversely, if other tasks use less than 25%, BOINC runs ... at 24%, 23%, 20%, 10%, 5%. Don't be confused, easy though it is, that the figure '0' (note: not a percentage) is used to represent 100%, to save typing. |
|
4327)
Message boards :
Questions and problems :
Message received - 'Suspending Computation-CPU usage is too high
(Message 32421)
Posted 27 Apr 2010 by Richard Haselgrove Post: But why this ridiculous low default? Isn't 50% more than 25%? What's "25%" anyway? One core at 25% or 1 core busy on a quad? Both - it's the same thing, as far as BOINC's concerned. |
|
4328)
Message boards :
Questions and problems :
BOINC and Skype Conflict on XP
(Message 32401)
Posted 26 Apr 2010 by Richard Haselgrove Post: Are you using a NVidia graphics card (CUDA) for your SETI@home tasks? |
|
4329)
Message boards :
Questions and problems :
[Idea / question] Trying to do 3d rendering via boinc?
(Message 32374)
Posted 24 Apr 2010 by Richard Haselgrove Post: Been there, Done that, ... And I hope you got a very pretty 3-D teeshirt. :D (Sorry, couldn't resist) |
|
4330)
Message boards :
Questions and problems :
[Idea / question] Trying to do 3d rendering via boinc?
(Message 32373)
Posted 24 Apr 2010 by Richard Haselgrove Post: Or even BURP. Check the date on the latest news item carefully before jumping to conclusions.... |
|
4331)
Message boards :
Questions and problems :
Boinc not working under OS X 10.6.3?
(Message 32346)
Posted 23 Apr 2010 by Richard Haselgrove Post: It depends which version of BOINC you're talking about. There were communications problems for some people in v6.10.43, solved (subject to final testing) with v6.10.44 |
|
4332)
Message boards :
Web interfaces :
virus at boincstats?
(Message 32343)
Posted 23 Apr 2010 by Richard Haselgrove Post: BOINCstats works for me. BOINC triumphs again! Many thanks for that link, Les. I got a call from a friend yesterday morning, describing some very curious symptoms I couldn't make sense of, but which had (among other things) cut him off from the internet. I expected a very long and frustrating virus-hunting session at his house out in the heart of the Dales, with no Google or other tools to assist. Instead, today's planning phone call confirmed he uses McAfee, I took the fix tool with me, and that sorted it in five minutes. Off for a pub lunch to celebrate, and an afternoon's walking in Wharfedale in the spring sunshine. Turned into a very satisfying day, thanks purely to this post. But be warned - the McAfee foul-up affects home users too, not just the corporates as implied in the linked article. |
|
4333)
Message boards :
Questions and problems :
BOINC mgr. GUI
(Message 32170)
Posted 13 Apr 2010 by Richard Haselgrove Post: Thank goodness that dialog is more resizeable than it looks, and isn't modal. But it's not ready for prime-time: it remembers its size, but opens in a fixed place on screen. So if you give it enough width to see the full messages, next time you open it the 'close' button is off the edge of the screen.... |
|
4334)
Message boards :
BOINC client :
Fermi Cards recognition
(Message 32167)
Posted 13 Apr 2010 by Richard Haselgrove Post: Reporting from GPUGrid: 13-4-2010 9:58:53 Starting BOINC client version 6.10.45 for windows_intelx86 |
|
4335)
Message boards :
Questions and problems :
BOINC mgr. GUI
(Message 32166)
Posted 13 Apr 2010 by Richard Haselgrove Post: "Client is processing results" Yes, I noticed that too. But my v6.10.36/.45 Managers are saying 5 tasks running. - no longer listing the projects or %ages reached. Still tells me more, in fewer words, than trunk, though ;-) |
|
4336)
Message boards :
Questions and problems :
BOINC mgr. GUI
(Message 32164)
Posted 13 Apr 2010 by Richard Haselgrove Post: Works on v6.10.36 |
|
4337)
Message boards :
Questions and problems :
BOINC mgr. GUI
(Message 32159)
Posted 13 Apr 2010 by Richard Haselgrove Post: I was unable to get Boincmgr76.exe to start, comes up with: Ran as advertised for me on XP32. Has a tendency to run micro-miniaturised, but that's easily sorted by updating my boincmgr670 shortcut. More oddly, it has the new 'notices' tab which Jord mentioned, and a button in readiness for an RSS feed from Berkeley: but it does not have the traditional "messages" tab to tell you what's going on inside your own machine. Jord, could that be an artefact of your build process, or are they really planning to hide that information from the general public? |
|
4338)
Message boards :
BOINC client :
6.10.45 BOINC issues
(Message 32120)
Posted 11 Apr 2010 by Richard Haselgrove Post: Yes, it was experimental code, introduced in [trac]changeset:21035[/trac] (no reason given), reverted in [trac]changeset:21144[/trac]. Only happens in v6.10.45 |
|
4339)
Message boards :
Questions and problems :
BOINC 6.10.43 - Runs two task on single gpu
(Message 32078)
Posted 9 Apr 2010 by Richard Haselgrove Post: You double-compressed the log file, which confused me for a moment. Yes, it starts with 09-Apr-2010 17:30:11 [SETI@home] Computation for task 21fe07ae.3786.14387.8.10.4_0 finished 09-Apr-2010 17:30:11 [---] [cpu_sched_debug] Request CPU reschedule: handle_finished_apps 09-Apr-2010 17:30:11 [---] [cpu_sched_debug] schedule_cpus(): start 09-Apr-2010 17:30:11 [SETI@home] [cpu_sched_debug] Result 13dc06ae.20481.890.15.10.196_2 projected to miss deadline. ... 09-Apr-2010 17:30:11 [SETI@home] [cpu_sched_debug] Project has 313 projected NVIDIA GPU deadline misses In theory, the allocation is right: 09-Apr-2010 17:30:11 [SETI@home] [coproc_debug] Assigning CUDA instance 0 to 13dc06ae.20481.890.15.10.196_2 09-Apr-2010 17:30:11 [SETI@home] [coproc_debug] Assigning CUDA instance 1 to 12ja07ae.7563.19295.13.10.254_0 09-Apr-2010 17:30:11 [SETI@home] [coproc_debug] Assigning CUDA instance 2 to 12ja07ae.7563.19295.13.10.252_0 09-Apr-2010 17:30:11 [SETI@home] [coproc_debug] Assigning CUDA instance 3 to 12ja07ae.7563.19295.13.10.249_1 09-Apr-2010 17:30:11 [SETI@home] [coproc_debug] Assigning CUDA instance 4 to 12ja07ae.7563.19295.13.10.248_1 09-Apr-2010 17:30:11 [SETI@home] [coproc_debug] Assigning CUDA instance 5 to 12ja07ae.7563.19295.13.10.247_0 but it may have been confused by 09-Apr-2010 17:30:12 [---] [cpu_sched_debug] coproc quit pending, deferring start 09-Apr-2010 17:30:12 [---] [cpu_sched_debug] Request enforce CPU schedule: coproc quit retry Roll on per device client DCF! |
|
4340)
Message boards :
Questions and problems :
Tasks running at high priority
(Message 32059)
Posted 9 Apr 2010 by Richard Haselgrove Post: Unfortunately, BOINC can't (yet) predict the future, so it can only derive assumptions from the past. It may well be assuming that, because you switched the computer off for six months, you might switch it off again before the tasks are finished. (Question: could we teach BOINC the concepts of 'summer' and 'winter'? Many volunteers have a seasonal variation in the level of contribution they are able to make). If you look at the 'Computer Summary' page for your machine on the website of any project you're attached to, you can see BOINC's estimate of "% of time BOINC client is running" (towards the bottom of the page). If that figure is very low, then it's not surprising if BOINC tries to make urgent use of whatever little time it expects to have available. |
|
4341)
Message boards :
Questions and problems :
Already attached - deleting project_init.xml
(Message 32035)
Posted 8 Apr 2010 by Richard Haselgrove Post: And now there's [trac]changeset:21144[/trac], which reverses that change. Edit - and a good thing too. Looking at one of my v6.10.45 boxes, I seem to be getting the 'deleting' message every two or three hours, totally at random - I can't relate it to anything else going on at the time. And it doesn't tell you which project_init.xml it might be deleting.... |
|
4342)
Message boards :
Questions and problems :
BOINC 6.10.43/6.10.44 no longer released for public
(Message 31981)
Posted 6 Apr 2010 by Richard Haselgrove Post: Beg your pardon, I did copy it from a working installation, but too many of mine are at 100%...... To put in a restriction: <suspend_cpu_usage>97.000000</suspend_cpu_usage> or no restriction at all <suspend_cpu_usage>0.000000</suspend_cpu_usage> |
|
4343)
Message boards :
Questions and problems :
BOINC 6.10.43/6.10.44 no longer released for public
(Message 31978)
Posted 6 Apr 2010 by Richard Haselgrove Post: In the BOINC manager's menu: Advanced -> Preferences _> processor usage Les, he said no GUI! It's not in boinccmd either. Only solution I can come up with is a global_prefs_override.xml file with: <global_preferences> <cpu_usage_limit>100.000000</cpu_usage_limit> </global_preferences> |
|
4344)
Message boards :
Questions and problems :
Download mirror broken for v6.10.43 (all platforms)
(Message 31977)
Posted 6 Apr 2010 by Richard Haselgrove Post: For my (UK based) system, the BOINC download pages are currently pointing to files held on Einstein's Glasgow University mirror - except the files aren't there. This is for the new 'recommended' versions only - development versions are still downloadable direct from Berkeley as usual. [I have has problems in the past with Einstein's own files being missing from that mirror, too] |
|
4345)
Message boards :
BOINC Manager :
linux 6.10.43: cherry picking bug
(Message 31958)
Posted 5 Apr 2010 by Richard Haselgrove Post: AFAIK, BOINC Manager follows the standard (or is it Windows only?) convention. If you hold down the 'Shift' key, the selection becomes "this one, the first one you clicked, and everything in between." If you hold down the 'Ctrl' key, you can select or deselect individual items in the list, without affecting the ones already selected/deselected. |
|
4346)
Message boards :
Questions and problems :
A bug in 'Upgrade Information Page'?
(Message 31925)
Posted 3 Apr 2010 by Richard Haselgrove Post: Correct. (Cosmetic) changeset [trac]changeset:20772[/trac] never made it into branches/boinc_core_release_6_10 for v6.10.43 It didn't make it into v6.10.40 either, as I reported to boinc_alpha on 19 March. It's one of the classic 'overlook' errors, where a client change is hidden under a server-only checkin note if you only look at the one-line version of the logs. I think the eventual decision was to let it pass, as it was a cosmetic bug only affecting non-default message logs, and I was the only nit-picker likely to report it :-( ! |
|
4347)
Message boards :
Questions and problems :
Separate schedules for separate GPUs
(Message 31899)
Posted 1 Apr 2010 by Richard Haselgrove Post: Set networking off Disable GPU Suspend proect Set networking on and reverse? |
|
4348)
Message boards :
Questions and problems :
Boinc Corporate Standards ed.1.1
(Message 31896)
Posted 1 Apr 2010 by Richard Haselgrove Post: I am pleased to see that your new President, with his well-known interest in BOINC (and exemplifying the trend which Mo's research reveals), is taking active steps to alleviate the supposed "problem". When the merger of the United States and China is complete, all the family will be able to participate in BOINC without shame. Scientific research will gain a massive influx of new participants, and a new golden era of mass-participation Citizen Science will be ushered in. 
|
|
4349)
Message boards :
Questions and problems :
Separate schedules for separate GPUs
(Message 31895)
Posted 1 Apr 2010 by Richard Haselgrove Post: There's a configuration directive available for <ignore_cuda_dev>, which allows you to deselect a particular device by ID. You could have a 'day' cc_config.xml file which ignores the 8600GT, and a 'night' file which allows it to be used - use a timed script to switch them over. [You'd need to upgrade BOINC to at least v6.10.19 for this to work]. That would help with the sluggishness and the movie-watching: haven't thought of a way of pinning the applications, though you could suspend/resume the PowerGrid project at the same times in the script. |
|
4350)
Message boards :
BOINC client :
Fermi Cards recognition
(Message 31874)
Posted 31 Mar 2010 by Richard Haselgrove Post: Hmmmmm. Maybe I take that back. Reading from "CUDA Programming Guide Version 3.0" (http://developer.download.nvidia.com/compute/cuda/3_0/toolkit/docs/NVIDIA_CUDA_ProgrammingGuide.pdf), I see: For devices of compute capability 1.x, a multiprocessor consists of: For devices of compute capability 2.0, a multiprocessor consists of: So maybe David's way is the only officially-sanctioned one. But if you have to derive it from the Compute Capability, shouldn't the test be (prop.major>=2)? The 'compute capability' for a Fermi is 2.0: it's the 'CUDA version' which has gone up to 3000 or above. That's a software number (from '__cuDriverGetVersion'), not hardware (from '__cuDeviceComputeCapability'). |
|
4351)
Message boards :
BOINC client :
Fermi Cards recognition
(Message 31872)
Posted 31 Mar 2010 by Richard Haselgrove Post: Jord, I still don't think you have it quite right. A DLL is a code library - a tool for doing something, not a repository of information in its own right. It will be part of the Application Programming Interface: so a better characterisation would be "BOINC uses the driver's DLL file to read information from the card" - not the peakflops directly, but lower-level information which can be used to calculate it. That new code is an even worse kludge than the previous one: let's hope it's just a temporary holding operation until David (as posted earlier in this thread) can get a proper answer back from NVidia. Maybe they made a similar assumption when writing the API, and forgot to include a function for "How many shaders does each multiprocessor have?". @ skgiven, I don't think this false assumption by BOINC is going to have any effect at all on actual computation. The 'Peak FLOPs' figure is only posted by BOINC to give people something to brag about, and it will help in work fetch calculations. But I don't think it will control the science application. Far more likely is that the application developers (perhaps led astray by BOINC's code) have similarly failed to find an API for "shaders per MP", and have taken the lazy way out with a hard-coded constant. GPUGrid don't seem to have cracked it yet, even with the cuda30 version of their Beta v6.22, but it's early days for Fermi. |
|
4352)
Message boards :
BOINC client :
Fermi Cards recognition
(Message 31859)
Posted 30 Mar 2010 by Richard Haselgrove Post: // each processor has 8 cores; BOINC bug. He's hardcoded a variable, not valid for Fermi: each processor has 32 cores. |
|
4353)
Message boards :
Questions and problems :
BOINC No longer getting work
(Message 31856)
Posted 30 Mar 2010 by Richard Haselgrove Post: Gundolf, I can't point you to exact wording just at the moment, because SETI's weekly maintenance has just started: but SETI doesn't have a 'use GPU' preference. It does have a 'use NVIDIA GPU' option, but the matching 'use ATI GPU' option is hidden - automatically - because SETI doesn't have any ATI applications yet. So in your configuration, and with that option set, BOINC will continue to check periodically to see if any ATI application, plus work for it, has materialised yet. It won't have. |
|
4354)
Message boards :
BOINC client :
Fermi Cards recognition
(Message 31851)
Posted 30 Mar 2010 by Richard Haselgrove Post: Re: GPU overclocking (pre-edit!) If you haven't seen any overclocked GPUs, then you haven't read Post your BOINC Startup 'CUDA' Info. Lots of weird and (not so wonderful) figures there. [And I see my last post killed it] In my case, I have one card which was overclocked by the manufacturer. But I'm no longer using the drivers from the (15 month old) manufacturer's CD: I'm using standard, but newer, NVidia drivers. They pick up the hardware overclock. Other people use software tools like (from memory) RivaTuner or EVGA Precision something. Apparently, they aren't manufacturer specific but can adjust the clockings of any NVidia-based card. The variation in speeds in that SETI thread mainly result from the use of such tools. There are also tools like GPU-Z which can get data the same way BOINC can: this is a reviewer's shot of a Fermi, re-posted from GPUGrid: 
|
|
4355)
Message boards :
BOINC client :
Fermi Cards recognition
(Message 31849)
Posted 30 Mar 2010 by Richard Haselgrove Post: All BOINC does is read the information about your GPU and translate that into values you see. Jord, I don't think that's the right reference. The link you gave is more concerned with displaying the information that BOINC already holds. BOINC actiually gets the information in the first place via coproc_detect.cpp. As you say, that relies on NVIDIA-supplied code in a DLL. But surely that code must actually query the underlying hardware? Not for the driver version, obviously, but things like current speed can't be hard-coded into a lookup table: otherwise overclocking could never be reported. WRT drivers: would it not be a plausible first assumption that driver version 19733 came off the CD supplied in the box with the card itself? Still nice to have that confirmed, or otherwise, though. Edit: From coproc.h inline double peak_flops() {
double x = attribs.numberOfSIMD * attribs.wavefrontSize * 2.5 * attribs.engineClock * 1.e6;
|
|
4356)
Message boards :
BOINC client :
Fermi Cards recognition
(Message 31836)
Posted 29 Mar 2010 by Richard Haselgrove Post: 3/29/2010 ... NVIDIA GPU 0: GeForce GTX 480 (driver version 19733, CUDA version 3000, compute capability 2.0, 1503MB, 194 GFLOPS peak) 194 GFLOPS isn't exactly exciting either - even my humble 9800GT are reported at 339 GFLOPs these days. One source of practical information at the moment is this GPUGrid thread. Their appication (not BOINC) is reporting the GTX 480 as # There is 1 device supporting CUDA # Device 0: "GeForce GTX 480" # Clock rate: 0.81 GHz # Total amount of global memory: 1576468480 bytes # Number of multiprocessors: 15 # Number of cores: 120 - that's from a host which is also displaying driver 19733 through BOINC v6.10.18/Windows 7. GPUGrid reckon the low core count is because they've coded 8 shaders per MP, where the Fermis have 32 - maybe that could account for the low speed report too. |
|
4357)
Message boards :
Questions and problems :
6.10: "No usable GPUs found"
(Message 31777)
Posted 26 Mar 2010 by Richard Haselgrove Post: With Windows (XP/W7) you can [if not already mentioned] and I have been using the delayed load service-start function set via the services.msc. Just run that in admin, right click BOINC then properties. Default is automatic, as drop down list offering automatic, with delay I'm using that setting in Vista, but I don't think it's available in XP. |
|
4358)
Message boards :
Questions and problems :
why no graphics?
(Message 31723)
Posted 23 Mar 2010 by Richard Haselgrove Post: ideas? Name of project(s) attached? Name of project running? |
|
4359)
Message boards :
Projects :
News on project outages
(Message 31719)
Posted 22 Mar 2010 by Richard Haselgrove Post: It looks like Einstein@Home took another dive off line. Seems OK from here - I was just reading Mike Hewson's message 103142. |
|
4360)
Message boards :
Projects :
News on project outages
(Message 31712)
Posted 22 Mar 2010 by Richard Haselgrove Post: Data communication has been restored to SETI, but the websites all seem still to be down. |
|
4361)
Message boards :
BOINC Manager :
6.10.43 bug? entire message buffer being duplicated
(Message 31695)
Posted 21 Mar 2010 by Richard Haselgrove Post: Just to be on the safe side, I checked my v6.10.43 Windows installation: the message log starts with 19-Mar-2010 19:11:10 [---] Starting BOINC client version 6.10.43 for windows_intelx86 and continues properly in date order without duplicates from then on. BB, be aware that the log files (such as stdoutdae.txt on disk) are created by the core client, not the Manager. So when you use a v6.10.43 Manager to connect across the network to a v6.10.32 client, you're not really testing v.43 at all - you're just testing that v.32 is still not creating duplicate messages (which I think we knew, but it's nice to be reassured :-) ). |
|
4362)
Message boards :
Questions and problems :
BOINC 6.10.36 Intel Mac Network Problems
(Message 31633)
Posted 17 Mar 2010 by Richard Haselgrove Post: denbarth at CPDN reports that he can't get either of the two test versions to work. T7400 @ 2.16GHz, Darwin 9.8.0 - MacBook? |
|
4363)
Message boards :
BOINC Manager :
ping Ageless: question about wiki RAC spreadsheet
(Message 31611)
Posted 16 Mar 2010 by Richard Haselgrove Post: The problem seems to be in the ROUNDDOWN cells (columns O, T, Y for PC #1, #2, #3 respectively). If you make a trivial edit in, for example, cell O18 - I just put a space at the end - and press enter, it gets recognised (in Excel 2002). Then just drag the corner handle down to replicate the formula down the rest of the column. |
|
4364)
Message boards :
Questions and problems :
BOINC 6.10.36 Intel Mac Network Problems
(Message 31603)
Posted 15 Mar 2010 by Richard Haselgrove Post: Charlie, you have another satisfied customer at CPDN: Installing the experimental version (6.10.36A) directly over 6.10.36 completely resolved the issue. DKD reports his host as "OS X 10.6.2 on a 4x3GHz (early) Mac Pro tower", but the rest of his details are hidden. |
|
4365)
Message boards :
BOINC Manager :
6.10.36 HTTP error on upload (SETI@home [file_xfer_debug] FILE_XFER_SET::poll(): http op done; retval -184)
(Message 31597)
Posted 15 Mar 2010 by Richard Haselgrove Post: I'm subscribed to the SETI project on several machines. Only one machine is having this problem and it's been weeks now. Indeed. The current SETI server upload problems started on February 15, and have continued (more or less continuously) ever since. But it would be better to discuss them on the SETI message boards. |
|
4366)
Message boards :
BOINC Manager :
6.10.36 HTTP error on upload (SETI@home [file_xfer_debug] FILE_XFER_SET::poll(): http op done; retval -184)
(Message 31594)
Posted 15 Mar 2010 by Richard Haselgrove Post: It's a SETI server problem. Nothing BOINC can do, nothing you can do - except wait. |
|
4367)
Message boards :
Questions and problems :
6.10: "No usable GPUs found"
(Message 31579)
Posted 14 Mar 2010 by Richard Haselgrove Post: Either that, or a slow/late loading video driver. |
|
4368)
Message boards :
Questions and problems :
Large team deleted - impossible?
(Message 31546)
Posted 11 Mar 2010 by Richard Haselgrove Post: Seti's done database restoration to the 16th of January. Welcome back, team Richard Dawkins Foundation for Reason and Science. That's the same Team ID 140509 that David found right at the beginning of this thread - it's the recreation, not the original. With a total credit of under ~11K cobbles and six active members, there wasn't much recovered from the backup either. But if they have got Jan 16 back, I did offer to pull a membership list out of the xml dump - still stands. |
|
4369)
Message boards :
Questions and problems :
Is it possible to chose between two CUDA devices?
(Message 31500)
Posted 9 Mar 2010 by Richard Haselgrove Post: Interesting to see that BOINC has chosen the nForce 750a as the 'better' device, when the only thing it has going for it is the compute capability 1.1 Memory is, as tgtcat6 has found, too small, and as for speed.... |
|
4370)
Message boards :
Questions and problems :
6.10.36 creates new host on stats sites?
(Message 31479)
Posted 9 Mar 2010 by Richard Haselgrove Post: I really have no idea!! Nor do I, which is why I haven't posted it. Some poking around in the archived logs (stdoutdae.txt, or even stdoutdae.old, in that data directory) would seem to be in order. See if you can find the bit that corresponds to the first re-start after the new version was installed. |
|
4371)
Message boards :
Projects :
Is SETI@Home website under attack?
(Message 31477)
Posted 9 Mar 2010 by Richard Haselgrove Post: Project administrators perhaps need to be aware of this BBC Technology News article from the end of last week: Fake drug scam hijacks UK college websites UK academic institutions have unwittingly become the accomplices of criminals selling fake drugs online. Sounds very similar to the exploit which got through SETI's defences the week before, attacking their .edu academic address. |
|
4372)
Message boards :
Questions and problems :
6.10.36 CUDA download but not active?
(Message 31475)
Posted 9 Mar 2010 by Richard Haselgrove Post: OK, I've been out of it for a while. So with CUDA 2.3 you need 384MB or more on the card? Was this enforced anywhere between 6.10.17 and 6.10.36 that you can recollect so quickly? (my low blood pressure headache is coming up again, I won't stay long behind the keyboard to scroll through change logs... sorry). I think all the important changes were server-side - see for instance changeset [trac]changeset:18887[/trac], which is where the memory requirement was added. That was last August (and by Eric Korpela, round about the time the Beta application was installed - so it's relevant to SETI). I presume all you need on the client side is the ability to report back the installed memory on the GPU, which we've had from the very beginning. Edit - Reading http://boinc.berkeley.edu/trac/wiki/AppCoprocessor may be easier on the headache than chasing through source changes. |
|
4373)
Message boards :
Questions and problems :
6.10.36 creates new host on stats sites?
(Message 31471)
Posted 9 Mar 2010 by Richard Haselgrove Post: Did the new (upgraded) installation find your old BOINC Data folder, and carry on crunching the tasks/projects you had on board before the upgrade? If you were asked to re-attach anywhere, BOINC will have created a new Host ID, which would have the effect you describe. The most common reason for losing the old BOINC data folder, from what I've read, is installing a 32-bit version of BOINC when you were previously running a 64-bit version, or vice-versa. I believe the detection code on the download page of this website doesn't detect the 32/64 bit OS properly with all browsers. I've upgraded three machines to pretty much every version ob BOINC released recently, including v6.10.36, and never had the stats split. But I'm 32-bit only. |
|
4374)
Message boards :
Questions and problems :
6.10.36 CUDA download but not active?
(Message 31470)
Posted 9 Mar 2010 by Richard Haselgrove Post: OK, I'll forward the download problem to the developers and in the mean time ask the Seti admin if something changed, memory wise, for CUDA at Seti. Jord, if you look at the SETI Applications page, you'll see that they have both CUDA v6.08, and CUDA23 v6.09 If David has done the plan_classes right, anything which has CUDA driver v190.38 or better, and 384 MB VRAM, should get <plan_class>cuda23 work allocated: if it has a lower driver, or 254MB < VRAM < 384MB, it should get <plan_class>cuda. Not 'insufficient memory'. |
|
4375)
Message boards :
Questions and problems :
BOINC 6.10.36 Intel Mac Network Problems
(Message 31423)
Posted 8 Mar 2010 by Richard Haselgrove Post: A similar Mac OS X v6.10.36 "Couldn't resolve host name" is being reported at SETI: message 976717. OK, SETI servers have been flakey recently, but this sounds like a local problem. SETI user has restarted Gateway (presume == router), so with this report, beginning to sound like a client problem? |
|
4376)
Message boards :
Questions and problems :
Scheduler request failed: Server returned nothing (no headers, no data)
(Message 31415)
Posted 8 Mar 2010 by Richard Haselgrove Post: As the most recent poster in that 'problems' thread, may I confirm that I am running v6.10.36 on the host which generated those logs and captured packets. Although there is an improved communications package (libcurl) in the new version as compared with v6.10.18, I don't think that has any bearing on this particular problem - it seems to be purely a server issue. |
|
4377)
Message boards :
Questions and problems :
No project datas found / run ?
(Message 31353)
Posted 4 Mar 2010 by Richard Haselgrove Post: I've got two on the go at the moment. About 100MB on disk in the project directory, 300MB - 350MB for each model in their own subdirectory. Each using about 215MB of RAM while running, graphics not possible on this Vista32 host. Ah - got one on XP too. About 25MB RAM for graphics. |
|
4378)
Message boards :
BOINC client :
BOINC 6.1 to 6.10 Versions Change Log
(Message 31314)
Posted 2 Mar 2010 by Richard Haselgrove Post: The word order at the end of the disclaimer is a little clumsy, to English eyes. I suggest: Disclaimer This is a development version of BOINC. Expect parts of it to be broken, or in the least to work in a different way than you are used to. Expect work failures, deadline misses and losing all your accumulated work in progress, or not getting credit for your work due to unknown and unforeseen circumstances. Only use it when you are willing to participate in the Alpha test program and do not mind sending (bug) reports in to the developers on the email list specially assigned for this purpose. |
|
4379)
Message boards :
Questions and problems :
Large team deleted - impossible?
(Message 31266)
Posted 1 Mar 2010 by Richard Haselgrove Post: I've discussed it with David, he is looking into things, but could really be helped if he had the original teamID, not the new one. I sometimes download the stat files for other research: I've had a look, but the only old one I can find at the moment is a year-old "hosts.gz", and it doesn't have any team IDs (they're not stored in that database table - checked the schema). We know that SETI have a mid-January dump lying around - it was on the web server backup image that they restored recently, and got downloaded by mistake. If David could make the team.gz file from that backup visible somewhere, I don't mind doing the search. Edit - and likewise, if the matching user.gz was accessible, I could create a membership list as at that date - it was 16 January. |
|
4380)
Message boards :
BOINC client :
BOINC 6.1 to 6.10 Versions Change Log
(Message 31263)
Posted 1 Mar 2010 by Richard Haselgrove Post: I am sad that you have had to reach this decision, but I understand why you felt this thread had out-lived its usefulness. The thing I will miss most is having a single page where I can do a text search to find which version a particular bug or feature was fixed in. That's really helpful when a user re-discovers an old bug: it makes it easy to give sensible advice about which version they need to upgrade to. As an alternative to this thread, perhaps the community could put some collective effort into updating the BOINC Version History: not with a blow-by-blow account of every typo fixed, but the major chenges that might be visible to end-users. |
|
4381)
Message boards :
Questions and problems :
Large team deleted - impossible?
(Message 31254)
Posted 28 Feb 2010 by Richard Haselgrove Post: Forwarded to development. I have archived both the BOINCstats pages, so there will be an independent copy available for reference after they update again. Jord, do you think there's any way you can persuade David to re-visit his, perhaps cursory, fly-by? |
|
4382)
Message boards :
Questions and problems :
Large team deleted - impossible?
(Message 31250)
Posted 28 Feb 2010 by Richard Haselgrove Post: ?? team is here: I believe some members have started re-creating the team from ground zero - see for example Einstein thread 7855. Where does SETI team 140509 fit in the auto-increment ID sequence? Does it have a creation date of several years ago, and several million credits? Edit - to answer my own question, the final three teams in the current (26 February 2010 13:05) team .gz dump are: 140513 Big McLargehuge 140512 BIONC BC 140509 Richard Dawkins Foundation for Reason and Science |
|
4383)
Message boards :
BOINC client :
6.10.34 System wide CPU use Throttle setting
(Message 31222)
Posted 26 Feb 2010 by Richard Haselgrove Post: If you have a task Suspended for a project, you can't get more tasks for that project. Oh, much longer than that - most of 5.* as well. But the "can't get" in your sentence refers to scheduler requests for new work to be allocated. I haven't seen reports before that already allocated work can't be downloaded because of task suspension. In fact, with big CPDN tasks, I regularly find that I get allocated more than I need, and I suspend one or more of them while they're still downloading: hasn't prevented the download completing. |
|
4384)
Message boards :
BOINC client :
6.10.34 System wide CPU use Throttle setting
(Message 31210)
Posted 25 Feb 2010 by Richard Haselgrove Post: Hmmm. Let me guess how that one works. The Manager (itself) creates - by overwriting - a new global_prefs_override.xml file, and then instructs - by GUI_RPC - the core client to read the new file. So anything the Manager doesn't explicitly know about gets splatted. I remember using GetPrivateProfileString and WritePrivateProfileString APIs under Windows 3.1 - in the early 1990s - to change single values within a complex .ini file. Have we really learned nothing about backwards and forwards compatibility since then? |
|
4385)
Message boards :
BOINC client :
6.10.32 linux: pct cpu not working with AQUA MT
(Message 31205)
Posted 25 Feb 2010 by Richard Haselgrove Post: The number of CPUs used by AQUA is set by the number of CPUs available at the time the task is allocated by the scheduler (there's a command line parameter in the workunit specification) - you'll probably find that the next AQUA task you download is set to 3 CPUs, and will remain that way even if you reset CPU usage to 100% |
|
4386)
Message boards :
BOINC client :
6.10.34 System wide CPU use Throttle setting
(Message 31200)
Posted 24 Feb 2010 by Richard Haselgrove Post: So what you're saying is that if 'Leave Applications in Memory' is set to "no", you would like BOINC to clear/release the space in RAM when preempting one science application in favour of another, but not to clear RAM when suspending BOINC itself in favour of a non-BOINC CPU-intensive task like Spybot? Sorry, I can't see that one flying - I certainly wouldn't want to write the help file explaining it, for starters! |
|
4387)
Message boards :
BOINC client :
6.10.34 System wide CPU use Throttle setting
(Message 31192)
Posted 24 Feb 2010 by Richard Haselgrove Post: How much CPU is Media Player using? The idea of the new control is that BOINC suspends only if other (non-BOINC) processes need 25% (or whatever figure you set) of - I think - the total power of all CPU cores combined. |
|
4388)
Message boards :
Projects :
Is SETI@Home website under attack?
(Message 31175)
Posted 22 Feb 2010 by Richard Haselgrove Post: I created an account at Google, purely to gain access to the Webmaster tools and report spam. Anyone here would be familiar with the procedure - give a valid email address, and respond to the check email they send. You don't actually have to own or register a website - I just twisted their tools a bit, so I could submit a report about Google's misrepresentation of SETI's site. The more complaints they receive, the more seriously they (should) take it. No response as yet, though I don't know if/how they would send any response (and the spam is still there). I tested BOINC itself, and some of the bigger projects, the same way, but SETI was the only affected one I saw. Perhaps all project managers should check their own sites, just to be sure. |
|
4389)
Message boards :
BOINC client :
6.10.32 client crashing under high CPU stress
(Message 31076)
Posted 17 Feb 2010 by Richard Haselgrove Post: I don't see how the Manager would come into this, but the Core Client handles its communications through libcurl: and there has been a recent update (bugfix) to libcurl, at about v6.10.22 Could this have crept in as a side-effect of the libcurl update? |
|
4390)
Message boards :
BOINC Manager :
Switch between applications between every
(Message 30921)
Posted 8 Feb 2010 by Richard Haselgrove Post: There's a very real discussion to be had about the best way to handle multiple tasks all under deadline pressure, but I don't think it has anything to do with TSI - the "Switch between applications between every". If you set that value to 1 minute or 1440 minutes, does it make the slightest difference to your beta tasks? |
|
4391)
Message boards :
BOINC Manager :
Switch between applications between every
(Message 30919)
Posted 8 Feb 2010 by Richard Haselgrove Post: Not quite. John is suggesting that deadline-pressure intra-project switches are also subject to the task switch interval. I think that's on the wish-list: to have a distinction between 'hard' deadline pressure, requiring an immediate switch, and a 'softer' version, allowing for a more relaxed switch at the next interval opportunity. I wasn't aware that had been implemented. |
|
4392)
Message boards :
Questions and problems :
All projects dumped ?
(Message 30903)
Posted 7 Feb 2010 by Richard Haselgrove Post: Likewise with Norton Sonar. In that case, the flagged behaviour (there were four, but I stupidly didn't note them. I think they included high CPU usage, no user interface and maybe high network activity) were exactly the characteristics that you would expect of a BOINC application. I sent Symantec a stinking complaint, and I'm running a test version to see if it crops up again nothing yet. Sonar has caught both CPDN and Einstein, and from what I can tell it can't be switched off, and it can't be taught the difference between 'good' and 'bad' apps until after it's caught them and prevented them from running. |
|
4393)
Message boards :
BOINC Manager :
Switch between applications between every
(Message 30886)
Posted 7 Feb 2010 by Richard Haselgrove Post: But I'm not sure it should say switch between projects instead of switch between applications. My public release Boinc on a dual-core often has apps from two different projects running simultaneously. I think think that switching apps is the only way it can be done to ensure the switch happens straight after app checkpoints and without allowing another core(s) to wait idle until its app is also ready for a switch. Been thinking about this, because I mainly run multicores (5 or 8 cores, counting CUDA). But I think 'project' is still good. The control that this label is describing is to do with routine preemption/rotation, to maintain resource share between projects. These switches do always involve a change of project: you'll never see an application from one project switched out, and another application from the same project taking its place, under TSI control. You will occasionally see intra-project switches, but only because of deadline pressure: those aren't affected by the 'switch interval'. |
|
4394)
Message boards :
BOINC Manager :
Switch between applications between every
(Message 30883)
Posted 7 Feb 2010 by Richard Haselgrove Post: The second 'between' is (correctly) missing from root/trunk/boinc/clientgui/DlgAdvPreferencesBase.cpp, as per [trac]changeset:19414[/trac] But root/branches/boinc_core_release_6_10/clientgui/DlgAdvPreferencesBase.cpp is still back at [trac]changeset:19322[/trac]. Before the next "Recommended" promotion, they really ought to audit the codebase for errors like that. They have a wonderful tool for keeping track of every jot and tittle, but it's being let down by the human factor when it comes to migrating changes from trunk to branch. On the other point, Switch between main projects every xxx reads fine in english English, but even the 'main' is possibly superfluous: 'Switch between projects every xxx' would be enough. |
|
4395)
Message boards :
Questions and problems :
Warns that WCG project "may not have work" incorrect
(Message 30870)
Posted 7 Feb 2010 by Richard Haselgrove Post: If you're doing a clean install, the initial "Attach to project" will be done using the all_projects.xml file distributed inside the installer: it's the only data file in the package. I only have the packages for v6.10.24 and v6.10.25 unpacked for immediate inspection, but they both have blank platform lists. On the other hand, v6.6.36 has a full set: <platforms>
<name>windows_intelx86</name>
<name>i686-pc-linux-gnu</name>
<name>i686-apple-darwin</name>
<name>powerpc-apple-darwin</name>
<name>x86_64-pc-linux-gnu</name>
<name>i386-portbld-freebsd</name>
</platforms>If someone has been using BOINC for a while, the 'Attach to project' wizard will use a fresh copy of all_platforms: that's got the blank list, as I showed you earlier. |
|
4396)
Message boards :
Questions and problems :
Warns that WCG project "may not have work" incorrect
(Message 30863)
Posted 6 Feb 2010 by Richard Haselgrove Post: 1) The 'Platforms' tag in the all_projects.xml file that is managed and circulated by BOINC is blank for WCG: <platforms>
<name></name>
</platforms>That would indeed seem to be a problem that BOINC should correct. 2) It is entirely up to each individual project what they do, or don't do. I used to crunch for LHC@home too, and I regret that it seems to have fallen into a black hole - at least, their administrative staff have retreated behind an event horizon: the website is still there, idle and covered in cobwebs. But that's not BOINC's responsibility. |
|
4397)
Message boards :
Projects :
News on project outages
(Message 30851)
Posted 6 Feb 2010 by Richard Haselgrove Post: SETI back up, and hit the ground running. Reckon it was a comms failure. |
|
4398)
Message boards :
Projects :
News on project outages
(Message 30845)
Posted 6 Feb 2010 by Richard Haselgrove Post: SETI is down again. No idea what the problem is this time. Looks like a campus-wide (or at least Space Science Lab) power outage: http://downforeveryoneorjustme.com/%20http://seti.berkeley.edu/ Edit - cancel that, BOINC lives in the SSL too, doesn't it. May just be a router/switch - they've been having trouble with those recently. |
|
4399)
Message boards :
Questions and problems :
Vacation - Will not be able to complete task
(Message 30819)
Posted 5 Feb 2010 by Richard Haselgrove Post: Let it run for a while, and watch how the estimate changes. No SETI task (not even the Astropulse ones, recently re-released) should actually take as long as that. But if it looks as if it's really going to over-run, set 'No New Tasks' (to avoid getting another one), and abort it - there are plenty of volunteers at SETI who will be glad of the resend. Click 'update project' (to tell the servers that you don't need the task anymore), and enjoy your holiday. |
|
4400)
Message boards :
Questions and problems :
Running CPU benchmarks
(Message 30801)
Posted 4 Feb 2010 by Richard Haselgrove Post: I've always thought this to be the most mind boggling thing about the benchmark and unloading the science, not by exception holding them in memory even if LAIM is off [Wrote about this several times]. My machine is working on several concurrent jobs that checkpoint every 2 hours or so. Would BOINC decide it's that time again, that in worst case is 4 hours crunching out the window. No, that's not quite the most mind-boggling thing. The most mind-boggling thing is unconditionally unloading all the GPU's VRAM, and restarting a CUDA (or, I presume, ATI) application from scratch - with or without benefit of checkpoint - to benchmark the CPU. At least the CPU application problem can be solved with the configuration change Gundolf describes. |
|
4401)
Message boards :
Server programs :
Segfault in uppercase sample application during boinc_finish(0)
(Message 30791)
Posted 4 Feb 2010 by Richard Haselgrove Post: Some deeper digging exposes a MEMSET!! on the APP_INIT_DATA object in question (aid)... Replacement code for the MEMSET was added at changeset [trac]changeset:20191[/trac], and a second instance was corrected at changeset [trac]changeset:20199[/trac]. |
|
4402)
Message boards :
Questions and problems :
Mac CUDA Tasks stuck in 'Ready to start'
(Message 30780)
Posted 3 Feb 2010 by Richard Haselgrove Post: This could be the problem - is the VRAM requirement something that can be set by the project to stop WU's being downloaded if they can't be run? It can be set somehow in the server plan_class for the application, but I'm not an expert on that. You might want to compare notes with Oliver Bock at Einstein, who has run into similar problems - read message 102091 and the discussion leading up to it. |
|
4403)
Message boards :
Questions and problems :
Mac CUDA Tasks stuck in 'Ready to start'
(Message 30768)
Posted 2 Feb 2010 by Richard Haselgrove Post: What's the memory requirement for Primegrid? Others have reported that Mac OS X desktop effects can use a lot of GPU RAM, and your 254 MB isn't a lot to start with. |
|
4404)
Message boards :
Questions and problems :
ATI GPU utilisation
(Message 30745)
Posted 31 Jan 2010 by Richard Haselgrove Post: There were reports in the early days that the hardware detection code used in the BOINC client could change speeds like this - nothing to do with projects at all. I'm afraid you'll have to wait for an ATI expert to come along and tell you if the behaviour is different with other ATI drivers or other BOINC versions. |
|
4405)
Message boards :
Questions and problems :
ATI GPU utilisation
(Message 30742)
Posted 31 Jan 2010 by Richard Haselgrove Post: What is the evidence you're seeing which suggests that the ATI GPU is being used? |
|
4406)
Message boards :
Server programs :
"Output file absent" when running "test" project-"uppercase" application
(Message 30727)
Posted 28 Jan 2010 by Richard Haselgrove Post: The "output file" in question is supposed to be generated by your scientific research application: at this early stage in testing, by the sample "uppercase" application included in the BOINC software suite. I don't actually know whether "uppercase" is designed to produce an output file, but I assume it is - if it's going to mimic the scientific research process, an output file is usually a significant part of the research. So your (temporary, surrogate) science application is failing to produce any science. That usually indicates an error condition: look in the message log of the workstation - not the server - that was supposed to be mimicing the cloud clients that will be doing your science when you go live. There should be an error number, either in the message log or in the stderr.txt file. From that error number, you will have to work out why the application failed to run. But remember: 'Output file absent' is always a symptom of an earlier failure, not a problem in itself. |
|
4407)
Message boards :
Questions and problems :
Out-of-BOINC tasks downloading and adding then files to BOINC.
(Message 30695)
Posted 26 Jan 2010 by Richard Haselgrove Post: I would nominate AQUA as a notably low-bandwidth project. The biggest hit - well above your limit - would be downloading BOINC itself, but AQUA is an order of magnitude below that. And if you download at the right time, their most demanding application can run for up to 10 days without refueling on a dual core. [Currently available tasks run shorter, and the mix changes frequently as the research develops - keep an eye on their web site]. |
|
4408)
Message boards :
BOINC Manager :
Bionc does not fetch any work from SETI
(Message 30651)
Posted 24 Jan 2010 by Richard Haselgrove Post: Not necessarily. Regulars will know that I have been testing BOINC pretty hard, with special reference to SETI and CUDA, for about a year now. I still don't think it's working properly. Last week, I had to mess around with my test rigs to investigate some AQUA/BOINC/CPDN issues. That involved dropping back to BOINC v6.10.21 to finish off some incompatible CPDN tasks. The last one finished on Friday morning: I ensured that all three of my test rigs were upgraded back to v6.10.29, SETI CUDA work enabled, and CUDA long-term (work fetch) debt zeroed. Then I went away for a couple of days. I left the hosts with a 12-hour cache setting (I keep it low-ish for testing), and freedom to fetch work as and when needed (always-on DSL router). I checked that all was running well at about 13:00 UTC Friday 22 Jan.: I got back home at 18:00 UTC today Sunday 24 Jan. to find all three CUDA cards idle. From logging, I can see that the CUDA queues ran dry at 10:05:16 24/01/10, 12:01:29 24/01/10, and 21:40:58 23/01/10 (recording the times before the log rotates). In all three cases, CUDA debt had built up to well over 100,000 seconds, and clearly inhibited work fetch when they passed the 86,400 second "overworked" mark. I zeroed CUDA LTD for SETI (only - I have a tool for that) and new work was downloaded immediately. I also have a tool which logs all BOINC debts even when I'm away. Tomorrow, when I have more time to investigate in detail, I will compare the "overwork" crossing points with the last work fetch times of 23 Jan 2010 10:53:21 UTC, 23 Jan 2010 5:51:52 UTC, and 22 Jan 2010 23:42:54 UTC (again recording for my own reference!). First conclusion: "overworked" is being given higher priority in work fetch considerations than "GPU idle", which I think is the wrong way round. Second assertion, which will need further monitoring and analysis before it becomes a conclusion, is that long-term debt is still being built up too rapidly. Await further reports on boinc_alpha later in the week. |
|
4409)
Message boards :
BOINC client :
no gpu use with boinx 6.10.21 mac
(Message 30561)
Posted 17 Jan 2010 by Richard Haselgrove Post: why when bencmarking my mac pro, no trace of GPU usage is listed ??? it's a 8800GT nvidia running on mac os x 10.5.8 I think you need to go up to BOINC v6.10.28 before the GPU use on Mac worked properly - and there are very, very few projects which have Mac GPU applications. Benchmarking is a CPU function only - you never see GPU usage during benchmark (indeed, all GPU applications are unloaded during benchmarks). |
|
4410)
Message boards :
BOINC Manager :
No CPU core used while there are available WUs!!
(Message 30526)
Posted 13 Jan 2010 by Richard Haselgrove Post: In BOINC Manager --> Advanced View --> Tasks tab: Drag the column divider after 'Application' as far as you need to the right, so you can see the entire application name. Cuda tasks show (cuda) or (cuda23) after the version number. |
|
4411)
Message boards :
BOINC Manager :
No CPU core used while there are available WUs!!
(Message 30508)
Posted 12 Jan 2010 by Richard Haselgrove Post: btw, i have been playing with the "additional work buffer" at Preferences... May it be the cause? Yes, it could be related - when you make changes on that page, every single setting on all three 'Preferences' pages get read and acted on. Look and see whether you've perhaps unticked "Do work ... while computer is in use" (processor usage tab). oh, and, the "restarted WU" shown in the last line, is restarted using the GPU, and none of the CPU cores are used :) Yes, I can see from the 'version 609' at the end :-) |
|
4412)
Message boards :
BOINC Manager :
No CPU core used while there are available WUs!!
(Message 30505)
Posted 12 Jan 2010 by Richard Haselgrove Post: I've not heard of Windows 7 causing any problems at SETI: there have been issues at Einstein, but they use an unusual application structure. One thing to check: do you have a file cc_config.xml in your BOINC Data folder, and if so, does it contain the line <ncpus>0</ncpus> in the Options section? If so, change it to <ncpus>-1</ncpus> (that's minus one) and restart BOINC. Otherwise, two more questions: 5) How long ago did you install v6.10.18, and has it ever processed SETI on CPU since then? 6) Could you please post the initial lines from the 'Messages' tab of BOINC after a restart, and any extra lines which show any sign of an error. |
|
4413)
Message boards :
BOINC Manager :
No CPU core used while there are available WUs!!
(Message 30497)
Posted 12 Jan 2010 by Richard Haselgrove Post: We can probably help you understand this better if you can give us more information. 1) Which (numbered) version of BOINC do you have installed? 2) Which project(s) are you attached to? 3) Which project(s) have work downloaded and available to run? 4) Which project is running a task on the GPU? |
|
4414)
Message boards :
Projects :
Orbit@home
(Message 30492)
Posted 11 Jan 2010 by Richard Haselgrove Post: Continuing from thread 3583 (could we have autolock disabled on this board, please?), Pasquale has posted Update & Plans for 2010. Highlights: For 2010 we plan to run our code on real data, primarily sky coverage data by real NEAs surveys, to produce two main products: |
|
4415)
Message boards :
Questions and problems :
Immediate BSOD when boinc.exe loads in XP on Mac
(Message 30483)
Posted 10 Jan 2010 by Richard Haselgrove Post: To which I would add: it would be helpful if you could get the details of the BSOD, at least the 'STOP' number and any associated data (such as any file name referenced). On ordinary PC hardware, you can call up a boot menu by pressing 'F8' after the BIOS initialises, but before any part of the Windows loader is visible on screen. I guess (hope!) that you can do the same thing on the Mac: you would have to select Windows as the operating system to load, and then immediately start tapping F8 until the boot menu appears. You may have to practice a bit to catch it in time. Once you have control at the menu, two choices (in particular) would be useful: 1) Disable restart after errors (or words to that effect). That would keep the BSOD on screen (just once, for the current session) so you could take notes. 2) Safe mode, which will stop the BOINC service loading and hopefully give you a stable desktop for further investigation. From here, you can right-click and choose 'Properties' for "My Computer" (start menu or desktop, depending on configuration). On the 'Advanced' tab of what is now showing as "System Properties", you can choose 'Startup and Recovery' settings to disable automatic restarts on system failure permanently. [Anyone care to translate that 'right-click' for the one-button Mac mouse environment?] |
|
4416)
Message boards :
Questions and problems :
Multiple CUDA cards, why no multiple CUDA processes?
(Message 30471)
Posted 9 Jan 2010 by Richard Haselgrove Post: You can overrule the default behaviour (use 'best' GPU only) with the <use_all_gpus> option. Details on the client configuration page. |
|
4417)
Message boards :
Documentation :
ATI Stream - wrong link
(Message 30463)
Posted 8 Jan 2010 by Richard Haselgrove Post: Thanks - the suggested page seems to be better then the current one. I had hoped it would be a Wiki entry, which I could edit - but it's a web php page, which I can't. It'll need Admin intervention, and as you say there's no 'notify Webmaster' link on those pages. The Admins rarely visit these message boards, so I'll try and email one to draw their attention. |
|
4418)
Message boards :
BOINC client :
Use all but primary ("this") GPU while computer is in use?
(Message 30372)
Posted 30 Dec 2009 by Richard Haselgrove Post: There's a recent feature which may help. From client configuration: <ignore_cuda_dev>, <ignore_ati_dev> |
|
4419)
Message boards :
Questions and problems :
Problem running 6.10.17 under Mac OSX 10.6.2
(Message 30365)
Posted 29 Dec 2009 by Richard Haselgrove Post: Any one else run into this or have any suggestions? Er, I think it means: "Project has no jobs available" From their front page news, dated today: We are now recovering from a planned power outage, during which all services were off line. The workunit storage machine is undergoing a RAID resync and no new workunits will be generated until this has finished. |
|
4420)
Message boards :
Questions and problems :
windows 7 Boinc wont install
(Message 29528)
Posted 17 Dec 2009 by Richard Haselgrove Post: Editing the HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion?\Explorer\User Shell Folders" with Data on others than local Drives or %USERPROFILE entries did the trick. thank you very much Good to hear it. Ticket updated. |
|
4421)
Message boards :
Questions and problems :
windows 7 Boinc wont install
(Message 29521)
Posted 17 Dec 2009 by Richard Haselgrove Post: Try the first workround from that trac ticket (bug report): Edit all entries in the registry: |
|
4422)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 29501)
Posted 16 Dec 2009 by Richard Haselgrove Post: It's very improbable, but could an AV security software of some kind be impairing inter process comms in this case? Highly unlikely. This is a home file/print server only, not used for any interactive tasks (I only access it by RDP): it has a very light-touch AV installed. As noted lower down the thread, this is a problem I have only seen with service installations: it's a BOINC v5, so I can run the exact same files in Manager/Client mode, and BOINC doesn't waste cycles then. The problem only occurs when there are security updates downloaded and waiting to be installed. Rebooting alone is not sufficient to clear it: installing updates and rebooting invariably does clear it. Okay, I've had BoincTasks up for 4 hours since upgrading to 0.32 with a refresh to monitor local and remote at 10 seconds intervals. It's used 13 seconds in that 4 hours, running on a quad. Did you check to see how much BOINC CC time was used servicing those BoincTasks data requests? |
|
4423)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 29499)
Posted 16 Dec 2009 by Richard Haselgrove Post: There are definitely many, reliable reports that BOINC Manager gets overwhelmed when the task list approaches 1,000 or 1,500 tasks. That would seem to be inevitable with a BM refresh rate - requesting RPC, parsing, sorting, displaying - of 1 second. There are also reports - perhaps fewer and less certain - of the core client itself clogging up under even heavier load - 5,000 or 6,000 tasks. Some of that overload has been attributed to the work involved in preparing, formatting and communicating the responses to RPC requests. That work will continue, whichever manager is used - though if BoincTasks can be configured to refresh RPC data less frequently, the demand on the core client will be correspondingly lower. But in the very rare and specific case I've been documenting in this thread, none of the above applies. 3 tasks, BOINC Manager is never opened (except for testing), and the only RPC requests come from BoincView once every 30 seconds. |
|
4424)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 29497)
Posted 16 Dec 2009 by Richard Haselgrove Post: BOINC (installed on a Windows Server, running as a daemon) is reminding me that it's time to install security updates - again. This time, I had a copy of Process Explorer handy:  Almost 50% CPU usage in ADVAPI32.DLL ??? What's that about? This is my home server, Windows 2000, single P4: one Einstein task running, one SETI task waiting to run (in memory), another Einstein ready to run. That's all. |
|
4425)
Message boards :
Questions and problems :
windows 7 Boinc wont install
(Message 29494)
Posted 16 Dec 2009 by Richard Haselgrove Post: Is this an upgrade to an existing BOINC installation, or are you trying to install BOINC for the first time on this machine? If it's an existing installation, where are your existing BOINC data files? BOINC itself will tell you in the block of messages it shows at startup. This is a bug which has been logged in the trac database for over a year ([trac]#815[/trac]), and I managed to work round a variant of it - but you probably need to find and change things in the Windows registry. Have you got experience of doing that? |
|
4426)
Message boards :
Questions and problems :
Windows 7 x64 no GPU detect
(Message 29480)
Posted 15 Dec 2009 by Richard Haselgrove Post: My eye always skips the stickies, because they so rarely change - took me some time to find it, too! You know where the trac tickets live - I believe 'enhancement' is an allowable category ;-) They could blink at different speeds, depending how long it was since the last post...... |
|
4427)
Message boards :
Questions and problems :
Windows 7 x64 no GPU detect
(Message 29477)
Posted 15 Dec 2009 by Richard Haselgrove Post: Have you read the "6.10: No GPUs Found" thread? Made clickable My eye always skips the stickies, because they so rarely change - took me some time to find it, too! |
|
4428)
Message boards :
Projects :
News on project outages
(Message 29406)
Posted 13 Dec 2009 by Richard Haselgrove Post: GPUGRID is offline from every aspect i've looked at.... switching to back up project collatz and seti for nvidia gtx 260 FOC. They had a powercut. They're back now, but warning of recovery problems - database server seems to be up and down every few seconds. |
|
4429)
Message boards :
API :
GUI_RPC and GPUs
(Message 29283)
Posted 6 Dec 2009 by Richard Haselgrove Post: But you only have to read it once. I don't expect my videocard/GPU to change every so many seconds, do you? So you'd only have to read it at app-start. I don't expect my CPU to change every few seconds, either. It just seems inconsistent to restrict get_host_info to CPU only in this day and age. It shouldn't be any more difficult to spit out the <coprocs> vector in response to a GUI_RPC, than it is when preparing a sched_request. |
|
4430)
Message boards :
API :
GUI_RPC and GPUs
(Message 29281)
Posted 6 Dec 2009 by Richard Haselgrove Post: Fred M and I are chatting with Jason about automating the flop-balancing for optimised SETI applications, while we're waiting for app_version specific TDCFs. (lunatics) GUI_RPC would be a more 'professional' way to go than scraping sched_requests - though knowing Jason, he'll probably end up querying the hardware directly. Having it in the GUI - I think leaving it out is just an oversight - would also expose it to remote managers like BoincTasks. |
|
4431)
Message boards :
API :
GUI_RPC and GPUs
(Message 29279)
Posted 6 Dec 2009 by Richard Haselgrove Post: Is it possible to retrieve information about installed GPUs via GUI_RPC? <get_host_info> only returns CPU data. So does <get_state>. |
|
4432)
Message boards :
Questions and problems :
Anyone have problems posting?
(Message 29218)
Posted 4 Dec 2009 by Richard Haselgrove Post: I'll give it a try.... Edit - no, no apparent problem here. |
|
4433)
Message boards :
Questions and problems :
Windows XP User Profile
(Message 29152)
Posted 2 Dec 2009 by Richard Haselgrove Post: I am not quite sure what you are talking about. Care to explain further or in better detail what you mean? Probably things like  or  showing up in the 'Documents and Settings' user area. |
|
4434)
Message boards :
Questions and problems :
Install problems on WIndows Server 2003 for Small Business
(Message 29104)
Posted 1 Dec 2009 by Richard Haselgrove Post: OK, guilty of speed-reading as charged. But you did mention a DC in the first sentence, and then provided a link.... And this thread specifically mentions Small Business in the thread title, which as previously mentioned is always a DC. |
|
4435)
Message boards :
Questions and problems :
Install problems on WIndows Server 2003 for Small Business
(Message 29101)
Posted 1 Dec 2009 by Richard Haselgrove Post: That post specifically says The server may be a domain member (but not a domain controller!). It's the word controller which matters here. |
|
4436)
Message boards :
Questions and problems :
Install problems on WIndows Server 2003 for Small Business
(Message 29095)
Posted 1 Dec 2009 by Richard Haselgrove Post: Yes, Small Business Server is always configured as a Domain Controller - I've set up several of them. |
|
4437)
Message boards :
Questions and problems :
cc_config.xml, unparsed tag
(Message 29085)
Posted 30 Nov 2009 by Richard Haselgrove Post: PS: And Richard didn't find the reason for your problem ;-) Got to leave some for everyone else! ;-) No, you're right: I missed it. Must be getting old. And well done for getting it yourself. |
|
4438)
Message boards :
Questions and problems :
cc_config.xml, unparsed tag
(Message 29078)
Posted 30 Nov 2009 by Richard Haselgrove Post: <cc_config> There may be hidden (non-printing) characters introduced by mistake during the editing process. You don't say which OS you use, or how you created the file: but going on the "90% use Windows" principle.... It's best to use the absolutely simplest editor, usually Notepad. Even then, I've found that the Notepad in Vista can hide errors which the Notepad in XP reveals. The text above looks 'clean': probably the easiest thing would be to start with a new (empty) file, and copy the text again from this message board - the board itself will have taken out non-printing characters. |
|
4439)
Message boards :
Questions and problems :
6.10.18 is now Recommended version for: Windows 2000/XP/Vista/7
(Message 29055)
Posted 28 Nov 2009 by Richard Haselgrove Post: Same problem with .18 as .17 on i7 920 running Vista 64 Ultimate. Checked status this morning - only one task running, instead of 8. GPU task running (GTX 250). Projects with work to run - AQUA, Einstein, SETI, WCG. Only Einstein was running. Known problem with both .17 and .18 - confirmed fixed in .19 Only happens because you're running AQUA - read the board there - or another multi-threaded application (if there is one). |
|
4440)
Message boards :
BOINC Manager :
High CPU Utilization for boincmgr.exe
(Message 29027)
Posted 26 Nov 2009 by Richard Haselgrove Post: I can do all that, but it sounds like my setup is the same as yours (even to the SP3). I take it you didn't see the spikes and extra threads that DigiK-oz had? So I'm on standby: can log if needed, but won't do a duplicate at this stage. |
|
4441)
Message boards :
BOINC Manager :
High CPU Utilization for boincmgr.exe
(Message 29020)
Posted 26 Nov 2009 by Richard Haselgrove Post: Just spent a few minutes looking at BOINC Manager v6.10.19 on WinXP 32-bit: didn't see any signs of excess CPU spikes, and it only had 4 threads. One thing struck me: if you open two copies of boincmgr, and change the layout in one of them, you could observe whether the second copy follows the first - either five seconds later, or when you close the first copy: that would confirm that it was re-reading the registry. Mine didn't, which might suggest a difference in either Win7 or 64-bit mode. |
|
4442)
Message boards :
BOINC Manager :
High CPU Utilization for boincmgr.exe
(Message 29012)
Posted 26 Nov 2009 by Richard Haselgrove Post: EDIT : Upon further investigation, when in advanced mode, boincmgr does hundreds of registry reads when the CPU spikes. In simple mode, this does not happen. Keys almost all start with: BOINC keeps all the grid layout information (column to sort on, column widths, etc.) for all the advanced view tabs there. A few reads at startup would be expected and legitimate, but it shouldn't need to reassure itself that nothing has changed several hundred times a second. |
|
4443)
Message boards :
Questions and problems :
Can't resolve hostname - bug
(Message 28991)
Posted 25 Nov 2009 by Richard Haselgrove Post: Sorry I don't know which directory it's in. It's "C:\Windows\System32\drivers\etc" on Vista32 (and that sounds very familiar from other NT-based Windows versions, too) |
|
4444)
Message boards :
Questions and problems :
Boinc Request GPU WU Only
(Message 28906)
Posted 21 Nov 2009 by Richard Haselgrove Post: BoincView has a (slightly clunky) way of creating and managing cc_config.xml files. It's a long time sine I used that particular feature (although I use BV itself constantly, I do the config files with notepad). IIRC, BV comes with a cc_config.xml template file, and if you edit even one value, you get default settings for all the others too. Since some tags have been renamed or deprecated since BV was written, you tend to get parsing errors (unrecognised tags) and defaults whose meaning has changed. Anyone round here been using BoincView? Ah - here's the culprit from BV's 'list_config.xml' file. Note that he's set a 'min' value: <item>
<tag>ncpus</tag>
<name>Override number of CPUs</name>
<text>Act as if there were N CPUs: run N tasks at once. This is for debugging, i.e. to simulate 2 CPUs on a machine that has only 1. Don't use it to limit the number of CPUs used by BOINC; use general preferences instead.</text>
<type>integer</type>
<default>0</default>
<min>0</min>
<max>255</max>
</item>
|
|
4445)
Message boards :
Questions and problems :
Boinc Request GPU WU Only
(Message 28889)
Posted 20 Nov 2009 by Richard Haselgrove Post: Have a good look at the messages BOINC puts out as it starts up. Often there's a clue in there, like "Preference limit number of CPUs to 0", in the case Jord mentions. |
|
4446)
Message boards :
BOINC client :
6.10.18 Bugs [to include a major bummer]
(Message 28879)
Posted 20 Nov 2009 by Richard Haselgrove Post: No, to be fair the GUI (was) (would have been, if I'd chosen to look) showing the settings that were in effect. The trouble is, those settings - for over four hours - bore no relation to my preferences, whether web preferences or local preferences. I'll leave it to somebody else to do the Account Manager tests! Ah - and of course, if I'd clicked 'OK' instead of 'cancel' to dismiss the GUI preferences dialog in the Manager, I'd have locked in the hardwired prefs set (including the unwanted settings on other tabs, like 'suspend GPU while computer in use') as my new local override set. |
|
4447)
Message boards :
BOINC client :
6.10.18 Bugs [to include a major bummer]
(Message 28876)
Posted 19 Nov 2009 by Richard Haselgrove Post: Yes! Almost a full six hours after 19-Nov-2009 15:57:51 [orbit@home] Detaching from project I finally got 19-Nov-2009 21:38:25 [CPDN Beta] [wfd] request: 0.00 sec CPU (0.00 sec, 0) NVIDIA GPU (0.00 sec, 0) 19-Nov-2009 21:38:25 [CPDN Beta] Sending scheduler request: To send trickle-up message. 19-Nov-2009 21:38:25 [CPDN Beta] Not reporting or requesting tasks 19-Nov-2009 21:38:30 [CPDN Beta] Scheduler request completed 19-Nov-2009 21:38:30 [---] General prefs: from http://orbit.psi.edu/oah/ (last modified 19-Nov-2009 15:36:13) 19-Nov-2009 21:38:30 [---] Computer location: school 19-Nov-2009 21:38:30 [---] General prefs: using separate prefs for school 19-Nov-2009 21:38:30 [---] Reading preferences override file and everything is back to normal - GUI shows 0.05 / 0.3, and I can use the mouse / keyboard without unloading CUDA tasks. And I can finally 19-Nov-2009 21:39:41 [SETI@home] Sending scheduler request: Requested by user. 19-Nov-2009 21:39:41 [SETI@home] Reporting 65 completed tasks, not requesting new tasks without interrupting the experiment. [OK. 'fess up time. Searching the log reveals 19-Nov-2009 20:05:21 [GPUGRID] Sending scheduler request: Requested by project. 19-Nov-2009 20:05:21 [GPUGRID] Not reporting or requesting tasks 19-Nov-2009 20:05:26 [GPUGRID] Scheduler request completed 19-Nov-2009 20:05:26 [SETI@home] General prefs: from SETI@home (last modified 04-Apr-2009 20:08:29) 19-Nov-2009 20:05:26 [SETI@home] Computer location: school 19-Nov-2009 20:05:26 [---] General prefs: using separate prefs for school 19-Nov-2009 20:05:26 [---] Reading preferences override file but I was out for a beer at the time, so it doesn't count.] |
|
4448)
Message boards :
BOINC client :
6.10.18 Bugs [to include a major bummer]
(Message 28871)
Posted 19 Nov 2009 by Richard Haselgrove Post: OK, second task switch completed successfully, so I think we have a winner. I sacrificed 5 minutes of CUDA work:  According to the log, the preferences override file hasn't been read since 15:54, so I've been running on hardwired for well over 2 hours. |
|
4449)
Message boards :
BOINC client :
6.10.18 Bugs [to include a major bummer]
(Message 28870)
Posted 19 Nov 2009 by Richard Haselgrove Post: Also, the fields in the local preferences GUI now show the hardwired defaults, matching neither my web preferences nor my override preferences. Normally, that page defaults to the set in use: for the purposes of this test, 0.01 days CI / 0.75 days additional (web, venue 'school'), or 0.05 days CI / 0.3 days additional (local override). After I detached Orbit, the GUI was defaulting to 0.1 / 0.25, matching that log from 7 June (I think - I don't actually want to touch the GUI again for at least another hour, to see how it handles AQUA task switches without "request reschedule CPUs - active state change", or that interruption to CUDA work. I deliberately chose the settings to potentially delay the 'Reading preferences override file' event for several hours, if my theory about it only happening at project update is correct). So far, it has handled the first few task switches from CPU to MT correctly, keeping all cores busy, although it has preempted an Einstein task after 4m 25s and a SETI (CPU) task after 2m 14s, both in favour of AQUA. |
|
4450)
Message boards :
BOINC client :
6.10.18 Bugs [to include a major bummer]
(Message 28868)
Posted 19 Nov 2009 by Richard Haselgrove Post: Installed v6.10.19, and carried out the 'detach from project' test - results much as previously described. I fiddled some general preferences at Orbit, updated that and then a few other projects to check that they replicated properly - they did, once I'd remembered to ensure the host was set on the right venue. Then, cleared local preferences (which had a smaller cache setting), refilled the cache according to venue settings, and re-established (restrictive) local prefs. Finally, detached from Orbit. Confirmed that global_prefs.xml file was deleted (though global_prefs_override.xml remains). Also, the fields in the local preferences GUI now show the hardwired defaults, matching neither my web preferences nor my override preferences. And to clinch it, the host suspended CUDA processing, although I always choose to allow CUDA while the computer is in use. Sure enough, CUDA restarted automatically 3 minutes later - it's surprising how long that is, when you're looking at the screen but can't allow yourself to touch the mouse! I've retreated upstairs, where I can monitor task switches etc. remotely (BoincView) without the computer being 'in use'. I'll examine the logs more closely when I next see that it's updated a project automatically. |
|
4451)
Message boards :
BOINC client :
What's the trick ?? Boincmgr handling GPU *AND* CPU WU's
(Message 28867)
Posted 19 Nov 2009 by Richard Haselgrove Post: This is a known problem, to do with Long Term Debt (or work fetch priority, as it is now known). If you look at the "properties" page for each of your CPU projects, you will probably find that each of them has a work fetch priority below -86,400 (seconds, = 1 day), and has the additional annotation "(overworked)". Under these circumstances, the projects are only contacted for a 'refill' when the work remaining in your cache drops below whatever setting you have for "Computer is connected to the Internet about every " - usually 0.1 days (2.4 hours), unless you've changed it. One cause of this problem has been identified, and the fix should be in the new version 6.10.19, which I started testing about 15 minutes ago. This is one of the things I was going to be checking anyway, so I'll report here what I find out - though in the nature of things, these several-day cache settings take - well, several days - to thoroughly test. In the meantime, be assured that current BOINC clients will indeed keep all your CPU cores busy, though it sometimes seems as if they leave it until the last minute before refilling. |
|
4452)
Message boards :
Questions and problems :
All of a sudden, no tasks running
(Message 28860)
Posted 19 Nov 2009 by Richard Haselgrove Post: Hey Grub- That was Gruß, the German for Greetings. That symbol at the end is another way of writing a long or double 's' (not sure of its official name, it's 40 years since I learned about it in school!) You have to be careful of things like that on a multi-cultural and intercontinental message board. |
|
4453)
Message boards :
BOINC client :
6.10.18 Bugs [to include a major bummer]
(Message 28856)
Posted 19 Nov 2009 by Richard Haselgrove Post: As you can see, it's re-reading the local preferences automatically at any detach. Well, it's re-reading at the next scheduler contact with another project, at least: might be a long time if you're running AQUA. (I was going to say CPDN, but the trickle upload might be enough). |
|
4454)
Message boards :
BOINC client :
6.10.18 Bugs [to include a major bummer]
(Message 28855)
Posted 19 Nov 2009 by Richard Haselgrove Post: That's why I posted (and luckily still had available) the Work Fetch Debug logs: that part, at least, was running on the hard-wired preferences until I reattached. Didn't look at the GUI while it was all going on: I only wanted to detach/reattach briefly so the server would know about the 100-odd tasks that had been trashed by a big bug in v6.6.34 a few days before. v6.10.19 is available for download (Windows only), and it should contain some important fixes I've been pestering for. I also want to do some serious research on debt accountancy in this version: one bug is among those fixed this time (subject to confirmation testing), but I suspect others remain. Part of my plan for starting debts with a clean slate includes detaching (permanently) from LHC and Orbit, since neither project is showing any signs of life. I'll watch to see how long it takes before it notices the prefs override file this time. |
|
4455)
Message boards :
Questions and problems :
Laptop crashed after BOINC install
(Message 28851)
Posted 19 Nov 2009 by Richard Haselgrove Post: I would also expect to have seen the output from a chkdsk Z: /F by now (or whatever your external drive is). |
|
4456)
Message boards :
BOINC client :
6.10.18 Bugs [to include a major bummer]
(Message 28850)
Posted 19 Nov 2009 by Richard Haselgrove Post: 2. When with multiple projects and local preferences entered and a project is detached, for whatever reason, the local prefs are lost/overwritten by the global prefs from another attached project (don't know which since 12 are attached]. The signs were that the LP connects was set to 0.01 and buffer 0.75. The values bacame 0.1 connect and 0.25 days buffer which may ring a bell. Network time changed from 12:45-13:00 to 00:00-00:00) This one at least isn't Windows 7 related. 07-Jun-2009 17:53:48 [---] [wfd] target work buffer: 864.00 + 0.00 sec 07-Jun-2009 17:54:38 [SETI@home] Detaching from project 07-Jun-2009 17:54:43 [---] [wfd] target work buffer: 8640.00 + 21600.00 sec ... (now this is odd) 07-Jun-2009 17:55:58 [SETI@home] [wfd] request: CPU (0.00 sec, 0) CUDA (30240.00 sec, 1) ... (before initialisation?) 07-Jun-2009 17:56:03 [SETI@home] Master file download succeeded 07-Jun-2009 17:56:08 [SETI@home] [wfd] request: CPU (0.00 sec, 0) CUDA (30240.00 sec, 1) 07-Jun-2009 17:56:08 [SETI@home] Sending scheduler request: Project initialization. ... (some comms errors) 07-Jun-2009 18:00:33 [SETI@home] New computer location: school 07-Jun-2009 18:00:33 [SETI@home] General prefs: from SETI@home (last modified 04-Apr-2009 20:08:29) 07-Jun-2009 18:00:33 [---] General prefs: using separate prefs for school 07-Jun-2009 18:00:33 [---] Reading preferences override file 07-Jun-2009 18:00:39 [---] [wfd] target work buffer: 864.00 + 0.00 sec Looking back at my report to boinc_alpha the following day ("target work buffer wrong after project detach"), that was with BOINC v6.6.31 running under Windows XP. I don't think it reverts to preferences from another project: I think those are hard-wired defaults in the client itself. |
|
4457)
Message boards :
Questions and problems :
All projects "waiting to run" after update to v.6.10.18
(Message 28825)
Posted 18 Nov 2009 by Richard Haselgrove Post: Thanks for you help, faster than lightning :) Don't mention it. Just happened to be passing at the critical moment, and it was one I recognised. |
|
4458)
Message boards :
Questions and problems :
All projects "waiting to run" after update to v.6.10.18
(Message 28823)
Posted 18 Nov 2009 by Richard Haselgrove Post: You possibly have a "cc_config.xml" file in your data directory, left over from previous experiments. And it possibly has the line <ncpus>0</ncpus> in it. That used to be the default for 'ignore me, use everything available'. Now zero means zero: use -1 (minus 1) for the default 'use everything' setting. |
|
4459)
Message boards :
Questions and problems :
UMTS / GPRS ... wrong kompressed bitmaps
(Message 28774)
Posted 15 Nov 2009 by Richard Haselgrove Post: The BOINC developers are aware of this problem. If you visit the BOINC preferences page for any project you're attached to, you can find an option Skip image file verification? Because this option exists, I don't think any project would be foolish enough to send crucial data needed for calculations in an image format (they mostly use .gz format for the important stuff). The bitmaps are purely for entertainment, so it doesn't matter if they get compressed. |
|
4460)
Message boards :
BOINC client :
6.10 not using all threads
(Message 28751)
Posted 14 Nov 2009 by Richard Haselgrove Post: It's been mentioned. While we're waiting for the next version to be built and tested, v6.10.13 works reasonably well and does use all threads. |
|
4461)
Message boards :
Questions and problems :
All my WU's end immediately with "Output file xx for task xx absent"...
(Message 28685)
Posted 12 Nov 2009 by Richard Haselgrove Post: Just a FYI in respect of the thread title: "Output file absent" is always a symptom, not a cause. The output file is absent because of some earlier crash (like the error messages you've also posted) - the application never gets a chance to do any useful work. |
|
4462)
Message boards :
BOINC Manager :
Rant-away.
(Message 28646)
Posted 10 Nov 2009 by Richard Haselgrove Post: In my part of the world, p y f o is a relatively mild exhortation to hurry up or get a move on. Thinking about it during my walk, and observing that the water level in my local canal has dropped 10cm since I walked that way last week, I remembered that it may have a rather more urgent connotation in the Dutch stereotype. Apologies if my English english idiom jangled a few cross-cultural nerves. In terms of product launches, I'm a great believer in 'launch when ready', but not before. Parhaps it dates back to the discipline I was put through with our own product launch..... We hired one of the ornate function rooms in a grand Victorian town hall (Wakefield, in our case), and hooked up a PC - probably a 286 or 386: this was 1992 - to the first big ceiling-mounted three-lens Barco projector any of us had ever seen. The project Director talked the great and the good - representatives of the national bodies who had pur up the development money which enabled the project to get off the ground in the first place - through the purpose and functions of the software: my role, as the one and only single-handed programmer, was to act as 'keyboards' for the day and press the right buttons. You do not want an error message to appear on the projection screen under those circumstances! But even with that level of preparation (and believe me, there was some thorough testing before we got anywhere near the Town Hall), we still expected problems and made sure we stood by the phone for the first few days after the initial orders and complementary copies were posted out. I've looked back through the original bug report forms I jotted down in the early days: "Data files [text] found to be in WordPerfect format - their printer settings" - "No access to LPT1: - uses Netware queue capture" - "Machine type unknown, but AT class - presume HD drive, but using double-density 96tpi 5.25inch disks - probably incompatible with drive" - "Not enough memory - config.sys installs non-standard RAM disk, which reports taking up 384KB". And so on. We probably learned more about the variety of computing environments in those few weeks than at any time before or since. And it was our job - my job - to make the software work in as many of those environments as possible. We accepted the responsibility to deal with the problems, even though many of them - as you can see from the list above - didn't stem directly from bugs in our software. And since the organisation I was programming for is stull running eighteen years later, we must have done something right. |
|
4463)
Message boards :
BOINC Manager :
Rant-away.
(Message 28637)
Posted 10 Nov 2009 by Richard Haselgrove Post: Oh dear. You won't mind changing that title to "mild exasperation", would you, while I go out for a walk to clarify my thoughts on the equivalence of 'politeness' and 'lying'? |
|
4464)
Message boards :
BOINC Manager :
Only one processor starts
(Message 28634)
Posted 10 Nov 2009 by Richard Haselgrove Post: Changeset 19522 might have fixed it. I'm ready to test when a new version is compiled - hasn't appeared yet. |
|
4465)
Message boards :
BOINC Manager :
Rant-away.
(Message 28633)
Posted 10 Nov 2009 by Richard Haselgrove Post: Please stay nice. If you know how to fix it, add the code yourself and let them know, else refrain from attacks. The nice version of that answer is available at Einstein - I find I write polite answers in the evenings, and grumpy answers in the mornings. Sorry, that's just the way I'm made :-) As you know, I reported this problem on 20 October, over a week before v6.10.17 was declared "recommended" on 29 October. I accept and appreciate that it isn't a show-stopper for the vast majority of BOINC users, and I can live with the decision to release regardless. But I have two comments on the process. 1) When a new version is released with known problems, even if they only affect a small number of users, I think there should be a caveat in the release notes: "Vx.yy.zz released and recommended. There's work still to do in [some obscure area] - hold off upgrading if you run [some special case]" 2) There is a very strong human tendency - I know, I've written software and had it released to the public - to ship it out of the door, heave a great sigh of relief, and turn your mind to something else - anything else - to recharge your batteries before the next bout of coding. But I've had to try and resist that tendency in my own work: the hours or days after a software release are the time when the new code is exposed to the widest spectrum of new environments in the shortest time, and almost inevitably untested wrinkles will reveal themselves. As far as possible, the helpdesk and debug teams should be geared up to handle this entirely predictable workload bulge. |
|
4466)
Message boards :
BOINC Manager :
Only one processor starts
(Message 28629)
Posted 10 Nov 2009 by Richard Haselgrove Post: It'll be the AQUA. BOINC has that problem currently with multi-threaded applications (see the AQUA message board). Downgrade to v6.10.13, or put up with it until the BOINC developers have pulled their finger out and fixed the (reported) problem. |
|
4467)
Message boards :
BOINC Manager :
How to delete a project
(Message 28471)
Posted 3 Nov 2009 by Richard Haselgrove Post: Is this one where Account Manager overrides kick in? If you attached via an account manager, I believe you have to detach the same way. |
|
4468)
Message boards :
Questions and problems :
A Question that's been bothering me
(Message 28461)
Posted 2 Nov 2009 by Richard Haselgrove Post: We had an intersting case at SETI today, in going green...well sorta.. - someone whose electricity bill had gone down when they started crunching for BOINC, even though the weather was colder than last year and the price of units higher. Turned out his PCs were more efficient heaters than his heater.... The condo I live in is 30+ years old. This system just has big wire heating coils that get air blown over them. So the computers waste heat as a heating device is probably more efficent... So if we could just get all those big ole wire coils ripped out of 30 y.o. condos.... |
|
4469)
Message boards :
Questions and problems :
BOINC does show Running CPU benchmarks message
(Message 28389)
Posted 30 Oct 2009 by Richard Haselgrove Post: 'Running benchmarks' and 'Starting computing' are two different functions. Because you've stayed with the same version of BOINC (there's no message about a version change in your log), BOINC is happy to continue with the benchmark results from last time. You can leave it like that until the next time benchmarks are scheduled to be run (which will be within the next five days), or you can request a manual run from the 'Advanced' menu in BOINC Manager. Either way, it'll make no difference to whether computing starts. The only project you're attached to is SETI@home. Could it be something as simple as making the move at a time when SETI is very low on available work? - the Arecibo telescope is undergoing maintenance, and it's taken longer then expected to get alternative data back from SETI's off-site archive. BOINC could have asked for new work, failed to get any, and backed off its server requests to save hammering the project servers. Look at the 'Projects' tab in BOINC Manager. Select SETI, and click the 'Properties' button. If communication is backed off, try updating the project - new work is available, but not much of it, and it may take a few attempts before you strike lucky. |
|
4470)
Message boards :
Questions and problems :
GPU projects
(Message 28371)
Posted 29 Oct 2009 by Richard Haselgrove Post: Is there currently any work for einstein, seti or hydrogen on CUDA? Einstein: CUDA only, very early Beta, needs to be installed manually, doesn't make much use of GPU. SETI: Production CUDA application, temporary shortage of work, hopefully more later today or next week at worst. SETI Astropulse: very early third-party ATI Beta, in closed testing only. Hydrogen: ??? |
|
4471)
Message boards :
Questions and problems :
exclusive app for gpu use
(Message 28351)
Posted 27 Oct 2009 by Richard Haselgrove Post: Perhaps an anhenacemnt request^^. As the originator of ticket #842, I resent that remark. I've come to expect that behaviour from David Anderson, but not you, Jord. Please go and re-read [trac]#842[/trac]. It started with two (related) enhancement requests: then a third was added - an enhancement to the enhancement, if you like. The first two requests were for user controls. It's only the third request which would require any auto-detection. Yet David has used the difficulty of #3 as an excuse for throwing #1 and #2 in the trash. It's a classic case (all too common in BOINC-land) of speed-reading and a LIFO in-tray. Only deal with the latest isuue (usually a symptom), don't bother to go back to the beginning and understand what the conversation is really about. Learning point: when talking to David Anderson, only talk about one thing at a time. Don't get sidetracked, don't waver from your original point at issue. If you ask a supplementary, you lose the whole argument. mahdia is asking for enhancement #2 from that trac ticket. David's answer gives no reason for ruling it out. </rant> |
|
4472)
Message boards :
Questions and problems :
DCF Integrator
(Message 28305)
Posted 24 Oct 2009 by Richard Haselgrove Post: Well I made the following changes in my app_info and we'll see how it goes. That sounds plausible: maybe even still too low. Those original calculations were made in March, remember: since then, we've had first Cuda 2.2 and later Cuda 2.3 For SETI, specifically (and this does not apply to other BOINC projects I've tested), each Cuda runtime upgrade improved speed by at least 30%. You could even need x 0.4 if you're running the 2.3 DLLs. But don't sweat it - it is impossible to reach a mathematically-perfect set of multipliers, because the ratio of the speed of the different applications is different for different datasets (ARs, in SETI terminology). Just set something which keeps things broadly under control, and let BOINC manage its own affairs from there. |
|
4473)
Message boards :
Questions and problems :
Does boinc utilize multi core cpus well?
(Message 28277)
Posted 23 Oct 2009 by Richard Haselgrove Post: I was moved to try out one or two of the other common European languages. It's still there in Franglais: 
|
|
4474)
Message boards :
BOINC client :
Client ohly running on task on Duo cpu machine.
(Message 28274)
Posted 23 Oct 2009 by Richard Haselgrove Post: Yes. I put in a bug report for v6.10.13, complaining that that version was too quick to pre-empt single threaded tasks in favour of multi-threaded (AQUA) tasks, to the extent that AQUA tasks were often starting the second they'd finished downloading. David Anderson put in a quick fix before he left for Europe and The 5th Pan-Galactic BOINC Workshop, but the cure is worse than the disease. As I started to type this post, I was watching a quad-core running a CUDA task, one CPDN Beta, and nothing else. [It's now switched fully to AQUA, which at least keeps (most of) all four cores occupied]. David needs to have another go at this when he gets back: in the meantime, I prefer v6.10.13, even though it makes AQUA look a bit of a bully. See also Weird Behaviour when AQUA is suspended on the AQUA boards. |
|
4475)
Message boards :
Questions and problems :
Does boinc utilize multi core cpus well?
(Message 28271)
Posted 23 Oct 2009 by Richard Haselgrove Post: BTW, why does the word "between" appear twice on the first line of that screenshot? It was a straight grab from v6.10.16, newly installed today. |
|
4476)
Message boards :
Questions and problems :
Does boinc utilize multi core cpus well?
(Message 28270)
Posted 23 Oct 2009 by Richard Haselgrove Post: The value is integer, meaning that 99% on a dual core system means 1 CPU. Just remember that. I wonder if it would help if BM could be trained to display that integer value for the current setting:  [Edit - spelling!] |
|
4477)
Message boards :
Questions and problems :
Moving Project Files to Different Drive
(Message 28263)
Posted 23 Oct 2009 by Richard Haselgrove Post: Actually, just 'Detach' doesn't report the tasks as available for new users. That only happens if you re-attach - which rather defeats the object of the exercise here. If you want to sort things out quickly, best to set NNT, then abort and report the new tasks. When you're ready to switch back to the old folder, it'll be best to copy the old BOINC Data folder to the new partition, and then re-run the BOINC installer to ensure that all the permissions are set correctly. |
|
4478)
Message boards :
Questions and problems :
DCF Integrator
(Message 28249)
Posted 22 Oct 2009 by Richard Haselgrove Post: Now my question is. After all this work to get the DCF down near the 0.2 range, why is that the desirable value? My fault. I chose it. No single overwhelming reason, really. I do like projects which (slightly) over-estimate their initial running times, and hence slowly edge the DCF downward. That means you start by downloading relatively few WUs, and gradually download more as BOINC gains confidence. Compare with a project which under-estimates the initial running time: you download too many before you know what's going on, and then have a struggle to complete them in time. Also, IIRC, at the time I first chose the figure, the stock SETI application running on a fairly standard Core2 processor under Windows tended to settle at around DCF=0.2 (optimised AK_V8 could reach 0.1, and stock AP was about 0.4). It seemed best to choose a figure which would seem not too outrageous compared with what people had seen before, and hence avoid frightening the horses. So that figure is SETI-specific, and related to the general downward drift in DCF as the stock application has improved, and as CPU architectures have improved. If I was starting with a clean sheet and none of that historical baggage, I'd probably have targeted 0.8 |
|
4479)
Message boards :
Questions and problems :
boinc.exe starting problem
(Message 28237)
Posted 22 Oct 2009 by Richard Haselgrove Post: Windows Updates simply waiting to install. Yes, that's exactly the situation I've been observing - sporadically, not every month. But then, I don't update servers every month, so the problem may only manifest itself once a critical mass has been reached. |
|
4480)
Message boards :
BOINC Manager :
BOINC Manager Won't Connect to Localhost
(Message 28227)
Posted 22 Oct 2009 by Richard Haselgrove Post: The thing here is that it's not a double starting BOINC Manager that's the problem, but a double starting boinc.exe Yes, but according to the book boinc.exe is never started twice, ever. So it's already doing something not anticipated by the BOINC developers. One possibility is a user-created shortcut to boinc.exe somewhere. Best to spend just a few seconds checking for that remote possibility. Always expect the unexpected. |
|
4481)
Message boards :
BOINC Manager :
BOINC Manager Won't Connect to Localhost
(Message 28225)
Posted 22 Oct 2009 by Richard Haselgrove Post: Can you double check in HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Run how many entries there are for boincmgr.exe /a /s ? Also might be a good idea to look in the Startup folders (both your own, and the 'All users' one), to see if there's anything left over from previous versions in there. Or even go the whole hog and look at all the possible auto-run startup options with msconfig. |
|
4482)
Message boards :
Questions and problems :
DCF Integrator
(Message 28204)
Posted 21 Oct 2009 by Richard Haselgrove Post: Aw shucks. Just as I'd finished looking up the first appearance of the flopcount in print. But it's still relevant to make the point I was going to make. DCF is a safety valve. It's designed to react quickly when things go wrong, and to reset gradually as they settle back down. But there are problems when, as with SETI at the moment (and ever since that first post on 8 March), where three different applications with vastly different characteristics are fighting for control of the same safety valve. The remedy is well known, and has been accepted in principle by the developers: separate DCF values for each app_version. But that's been fighting for developer time with every other item on every wish list on every project. In the meantime, there is a mechanism which can be fine-tuned (even coarse-tuned will do) to help smooth things over pro tem. But it does require care and attention to detail. There are step-by-step posts by MarkJ and Pappa on the main SETI board: use those, not the values in that Beta post I linked - life was different then. And don't be afraid to post here - but with actual figures, copied and pasted from your computer, please: just re-typing what you believe the figures to be can add even more confusion. |
|
4483)
Message boards :
Questions and problems :
DCF Integrator
(Message 28192)
Posted 21 Oct 2009 by Richard Haselgrove Post: My flops estimates are correct as all predicted crunch times are within 60 seconds of actual. Including the predicted crunch times of the VLAR work on the CPU's. One of my computers just completed a VLAR CPU work unit. The cache size instantly doubled the estimated time to crunch the cache. These two statements are incompatible. Either the flops estimate in your app_info.xml file is accurate, in which case the cache estimate will remain unchanged on completion: or the cache estimates doubled because of a DCF change, in which case your flops value is wrong (by definition, because the sole purpose of the flops value is to normalise DCF). |
|
4484)
Message boards :
Questions and problems :
Does boinc utilize multi core cpus well?
(Message 28191)
Posted 21 Oct 2009 by Richard Haselgrove Post: I'll be installing Win7 soon enough, but in the meantime I'd like to know why AQUA@Home does not recognize my dual core CPU. Work from the bottom up. First, does Windows recognise the dual cores? Open Task Manager, Performance tab. How many CPU usage graphs do you see? If fewer than two, you need to sort that out first - could be a BIOS setting. Second, if Windows is OK, does BOINC recognise the dual cores? Find the messages listed by BOINC at startup (or restart BOINC and get a new set). How many CPUs, any sign of a "Preferences limit....."? If in doubt, post the messages here and ask for help in interpretation. |
|
4485)
Message boards :
Questions and problems :
boinc.exe starting problem
(Message 28140)
Posted 20 Oct 2009 by Richard Haselgrove Post: Hi Cleaner, I've just been round an update cycle on three machines with Win XP, BOINC installed as a service: 1) Set BOINC service start to 'disabled' (but leave it running) 2) Install Windows updates - all offered 3) Allow Windows Update to restart computer 4) Install BOINC upgrade while idle, v6.10.13 --> v6.10.15, settings unchanged 5) Open BOINC Manager to re-start daemon All worked OK, no error messages - but v6.10.15 is seriously messed up for multi-threaded tasks, the cure is worse than the disease. Will prepare logs, screenshots and write-up for boinc_alpha. |
|
4486)
Message boards :
Questions and problems :
boinc.exe starting problem
(Message 28129)
Posted 20 Oct 2009 by Richard Haselgrove Post: It sounds similar to an experience I documented in BOINC cc using excess CPU as service, and I've seen a couple of times since. It seems only to affect servers, and only to affect BOINC itself if installed as a service. In that thread, I showed that the same installation of BOINC, when started by BOINC Manager as a user program, behaved absolutely properly, but when started as a service (from the service control panel, services.msc) hit that excess cpu problem, and everything follows from that. My experience was with v5.10.45 - can't go above that on a domain controller - but you make it sound the same issue is still in v6.6.36 The only thing that's cured it for me is to reboot the computer - not always easy to schedule on a server, but it has to be done sometime to install security patches. In fact, I see some correlation with automatic updates being downloaded and ready to install: that might explain why all four of your machines developed the same behaviour at around the same time. 'Update Tuesday' was last week. |
|
4487)
Message boards :
Questions and problems :
Not getting new work automatically
(Message 28077)
Posted 17 Oct 2009 by Richard Haselgrove Post: Given that they've just made the ABP1 v3.11 application the mainstream release that will be downloaded by everyone, there's no need to run an app_info.xml file at all (why did you go down that route in the first place?). Assuming you don't want to continue with it, the advice is as you quoted from Gary: Run down your cache (set 'No New Tasks' to be on the safe side). Delete app_info.xml Reset the Einstein project Restart BOINC Allow new tasks Relax |
|
4488)
Message boards :
Questions and problems :
Not getting new work automatically
(Message 28064)
Posted 17 Oct 2009 by Richard Haselgrove Post: I'm pretty sure this is just an Einstein server quirk. Einstein went through a fairly intensive phase of software testing (many months ago), and lots of people were using Beta applications with app_info.xml files. Then, they got settled on a stable set of production applications, and wanted to wean people off the Betas. So they set that 4-hour comms backoff, requested by the server, for everyone using anonymous platform. Now they are testing a new set of Beta apps, but haven't taken off the 4-hour backoff. It really doesn't matter, because 1) Einstein tasks take more than 4 hours on all but the most exotic hardware. 2) You can cache more than one task anyway. 3) It's a bad idea to be using anonymous platform just at the moment, because they're in the transition to S5R6 and you won't get any. 4) If all else fails, a manual update still works before the end of the 4 hours. |
|
4489)
Message boards :
BOINC Manager :
BOINC Manager Won't Connect to Localhost
(Message 27990)
Posted 14 Oct 2009 by Richard Haselgrove Post: Wow, that's a comprehensive list. Congratulations on your perseverance! There's one slight doubt that's been nagging me reading these series of problems about connecting to localhost. Have we thoroughly and completely distinguished between two cases: 1) There is a core client running, but BM can't connect to it. 2) The core client is not running, so there's nothing to connect to. Case (1) might point to a firewall issue: alternatively, I've sometimes seen BOINC.EXE spending so much time examining its own navel that it forgets to listen for BM RPCs (usually on servers which haven't been restarted since Update Tuesday - which reminds me....) But I'm beginning to be suspicious about case (2). Could you bear to go back to the machine (without being tempted to throw a brick through the screen), and use Task Manager to examine what BOINC or boinc-project tasks might be running, and how much CPU they're using? |
|
4490)
Message boards :
Questions and problems :
BoincTasks alternative BOINC manager
(Message 27955)
Posted 13 Oct 2009 by Richard Haselgrove Post: Sorry, see no log entry, nor a field which can be set to blank or zero. Is that in cc_config.xml or global_prefs/global_prefs_override.xml where that field is stored? I think he means in BoincTasks. |
|
4491)
Message boards :
BOINC Manager :
Best cards for CUDA
(Message 27945)
Posted 12 Oct 2009 by Richard Haselgrove Post: I read in a computer magazine that adding a second video card in SLI mode will not double the performance. You would get about 20-30 percent improvement. With two same GeForce video cards in SLI mode, you would get one cuda credit. If you disable the SLI then you will get two cuda credits. No longer true with the newer 190.xx and later drivers. |
|
4492)
Message boards :
Questions and problems :
BOINC won't use my ATI GPU
(Message 27842)
Posted 9 Oct 2009 by Richard Haselgrove Post: 9.10.2009 16:34:27 Milkyway@home Sending scheduler request: To fetch work. You've done your bit. As to why you got nothing back..... dunno. That's a project question, not a BOINC question. I suggest you ask on the project's forum. |
|
4493)
Message boards :
Questions and problems :
BOINC won't use my ATI GPU
(Message 27840)
Posted 9 Oct 2009 by Richard Haselgrove Post: 9.10.2009 15:32:45 Starting BOINC client version 6.6.38 for windows_intelx86 BOINC v6.6.38 doesn't have ATI support. You need to be running v6.10.xx to use ATI. v6.10.13 from the download - all versions page seems pretty stable so far, and probably the best bet. |
|
4494)
Message boards :
Questions and problems :
BoincTasks alternative BOINC manager
(Message 27833)
Posted 8 Oct 2009 by Richard Haselgrove Post: Regrettably no, the Help Cure Muscular Dystrophy - Phase 2 6.14 as it shows in BOINC manager is listed in BT both for active and the contracted Ready to Start as 6.14 Beta Testing - Help Cure Muscular Dystrophy - Phase 2 They've done that at Einstein too. |
|
4495)
Message boards :
Questions and problems :
Cuda on Geforce 9400
(Message 27776)
Posted 5 Oct 2009 by Richard Haselgrove Post: My BOINC... Which BOINC would that be? Version number? How installed? [Edit - removed reference to Vista/Win7. Must learn to read!] |
|
4496)
Message boards :
Projects :
News on project outages.
(Message 27744)
Posted 4 Oct 2009 by Richard Haselgrove Post: SETI@home seems to be back (they must have rescued Matt from under that server), but things will be slow and flakey for a while. |
|
4497)
Message boards :
Questions and problems :
Boinc links
(Message 27715)
Posted 1 Oct 2009 by Richard Haselgrove Post: Rather than just re-installing BOINC, there are some othe tests you could try: 1) Stop BOINC entirely, then restart without reinstalling 2) Re-boot the computer (assuming this doesn't get in the way of your other work), then retry. 3) Use Task Manager to see what else might be getting in the way - look in particular for applications which use a huge amount of memory. BTW, which projects (apart from MilkyWay) are you running? And how much memory (RAM) do you have in the computer? |
|
4498)
Message boards :
Questions and problems :
CUDA enabled projects list?
(Message 27696)
Posted 1 Oct 2009 by Richard Haselgrove Post: AQUA have temporarily suspended development of their CUDA apps (which were much slower than their CPU apps). They show no sign of releasing new work targeted at CUDA, so the only stuff available will be the occasional (very rare) resend. They plan to start testing a new CPU app 'next week' (it's been 'next week' for several weeks now), and at that point will cancel all existing work from the database, so even the limited CUDA resends will stop. Whether/when they will restart CUDA testing is anybody's guess. |
|
4499)
Message boards :
BOINC Manager :
Can't get Attach to Project to work
(Message 27656)
Posted 29 Sep 2009 by Richard Haselgrove Post: That was a lucky guess on my part ;-) But a nasty bug on BOINC's part - request a new copy of all_projects, and stuff whatever HTML comes back into a file, with no sanity check. Somebody could program a spoof there. |
|
4500)
Message boards :
BOINC Manager :
Can't get Attach to Project to work
(Message 27652)
Posted 29 Sep 2009 by Richard Haselgrove Post: The list of projects that should appear in that drop-down is a file - all_projects_list.xml a) Do you have a file of that name in the BOINC Data directory on your laptop (with the problem)? If so, what's the size/datestamp? b) If found, take backup of dodgy file for later analysis c) Likewise, do you have the file on the working desktop? Size/datestamp? d) Transfer file from working machine to problematic machine. Does it help? e) If yes, please report differences between good file and bad file. |
|
4501)
Message boards :
Questions and problems :
boinc sometimes will only request GPU work.
(Message 27636)
Posted 28 Sep 2009 by Richard Haselgrove Post: Agreed - this is going to be a chicken-and-egg alpha test. I'm still working my way through the double negatives, but I think the code as written will 'ping' legacy (CPU-only) BOINC servers for GPU work as before. If that checks out after more coffee, I think I may write DA to suggest no_cuda and no_ati default 'true' in the final (release) client. Can't do that yet, of course, or we'll get no work to test from current servers! But it will be easier to monitor and confirm that all the active GPU-enabled projects have updated, and change the defaults then, than to persuade dormant projects to update their server code. LHC and Orbit, anyone? Talking of updated server code, the newest I've seen is AQUA: check out the new task listing. |
|
4502)
Message boards :
Questions and problems :
boinc sometimes will only request GPU work.
(Message 27634)
Posted 28 Sep 2009 by Richard Haselgrove Post: David Anderson has been busy over the weekend: changeset [trac]changeset:19194[/trac] - client/scheduler/web: add per-project preferences for whether to accept CPU, NVIDIA and ATI jobs. It's not in any downloadable version yet (unless someone is willing and able to make a private build), but it sounds good to me: I imagine the participants in this conversation will be queuing up to test it once there's an alpha build and a project with the necessary server code installed. |
|
4503)
Message boards :
Questions and problems :
boinc sometimes will only request GPU work.
(Message 27622)
Posted 26 Sep 2009 by Richard Haselgrove Post: Still, I've one unanswered question in particular, why, in the middle of a 'CPU fetch' routine did I get a Fetch ATI GPU call (to a project with no GPU support) on a 9800GT system? That seems at best just silly. Almost certainly just a random backoff timer decrementing to zero - one of the twice-daily pings to see if they've written an ATI app in the last 12 hours :-) But an ATI ping on an nVidia system? You've got me there. Sounds like a bug report to boinc_alpha is indicated if you can track down the circumstances. [In particular - and being serious, for a moment - someone needs to enable [wfd] or [sched_op_debug], to see if it's really asking for work or just issuing a 'request' for 0.00 seconds, like the Einstein case I documented in April] |
|
4504)
Message boards :
Questions and problems :
boinc sometimes will only request GPU work.
(Message 27616)
Posted 26 Sep 2009 by Richard Haselgrove Post: Interesting explanation -- so, (I think this is what you are saying) when a particular workstation 'senses' it needs more CPU work, it broadcasts its request to each and every attached project until it completes its mission. The thing is, I had more than those two projects on that workstation including a couple of CPU only projects which were going 'short queue' and they were not pinged. Instead, two GPU only projects were pinged. Did you pick up on the reference to "overworked" (analogy: too much fish)? If any of your CPU projects are in this state, they won't get "pinged" for work until the last possible moment - about five minutes before the last CPU task is expected to finish. If you don't want to go down the full "work fetch debug" route, you can get a snapshot of what's going on by looking at the 'projects' tab of BOINC Manager (advanced view), and clicking the 'properties' button for each project in turn. You'll see 'Work fetch priority' for both CPU and GPU. If these are zero, it'll ping for work in the first cycle (unless deferred: that will also be shown). If the priority is negative but small, it'll ping for work as soon as the obvious candidates have been tested and failed. If it's below about -100,000, it's "overworked" and won't be pinged until the last moment. You may be interested to know that the parallel discussion with Paul Buck started on the boinc_dev mailing list has unearthed some interesting insights. I had assumed, from observation, that the current 'use CPU' and 'use GPU' settings were being applied on project servers only, and not transmitted to clients. It turns out that they are being sent out to the clients, but included in a "project specific" bundle with things like screensaver colours and movement rates, which BOINC itself doesn't get involved in but simply passes on to the project science application. Largely as a result of this conversation, I have argued that more settings should be moved to a grouping analogous to "Resource Share": known and understood by the BOINC client across all projects, but with individual values applied separately to each individual project. That seems to have found favour: I agree. I'll do this soon (just the "no CPU" and "no GPU" prefs for now). -- David Don't hold your breath - it'll be complicated, and take a while - but there is hope on the horizon. As to pinging the server extra times -- some day you might tramp thru the swamp of traffic over in SETI regarding Server I/O issues and that does to that project -- one which has far more resources than most other projects. Been there, done that. Did you see my post in Technical News yesterday, reporting 6,377 upload failures on 1,336 finished tasks over a 102 hour period? :-P |
|
4505)
Message boards :
Questions and problems :
boinc sometimes will only request GPU work.
(Message 27599)
Posted 26 Sep 2009 by Richard Haselgrove Post: Have you ever tried using "Work Fetch Debug"? You get something like this - it's an old one, but still instructive, and you'll see how much information you can get. 07-Jun-2009 17:56:08 [---] [wfd] ------- start work fetch state ------- 07-Jun-2009 17:56:08 [---] [wfd] target work buffer: 8640.00 + 21600.00 sec 07-Jun-2009 17:56:08 [---] [wfd] CPU: shortfall 0.00 nidle 0.00 est. delay 0.00 RS fetchable 200.00 runnable 200.00 07-Jun-2009 17:56:08 [climateprediction.net] [wfd] CPU: fetch share 0.00 debt 0.00 backoff dt 0.00 int 0.00 (no new tasks) 07-Jun-2009 17:56:08 [CPDN Beta] [wfd] CPU: fetch share 0.00 debt 0.00 backoff dt 0.00 int 86400.00 (no new tasks) 07-Jun-2009 17:56:08 [Einstein@Home] [wfd] CPU: fetch share 0.50 debt -51614.93 backoff dt 0.00 int 120.00 07-Jun-2009 17:56:08 [lhcathome] [wfd] CPU: fetch share 0.00 debt 0.00 backoff dt 79785.40 int 86400.00 07-Jun-2009 17:56:08 [SETI@home Beta Test] [wfd] CPU: fetch share 0.00 debt -244776.66 backoff dt 0.00 int 7680.00 (no new tasks) (overworked) 07-Jun-2009 17:56:08 [SETI@home] [wfd] CPU: fetch share 0.50 debt 0.00 backoff dt 0.00 int 0.00 07-Jun-2009 17:56:08 [---] [wfd] CUDA: shortfall 30240.00 nidle 1.00 est. delay 0.00 RS fetchable 100.00 runnable 0.00 07-Jun-2009 17:56:08 [climateprediction.net] [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 0.00 int 86400.00 (no new tasks) 07-Jun-2009 17:56:08 [CPDN Beta] [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 0.00 int 61440.00 (no new tasks) 07-Jun-2009 17:56:08 [Einstein@Home] [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 2679.27 int 3840.00 07-Jun-2009 17:56:08 [lhcathome] [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 37745.51 int 86400.00 07-Jun-2009 17:56:08 [SETI@home Beta Test] [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 0.00 int 15360.00 (no new tasks) 07-Jun-2009 17:56:08 [SETI@home] [wfd] CUDA: fetch share 1.00 debt 0.00 backoff dt 0.00 int 0.00 07-Jun-2009 17:56:08 [climateprediction.net] [wfd] overall_debt 0 07-Jun-2009 17:56:08 [CPDN Beta] [wfd] overall_debt 0 07-Jun-2009 17:56:08 [Einstein@Home] [wfd] overall_debt -51615 07-Jun-2009 17:56:08 [lhcathome] [wfd] overall_debt 0 07-Jun-2009 17:56:08 [SETI@home Beta Test] [wfd] overall_debt -244777 07-Jun-2009 17:56:08 [SETI@home] [wfd] overall_debt 0 07-Jun-2009 17:56:08 [---] [wfd] ------- end work fetch state ------- In your case, you'll have had a CPU shortfall and no CUDA shortfall - the other way round - but the principle is the same. What it's doing is saying 'my CPU is hungry - where can I get some nibbles?', and then going through *every* project to consider whether it's allowed even to ask. It *won't* ask if comms are deferred, NNT set, or backoff dt > 0.00 (mybe others too). Anything else is 'fetchable'. So, it tried Collatz. Any nibbles? No? *shrugs and moves on*. It tried GPUGrid. Any nibbles? No? *shrugs and moves on*. And so on, in debt order. It'll keep asking until it finds one with the right sort of nibbles - though there is the concept of "overworked" ('oh no, not fish *again* - I'm fed up with fish - I won't ask there until I'm really starving') Think of it as a shambolic bear of very little brain. It'll keep turning over the trashcans, even the ones that didn't have any food scraps yesterday, and won't have any tomorrow. Meanwhile, back with BOINC, if there is no CUDA shortfall (cache full), you won't see any CUDA requests. The question is, does any of this matter? Rom's famous "The evils of returning results immediately" blog implied that the database overheads involved in processing a scheduler RPC were high, and to be avoided weherever possible. But that was almost three years ago, and a lot of code has flowed under the bridge since then. Maybe the simple 'Got CUDA? No." interaction can happen without touching the user / host / workunit / result tables in the database (though I doubt it - at least the RPC count and 'last contact' fields for the host get updated). As I said yesterday, it's really a question for the projects: can their servers handle four extra 'pings' per host, per day? If that question has been asked, and answered in the negative, then it really shouldn't be a problem for us. But I suspect it hasn't been asked. |
|
4506)
Message boards :
Questions and problems :
boinc sometimes will only request GPU work.
(Message 27580)
Posted 25 Sep 2009 by Richard Haselgrove Post: BarryAZ, Sorry if I came over a bit heavy last night - it was my last post before going to bed, and I was probably in too much of a rush and more tired than I realised. One of the problems with BOINC development recently (as some of us see it) has been the "dumbing down" of the default message log. It used to display the exact number of seconds of work requested: then for a while it just said "to request work": finally, after much pressure from mo.v of CPDN and others, we got the present indicator for which resource (CPU, CUDA, ATI) is being 'topped up'. I have been trying to explain that the behaviour is much older, and it was only the information that was (re-)introduced with v6.6.36 (message 26309, 25520): I even used that line "the only thing you have gained by downgrading is ignorance" back in June! So to see it described as a 'bug ... starting with the 6.6.36 client' touched a raw nerve. Again, I'm sorry if it came over too heavy, but there's a danger that the 'bug' could become an urban myth and divert attention from the more fundamental re-working required in the client. For what it's worth, I've also expressed concern about meaningless 'pings' to project servers: message 24379, trac [trac]#896[/trac]. But we really need feedback from project administrators who are prepared to examine their server logs, to see if the extra traffic from inappropriate work requests makes a significant difference to their server workloads. If it doesn't concern the projects, then - however messy it looks - it needn't concern the users either. The theory is that work requests for inappropiate resources rapidly back off to an average of two per day from current clients, and with recent code changes (not yet widely deployed) that can be backed down to one request every 28 days by pro-active configuration changes to project servers. Given that I've recently come across several project admins who don't yet understand or know about the existing configuration tools (see the discussion about 'resend lost results' at GPUGrid!), I think that adding yet more new tools is a triumph for hope over expectation, but we'll have to see how it develops. The 'work fetch backoffs' for active resources are reset to zero when a task finishes, but only for that resource, so (for instance) a CUDA task finishing at GPUGrid won't trigger a CPU request. But manually updating the project resets the backoff for all resources, which may be counter-productive. You made an interesting point over at GPUGrid about projects (POEM, Spinhenge) which impose a long communications deferral after any scheduler contact. That's another way that clients can starve themselves by asking for the wrong resource first, to add to 'the client which cries wolf' I described in 24379. I don't know how long PEOM makes you wait between requests, or why they've chosen to introduce the delay: are the delays following the twice-daily GPU ping (actually, presumably now four times a day, twice for CUDA and twice for ATI) sufficient to explain your work drought with v6.10.7 on their own? Or could their be other mechanisms at play, like the changing definitions of long term debt and 'overworked'? You'll probably need to get deep down and dirty with "work fetch debug" logging flags before you can explain exactly where the current mechanism is breaking down, and that's a necessary first step before fixing it. Overall, I agree with your general thrust that use/don't use resource switches should operate at the client level, project by project and under user control (I'm a great believer in giving users choice over how the resources they're donating are used), but it will take a concerted and well-documented effort to persuade David Anderson that this is the way forward. |
|
4507)
Message boards :
Questions and problems :
boinc sometimes will only request GPU work.
(Message 27575)
Posted 24 Sep 2009 by Richard Haselgrove Post: .... (which has been there starting with the 6.6.36 client and appears to have gotten significantly worse with the 6.10.x series). No, no, and again no. You're referring to the logged request for CPU/GPU work on projects which supply (for the time being) only GPU/CPU work. Whether you feel that the request "Have you, perchance, installed a new application since the last time I asked?" is a bug or a feature - and that's a fair question, which is worthy of debate - the fact remains that BOINC v6.6.xx clients have always made these requests. The only thing new "starting with the 6.6.36 client" was that details of the request were logged by default to the messages tab: previously, you had to enable debug logging to find out whether the work requested was for CPU, GPU or (notoriously) neither. As I've said before, the only thing you gain by degrading to v6.6.33 or earlier is ignorance (lack of information in the logs). And the only thing you gain by degrading to v6.4.5 is the inability to run both CPU and GPU tasks for the same project/application. |
|
4508)
Message boards :
Questions and problems :
Boinc 6.10.7 zero cpus
(Message 27560)
Posted 24 Sep 2009 by Richard Haselgrove Post: The <ncpus> flag (debug purposes only) has been available for longer than any of us needs to worry about. As Pepo says, the description in the Wiki was changed by Dave A on 9 February 2009, which was fairly early in the intensive CUDA fine-tuning that led to the functional v6.6.20 and later clients. The Wiki note reflects changeset [trac]changeset:17160[/trac] from four days earlier, which is where the underlying function was changed: - client: if ncpus is zero in cc_config.xml, honor it. Unfortunately, this is yet another of the changesets which never got copied into the boinc_core_release_6_6a stable branch, which perhaps explains why v6.6.36 doesn't behave as described in the Wiki..... Perhaps the lead check-in note explains why Rom overlooked it: - web: add another team stats link |
|
4509)
Message boards :
Questions and problems :
Boinc 6.10.7 zero cpus
(Message 27556)
Posted 24 Sep 2009 by Richard Haselgrove Post: Will change the parameters in cc to -1 as suggested and reload 6.10.7. To be honest, I thought the change came into effect quite some time ago, and I'm surprised v6.6.36 worked. The full list is at Client configuration. There are several new(ish) tags, but I think ncpus is the only one where the default value has been changed to mean something completely catastrophic. But unless you have an explicit need to use a non-standard setting, just delete the whole file. BOINC doesn't actually need any of it. |
|
4510)
Message boards :
Questions and problems :
Boinc 6.10.7 zero cpus
(Message 27552)
Posted 23 Sep 2009 by Richard Haselgrove Post: 9/23/2009 1:34:07 PM Reading preferences override file Double-check <ncpus> in cc_config.xml. Back in the day, 0 meant 'ignore me, use whatever the OS tells you'. Now, 0 means zero, and you have to use -1 for 'whatever the OS says'. |
|
4511)
Message boards :
Questions and problems :
Cuda question
(Message 27504)
Posted 20 Sep 2009 by Richard Haselgrove Post: It will make no difference at all to the speed of running of BOINC. It will make a considerable difference to the speed of running of project science applications, but only if the project has a CUDA application compatible with Snow Leopard - and I haven't heard of any yet. You will need to research what applications are available at each of the projects you are attached to: I suggest that you do that before spending any money on the replacement graphics card. |
|
4512)
Message boards :
Questions and problems :
benchmarks
(Message 27408)
Posted 15 Sep 2009 by Richard Haselgrove Post: Benchmarks are certainly unreliable, and should be taken with a large pinch of salt. But in this case, the Windows benchmarks are near enough double the Linux benchmarks. Linux, particularly Ubuntu, has a tendency to run the CPU in low power, low speed mode when there are no high priority tasks running, and BOINC is low priority. I can't remember the keywords we used last time it was discussed on this board, but it would be helpful if someone could post the linux commands to: a) Check current CPU speed b) switch the power profile to 'full speed at all times' (assuming that's how Rein would like it to run). |
|
4513)
Message boards :
Questions and problems :
Deadline madness
(Message 27277)
Posted 11 Sep 2009 by Richard Haselgrove Post: WU 'Ready to start' are normally displayed in FIFO order unless you sort a column to change the order..... One of my pet peeves is that I like to view BOINC manager tasks in 'native' FIFO order - but once you hit a column header and invoke sorting, you can't return to 'natural' order without a registry hack. AB is right that even without a sort, BOINC starts up with the projects pre-sorted into alphabetic order, and the tasks displayed FIFO within projects. But if you leave BOINC running for extended periods, new tasks are added to the list in FIFO order, and not sorted by project, so you get a clearer view of what is going on. Until very recently, the data/time a task was received from the server wasn't even recorded by the BOINC client, so there was no possibility of sorting by it. The field now exists, but I don't know if it's exported by the RPC process and available to the Manager. If it is, it would be a relativly simple change to add an extra column to the Manager display and replicate the sort function: that would supply the information that Jave is interested in, and solve my 'unsort' problem, in one go. |
|
4514)
Message boards :
BOINC client :
Boinc V6.10.4 bug fix annoyance
(Message 27226)
Posted 9 Sep 2009 by Richard Haselgrove Post: Afterthought: Facebook users will have got the full benefit of this one, and it hasn't been corrected yet: http://boinc.berkeley.edu/trac/browser/branches/boinc_core_release_6_8/client/time_stats.cpp 169 if (dt > 100) return; 170 // client must have been suspended; skip |
|
4515)
Message boards :
BOINC client :
Boinc V6.10.4 bug fix annoyance
(Message 27212)
Posted 9 Sep 2009 by Richard Haselgrove Post: For a change, David A's erratic approach to coding and documentation saved us from this one. The time stats bug was created, according to the checkin notes for the fix in [trac]changeset:19022[/trac], by [trac]changeset:17166[/trac]. That changeset, according to the checkin, was just about "web: fix bug in cached user page stuff from 12/29/08" - even though it also contained a client update. So, presumably, Rom didn't notice it, and didn't copy it into any release build. In any event, the code doesn't appear in the branches/boinc_core_release_6_6a version of time_stats.cpp, so those of us who have religiously ploughed through all the Beta releases since v6.6.14, but have avoided the Alpha stuff, don't know what all the fuss is about ;-0 |
|
4516)
Message boards :
Questions and problems :
Seti CUDA Likes To Hog The GPU
(Message 27161)
Posted 8 Sep 2009 by Richard Haselgrove Post: I'm one of those people that doesn't forget anything I've learn, even if i was gone for a long time. The danger with that - especially with BOINC - is that quite often the things you've learned turn out to have changed while your back was turned, and hence not to be true any more! ;-) |
|
4517)
Message boards :
Questions and problems :
More GPU options
(Message 27131)
Posted 7 Sep 2009 by Richard Haselgrove Post: There are only a handful of projects with CUDA applications. Most of the 75+ projects still use CPU applications, which makes it really a bad option to have a BOINC preference for not using the CPU. People are forgetful enough as it is, but they're oh so quick to jump onto forums (these and project forums) to blame something in BOINC is broken over a preference they themselves set and forgot about. Isn't the project-level CPU on/off switch actually a task allocation control (send/don't send work intended for CPU processing), rather than a 'suspend while in use' switch, where we currently have 'suspend BOINC entirely' or 'suspend GPU only', but not 'suspend CPU only'? |
|
4518)
Message boards :
BOINC client :
v6.10.3 ATI request gets lost and only CPU request occurs
(Message 27060)
Posted 4 Sep 2009 by Richard Haselgrove Post: Probably the same (or a resurrected version of the) problem which occurred with CUDA cards: Not getting new WU's for CPU projects after upgrade to 6.6.20. I believe that one's been fixed (haven't seen it for a while, but to be honest I haven't been looking), so if David can remember what he did, something similar could be done for ATI. Though come to think of it, the symptoms weren't quite the same: in that case, a CUDA shortfall existed, but a CPU request was issued for the true CPU shortfall - which happened to be zero. You are at least getting a request for the value of the ATI shortfall, just in the wrong request slot. |
|
4519)
Message boards :
Questions and problems :
ATi GPU Supporting?
(Message 26976)
Posted 1 Sep 2009 by Richard Haselgrove Post: That exercise proved that, although it's hard to fully test a client without an app, or vice-versa, nevertheless such testing is vital: it took a further month to get the science app working without crashing, and even longer to get the bugs out of the client. What I said - even longer. Eight months and counting. And as fas as I'm concerned, there were bugs before, there are bugs now, and there always will be bugs. BOINC is that sort of project. The ratio of bugs to features varies from time to time, and maybe the GPU developments (plural) have added bugs even faster than they have added features - though I find the version I'm currently using (v6.6.37 with Thyme Lawn's [trac]#139[/trac] extensions) is fairly predictable if you don't push the boundaries too hard. [That reminds me - I must write up a work fetch bug report for multi-cores, long tasks, short cache, and a big multi-threaded app in the mix. It doesn't fetch enough tasks to keep all cores active with single-threaded apps, so it runs the MT as a spare single-threaded app. Messy.] |
|
4520)
Message boards :
Questions and problems :
ATi GPU Supporting?
(Message 26965)
Posted 1 Sep 2009 by Richard Haselgrove Post: The OP may be thinking of the NVidia/CUDA release in December last year, when both the BOINC client and a science application were made available at the same time - first as Beta, then a week later as a public recommended release. That exercise proved that, although it's hard to fully test a client without an app, or vice-versa, nevertheless such testing is vital: it took a further month to get the science app working without crashing, and even longer to get the bugs out of the client. I would imagine that private alpha testing has already started at some projects: later, there may be a call for suitably-motivated volunteers to participate in more open Beta testing. It would be in everybody's best interests for that testing to be allowed to run its course, so that the applications when released don't suffer the problems that the premature CUDA release suffered. |
|
4521)
Message boards :
BOINC Manager :
Resource share question
(Message 26956)
Posted 1 Sep 2009 by Richard Haselgrove Post: Elapsed. |
|
4522)
Message boards :
Projects :
News on project outages.
(Message 26936)
Posted 31 Aug 2009 by Richard Haselgrove Post: It looks as if Einstein@Home has gone down again. IIRC, both of the two previous major filesystem outages started when SETI was having problems (Einstein is a popular backup project for SETI users). SETI ran out of raw data loaded for workunit generation about four hours ago, and is likely to remain dry for another five or six hours until the start of the working week in Berkeley. Could/should there be any sort of cross-project early warning system to prevent this 'domino' or cascading effect of project failures? |
|
4523)
Message boards :
Questions and problems :
Getting BOINC to attach to projects
(Message 26844)
Posted 28 Aug 2009 by Richard Haselgrove Post: Well, that brings PPTP tunneling into the equation too - do I presume that you're not even on the same site as the proxy server? This is getting too complex for an outsider to diagnose - I think you need: a) to sweet-talk IT into telling you how it's done b) to find one of those 'other people' you mention in your first post as having run BOINC, and ask them c) nip up the A34 to Oxford, and enrol on an advanced networking course d) wait until one of the real networking gurus here comes on line, and can talk you through the http_debug route. I think I've taken you as far as I can - good luck. |
|
4524)
Message boards :
Questions and problems :
Getting BOINC to attach to projects
(Message 26841)
Posted 28 Aug 2009 by Richard Haselgrove Post: Well, rl.ac.uk is a public domain registered to the Rutherford Appleton Laboratory - so if you are not a publically-funded reasearch institution (either RAL or another part of the Science and Technology Facilities Council), you/they shouldn't be using it. But if you are a research lab, my comments about factsheets/FAQs still apply. This isn't a BOINC question, it's a networking question. If you were allowed to bring a personal laptop into work, and use the rl.ac.uk network to browse external websites, you would need exactly the same proxy information to configure your browser - hence my suggestion to see if the information is visible in your work computer's browser configuration. |
|
4525)
Message boards :
Questions and problems :
Getting BOINC to attach to projects
(Message 26839)
Posted 28 Aug 2009 by Richard Haselgrove Post: Those are the sort of questions only your IT department can answer. I would have thought that a university (going by the .ac.uk suffix) would maintain a supply of factsheets or FAQs for research staff to refer to in situations like this. |
|
4526)
Message boards :
Questions and problems :
Getting BOINC to attach to projects
(Message 26837)
Posted 28 Aug 2009 by Richard Haselgrove Post: Looking at the BOINC proxy dialogs (advanced -> options, third and fourth tabs), you also need to know: Whether it is a HTTP or Socks proxy The address to use The port number to use Whether the proxy requires authentication Similar information will be needed for your web browser (Internet Explorer, Firefox or whatever) to work, but the IT department may have configured their network to supply the relevant details automatically to those programs. You may be able to read the values you need from the configuration pages of your browswer (tools -> internet options, connections tab, click 'LAN settings' for IE). I believe there are attempts afoot to teach BOINC how to detect proxy servers automatically, but I don't think they're generally available until the v6.10 range - and you don't want to add testing an unstable version to all your other tasks. So I think you'll have to copy the information to BOINC by hand. |
|
4527)
Message boards :
Questions and problems :
Getting BOINC to attach to projects
(Message 26835)
Posted 28 Aug 2009 by Richard Haselgrove Post: It would probably also help if you could post the technical details your IT department gave you about their proxy server so you could set BOINC to use it (not the security stuff, obviously). Some versions of BOINC have problems with some sorts of proxies, or need special settings (HTTP 1.0, for instance). |
|
4528)
Message boards :
Questions and problems :
Multiple accounts on Boincstats?
(Message 26826)
Posted 27 Aug 2009 by Richard Haselgrove Post: As just posted at GPUGrid: Anyone know how to combine all projects into one "account" on boincstats? |
|
4529)
Message boards :
Projects :
News on project outages.
(Message 26606)
Posted 14 Aug 2009 by Richard Haselgrove Post: Einstein Update on their temporary front page: Fri Aug 14 17:01:20 UTC 2009 |
|
4530)
Message boards :
BOINC Manager :
With nVidia driver 190.38 now SLI should be enabled?
(Message 26550)
Posted 9 Aug 2009 by Richard Haselgrove Post: Yes, but in the other posts wasn't mentioned that now SLI must be enabled. It's becoming apparent that this is only needed for some OSes - Vista x64 has been mentioned. I expect the 'Vista' bit is more important than the 'x64'. |
|
4531)
Message boards :
BOINC client :
Reading WU report file after uploading
(Message 26528)
Posted 7 Aug 2009 by Richard Haselgrove Post: A backup was something I had in mind but there are some potential problems. Knowing some BOINC quirks may help you here. If the application exits with an error, I think you get an automatic comms backoff of 60 seconds, so no immediate sched_request. There's usually no upload file for an exit_error, so the error text will definitely be in the next sched_request, which as you say is likely to be pretty soon after the 60 seconds are up. If the application exits early, but reports success - e.g. a SETI dash-9 overflow - then as you say the cache/EDF recalculation may trigger an immediate work fetch: but there will be a file to upload, and the upload won't be complete when the sched_request goes in. So the error text won't appear until the next sched_request, or possibly even later if the upload handler is playing up. |
|
4532)
Message boards :
BOINC client :
Reading WU report file after uploading
(Message 26523)
Posted 7 Aug 2009 by Richard Haselgrove Post: The same information goes out in the sched_request_*.xml file, in very useful <stderr></stderr> tags, which by any means is always much smaller than the client_state.xml file is. You can read it from there. Certainly does, and the files are persistent - can be read at any time until the next scheduler request. But those requests can come in pretty quick succession - typically ~20 seconds at SETI - and no backup of the previous file is kept. |
|
4533)
Message boards :
BOINC client :
Reading WU report file after uploading
(Message 26517)
Posted 7 Aug 2009 by Richard Haselgrove Post: Using the still very functional BOINCview it Logs and even provides an exit status code in the Completed window, so yes think it should be possible, then for the detail visit the Result Status page on the project website. Fred is writing his tool in a SETI context. The SETI admins are being quite tough on web-scraping at the moment (to the extent of blocking IPs), and Fred has been advised to avoid it. |
|
4534)
Message boards :
BOINC client :
Reading WU report file after uploading
(Message 26515)
Posted 7 Aug 2009 by Richard Haselgrove Post: Not using BOINC itself. BOINC apps write the stderr messages into the slot directory while running, but they get transferred to client_state.xml when the application terminates. The slot directory is immediately cleaned for re-use - nothing available there. Similarly, science output data files are cleaned up and unavailable as soon as the 'file upload' has completed successfully - usually a matter of seconds after the application terminates. You only get a breathing space if no internet connection is available, or networking is disabled in BOINC. The stderr messages remain in client_state.xml, and are accessible via standard RPC calls, until the result is reported: but they are removed as soon as the report ack is received from the server. And they need to be: client_state.xml is an awkward enough construct already (and BOINC can barely cope with it): reclaiming the space is vital. The only way of doing what you want is to use, or re-create, a tool like BoincLogX, which would monitor the files and messages in real time and transfer them to your own storage structure for subsequent inspection. |
|
4535)
Message boards :
Questions and problems :
What did I do Wrong ??
(Message 26440)
Posted 2 Aug 2009 by Richard Haselgrove Post: -- however, system will not upgrade. There are plenty of similar problems, but they usually include a slightly fuller description of what went wrong :-) The commonest: can't find boinc...msi This isn't a BOINC problem, but a Windows (you do use Windows, I presume?) installer problem. What it's actually looking for is the old (v6.2.18) installer file, so it can find out what to uninstall as part of the upgrade process. Re-run the installation for v6.2.18 in 'repair' mode, and then try again. |
|
4536)
Message boards :
BOINC client :
Why waiting?
(Message 26428)
Posted 2 Aug 2009 by Richard Haselgrove Post: There is a debug command that will cause work to report immediately after completion, but I've never used it and having been up all night, I'm to sleepy to look it up. I know where it is, but I'll only bother posting it if someone can demonstrate a real need: as AB says, it's intended for debug use only, and can cause unwanted extra strain (and even validation errors) on projects where speed is not important. Altogether, it's best to leave these issues to the projects: like GPUGrid, if they need their own results back quickly, they have the tools available at their end to make it happen - without imposing a global solution on all the other projects attached to your computer. |
|
4537)
Message boards :
BOINC client :
Why waiting?
(Message 26419)
Posted 1 Aug 2009 by Richard Haselgrove Post: GPUGrid is currently offering a 25% 'bonus' for reporting CUDA tasks within two days, instead of the five days allowed by their deadlines. But they have also implemented regular scheduler (server) connections - item (9) in John's list - so no special action is needed by the user to get the results in quickly. |
|
4538)
Message boards :
BOINC Manager :
Boinc 6.6.36 scheduling strangely
(Message 26309)
Posted 28 Jul 2009 by Richard Haselgrove Post: i'll go to .38 on that pc..., no bugs in .38 other than a gpu request on projects with no gpu app out... Yet again - every BOINC V6.6 has always asked for GPU work from every available project, just in case a GPU application may have been deployed since last time. v6.6.38 has a bug fix that puts information back into the message log so you can see what it's doing. |
|
4539)
Message boards :
Questions and problems :
Is it possible to run only CUDA WU ??
(Message 26217)
Posted 22 Jul 2009 by Richard Haselgrove Post: What applications do you use? If you use the standard project-supplied applications, then the website prefences you've listed should work. If you use third party or optimised applications (as many at SETI do), then I think the web preferences get overtaken by the app_info.xml specified apps. If you don't want to use the CPU, don't soecify a CPU application in app_info.xml |
|
4540)
Message boards :
Questions and problems :
Windows XP Pro Agonizingly Slow w/BOINC 6.6.36
(Message 26068)
Posted 17 Jul 2009 by Richard Haselgrove Post: Not a guess, bitter experience. As well as the Dell, I run three quads with CUDA cards - two of them are currently in a similar state because of badly-behaved CUDA tasks (AQUA, in these two cases). Anyway, glad the machine is working fast enough for you to finish off the detective work. |
|
4541)
Message boards :
Questions and problems :
Windows XP Pro Agonizingly Slow w/BOINC 6.6.36
(Message 26066)
Posted 17 Jul 2009 by Richard Haselgrove Post: I run an earlier Dell Precision 490, also with 8 cores. It started BOINC under Windows XP, and continues under Vista. No problem, though saturation of the quad-channel FB-DIMM memory sub-system means that with the wrong mix of tasks, it doesn't scale well compared to two separate quads. I suspect the slowness is due to your CUDA card. What projects/tasks are you running on it? Some (AQUA, SETI/VLAR) aren't very well behaved at the moment. |
|
4542)
Message boards :
Questions and problems :
BOINC daemon hangs when internet not available
(Message 26060)
Posted 16 Jul 2009 by Richard Haselgrove Post: I'm wondering whether to add a cross-reference to BOINC cc using excess CPU as service, which bit me again today. Internet access was available, but the problems I've documented in that thread tend to occur when Microsft Automatic Update has downloaded patches but not yet been given permission to install them - as, obviously, today. Note. My observations relate to BOINC v5.10 - but before you patronise me :-), today's was on a MS Small Business Server, which has to be a Domain Controller. So no v6 version can be installed. Similar observations - BM can't connect to the daemon, even though the daemon is running (and using excess CPU), according to Task Manager. And because it's v5, there is an additional test I can perform which Raistmer can't. If I stop the service using the service manager snap-in, I can start BOINC in user mode (v6 just re-starts the daemon). In user mode, the exact self-same boinc installation uses zero CPU (unless asked to do something, like a manual project update), and the Manager is responsive and fully functional. Close the Manager, restart the service, and the Manager stalls at start-up (with two processes showing in Task Manager). Kill one process, and they both stop, and BM closes. So there is something specific to running in daemon mode in all this. |
|
4543)
Message boards :
Questions and problems :
Alpha Account
(Message 26038)
Posted 15 Jul 2009 by Richard Haselgrove Post: I never knew Richard is my email database. ;-) No, it was just his way of saying that he thinks I'm obsessive. Must get out more Must get out more Must get out more Must get out more Must get out more Must get out more Must get out more Must get out more Must get out more Must get out more Must get out more Must get out more :-) |
|
4544)
Message boards :
Questions and problems :
Alpha Account
(Message 26030)
Posted 15 Jul 2009 by Richard Haselgrove Post: You can't. The scheduler at the Alpha server is down, so it won't be able to communicate with your BOINC. From: "David Anderson" To: "Nicolás Alvarez" Cc: <boinc_alpha@ssl.berkeley.edu> Sent: Friday, June 12, 2009 7:01 PM Subject: Re: [boinc_alpha] Cannot upload results to BOINC Alpha project I've shut down the Alpha project for the time being; uploads were causing performance problems on the server. I may start it later with smaller files -- David |
|
4545)
Message boards :
Questions and problems :
Can't install seti files hang in downloading
(Message 26026)
Posted 15 Jul 2009 by Richard Haselgrove Post: However, David Andseron is also listed on the SETI 'project personnel' page as a Director of SETI@home. He certainly has the power (= access codes) to work directly on SETI's servers, updating code and so on - though whether he has the authority to do so is a moot point. I would have thought that the originator of BOINC as a robust infrastructure would at least take an interest in its behaviour at the extremes of the envelope. |
|
4546)
Message boards :
Questions and problems :
Can't install seti files hang in downloading
(Message 26018)
Posted 15 Jul 2009 by Richard Haselgrove Post: Sekerob's suggestion was posted to address a download problem - nothing to do with uploads. You cannot upload to SETI at the moment, because their upload server has been disabled by staff - as is clearly visible on the project's Server Status page (though no explanation of the need for the extended outage has been posted yet, AFAIK). No amount of fiddling with the client at your end can restart that server at Berkeley. |
|
4547)
Message boards :
Projects :
AQUA grants insane amount of credits
(Message 25992)
Posted 13 Jul 2009 by Richard Haselgrove Post: Hmmm... someone over there with a couple of computers that consistently request 3 to 12 times the amount of credit anyone else does? No, two separate computers, operated by separate people (but teammates), each of which returned one single rogue result after previously (I believe) working normally. One of the rogue results got awarded 11 million credits, the other 569,600,074.02 credits. That's on the Zimbabwe scale of hyper-inflation, and around double the usual daily credit of all BOINC users on all BOINC projects, combined! No wonder other users, and other teams, smelled a rat. |
|
4548)
Message boards :
Projects :
AQUA grants insane amount of credits
(Message 25990)
Posted 13 Jul 2009 by Richard Haselgrove Post: There's a simple rule of mine: If you don't like or trust the project itself, or don't like or trust the credits on a project, then do not crunch for that project. AQUA was actually doing quite well, and credits - while high for CUDA work - were not completely outrageous for an Alpha project. This particular problem has blown up suddenly and quickly. The first sign of trouble was posted 8 Jul 2009 21:41:02 UTC: I don't think it invalidates any decision people might have made to crunch for the project before that time - and as I said, the 'trigger' event came from outside the project. The subsequent mistakes made by project management were during their 'rapid response' phase to those external triggers. Reports suggest that many people have aborted work or simply gone elsewhere over the weekend. Perhaps that is a useful lesson that should be written into the BOINC (server) FAQs: no matter what it says on the front page, the BOINC server software is complex and needs fine-tuning of many configuration parameters, by a skilled and well-trained administrator, before it is safe to expose it to the public at large. BOINC may be cheap to install and run, but it's not that cheap. |
|
4549)
Message boards :
Projects :
AQUA grants insane amount of credits
(Message 25982)
Posted 12 Jul 2009 by Richard Haselgrove Post: They were the recipients of a couple of anomalous results. Debate is fierce over whether they were computer errors, deliberate demonstrations of systemic weaknesses, or cheating. All three points of view have been expressed, sometimes more than one by the same person. As a result, they changed their credit system from modified benchmark*time to fixed credit at extremely short notice, without thinking it through properly. At the same time, they had just launched new applications, which turned out to be massively better optimised than they expected - big CPUs (i7 running x64) are returning results in 4 hours that it takes a big CUDA card 5 days to complete (Personally, I suspect a bug). So CPU credit rates have gone massive, but CUDA are reasonable. Personally, I suspect cock-up rather than conspiracy: a combination of inexperienced management (in BOINC system terms), external ?attack?, and unexpected application behaviour. I hope they can be offered support and guidance on recovery procedures, at least for the first part of next week: only bring out the big stick if they don't respond responsibly when the office opens tomorrow (they're another California project, so 'tomorrow' is still a lot of time-zones away). |
|
4550)
Message boards :
BOINC Manager :
Another "Not Connected to a Client" Issue
(Message 25847)
Posted 1 Jul 2009 by Richard Haselgrove Post: When I suspend network activity it doesn't diconnect. As soon as I "unsuspend" it will disconnect, regardless if I suspended the individual projects. That sounds like a clue. We need the networking gurus, but I've heard phrases like "blocking dns" in this context. Basically, the BOINC Manager talks to the BOINC daemon by TCP/IP, on localhost (127.0.0.1). The BOINC daemon talks to the projects by TCP/IP, and finds them by DNS. So if DNS takes ages to resolve the project names, TCP/IP stalls, and you lose the connection to the manager in the traffic-jam. That could be the problem. Unfortunately, I can't guide you to the solution, but I'm sure someone here can. |
|
4551)
Message boards :
BOINC client :
Bugs in 6.6.31/WinXP
(Message 25807)
Posted 30 Jun 2009 by Richard Haselgrove Post: Minor, but given the file is nowhere to be found, a dead feature or a future function? Future function. |
|
4552)
Message boards :
Questions and problems :
Had to reinstall Boinc manager... am I doing something wrong?
(Message 25794)
Posted 29 Jun 2009 by Richard Haselgrove Post: If you don't have a backup of the BOINC data folder from your old HDD, and you don't use one of the thirdy-party account managers like 'BAM' from the BOINCstats stable, then yes, you will have to attach to the projects all over again. But do remember that you are a 'returning' user, so you can attach to your existing personal account at each project: that saves a bit of time, and keeps your statistics together. |
|
4553)
Message boards :
Questions and problems :
BOINC manager one click remote connection
(Message 25740)
Posted 27 Jun 2009 by Richard Haselgrove Post: I open a remote desktop, and then run the native BOINC Manager on the remote machine. That way you do the authentication/password stuff as you open the desktop, before you get to the Manager stage. Won't work for Vista/CUDA, according to reports - you'd need one of the many and various remote-control programs, like some flavour of VNC. Edit - that's for the (very few) remote management jobs I can't do with BoincView, which is always open. |
|
4554)
Message boards :
Questions and problems :
Stop switching between WU
(Message 25698)
Posted 25 Jun 2009 by Richard Haselgrove Post: 6.6.12: If you can prove that, it'd be a reportable bug. |
|
4555)
Message boards :
Projects :
News on project outages.
(Message 25690)
Posted 25 Jun 2009 by Richard Haselgrove Post: Einstein seems to have gone again. No web site, no file uploads. |
|
4556)
Message boards :
Questions and problems :
Stop switching between WU
(Message 25640)
Posted 23 Jun 2009 by Richard Haselgrove Post: * first schedule jobs projected to miss deadline in EDF order And if EDF order is undefined (because multiple jobs have identical deadlines?) ;-) |
|
4557)
Message boards :
Questions and problems :
Stop switching between WU
(Message 25638)
Posted 23 Jun 2009 by Richard Haselgrove Post: OK, I can buy that. 'Preempt before checkpoint' has a "by design" alternative handling mechanism that doesn't involve removal from memory (for the benefit of science apps that don't checkpoint. Question: are there any CUDA science apps in this day and age which don't checkpoint? And if so, why?). So the next question is - why, and by what mechanism, would a CUDA app be preempted before first checkpoint? I can manage once: app exits on DCF downslope - triggers work fetch - new (cached) task starts - work is allocated and downloaded - turns out to be VHAR - pre-empt on completion of download. But I can't explain seven, let alone hundreds. Do these multiple pre-empts coincide with downloads? or are they previously cached tasks? Need data. |
|
4558)
Message boards :
Questions and problems :
Stop switching between WU
(Message 25635)
Posted 23 Jun 2009 by Richard Haselgrove Post: 1) If "more than 100" are suspended at the same time, then that can't be "equal deadline". The usual maximum is 20, but even with recent feeder changes, the absolute maximum per scheduler request would be 97. And successive sheduler requests at SETI usually mean that the deadlines are separated by at least 20 seconds. 2) I think we need some clarification on 'memory'. a) "CUDA exe is kept in memory". Surely that must refer to the main CPU application (which later launches the CUDA execution kernels), and system RAM. I don't know of any tool which can show the name of any program, and the size of its memory footprint, inside the CUDA/video RAM system. Ideas anyone? My understanding is that a SETI task requesires of the order of ~150MB - ~200MB of free graphics RAM: two can fit in my 512MB cards, but not three. Having larger numbers in memory, other than main system memory, would require a very large graphics card indeed. b) After a couple of bodged starts, and a "nasty bug", I thought we had reached the point in v6.6.31 that grahics RAM (but not main system RAM) was invariably cleared by BOINC when pre-emtping CUDA tasks, whatever the state of the 'leave apps in memory' flag. I didn't find any problems with graphics RAM over-filling when testing v6.6.31 and later: if there are situations where apps remain in graphics memory (suggested by that 'going into fallback'), then they need to be enumerated, the circumstances identified, and the cause reported as a bug. |
|
4559)
Message boards :
Questions and problems :
Stop switching between WU
(Message 25633)
Posted 23 Jun 2009 by Richard Haselgrove Post: I've been following this thread and the related 'BOINC Scheduling Issue' from the sidelines for a while, without being much the wiser about what, if anything, the problem is, and what, if anything, to do about it. It hasn't even always been clear which projects people are talking about: but this is about SETI, so I hope I can add something constructive. First, some facts about SETI, and then some speculation about how relevant they may be.
|
|
4560)
Message boards :
Questions and problems :
CPU Time column missing from Tasks in 6.6.36
(Message 25623)
Posted 22 Jun 2009 by Richard Haselgrove Post: And they also don't have a strictly-enforced policy about updating the documentation that exists in step with changes in released clients (even when new versions are made 'recommended' releases across all platforms). I've had to make myself a Wiki account and do some of the corrections/clarifications myself. |
|
4561)
Message boards :
Questions and problems :
Stop BOINC Requesting GPU/CUDA Work From Particular Projects
(Message 25581)
Posted 21 Jun 2009 by Richard Haselgrove Post: That is the point. CP does not do CUDA, therefore I would like to be able to tell BOINC to stop asking CP (and other projects that don't do CUDA) for CUDA work units. Does it really matter? Is it a measurable overhead for either your processing or your comms bandwidth? On the other hand, many years ago we were told that contacting a project scheduler (and hence accessing the BOINC database) was a costly operation in server resource terms. Now we seem to be happy throwing these redundant RPCs all over the place (even if only twice/host/project/day). I still worry that this may be a contributing factor in the spate of server problems we've seen recently, but we'd need input from a BOINC project server administrator to evaluate that. |
|
4562)
Message boards :
Questions and problems :
CPU Time column missing from Tasks in 6.6.36
(Message 25580)
Posted 21 Jun 2009 by Richard Haselgrove Post: Since I installed 6.6.36 (WinXPPro SP3), the CPU Time column is missing from the Tasks display. It was replaced by an 'Elapsed' time column in BOINC v6.6.16 I think the developers intended this as a feature, but YMMV. |
|
4563)
Message boards :
Questions and problems :
cpu usage
(Message 25549)
Posted 19 Jun 2009 by Richard Haselgrove Post: hi Jord When you hit clear to reset all local preferences, it goes back to the global settings defined for your account on project websites, not to any setting hard-coded into BOINC itself. |
|
4564)
Message boards :
Questions and problems :
6.6.36
(Message 25528)
Posted 18 Jun 2009 by Richard Haselgrove Post: Then, if I understand you right, the client will also ask for CPU work, so this is really no problem? Yes, exactly. That's the big difference of v6.6.xx above v6.4.x: it separates the CPU and GPU much more completely. Sometimes it will ask for GPU work, sometimes it will ask for CPU work, and sometimes (rarely) it will ask for both together. You probably noticed the GPU messages because they will come quite frequently at first (in case there is a CUDA application, but the server has just gone to sleep), but it will quickly calm down to the frequency I suggested previously. |
|
4565)
Message boards :
Questions and problems :
6.6.36
(Message 25520)
Posted 18 Jun 2009 by Richard Haselgrove Post: i've downgraded back to 6.31 and it's correct now., not getting request work for gpu from a project that doesn't have gpu app.. v6.6.31 (and in fact the entire v6.6.xx range) will ask for GPU work 'periodically'. The very earliest ones would, after an initial flurry, settle down to asking once per day: later ones will ask after random time intervals, averaging twice per day. The only relevant change between .31 and .36 is that BOINC is now, at long last, explaining a little bit more about what it's doing. So the only thing you have gained by downgrading is ignorance ;-) The BOINC team do know that these GPU requests are being sent - they wrote the code, after all! I think the idea is that if/when a project launches a new CUDA application, BOINC will notice and start using it without you having to make any changes by hand. |
|
4566)
Message boards :
BOINC Manager :
How to stop all projects from 6.6.31?
(Message 25321)
Posted 10 Jun 2009 by Richard Haselgrove Post: it take too long to suspend dozens of WUs. You could suspend the projects instead - I imagine you have fewer projects than WUs. And on both tabs, you can multi-select them all and suspend them with one click. But if you do that, it tends to mess with your debt values and everything. Better to shut down connected computer. Glad it worked, and you liked it. |
|
4567)
Message boards :
BOINC Manager :
How to stop all projects from 6.6.31?
(Message 25204)
Posted 4 Jun 2009 by Richard Haselgrove Post: As Jord says, there are reported problems. A workround seems to be Advanced view, advanced menu --> 'Shut down connected client' before exiting. But the more information to help the developers track down the problem, the better. |
|
4568)
Message boards :
Questions and problems :
6.6.31 = 6.6.28?????
(Message 25168)
Posted 2 Jun 2009 by Richard Haselgrove Post: Tullio is reporting a similar problem at SETI - message 902837. |
|
4569)
Message boards :
Questions and problems :
BOINC 6.6.20 hung totally after GPU card change
(Message 25137)
Posted 31 May 2009 by Richard Haselgrove Post: What was logged by BOINC in stdoutdae.txt (and stdoutdae.old) around the time of the change? |
|
4570)
Message boards :
Questions and problems :
Time to completion all over the place
(Message 25114)
Posted 29 May 2009 by Richard Haselgrove Post: Thanks Richard, but that only solves the startup after that you get the same problem very fast... No, if you're using an app_info it's a permanent fix: well, permanent insofar as I haven't seen these wild fluctuations since installing the flop correction values in early March: though, of course, you do have to do something to keep the VLARs at bay. Nothing can compensate for bad estimates within a single app-version. This can only be solved by threating CPU and GPU differently. Or a way to lock the time, because the time is rather constant. Sure, a true 'solution' needs better programming, but flop compensation is a highly effective surrogate. And the GPU scheduler has some rather odd behavior. I have about 10 Cuda tasks stopped at 3 seconds before the end, go figure. That's a real waste of time. A task that takes about 6 minutes is hardly worth stopping at all. Be careful with that. All recent BOINC versions have problems with pre-empted tasks: a) Up to v6.6.20 - preempted tasks stay in memory. If you have too many tasks stacked up in graphics memory, the next one to start can't squeeze in, and fails with a malloc error. Once you reach that stage, you need to reboot the computer. b) v6.6.23 onwards - preempted tasks are supposed to be given time to vacate memory and clean up behind themselves - but if that takes more than a second (as it can), the new task isn't initialised properly and errors. A fix for bug (b) was coded yesterday, and Rom has just told us the fix will be released on Monday (hopefully there'll be time for some testing in between...). If that timetable sticks, it'll be worth upgrading to the new version next week. |
|
4571)
Message boards :
Questions and problems :
Time to completion all over the place
(Message 25108)
Posted 29 May 2009 by Richard Haselgrove Post: Because they all interact with a single project-wide "Durection Correction Factor". BOINC tries to compensate for the differing processing efficiencies of four different applications - in the SETI case, MB/CPU, MB/CUDA, AP, and AP_v5 - using a single variable. It just can't be done. In the long term, the answer for BOINC is to switch to using DCF variables for each separate application version. This has been requested, mulled over, and at one stage semi-promised for v6.8: but with yesterday's announcement that v6.8 has been hijacked by GridRepublic, I guess this has been punted into the long grass again. In the short term, there are well-tested techniques for coping with the SETI manifestation of the issue. Go back to the SETI boards, and read e.g. message 894765. The flops multipliers were originally calibrated on my Q9300, and should ideally be re-worked for your own hardware - sounds like you have plenty of raw data - but they work as a quick'n'dirty approximation. |
|
4572)
Message boards :
Questions and problems :
Is this good enough?
(Message 25037)
Posted 25 May 2009 by Richard Haselgrove Post: It should be cheap and easy to find at least another 256MB of RAM for a machine like that, and it would certainly give you better general performance with WinXP and a wider choice of BOINC projects. |
|
4573)
Message boards :
Questions and problems :
Error 2718
(Message 24978)
Posted 21 May 2009 by Richard Haselgrove Post: Sounds a rather similar episode to the error 1327 described in [trac]#815[/trac]. |
|
4574)
Message boards :
Questions and problems :
When reporting CUDA errors...
(Message 24906)
Posted 17 May 2009 by Richard Haselgrove Post: I don't know if this is a boinc or cuda app function, but how can I force boinc/cuda to use only my first GPU and totally ignore the 2nd one? Boinc reports 2 cards, but since it cannot use the 2nd card due to out of memory errors (256mb card I use for my default video), it promptly starts 2 apps and feeds 2 workunits to the 1st device. I wish to stop that behavior and feed only 1 workunit to the first device ignoring anything about a 2nd device. I think you're going to be stuck for a little while on this one. The short answer would be 'upgrade to BOINC v6.6.28', which would do exactly what you want by default - use the higher-spec CUDA card, and ignore the lower-spec one. Hoever, for some as yet unknown reason (people keep asking), recent BOINC development builds haven't been made available - even for testing - for either Linux_x86 or Linux_x64. I'll ask again on the mailing list. |
|
4575)
Message boards :
Web interfaces :
Drupal migration
(Message 24895)
Posted 15 May 2009 by Richard Haselgrove Post: Will Drupal allow for a finer-grained allocation of additional user rights? At the moment, apart from the staff super-user categories (Administrator, Developer etc.), I think I'm right in saying that the BOINC forum software only allows priveledged access and editing rights to one class of volunteers: Forum Moderators. In most of the projects I have experience of, these posts are filled by people with a deep interest in, and knowledge of, the project and its scientific needs (witness the wonderful Skygrid file analysis thread at Einstein, almost exclusively written by forum moderators). But SETI has the unfortunate reputation of needing the other sort of 'etiquette' moderation. I feel that SETI loses out by being unable to recruit a separate, different cadre of 'technical' moderators - with elevated rights but separated from the policing role. If Drupal can offer at least some of the extra facilities that 'full feature' message board software like phpBB can offer, then that would be a great advantage: it should be implemented with those additional facilities. |
|
4576)
Message boards :
BOINC Manager :
Can't launch IE8 from Boinc 5.10.13
(Message 24894)
Posted 15 May 2009 by Richard Haselgrove Post: No, it's much easier just to unpack a precompiled one and hook it up. That way, I get to test whether the setup /a bug I described in ticket [trac]#847[/trac] is still present in v6.6.28: it is. |
|
4577)
Message boards :
Questions and problems :
Misleading info in BOINC log
(Message 24893)
Posted 15 May 2009 by Richard Haselgrove Post: Mac OS X has been promoted as well (v6.6.29), but Linux is still stuck back at v6.4.5 |
|
4578)
Message boards :
BOINC Manager :
Can't launch IE8 from Boinc 5.10.13
(Message 24889)
Posted 15 May 2009 by Richard Haselgrove Post: You know us old luddites, we stick with v5.10.13 where we're comfortable. But it is nice to drive it from a later v6 Manager, what with multi-select and all. |
|
4579)
Message boards :
BOINC Manager :
Can't launch IE8 from Boinc 5.10.13
(Message 24880)
Posted 15 May 2009 by Richard Haselgrove Post: Following a report at Einstein by user 'Nothing But Idle Time', I checked this out on a host running 32-bit Vista, IE8 and BOINC v5.10.13 Sure enough, none of the project url buttons in the v5.10.13 Manager would open IE8 and go direct to the required page. Neither would BOINC's own links in the help menu. Hoever, if IE8 is already open, the links do actually work. I have tested a variety of BOINC v6 Managers on the same host (still running CC v5.10.13), and they all opened IE8 and the requested page with no problem - I tested v6.2.19, v6.4.5, and a variety of v6.6.xx including .20 and .28 So it's a historical curiosity rather than a bug anyone will ever fix, but it might come up as an issue on other project message boards. |
|
4580)
Message boards :
BOINC Manager :
BOINC Manager incorrectly reports CUDA version
(Message 24818)
Posted 10 May 2009 by Richard Haselgrove Post: On that line, BOINC is reporting the 'compute capability' of your hardware, rather than anything to do with software. The definition is a bit hard to come by, but it's in Appendix A of the CUDA programmer's guide - try http://developer.download.nvidia.com/compute/cuda/2_2/toolkit/docs/NVIDIA_CUDA_Programming_Guide_2.2.pdf if you're interested. Appendix A starts on page 99 according to the document page numbering, or page 107 according to Acrobat Reader. |
|
4581)
Message boards :
Questions and problems :
BOINC 6.6.20 not reporting WU elapsed running time on remote hosts
(Message 24795)
Posted 8 May 2009 by Richard Haselgrove Post: It was fixed in v6.6.21: Changes for 6.6.21 v6.6.21 was supposed to be the released version, but something went wrong with the deployment and they dropped back to v6.6.20 rather than fix it. |
|
4582)
Message boards :
Questions and problems :
I can't download new Boinc version
(Message 24696)
Posted 2 May 2009 by Richard Haselgrove Post: The BOINC home page seems to be empty - no HTML 'body'. But the download page seems to be present and working. |
|
4583)
Message boards :
Questions and problems :
boinccmd --quit returns before BOINC stopped
(Message 24631)
Posted 28 Apr 2009 by Richard Haselgrove Post: The changes did not persist though, as they were replaced when the running state was written out to disk at the next file update. And that's precisely why we need to stop - and indeed restart - BOINC under script control. |
|
4584)
Message boards :
Questions and problems :
temperature increase
(Message 24625)
Posted 28 Apr 2009 by Richard Haselgrove Post: Have you noticed any change in your application processing times? Some versions of Ubuntu ran low-priority applications - like the science apps spawned by BOINC - in a low-power, low-speed CPU mode if nothing else on the system required waking the CPU up to full speed. The new version may be running at full speed all the time. That would consume more input electrical power, and churn it back out again as heat energy for your air-conditioning to deal with. That would put the temperatures up. |
|
4585)
Message boards :
Questions and problems :
boinccmd --quit returns before BOINC stopped
(Message 24621)
Posted 28 Apr 2009 by Richard Haselgrove Post: I have to admit that I'm mildly curious as to why you would want to edit the client_state.xml file in the first place. Most are content just to read it... It's a workround for a badly-behaved application (CUDA version of the SETI_enhanced app), which causes sluggish graphics performance on the host computer for some tasks. The desire is to re-allocate these tasks for processing with a traditional CPU application. For proper file integrity, client_state.xml is only ever read, and a parallel file is created with the desired changes (<version_num> and <plan_class>). On successful completion of the script, the files can be swapped with copy/rename. I've never found the script's access to client_state.xml to be blocked by the file locking you decribe, even while BOINC is fully active. |
|
4586)
Message boards :
Questions and problems :
BOINC 6.6.20 released for Windows, Windows x64, Linux, Linux x64 and MacOS X
(Message 24571)
Posted 26 Apr 2009 by Richard Haselgrove Post: I do not care how long the work units I get are, but I doubt that any SETI CUDA WUs should have an estimated competion time of 365 hours on a Q6600. Are the estimates any less 'wrong' now than when you first posted, over 12 hours ago? In other words, is BOINC starting to self-correct, as it is designed to? - though you would need to complete several dozen tasks on each project to get anywhere near complete correction. Also, are you allowing for the fact that v6.6.20 is now displaying (and estimating) "elapsed" time instead of "CPU" time previously. This makes a big difference for CUDA tasks, and it may make a difference whether you let BOINC compute all the time, or have it set to stop computing when the computer is in use for other tasks. |
|
4587)
Message boards :
Questions and problems :
BOINC 6.6.20 released for Windows, Windows x64, Linux, Linux x64 and MacOS X
(Message 24568)
Posted 26 Apr 2009 by Richard Haselgrove Post: I do not care how long the work units I get are, but I doubt that any SETI CUDA WUs should have an estimated competion time of 365 hours on a Q6600. They actually complete in 5 to 6 hours. SETI CUDA WUs normally have a completion time in minutes - down to about 6 minutes per, under ideal circumstances, on my rigs. Even the really, really horrible ones take no longer than 90 minutes. But they don't run on a Q6600. They run on a computer which has a Q6600 on the motherboard in my case too, but they run on the nVidia card. 5 to 6 hours sounds plausible, but still slow, for ordinary SETI tasks running on the CPU. You need to be clear in your own mind whether you are seeing slow computation, or a poor initial estimate of what the computation time will be. If it's just the estimate, it should correct itself over time using BOINC's inbuilt mechanisms such as "Duration Correction Factor". When you upgraded, did you find that all your accounts, projects, tasks etc. were immediately visible when the new BOINC started, or did you - like a few others - have to start with a clean slate? If you lost work and started again - which shouldn't happen, but sometimes does - then all estimates such as DCF get lost too, and have to be re-estimated from scratch. |
|
4588)
Message boards :
Questions and problems :
Boinc 6.6.20 ends in repeated "Error while computing"
(Message 24556)
Posted 25 Apr 2009 by Richard Haselgrove Post: I agree the workunit looks bad, but I don't think it's the cause of darkmoebius's problem. The three download errors are a well-known proxy server problem (SETI uses one to spread the load of distributing new applications, and some ISPs block it) in late February. The compute error after 0 seconds seems to be a consistent problem on that particular host. Then there are two errors in running (one of which we're discussing here), and a single 'home run' all on their lonesome with a goose egg. Unfortunately, not an unusual outcome for an Astropulse task. |
|
4589)
Message boards :
Questions and problems :
Boinc 6.6.20 ends in repeated "Error while computing"
(Message 24546)
Posted 25 Apr 2009 by Richard Haselgrove Post: Consistently <core_client_version>6.6.20</core_client_version> Stock application, current BOINC. |
|
4590)
Message boards :
Questions and problems :
NVIDIA 9800GX2
(Message 24522)
Posted 24 Apr 2009 by Richard Haselgrove Post: Does (could) the GX2 operate in SLI mode in the driver? If so, disable SLI for SETI use. |
|
4591)
Message boards :
Questions and problems :
BOINC 6.6.20 released for Windows, Windows x64, Linux, Linux x64 and MacOS X
(Message 24512)
Posted 24 Apr 2009 by Richard Haselgrove Post: It's also worth considering the possibility that a failed migration didn't ***wipe*** the files, it may just have left them behind in an unexpected place. A thorough search (including hidden files/folders) may retrieve the situation. |
|
4592)
Message boards :
Projects :
Seti down
(Message 24505)
Posted 23 Apr 2009 by Richard Haselgrove Post: And already back up. |
|
4593)
Message boards :
Questions and problems :
Not getting new WU's for CPU projects after upgrade to 6.6.20
(Message 24490)
Posted 22 Apr 2009 by Richard Haselgrove Post: Might also want to consider trying to get the info from MW too ... my bogus GPU calls to them may be part of the problem that they are seeing also. Could you ask at MW? I'm not a member of that project, and a complicated question like that from a total stranger isn't going to get much attention - especially if they're in the middle of a crisis. The mind set is that if the cost appears low there is no reason to avoid doing stupid things over and over and over again. I have still to hear why we want to reschedule CPUs once every 60 seconds much less more often than that. Yet that at its root what is causing the issue we have been trying to get them to recognize ... As I said on boinc_alpha, I don't mind if the client spends a few milliseconds checking itself over - "Everybody OK? No problems? OK, Carry on as you were". The problem only arises if the client takes wasteful action as an improper result of the self-check - preempting tasks and so on - and there are problems in spades if the client fools itself into sending DDOS 'content free' RPC calls to servers. But it's a big call to make, to claim that the new "recommended" client is crippling the BOINC server network, and I must have confirmation of my facts before I call in anything louder than the question I have already put. I'm doing my damnedest to get that confirmation. |
|
4594)
Message boards :
Questions and problems :
Not getting new WU's for CPU projects after upgrade to 6.6.20
(Message 24488)
Posted 22 Apr 2009 by Richard Haselgrove Post: There's at least one bug still in work fetch, which can look like the problem reported here. I've reported it as a v6.6.23 bug, but the relevent section of code hasn't changed since v6.6.20 It appeared on boinc_alpha - my name, subject 'Work fetch bug in v6.6.23', first line "I'm afraid we still have a problem with v6.6.23." My confirmation copy back from the boinc_alpha server is timed at 17 April 2009 11:51 (that's BST, or UTC +1 - or about 23 minutes before the post on this board that you quoted). No replies on the mailing list yet, but I'm waiting for fuller particulars from server administrators - I've asked at Einstein and CPDN, two projects I know well which have had recent server overload problems, if they feel that they're suffering from what I'm calling "content free" schedulewr RPC calls from clients. Thyme Lawn has guided Milo in what to look for in the server logs - Jord has access to the area where the question was put, and where any answer will appear. None yet. |
|
4595)
Message boards :
Questions and problems :
Internet Trasfert limit
(Message 24477)
Posted 22 Apr 2009 by Richard Haselgrove Post: You can set up how much room BOINC can use, but 20MB is not worth the effort in my opinion. Depending on the project you want to run and with applications downloaded, that leaves even less for tasks to be stored. Don't you have another drive you can install BOINC its data directory to? umts means he's talking about a bandwidth (communications) limit - using a mobile phone or similar - not a storage limit. akenathos, you might be well advised to try and use a wired connection or wifi, if available, to do the initial setting up and project application downloads: then ask Jord again if he has a handy reference list or FAQ for projects with a low communication bandwidth requirement. |
|
4596)
Message boards :
Projects :
Seti down
(Message 24393)
Posted 18 Apr 2009 by Richard Haselgrove Post: SETI now has a "Our database is temporarily offline" note on the front page. Einstein also has a "Our database server is temporarily shut down for maintenance" message on the important pages. This is beginning to look like a trend. I wonder why so many projects are having database problems at the same time. What could the common factor possibly be? Not BOINC, surely? |
|
4597)
Message boards :
Questions and problems :
Not getting new WU's for CPU projects after upgrade to 6.6.20
(Message 24383)
Posted 18 Apr 2009 by Richard Haselgrove Post: This may also be the "automated call-back to home to update statistics/preferences" thingy ... Just for giggles, I did a manual update on the same v6.6.23 host, and got the expected "not requesting new tasks": 18/04/2009 13:38:43 Einstein@Home Sending scheduler request: Requested by user. 18/04/2009 13:38:43 Einstein@Home Reporting 1 completed tasks, not requesting new tasks 18/04/2009 13:38:48 Einstein@Home Scheduler request completed: got 0 new tasks 18/04/2009 13:38:48 Einstein@Home Message from server: Server can't open database I think the call-back thingy would also show as "not requesting new tasks", and that doesn't worry me at all: it's the "Requesting new tasks" which turns out not to be a request which is confusing and - as I argued - potentially dangerous. That's where I call 'bug'. |
|
4598)
Message boards :
Questions and problems :
Not getting new WU's for CPU projects after upgrade to 6.6.20
(Message 24382)
Posted 18 Apr 2009 by Richard Haselgrove Post: but "Requested 0.00 seconds of work" - while, in computer geek terms, still technically a request - translates in human terms into "Didn't request work", which takes you down a different trouble-shooting path. It could have been, but in this case it wasn't. 17/04/2009 11:21:51 [work_fetch_debug] Request work fetch: Backoff ended for Einstein@Home 17/04/2009 11:21:51 Einstein@Home chosen: CPU minor shortfall 17/04/2009 11:21:51 [wfd] ------- start work fetch state ------- 17/04/2009 11:21:51 [wfd] target work buffer: 173664.00 sec 17/04/2009 11:21:51 [wfd] CPU: shortfall 0.00 nidle 0.00 est. delay 0.00 RS fetchable 100.00 runnable 500.00 17/04/2009 11:21:51 climateprediction.net [wfd] CPU: fetch share 0.00 debt 0.00 backoff dt 0.00 int 0.00 (no new tasks) 17/04/2009 11:21:51 CPDN Beta [wfd] CPU: fetch share 0.00 debt 1267472.66 backoff dt 0.00 int 0.00 (no new tasks) 17/04/2009 11:21:51 Einstein@Home [wfd] CPU: fetch share 1.00 debt 0.00 backoff dt 0.00 int 120.00 17/04/2009 11:21:51 lhcathome [wfd] CPU: fetch share 0.00 debt 0.00 backoff dt 5213.54 int 86400.00 17/04/2009 11:21:51 orbit@home [wfd] CPU: fetch share 0.00 debt 72954.31 backoff dt 0.00 int 0.00 (no new tasks) 17/04/2009 11:21:51 SETI@home [wfd] CPU: fetch share 0.00 debt -284.69 backoff dt 1373.18 int 15360.00 (comm deferred) 17/04/2009 11:21:51 SETI@home Beta Test [wfd] CPU: fetch share 0.00 debt -596125.25 backoff dt 0.00 int 0.00 (no new tasks) (overworked) 17/04/2009 11:21:51 [wfd] CUDA: shortfall 21420.16 nidle 0.00 est. delay 0.00 RS fetchable 0.00 runnable 300.00 17/04/2009 11:21:51 climateprediction.net [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 0.00 int 86400.00 (no new tasks) 17/04/2009 11:21:51 CPDN Beta [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 0.00 int 0.00 (no new tasks) 17/04/2009 11:21:51 Einstein@Home [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 48874.10 int 86400.00 17/04/2009 11:21:51 lhcathome [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 61413.46 int 86400.00 17/04/2009 11:21:51 orbit@home [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 0.00 int 0.00 (no new tasks) 17/04/2009 11:21:51 SETI@home [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 0.00 int 0.00 (comm deferred) 17/04/2009 11:21:51 SETI@home Beta Test [wfd] CUDA: fetch share 0.00 debt 0.00 backoff dt 0.00 int 0.00 (no new tasks) 17/04/2009 11:21:51 climateprediction.net [wfd] overall_debt 0 17/04/2009 11:21:51 CPDN Beta [wfd] overall_debt 1267473 17/04/2009 11:21:51 Einstein@Home [wfd] overall_debt 0 17/04/2009 11:21:51 lhcathome [wfd] overall_debt 0 17/04/2009 11:21:51 orbit@home [wfd] overall_debt 72954 17/04/2009 11:21:51 SETI@home [wfd] overall_debt -285 17/04/2009 11:21:51 SETI@home Beta Test [wfd] overall_debt -596125 17/04/2009 11:21:51 [wfd] ------- end work fetch state ------- 17/04/2009 11:21:51 Einstein@Home [wfd] request: CPU (0.00 sec, 0) CUDA (0.00 sec, 0) 17/04/2009 11:21:51 Einstein@Home Sending scheduler request: To fetch work. 17/04/2009 11:21:51 Einstein@Home Requesting new tasks 17/04/2009 11:21:56 Einstein@Home Scheduler request completed: got 0 new tasks Note that in this case, because of a combination of backoffs, NNT and 'comm deferred' (SETI was down at the time), the Einstein/CPU resource was the only combination in 'fetchable' state. So it went ahead with the fetch, even though it was fetching nothing. And also note that the Einstein/CPU resource has just ended a 120 second backoff, implying that it has already been round the cycle a couple of times (while I was getting [wfd] set up to see what the heck was going on). |
|
4599)
Message boards :
Questions and problems :
Not getting new WU's for CPU projects after upgrade to 6.6.20
(Message 24379)
Posted 18 Apr 2009 by Richard Haselgrove Post: There's at least one bug still in work fetch, which can look like the problem reported here. I've reported it as a v6.6.23 bug, but the relevent section of code hasn't changed since v6.6.20 With <work_fetch_debug> on, it looks like this: 17/04/2009 11:21:51 Einstein@Home [wfd] request: CPU (0.00 sec, 0) CUDA (0.00 sec, 0) 17/04/2009 11:21:51 Einstein@Home Sending scheduler request: To fetch work. 17/04/2009 11:21:51 Einstein@Home Requesting new tasks 17/04/2009 11:21:56 Einstein@Home Scheduler request completed: got 0 new tasks In that particular case, the host actually had enough CPU work (no CPU shortfall), but wanted more CUDA work. Somehow, this got translated into a CPU request for Einstein - the CPU shortfall figure was copied to the work request, and you see the result above. I'm worried about this bug, on a number of levels. 1) Without [wfd], it's extremely confusing for users and project helpdesk volunteers. "Requested work, got nothing" sounds like a project server problem: but "Requested 0.00 seconds of work" - while, in computer geek terms, still technically a request - translates in human terms into "Didn't request work", which takes you down a different trouble-shooting path. 2) Getting no work (because you didn't ask for any) doesn't change the machine state, which means that it's going to make the same request the next time the work fetch algorithm is run. So that's an awful lot of unnecessary scheduler contacts. Einstein - I believe - still has 'resend lost results' enabled, so each scheduler RPC is going to entail a LOT of server work. Could this explain why Einstein - and seemingly a lot of other projects - have been "experiencing load problems on the database server" (front page) recently? 3) We now have a client which instigates a per-resource backoff (not visible in the message log without [wfd] - you can look on the new 'properties' page for the project) when a work request results in no tasks being issued. Without considering the reason. So these "got 0 new task" replies will back off CPU work requests. That goes some way towards mitigating the server workload, but introduces another problem. Eventually, the point will come where the project really does need new work - but by then, it could be backed off by many hours. In effect, we have "the client which cries wolf" and ends up starving itself. And this is now the recommended download. I hope we can identify and cure the remaining bugs, and replace v6.6.20 with v6.6.24 or whatever as soon as possible. |
|
4600)
Message boards :
Projects :
Seti down
(Message 24355)
Posted 17 Apr 2009 by Richard Haselgrove Post: they'll be getting 200 mb cuda from me when they come back up. The upload server has never been offline - slow and congested, maybe, but not offline. The rest of the project is back now (including message boards) - all done bar the sweeping up. |
|
4601)
Message boards :
Projects :
Seti down
(Message 24348)
Posted 17 Apr 2009 by Richard Haselgrove Post: Looks like they're working on it, and have got the fundamentals back in place. I've had uploads, reporting, new work issued and downloads: just all very slowly, in standard network congestion ("outage recovery") mode. Don't believe what you read on the Server Status page, and the message boards are still down. |
|
4602)
Message boards :
Projects :
News on project outages.
(Message 24341)
Posted 17 Apr 2009 by Richard Haselgrove Post: SETI@home seems to have lost contact with its BOINC database: "Unable to connect to database - please try again later Error: 2013 Lost connection to MySQL server at 'reading initial communication packet', system error: 111" Friday is not a good day for this to happen, because they are even more short-staffed than usual, and in any event their working day doesn't start for another couple of hours. Patience may be needed. |
|
4603)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 24268)
Posted 13 Apr 2009 by Richard Haselgrove Post: No more than 5? Your first screenshots showed quite more than that :D Different machine - first was a dual core SBS 2003, second a single core Standard Server 2000. |
|
4604)
Message boards :
BOINC client :
What am I missing for CUDA?
(Message 24143)
Posted 6 Apr 2009 by Richard Haselgrove Post: And don't just say "the latest" - that'll be wrong in a week or two. Give us actual numbers (for BOINC and CUDA drivers), the same as you did for your motherboard. |
|
4605)
Message boards :
Questions and problems :
Boinc Only Recognizing 2 of 4 Cores
(Message 24028)
Posted 31 Mar 2009 by Richard Haselgrove Post: Windows XP Pro should be fine - I ran 8 cores under it for a while. But check it out, just to be sure - how many CPU graphs do you see on the 'Performance' tab in Task Manager? |
|
4606)
Message boards :
Questions and problems :
remove obsolete host
(Message 23996)
Posted 29 Mar 2009 by Richard Haselgrove Post: I don't see any computer drop-down lists on BOINC websites ;-) For Windows, the list seems to be kept in the registry at [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager\ComputerMRU] |
|
4607)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23914)
Posted 25 Mar 2009 by Richard Haselgrove Post: Yet again, BoincView solves it: allow the user a choice. Sorry, sorry. Misinterpreted again. I really must learn to write one of these days. ;-) <-- smiley BoincView is no longer in existence. BoincView never was available except for Windows. And it never was perfect, even though it was good enough to let me stop looking for anything better. So the time has come to steal the best bits and build them in to something else All I wanted to convey was that BV had a solution to the question you were discussing. Should we show technical names for tasks, or should we show friendly names? My answer (unashamedly stolen from BV) is that, if the answer isn't immediately obvious, then make both available but let the user choose which one not to clutter their screen with. Now, if you really want to build a better mousetrap, the feature I'd really like someone to steal (this time from M$ Access) is "Remove Filter/Sort" - whichever column heading you've clicked on, forget it, and go back to showing the order the tasks arrived in. |
|
4608)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23912)
Posted 25 Mar 2009 by Richard Haselgrove Post: Yes, but many will like the application column {snip}. Yet again, why have this argument about what (unspecified numbers of) (unspecified) people may or may not prefer, depending whether there is an 'R' in the month or what direction the wind is blowing from? Yet again, BoincView solves it: allow the user a choice. This all reminds me why BoincView is by so much a better program than BOINC Manager: it was designed from the beginning with the user in mind, to show the user whatever that user might want to know. It doesn't try to pre-judge what the user 'ought' to want to know. |
|
4609)
Message boards :
Questions and problems :
Delay starting
(Message 23873)
Posted 24 Mar 2009 by Richard Haselgrove Post: According to the documentation for <start_delay>, the delay is introduced after the core client starts, and before the science applications start - which isn't much help with a network card issue, since it's the core client which does network comms, not the science application. In Windows Vista, there's a useful startup option for services: "Automatic (delayed start)", but I don't know how to do that in earlier versions. (Last time this question came up, Jord posted a link to How to delay loading of specific services, but I haven't needed to try it myself). |
|
4610)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23851)
Posted 22 Mar 2009 by Richard Haselgrove Post: BOINCview is dead in development come next month 2 years. That interface has allot that could be adopted. Add a column for efficiency. That will signal people who are not throttling that something is not right and those that are throttling an indication that 60% is ~60%. 2 birds with 1 stone. And also there's an awful lot of really useful design in BoincView which BOINC has failed to imitate in the two years since development stopped: User-selectable columns (have both CPU and GPU available, let the user decide which one to hide/show - or keep both if their screen is large enough). Totals/counts/averages for columns. Filtering by project (recently introduced - in a less sophisticated way - for the messages tab only). |
|
4611)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23841)
Posted 22 Mar 2009 by Richard Haselgrove Post: Wallclock, hmmm, yet to find the reasoning of why this change, but a strong downside is that it makes any slow running highly obscured, since few will actually drill into the job properties to find out the CPU time. At any case, last night on a quad/Vista with 6.6.15 client, I remotely looked in BOINCview and saw that 3 of 4 processes were running at ~32% CPU efficiency. Walking over and checking what else was active found that somehow 3 of the 4 sciences were running of 1 core, with 2 cores idle ?!? Whether BOINC allowed this or the OS did this is entirely outside my ability to diagnose. The story is, that showing elapsed time aka wallclock is very misleading. The change arose out of CUDA development, where the opposite problem applies: showing the CPU time of an application running mainly on a GPU is likewise misleading. The trouble is, the quick change (and it was an extremely quick, spur of the moment, decision, if you read the boinc_alpha mailing list) means that no consideration was given to the side-effects such as the one you've described. Another consequence of the BOINC 'new feature' development model. |
|
4612)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23763)
Posted 18 Mar 2009 by Richard Haselgrove Post: Funnily enough, I was just musing on a similar theme. As an utterly, utterly trivial example: I claim bragging rights to changeset [trac]changeset:17606[/trac], following a report I made to BOINC_alpha on Monday. Now I don't expect my name in lights, or bonus cobblestones, or anything like that: fixing a typo is perfectly reasonably a silent process. But I still think it would be better if error reports (even this trivial) had some sort of reply. One word would do. I don't expect "Thanks": "Noted", "Fixed", or even "Ooops" are plenty. But there is a severe lack of 'ack' packets in the communication link between users/testers and development/administration. |
|
4613)
Message boards :
Questions and problems :
app_info.xml question
(Message 23747)
Posted 16 Mar 2009 by Richard Haselgrove Post: Thank you, Ah, so that's what you're trying to do. There's a way of fooling it. If you add a <file_ref> entry in the latest, current app_version (608 in your case), with the file name of the old (603) application - though not as main_program, obviously - that will keep the application file from being deleted from disk. Don't worry about client_state. I'm pretty sure that if BOINC finds a 603 reference in a workunit/result as it starts up, and there's a matching 603 entry in app_info and the necessary file on disk, it'll sort itself out without complaint. If you're writing the script I think you're writing (and somebody else writing the same script has already consulted me about this), then for BOINC v6.6.15 you also need to remove the plan_class line from the result block. Otherwise the task errors on restart. |
|
4614)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 23742)
Posted 16 Mar 2009 by Richard Haselgrove Post: No ideas? Here it is again:  Again on a server, this time Windows 2000 and not a domain controller. There has been some discussion at SETI recently about BOINC showing excessive usage, but that has been identified as a side-effect of running with cached task lists running into the hundreds or even thousands. This is my home server, with just a humble P4 Northwood (single core) and usually no more than 5 tasks on board. I noticed it because BOINC was too busy doing something to respond to my BoincView monitor. BV now reports it running Einstein at 99.32% efficiency, so problem solved. I think both these cases may relate to the length of time that these small servers may be left running unattended without rebooting. This morning, I finally got round to installing a set of Windows Updates that had been hanging around for, probably, months: it was the reboot which did the trick. BOINC as a Windoze Update laziness monitor? That's another selling point, alongside recovering stolen laptops. |
|
4615)
Message boards :
Questions and problems :
app_info.xml question
(Message 23741)
Posted 16 Mar 2009 by Richard Haselgrove Post: Thanks. Jord Those sound like SETI numbers. In all BOINCs up to v6.4.7, the version numbers refer to the application - in this case, I'm guessing 'setiathome_enhanced'. So, even if the version sections actually refer to different programs (say, a CPU program and a CUDA program), BOINC thinks they're all talking about the same thing, assumes they're functionally equivalent, and the only one you want is the newest, latest and greatest, right? So in your example case, it'll hang on to 608, but throw away 602 and 603 as soon as it thinks you've finished with them - which, as Jord says, will be when you've finished the last work for them. And since it will never fetch any new work for older versions, that'll be quite soon. BOINC v6.6.14/15, on the other hand, considers both the application and the plan class - e.g. CUDA. So you can have a version 603 without a plan class, and a version 608 with plan class CUDA, it will keep/use both versions. But the downside is you have to include much more detail in the app_info file if you want it to work with v6.6.15: if you can do that (there's an example posted on the SETI boards), then v6.6.15 is worth upgrading to. |
|
4616)
Message boards :
Questions and problems :
Show graphics on Linux
(Message 23686)
Posted 15 Mar 2009 by Richard Haselgrove Post: I heard that BOINC version 6 will run version 5 work units, but not display version 5 graphics. That is part of the problem, as Astropulse is still in version 5. (If I want to see graphics, maybe I should downgrade to BOINC version 5.) No, don't do that. There are versions, and versions, and versions. Astropulse is BOINC v6 compliant. That's the serious point. Hold there if you don't want the laughs. Astropulse is only the right description for vesions up to v5.00 Everything above v5.00 is Astropulse_v5 And for Linux, the graphics should work for x86, but they never got x64 graphics to work. Apart from that, it's a great app. |
|
4617)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23673)
Posted 13 Mar 2009 by Richard Haselgrove Post: One thing noticed on various releases starting 6.6.9, and currently on 6.6.11 client is that the dates (Vista 32), randomly show in alternation 24 hour or 12 hours am /pm notation. When hitting the message filter, all message lines fall in line. Note that the system is set to 24 hour notation, but the log shows then in 12 hour am/pm. And continuing through 6.6.14 with the interim, pre-release, v6.6.15 (ish) manager. |
|
4618)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23617)
Posted 11 Mar 2009 by Richard Haselgrove Post: And having done that, now I find that BOINC doesn't - for CUDA tasks - start them in order of receipt any more, but always starts them Earliest Deadline First. Welcome back EDF. 20 minutes. BOINC's so obsessive about EDF, it even pre-empts running tasks. You'd think that, even with a 7-day deadline for 'shorty' tasks (and 5 days slack in the cache), it could afford to wait 20 minutes for completion? |
|
4619)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23614)
Posted 11 Mar 2009 by Richard Haselgrove Post: The scheduler is indeed working well. On the two machines where I've stopped fiddling and changing things, and where I have got DCF nicely balanced across all four sub-projects, I'm seeing a perfect procession of 11/03/2009 10:55:42 SETI@home Sending scheduler request: To fetch work. 11/03/2009 10:55:42 SETI@home Reporting 1 completed tasks, requesting new tasks 11/03/2009 10:55:47 SETI@home Scheduler request completed: got 1 new tasks Trouble is, this is a CUDA machine, so those requests go through about once every 20 minutes: and even with long-running tasks on other projects, and a reasonable cache size, the sched_request file is 60KB. That's an awful lot of administrative network traffic. |
|
4620)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23613)
Posted 11 Mar 2009 by Richard Haselgrove Post: And having done that, now I find that BOINC doesn't - for CUDA tasks - start them in order of receipt any more, but always starts them Earliest Deadline First. Welcome back EDF. LOL |
|
4621)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23611)
Posted 11 Mar 2009 by Richard Haselgrove Post: Indeed 6.6.15 fixed that issue of empty grid lists and status bar. Showing tasks in order of receipt is the 'natural', unsorted, display order for tasks. But once you have applied a sort (of any description), there is no 'unsort' option in BOINC: I have to resort to editing the Registry, which is hardly an elegant solution. And having done that, now I find that BOINC doesn't - for CUDA tasks - start them in order of receipt any more, but always starts them Earliest Deadline First. Welcome back EDF. |
|
4622)
Message boards :
Questions and problems :
6.6.12:Hostname missing in boincview
(Message 23586)
Posted 10 Mar 2009 by Richard Haselgrove Post: I think some (but not all) BoincView bugs are eliminated either in BOINC v6.6.14, or in postfixes which should appear in v6.6.15 My remaining problem with BV (not yet reported to boinc_alpha, but will do) is inaccurate reporting of 'To completion' times for tasks which haven't started running yet - once they start, the remaining 'To completion' time is accurate, for both CPU and CUDA work. This relates to SETI running with CUDA, app_info.xml, optimised applications, etc. etc. - other projects are shown accurately, as before. In BB's screenshot, I expect the est. credits are shown as 0.00 because BV hasn't been able to get a speed (benchmark) figure for the host either. BV only calculates an estimated credit for the old (deprecated) benchmark*time credit system: always interesting to see how far that varies from other credit award schemas! |
|
4623)
Message boards :
Questions and problems :
Point me towards documentation on where BOINC project data is kept in OSX?
(Message 23450)
Posted 4 Mar 2009 by Richard Haselgrove Post: Or any system for that matter? I'm not seeing this in any of help or FAQs. Seems a pretty simple thing to want to know, so either I'm being very dumb or this is very well hidden. The locations are listed in Jorden's The Big BOINC 6 Answer Thread in the FAQ Service. |
|
4624)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23383)
Posted 1 Mar 2009 by Richard Haselgrove Post: There are reports from SETI that work requests for CUDA plan class are not being filled, and instead are being met with a 86400 second (1 day) backoff. Not clear from the log whether this is server-mandated or a client response: 3/1/2009 6:14:18 AM SETI@home [wfd] request: CPU (0.00 sec, 0) CUDA (371520.00 sec, 1) SETI multibeam (practicable for CUDA) work was available around that time - request not made during a project outage. References: SETI Will not fetch new work! Lunatics pre-release Problem downloading Seti Enhanced - Astropulse ok |
|
4625)
Message boards :
Questions and problems :
one project on dual core
(Message 23359)
Posted 28 Feb 2009 by Richard Haselgrove Post: "System clock was turned backwards": probably has very poor time metrics, and possibly running in EDF. I would expect a second task to be downloaded and to start running as the current task approaches 100% and the time "To completion" approaches 2 hours 24 minutes (default cache size for an unmodified project 'BOINC settings' page). But to be sure, we still need to know the name of the application/project which is running. |
|
4626)
Message boards :
Questions and problems :
nvlddmkm.sys bsod when running BOINC tasks
(Message 23357)
Posted 28 Feb 2009 by Richard Haselgrove Post: Ok...no luck. Even with GPU use turned off, I am still getting the nvlddmkm.sys bsod. But still only when BOINC is running tasks. You never told us what non-GPU task were running when you got these driver crashes. The answer isn't 'BOINC' - as Ageless said a while back, BOINC doesn't do any computing, and very little graphics (unless you call up the statistics charts). The 'tasks' come from any one of 70+ science projects, with up to four different experiments (science applications) each. Which were you running? |
|
4627)
Message boards :
Questions and problems :
6.4.6 localization broken?
(Message 23336)
Posted 27 Feb 2009 by Richard Haselgrove Post: Report from SETI boards that installer still broken at ~08:45 UTC today. |
|
4628)
Message boards :
BOINC Manager :
boinc and screenreaders
(Message 23322)
Posted 26 Feb 2009 by Richard Haselgrove Post: Comment by Richard Haselgrove: "A verbal (dis-)agreement does not satisfy the requirements of anti-discrimination legislation." |
|
4629)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23314)
Posted 26 Feb 2009 by Richard Haselgrove Post: There's been another report of (so far) unexplained preempting for 1 second with BOINC v6.6.10 - this time without the 'CPU - throttle' explanation. http://lunatics.kwsn.net/gpu-crunching/it-works.msg14850.html#msg14850 (SETI third-party site - CUDA application) |
|
4630)
Message boards :
BOINC client :
6.6.9 Windows Notes & Bugs
(Message 23303)
Posted 25 Feb 2009 by Richard Haselgrove Post: It looks as if Rom is clearing the decks ready for an official release - he's just bumped the milestone on a couple of my recent tickets (a bugfix from 6.6 to 6.8, and an enhancement from 6.6 to Undetermined). I also saw that he'd fixed a couple of older ones today, but none of mine - boo-hoo. Get your version 6.6 test reports in now - hurry, while stocks last. |
|
4631)
Message boards :
Questions and problems :
nvlddmkm.sys bsod when running BOINC tasks
(Message 23254)
Posted 23 Feb 2009 by Richard Haselgrove Post: Don't forget also that BOINC/SETI will be putting a sustained, 100%, continuous load on the power supply and all GPU components for hours, perhaps days, continuously. That's likely to be a different working envelop than even the most intensive games. |
|
4632)
Message boards :
Questions and problems :
Task Uploading Issues..
(Message 23232)
Posted 22 Feb 2009 by Richard Haselgrove Post: Are you sure the problem is at your end? For the last 29 hours, SETI's communications bandwidth has been stuck at absolute maximum download rate. When this happens, uploads fail for everyone (I have 193 queued), because BOINC doesn't have adequate back-off and load balancing facilities. The SETI problem is being caused because they can't distibute a new science application - Astropulse_v5 5.03 - quickly enough. Most tasks are failing with a download error -200 on the executable file. Yet BOINC immediately tries again, gets another task, clogs the available bandwidth with the associated 8MB datafile, and fails again on the absent .exe file. Tasks which should take 3 days are being recycled after 5 minutes. The only available tool that BOINC has to limit this runaway bandwidth consumption is the daily quota. But at SETI, this has been set at 100/day to suit the much shorter (and less data-intensive) multibeam tasks. Quota, like other BOINC parameters, needs to be moved to the application level, instead of the project level. |
|
4633)
Message boards :
BOINC client :
6.6.5 still ignoring no-new-work
(Message 23051)
Posted 11 Feb 2009 by Richard Haselgrove Post: and (b) we haven't come up with any ideas yet about what we, as client operators, can supply by way of evidence to help him track it down. Yes, I saw that one, but surely <work_fetch_debug>, <cpu_sched_debug> and <debt_debug> collectively can only reveal the decision-making process by which the local client came to its conclusion: "not requesting new tasks". The server shouldn't, AFAIK, be responding to that reasoning: it should only be responding to the actual sched_request...xml. Or are you suggesting, in light of Dj Ninja's "bug or feature?" thread on the mailing list, that the server is in effect saying "hehe - I've checked your calculations on the list of all the work you've got on hand, and you got it wrong: you didn't ask for any work, but you should have done, so I'm sending some to you anyway - so there". Still, it never hinders the investigation process to have additional information on hand, and possibly having those flags turned on will reveal something relevant. Of course, the one logging tool that would really help our understanding here would be the server log: if the problem was being reported at Einstein, we could click through from the host record to the relevant server scheduler log for the host: but that isn't available at SETI Beta, and I'm not going to email Eric and ask for it while he's in the middle of an application launch. |
|
4634)
Message boards :
BOINC client :
6.6.5 still ignoring no-new-work
(Message 23047)
Posted 11 Feb 2009 by Richard Haselgrove Post: Unfortunately, it doesn't seem to be as simple as that. Both you and BB were "not requesting new tasks". You didn't get any, as is correct: BB got 2 new tasks anyway. That isn't a NNT problem, it's a server problem. It happens very rarely, which makes it very difficult to isolate and fix (if everyone was getting work they hadn't asked for, all day every day, I think it would have been solved by now!) All we can say to David Anderson is (a) the problem isn't solved yet: it still happens, though rarely, and (b) we haven't come up with any ideas yet about what we, as client operators, can supply by way of evidence to help him track it down. All I can suggest is that anyone who actually spots it happening should disable network activity (to prevent files being over-written) and take a copy of the sched_request_ and sched_reply_ XML files for the project. |
|
4635)
Message boards :
BOINC client :
6.6.5 still ignoring no-new-work
(Message 23041)
Posted 11 Feb 2009 by Richard Haselgrove Post: This email exchange took place on the BOINC Development mailing list on 30 January. From: "David Anderson" David often does the SETI server updates himself (he has a dual role as Director of SETI, as well as lead developer for BOINC - see About SETI). BB's screen-shot suggests that the problem, identified prior to 30 January, has not been cured by whatever upgrade David applied. Which client-side logs (reportable by project participants) would help the BOINC developers track down this server bug? |
|
4636)
Message boards :
Questions and problems :
Fun with versions 6.4.5 and 6.6.3/4
(Message 23010)
Posted 9 Feb 2009 by Richard Haselgrove Post: Another interesting post at SETI: message 863921. This one seems to suggest that the v6.6.4 Manager is OK when showing only Astropulse and v6.08 (plan_class CUDA), but crashes when SETI v6.03 (no plan class) are mixed with v6.08 Since the precise benefit of v6.6.x is the ability to mix v6.03 and v6.08, this leaves us in a "damned if you do, damned if you don't" situation wrt upgrading. |
|
4637)
Message boards :
Questions and problems :
Another 6.4.5 upgrade problem.
(Message 23004)
Posted 9 Feb 2009 by Richard Haselgrove Post: Ok, I don't know what version of Windows you're running, but the advice was: "Use anything *except* .\program files\anything\anything else". It's the "program files" tree which is protected in Vista, Windows 7, and presumably anything later than that. You don't explain why you can't use C:\Documents and Settings\All Users\Application Data\BOINC\ (the Windows XP default), but you should still be able to make C:\ProgramData\BOINC\ (the Vista default) even if you're not running Vista. |
|
4638)
Message boards :
Questions and problems :
Fun with versions 6.4.5 and 6.6.3/4
(Message 22986)
Posted 9 Feb 2009 by Richard Haselgrove Post: (what's the plural of BOINC?) BOINCen? ... maxing out the amount of work from Milkyway (24 tasks, 12 per CPU) ... That could be where the 24 hours is coming from - daily quota limit reached, backoff set by project. You would need to look back in the logs (stdoutdae.txt, if the manager is playing up) and see what happened just before the 24-hours delay was imposed. There'll be a work fetch request for Milkyway, and a line about quota, if I'm right. |
|
4639)
Message boards :
Questions and problems :
Fun with versions 6.4.5 and 6.6.3/4
(Message 22982)
Posted 9 Feb 2009 by Richard Haselgrove Post: .... but when a program goes from stable to inoperable and all that has changed is the downloading of WUs from a new project, .... That's an interesting comment. I doubt that Milky Way made the difference, but downloading SETI tasks would load the GPU and try to display plan_class data into the tasks pane. That's the point where the developers believe the fault lies. Do we have any verifiable reports for v6.6.3/4 that might suggest that the tasks pane is stable and usable if only CPU tasks are loaded, but crashes as frequently reported when a plan_class is used? That would be a pretty powerful clue. Also, do we know whether the developers' pre-release test machine (I presume they have one?): a) is fitted with a CUDA card? b) is actually loaded with CUDA tasks? |
|
4640)
Message boards :
Questions and problems :
Fun with versions 6.4.5 and 6.6.3/4
(Message 22977)
Posted 9 Feb 2009 by Richard Haselgrove Post: I have to ask - what were you hoping to achieve by downloading an application clearly marked (MAY BE UNSTABLE - USE ONLY FOR TESTING) It sounds as if these versions are exactly as described: UNSTABLE - lol. If your intention is to help the developers uncover the bugs - welcome aboard. Please post your error messages and debug logs here. If you would prefer to continue stable production crunching for your science projects, I suggest you stick with the recommended stable version. |
|
4641)
Message boards :
BOINC client :
who is mishandeling the "no new work"
(Message 22968)
Posted 9 Feb 2009 by Richard Haselgrove Post: Of course it's BOINC. But I think it would be helpful if we could work out whether it's the BOINC client (which the user can do something about - enable logging flags and report behaviour, in detail, to the developers: or just downgrade to a stable version), or the BOINC server (where all the user can do is send advice to project admins to update their server software, if we can identify which changes the admins need to apply). FWIW, I think I've seen both client problems and server problems reported on the boards. I even got caught myself by a server problem at SETI (Work fetch anomaly). In that case, we were told that the server issue had been fixed: we looked for, but couldn't find, any relevant changes in the server code base: but the problem seems to have gone away. It's difficult to debug "Server sends too much work" problems at SETI, because often enough they can't generate enough work to fulfil legitimate requests - but work consumption has entered a quieter phase recently, so we might be able to find out whether this particular problem has been permanently laid to rest. |
|
4642)
Message boards :
Questions and problems :
Domain controller support
(Message 22919)
Posted 6 Feb 2009 by Richard Haselgrove Post: You have a CUDA graphics card on a domain controller? ;-) That's the only reason for installing v6.4.5 that I know of: just stick with the v5.10.45 if you don't need CUDA. But it's good to keep reminding them of the promise they made with the release of v6.2.4: - WINSETUP: Block installing on domain controllers for this release. |
|
4643)
Message boards :
BOINC client :
Are you Running as a Service? Can't get it to work on new versions
(Message 22910)
Posted 5 Feb 2009 by Richard Haselgrove Post: There should be a service visible in service manager - drill down through 'Control Panel' to 'Administrative tools'. There have been some reports of a corrupted registry entry preventing installation as a service - let us know if BOINC doesn't appear in service manager after a 'Protected Application Installation', and we'll go look for it. |
|
4644)
Message boards :
BOINC client :
Vulnerability in BOINC 6.4.5
(Message 22904)
Posted 5 Feb 2009 by Richard Haselgrove Post: Are developers aware of this published from Secunia vulnerability of BOINC 6.4.5? The Secunia advisory references BOINC's own trac [trac]#823[/trac], where a fix has already been added to the code. (12 January 2009) So the developers are aware of the problem: full marks for that one. But the "recommended" download is dated 9 December 2008 (and I have re-downloaded it tonight - the executables are also datestamped 9 December 2008): no sign of a recall and re-issue. Not so good. |
|
4645)
Message boards :
BOINC Manager :
boinc and screenreaders
(Message 22824)
Posted 1 Feb 2009 by Richard Haselgrove Post: Comment (strong comment) added to still-open trac ticket [trac]#147[/trac] At least the registry key values are the same in BOINC v6.4.5 |
|
4646)
Message boards :
Questions and problems :
CUDA makes computer really slow
(Message 22814)
Posted 1 Feb 2009 by Richard Haselgrove Post: Please join the many discussions on the SETI message boards for the answer to what is essentially a SETI problem, not a BOINC problem. Message 860285 is relevant. But in brief: the current SETI application is under-developed. It struggles on certain tasks (VLAR), with the symptoms you describe. The developers are aware of the problem, but there is no timetable for a fix yet. |
|
4647)
Message boards :
BOINC Manager :
boinc and screenreaders
(Message 22812)
Posted 1 Feb 2009 by Richard Haselgrove Post: Agreed. The 'Simple' view should respond to accelerator keys - Alt-D for consistency with the Advanced menu in Advanced view - or at the very least, allow tabbing between buttons. But the state is persistent (as you have found!). And if it's persistent, it's stored somewhere. And if it's stored, it can be found and changed. I have found the solution for BOINC v5.10 running on Windows XP. If your combination is wildly different, ask again. Here's a little RegEdit script. Save the text between the 'code' tags which follow as a file with the extension .REG, then open it with the default handler. Windows Registry Editor Version 5.00 [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager] "GUISelection"=dword:00000001 If you have other tools for editing the registry, search for the "BOINC Manager" key, and change the data for "GUISelection" from 2 (simple) to 1 (advanced). BTW, with the v5.10 default 'simple' view, it's probably illegal to distribute BOINC in the UK under the terms of s19 of the Disability Discrimination Act 1995 |
|
4648)
Message boards :
BOINC client :
6.4.5 unnecessarily assigning high proiority CUDA
(Message 22790)
Posted 31 Jan 2009 by Richard Haselgrove Post: Now you can see it all on one page:  (direct link) Note the six running tasks, and the Astropulse at the top (oldest tasks in cache except CPDN, still unstarted, wrong time estimate). Strangely, it's now decided that the least-processed Einstein task from last time is now safe to pre-empt, even though the most-processed one is still in EDF. |
|
4649)
Message boards :
BOINC client :
6.4.5 unnecessarily assigning high proiority CUDA
(Message 22789)
Posted 31 Jan 2009 by Richard Haselgrove Post: I think I've worked out one of the reasons why we get these 'high priority' complaints about v6.4.5 Refer to  (direct link) This is from a quad (one of my Q6600s) with a single 9800GT. A CUDA task is running, and there are other projects not visible at the top of the screen (2x CPDN Beta, 1 x CPDN, 2 x Astropulse). I'm running a cc_config.xml with <ncpus> set to 5 - the SETI app doesn't use enough CPU to need a full core, and now that the thread priorities (as opposed to the BOINC priorities) have been sorted out, it runs fast enough even when just scavenging CPU cycles from other BOINC tasks. Task switch interval is the default 60 minutes, and cache size is currently set to 2 days. The interesting project is Einstein. There are four tasks running, high priority: plus a fifth task waiting to run. I find that when running multiple projects like this, BOINC v6.4.5 gets itself into two minds about task switching. First it wants to pre-empt SETI (big STD at the moment), so it starts five CPU tasks (as per cc_config), and then it realises that the GPU is idle, so it starts a sixth task to utilise the CUDA resource. Then it realises that it's got too many cores running, and suspends the CPU task it's just started. The particular problem with Einstein is that tasks take a lot of preparation time before they get started (much longer than the 9 seconds BOINC allowed this task to run). So we have a task with positive CPU time, and zero progress - BOINC is worried that the runtime will be infinite, and that's why it runs the rest of the Einstein tasks in EDF. At five hours per task, and a fortnight before the first deadline, there's no other reason for the panic that I can see. I've looked through the change logs since v6.4.5, and I can't see any attention being given to task switching: anyone spot any flaws in my logic before I raise a trac ticket? Afterthought: a nasty interaction between resource share, sub-projects, TDCF and deadlines. At the moment, my SETI resource share is below 20%, so SETI (my only CUDA project) is pushed into massive STD by the imperative to keep the CUDA card loaded. That means that my Astropulse tasks, which are also classed as part of the SETI project, never get a chance to run. At the same time, of course, my SETI TDCF is being stabilised at the appropriate value for the CUDA app - currently 0.0423. At this value, BOINC is estimating that my Astropulse will complete in 5:01:57 Yet even using the fastest-available third-party optimisations for Astropulse (as I am), they will take at least 10 hours - TDCF should be around 0.1 for accurate estimation of optimised AP tasks, or 0.4 for SETI's stock AP application. BOINC will eventually get round to running AP in high priority, of course, but because of the TDCF basis error, it may well not start them in time - it certainly wouldn't give them the full 42 hour runtime that would be needed if I was using stock AP apps. There was talk on the mailing list of getting TDCF split out into a 'per application' record: this example shows why that development work should be "high priority". {But I can micro-manage my way out of the current problem} |
|
4650)
Message boards :
BOINC client :
BOINC 6.6.2: work fetch
(Message 22743)
Posted 28 Jan 2009 by Richard Haselgrove Post: Report copied from SETI Beta: Fetching had been repaired in BM 6.6.3? No... Again (as in 6.6.2), pressing update at beta project leads to 10-WUs request with already full queue, even with NNT. (BM 6.6.3, Windows Server 2003 x86) Again time format of displayed time in BM's tasks window at computer with non-English localization differs sporadically (but all selected lines with different time format being copied into clipboard are looks in similar locale-specific format). |
|
4651)
Message boards :
Questions and problems :
Boinc 6.6.2
(Message 22733)
Posted 28 Jan 2009 by Richard Haselgrove Post: ... You are trying to use a Beta version which has serious bugs. If you want to test and ennumerate the bugs for the developers to work on, then go ahead. If you want to run science applications in a stable BOINC environment, use an earlier version. |
|
4652)
Message boards :
BOINC client :
BOINC 6.6.2: work fetch
(Message 22705)
Posted 26 Jan 2009 by Richard Haselgrove Post: Yes please - other people may be interested in helping to try and disgnose the problem too, and posting it here would save you sending a separate PM/email to each of us. |
|
4653)
Message boards :
BOINC client :
scheduler choking on seti GPU WUs
(Message 22686)
Posted 25 Jan 2009 by Richard Haselgrove Post: I dont think that seti or seti beta have a quota limit. Yes they do, but it's probably bigger than you're used to seeing: From message 856650: In another post, someone worked out that with their multi-core CPUs and multiple graphics cards, their 'per computer' daily quota would be 2800 (with the newly-increased quota limits counting a CUDA card as 5 CPU cores for quota purposes). |
|
4654)
Message boards :
Questions and problems :
Boinc 6.6.2
(Message 22676)
Posted 25 Jan 2009 by Richard Haselgrove Post: Ulf That 'feature' can be overcome by using the ncpus parameter in a cc_config.xml file. |
|
4655)
Message boards :
BOINC client :
BOINC v6.6.2 Work fetch: client/server issues
(Message 22634)
Posted 24 Jan 2009 by Richard Haselgrove Post: Using 6.2.18, and running cpdn, when I report a task with NNW set, NO new work is downloaded. Yes, I'm sure it is. But it's the frantic tinkering with the BOINC code base (to enable BOINC to support CUDA) which has allowed bugs like this to creep out into the wild. Coming soon to a server near you? Hopefully not. |
|
4656)
Message boards :
Questions and problems :
BOINC and dual boot systems - is it possible to use one BOINCData folder?
(Message 22624)
Posted 23 Jan 2009 by Richard Haselgrove Post: I want to create dual-boot system Vista x86/Server 2008 x64. I've not actually tested it, but I think that it should be OK. You may have poblems, though, for those (very few) projects which automatically download 64-bit project applications on 64-bit OSs. Because the project applications go in the (shared) data area, and are listed explicitly in client_state.xml, you might get BOINC confused if such a project had to fetch new work when running in 64-bit mode - you might get apps which were incompatible with later 32-bit running. To avoid that, you could lock the project down to 32-bit running with an app_info.xml - though that might defeat the purpose of running 64-bit in the first place. |
|
4657)
Message boards :
BOINC client :
BOINC v6.6.2 Work fetch: client/server issues
(Message 22611)
Posted 23 Jan 2009 by Richard Haselgrove Post: Is anybody monitoring the work-fetch reports at SETI (likely high take-up of v6.6.2 because of their CUDA application)? 1) Server somestimes sporadically issues far more work than requested - SETI message 855982 and confirmations. 2) Server issues work when 'No New Tasks' in effect - reported here at BOINC Dev. 3) CUDA work requests don't take account of cached work already received - keep requesting the same amount until daily quota reached - SETI messages 856650, 856622 4) Server won't issue plain, bog-standard CPU work to a plain, bog-standard BOINC client (v5.10.45) - SETI message 856753. Backward compatibility, anyone? v6.6.2 definitely should never be upgraded beyond Beta. |
|
4658)
Message boards :
Questions and problems :
Most of my projects exit with zero status but no 'finished' file
(Message 22593)
Posted 22 Jan 2009 by Richard Haselgrove Post: I am getting very frustrated that most of the work my computer is doing seems to be for naught. I have a brand new Compaq CQ50 notebook with 2GB RAM, 2GHz Intel Celeron single-core CPU, 160 GB hard disk, and running Windows Vista Home Basic Edition, so I am relatively confident that my system meets the minimum specs for most projects. Yet, almost all of my work units have been exiting with a zero status, but no 'finished' file, according to my message log. This seems like a monumental waste of computing time -- not just for me, but for the projects to which I am attached. Norton Internet Security and Windows Defender were preinstalled on my system, so I am unclear if they might be the culprits here. Can anyone provide some insight into what I may perhaps be doing wrong, or where I might look to get back on track? Thanks... Also, please be aware that in the vast majority of cases, BOINC will automatically re-start any task that exits with zero status but no 'finished' file, and it should re-start from the last stored checkpoint. So normally, all you lose is a few seconds (or minutes, in the case of CPDN) of processing time - nothing like as bad as you seem to be fearing. The real problem happens if the exits are so frequent that the application never has time to progress to the next checkpoint, or so numerous that they reach BOINC's limit of 100 exits per task - then, the task is permanently abandoned, and the time is indeed wasted. The only time that is likely to happen is if the exits are caused bu CPU throttling, as Gundolf says. |
|
4659)
Message boards :
Questions and problems :
Boinc on a dual-boot system
(Message 22579)
Posted 21 Jan 2009 by Richard Haselgrove Post: I was wondering if there was some way to get the windows/ubuntu instillations to process the same projects? Not strictly true. Work downloaded to a host with a particular host ID has to be uploaded from a host with the same host ID - but since the host ID can travel between different physical hardware in the BOINC Data folder, that in itself isn't a problem. BeemerBiker could set up 2 copies of BOINC, with the program file installations in the appropriate program folders for each of the two Windows installations, but the data folder on a shared drive/partition visible to both OSs - perhaps a FAT32 disk, so that no pesky security descriptors could prevent access. That won't work for Bryan, because he can't use the same applications in Linux and Windows: client_state.xml would have to be rebuilt every time the OS changed. The 'shared (multi-visible) data folder' idea can only work for compatible platforms: I wouldn't even use it for win-32 + win-64 unless I knew that every project in the shared installation sent 32-bit apps to 64-bit clients. |
|
4660)
Message boards :
Questions and problems :
6.6.2 scheduler: still better off with 1 less CPU
(Message 22573)
Posted 21 Jan 2009 by Richard Haselgrove Post: The WU's were the 6.06 as shown here. Since last night seti has tried to send me 6.08 but the download is failing as shown here However, I am surprised that they even showed up as a failure as I have had "no new tasks" specified since late yesterday. I assume the download failed because of the checksum and not because the "no new tasks" Yes, SETI seem to have borked the v6.08 installation - Murphy supplied a faulty checksum. That's almost certainly the reason for recent SETI work errors. Once SETI has released a new app, all new work will be tagged for computation with that app. So you'll be getting allocated the same tasks as before, but they will have that '6.08' designation beside them, and they will all fail - until you reach that daily quota. If you're sticking with BOINC v6.2.2, I strongly suggest you disable network activity - perhaps a more reliable way to avoid these supernumerary work requests under current circumstances. |
|
4661)
Message boards :
BOINC client :
Result reporting not respecting 60 second delay
(Message 22569)
Posted 21 Jan 2009 by Richard Haselgrove Post: That's why I linked to the documentation (Wiki), where the 60-second delay is mentioned. I find, in general, that it's better if software documentation matches the software itself ;-) Looking at it (now borrowing a client's machine, waiting while their new PC downloads the initial 130MB security updates for a new Vista deployment :-( ), the issue overlaps with trac [trac]#728[/trac], where report_results_immediately is explicitly requested to adhere to the proposed delay. Unfortunately, both in that ticket and in the related thread 3092, people - including David Anderson - misunderstood the question, thinking that people were talking about a scheduler contact to report the current task, before the current task upload had completed. They weren't. The problem in that ticket/thread concerned a scheduler contact to fetch work, happening in parallel with an upload for a task just finished. Because the scheduler contact is initated by the client, as a new request, project-mandated backoffs don't (yet) apply, contrary to what some people were trying to say. But now that David has written it off, it might be easier to get this RRI delay accepted as a new enhancement. Or would that finally destroy any chance of getting the more general delay accepted? Suggestions? |
|
4662)
Message boards :
BOINC client :
Result reporting not respecting 60 second delay
(Message 22564)
Posted 21 Jan 2009 by Richard Haselgrove Post: That's why I linked to the documentation (Wiki), where the 60-second delay is mentioned. I find, in general, that it's better if software documentation matches the software itself ;-) In this case, I think the documentation is the better option. You're right: Crunch3r put RRI into his v6.1.0 client (which is actually a v5.x.x client - can't be installed over a v6 BOINC. But I digress). There were actually, I think, up to 5 different Crunch3r releases all under the v6.1.0 banner. The early ones did indeed report results immediately, and this caused problems with validate errors (the report arrived before the servers had digested the upload). Crunch3r was prevailed upon to build the 60-second delay into later iterations of the v6.1.0 line, and things have gone much more smoothly since then. I think it would be helpful if the stock client could follow Crunch3r's example, and its own documentation, in this case. |
|
4663)
Message boards :
BOINC client :
Result reporting not respecting 60 second delay
(Message 22562)
Posted 21 Jan 2009 by Richard Haselgrove Post: I might make a trac ticket of this later, when the BOINC server isn't so perishingly slow. OK, when I get back from my on-site callout, then. Got to go now. |
|
4664)
Message boards :
BOINC client :
Result reporting not respecting 60 second delay
(Message 22559)
Posted 21 Jan 2009 by Richard Haselgrove Post: I'm currently (and briefly - for CUDA testing only) using the deprecated "<report_results_immediately>" option. According to the documentation, reporting should happen "with an inbuild 60 second delay from completion of result upload". This isn't happening: 21-Jan-2009 10:15:57 [SETI@home Beta Test] Computation for task 03no08aa.22996.277095.7.11.62_1 finished 21-Jan-2009 10:16:00 [SETI@home Beta Test] Started upload of 03no08aa.22996.277095.7.11.62_1_0 21-Jan-2009 10:16:05 [SETI@home Beta Test] Finished upload of 03no08aa.22996.277095.7.11.62_1_0 21-Jan-2009 10:16:09 [SETI@home Beta Test] Sending scheduler request: To report completed tasks. Requesting 0 seconds of work, reporting 1 completed tasks 21-Jan-2009 10:16:14 [SETI@home Beta Test] Scheduler request completed: got 0 new tasks 21-Jan-2009 10:14:43 [SETI@home] Computation for task ap_01dc08ab_B2_P1_00057_20090107_28452.wu_3 finished 21-Jan-2009 10:14:46 [SETI@home] Started upload of ap_01dc08ab_B2_P1_00057_20090107_28452.wu_3_0 21-Jan-2009 10:14:51 [SETI@home] Finished upload of ap_01dc08ab_B2_P1_00057_20090107_28452.wu_3_0 21-Jan-2009 10:14:53 [SETI@home] Sending scheduler request: To report completed tasks. Requesting 0 seconds of work, reporting 1 completed tasks 21-Jan-2009 10:14:58 [SETI@home] Scheduler request completed: got 0 new tasks 21-Jan-2009 08:45:02 [SETI@home Beta Test] Finished upload of 03no08aa.22996.276686.7.11.236_1_0 21-Jan-2009 08:45:06 [SETI@home Beta Test] Sending scheduler request: To report completed tasks. Requesting 0 seconds of work, reporting 1 completed tasks 21-Jan-2009 07:13:47 [SETI@home Beta Test] Finished upload of 03no08aa.22996.276686.7.11.198_0_0 21-Jan-2009 07:13:49 [SETI@home Beta Test] Sending scheduler request: To report completed tasks. Requesting 0 seconds of work, reporting 1 completed tasks and so on. More like 2 to 4 seconds delay, instead of the advertised 60. BOINC v6.4.5 I might make a trac ticket of this later, when the BOINC server isn't so perishingly slow. |
|
4665)
Message boards :
Questions and problems :
6.6.2 scheduler: still better off with 1 less CPU
(Message 22558)
Posted 21 Jan 2009 by Richard Haselgrove Post: Which project, which version of the project's science application, and which data file (workunit)? From the screenshot in your last thread, you were working on SETI (main project) at that time. Again, at that time, the SETI project was issuing v6.06 of their CUDA-enabled application. In Beta testing, that version exhibited the same variation you've described between 3+1 and 4+1 operation. Overnight, SETI have released v6.08 on the main project. I think you may find that it handles 4+1 rather better: but I don't know anyone who's performed that particular combination experiment (v6.2.2 + v6.08, 3+1 vs. 4+1) to confirm. When you get v6.08, it would be good to find out - but be careful: according to reports at SETI, the project messed up the checksum for the downloads of the new files, so new tasks are erroring out. With the new CUDA quotas, that could be a very big waste of bandwidth. Best to set NNT for the time being ...... oh. The other variable in the equation (as if that little lot wasn't enough) is task variability. When you get v6.08, you'll be able - for the first time - to run SETI's "VLAR" tasks. But they run very slowly: more like your 4cpu figure. So it wouldn't be helpful to do the 3+1/4+1 test while a VLAR is going thropugh the works: wait for a more 'normal' task to test. |
|
4666)
Message boards :
Questions and problems :
6.6.2: ignoreing "Wont't get new tasks"
(Message 22557)
Posted 21 Jan 2009 by Richard Haselgrove Post: You'll find Claggy's message here on the forums as well and I answered him separately. :-) Well, if the forum had been working the last few days, I might have seen the posts as they came in ;-) But I still think Claggy's Astropulse allocation (in the other thread) rather scuppers your 'CUDA disobeys NNT' theorem (in this thread). No AP for CUDA yet. If David is looking for a needle in a haystack again, best to have all the evidence in one place. |
|
4667)
Message boards :
Questions and problems :
6.6.2: ignoreing "Wont't get new tasks"
(Message 22548)
Posted 20 Jan 2009 by Richard Haselgrove Post: This issue has been independently reported by Claggy at SETI Main: message 855551. Note that Claggy got Astropulse (i.e. non-CUDA) tasks, so it couldn't have been the CUDA request which failed to respect NNT. |
|
4668)
Message boards :
BOINC Manager :
Suggestion: add wall time to Tasks grid
(Message 22457)
Posted 15 Jan 2009 by Richard Haselgrove Post: ... a time stamp in the cuda results, elapsed time (wall clock like you mentioned) ... That would be a project-specific request. Eric Korpela (project developer for SETI/SETI Beta) has already accepted the idea in principle (see here), but he's rather tied up in getting the application working properly at the moment - he would probably say that the cosmetic niceties can be added later. We should remind him nearer the time. |
|
4669)
Message boards :
Questions and problems :
understanding overworked projects and LTD with CUDA
(Message 22450)
Posted 15 Jan 2009 by Richard Haselgrove Post: First, let's get rid of a couple of red herrings. 1) BoincView time 'To completion' of 13h45m I'm pretty sure this is a minor display problem in BoincView, given that it's a very old program and hasn't been updated to match all the latest developments in BOINC (but it still does a great job, and is the most-used program on my system). I *think* the display problem - since, as you have noticed, it only affects tasks 'Ready to start', and doesn't affect tasks 'Running', is an RPC bug in the latest CC - see my 'Time to completion' in BoincView for CUDA thread in API. 2) Messing with the task switch time slice. Will be a minor workround at best. Don't waste time on it. No, the real problem is with the management of debt on mutli-resource BOINC platforms. The BOINC developers forgot to think about this when rushing out the new version to catch the holiday graphics-card sales opportunity, and are now paying the price with some hurried reconsideration. I have exactly the same problem with a quad-core+CUDA Mine is currently running three projects, all with exactly equal resource share. So I have two cores running Astropulse, two cores running Einstein, and one core running SETI Beta CUDA. So in terms of seconds (the only currency debt is measured in round here), Astropulse and Einstein are each getting twice the share of CUDA. They process 2 seconds (one on each core), for every one second that CUDA gets. It doesn't matter that my CUDA card is doing vastly more FLOPS or cobblestones than my CPU, the only thing BOINC understands is seconds, and all seconds are created equal. For the time being. Normally, BOINC would solve this problem by suspending a project with negative STD on one core, and starting a second task on the project with positive STD on the spare core thus freed up. But here it can't - neither of us has got a spare CUDA core to give extra SETI Beta tasks to. And if BOINC is processing SETI Beta in CUDA mode, it won't assign a CPU core to it, even though the SETI Beta application is perfectly capable of dropping back to CPU-mode operation (well, in a rudimentary sort of way - they stripped out all the CPU optimisation code). Your display is trivial - I'm showing SETI Beta with +ve STD of 50,467: Astropulse -ve STD 21,826: and Einstein -ve STD 28,640. Astropulse is also showing a LTD of -3,958,122 - but it still fetches work if I massage it hard enough. For the time being, the only way to allow BOINC to manage debts automatically would be to set the sum of your CUDA-enabled projects (SETI and GPUgrid, if you wanted to add that back into the mix) to be exactly 20% of your overall all-project resource share. If you go any lower than 20%, except for a short-term re-balancing period, you'll go into -ve STD for CUDA. I don't know whether BOINC would be stupid enough to idle your GPU under those circumstances, and I probably won't bother to find out. For the moment, resource share is the key to your question, and remember that it's only measured in seconds - until at least BOINC v6.8, I believe. |
|
4670)
Message boards :
API :
'Time to completion' in BoincView for CUDA
(Message 22409)
Posted 13 Jan 2009 by Richard Haselgrove Post: Can I use the API board for an RPC query? If not, please move to a more appropriate place. I've just started using the CUDA features in BOINC v6.4.5 for SETI Beta testing. When I was using a third-party build with an app_info.xml file, BoincView converged under DCF control to showing reasonably accurate wall-time estimates - around 25 minutes on my rig. The BoincView display matched the local BOINCManager display. I have now switched to using the latest v6.07 Beta, under automatic project download control (no app_info.xml). BoincView is still showing a roughly accurate estimate of the remaining 'to completion' time for tasks that are running, but for cached tasks that are ready to start, BoincView is showing an estimate of 8 hours 38 minutes (total work buffer for the poject over 35 days), instead of the 22 minutes 34 seconds currently shown by BOINC Manager. Can anyone explain this roughly 23::1 discrepancy? TDCF for the project is currently 0.964897 and stable, according to both the project website and BoincView. NB I've written this note with reference to BoincView, but that's just because it happens to be my RPC viewer of choice: in all other respects, BoincView is displaying credible information. It is only the 'Ready to start' estimate for work without an app_info.xml which is ludicrous, which is why I suspect an RPC response bug. It would be helpful if somebody could cross-check this observation with an alternative RPC viewer, or suggest one (for Windows) I can test myself. |
|
4671)
Message boards :
Questions and problems :
Questions about process priorities
(Message 22336)
Posted 7 Jan 2009 by Richard Haselgrove Post: Now I must confess that I didn't see Richard's message about BOINC being responsible for setting the priority of the science apps. Maybe he is right, maybe Ageless and I are wrong. All I know for sure is that all the projects I am attached to run at lowest priority. I've had a quick look at the source code. Note the comment at line 366 of /client/app_start.cpp: // if this job uses a GPU and not much CPU, run it at normal priority and the parameter prio_mask in CREATE_NEW_PROCESS_GROUP. I think it's BOINC wot does it. edit - also see the Trac Revision Log for that same file (requires trac login). |
|
4672)
Message boards :
Projects :
Einstein project
(Message 22330)
Posted 7 Jan 2009 by Richard Haselgrove Post: http://einstein.phys.uwm.edu/ is working for me, and has done so consistently on a timescale of years and a timescale of days. The project did have a server crash just before Christmas (shown as 23 December on their front page), and service was a bit spotty for a couple of days after that, but everything has been normal for at least the last ten days. |
|
4673)
Message boards :
Questions and problems :
Questions about process priorities
(Message 22324)
Posted 7 Jan 2009 by Richard Haselgrove Post: Judging by yesterday's discussion "Power saving vs Boinc vs user control & confusion" on the BOINC Dev mailing list (sparked by Martin's suggestion, first posted at Einstein), BOINC *does* set the process priority levels. Note one possible source of confusion: the BOINC log message "Running, high priority" does *NOT* refer to thread or process priority as discussed here. It merely means what used to be described as "Earliest Deadline First" or EDF mode: this affects the sequence tasks are processed in, but does not affect the priority ('niceness') at the CPU level. There have been observations at SETI that if the CPU stub which feeds a CUDA application runs at lowest priority, then the CUDA part of the application is starved of data and runs extremely slowly. Using a tweak in the science app to over-ride BOINC's priority setting seems to help out. All of which suggests that more attention needs to be given to process priority settings as BOINC v6.4 / v6.6 / v6.8 is developed for CUDA / OpenCL support. |
|
4674)
Message boards :
Questions and problems :
Boinc Manager won't connect to localhost
(Message 22316)
Posted 6 Jan 2009 by Richard Haselgrove Post: You call H2 rogue? ;-) Well, I don't run it, so I can't call it from personal experience. But the posts in the thread I linked, and the matching Hydrogen@home killing BOINC Manager? that you cross-linked (in both directions), suggest at least the possibility that a project or projects might be implicated. Hence my question to GLRumble. Seemed as least as good a bet (well, a better bet, as Gundolf pointed out) as the firewall suggestion. Edit - maybe this will help. |
|
4675)
Message boards :
Questions and problems :
Boinc Manager won't connect to localhost
(Message 22314)
Posted 6 Jan 2009 by Richard Haselgrove Post: wasn't it a rogue project which caused BOINC to crash on startup? Actually, I was thinking of Hydrogen@Home. |
|
4676)
Message boards :
Questions and problems :
Boinc Manager won't connect to localhost
(Message 22312)
Posted 6 Jan 2009 by Richard Haselgrove Post: Last time we had one of these, wasn't it a rogue project which caused BOINC to crash on startup? GLRumble, what project(s) is that host attached to? |
|
4677)
Message boards :
BOINC Manager :
Any links to optimizing Boinc
(Message 22071)
Posted 25 Dec 2008 by Richard Haselgrove Post: Great guys thanks alot for the new information, I am really looking forward to contributing. I would love the resources just to monitor a whole room of computers crunching numbers. I find BoincView (http://boincview.amanheis.de/) a good monitoring tool. |
|
4678)
Message boards :
BOINC Manager :
Any links to optimizing Boinc
(Message 22067)
Posted 24 Dec 2008 by Richard Haselgrove Post: Would like to make the Boinc Client obviously complete WU's as fast as possible but I would like links to a decent guide because I can't find too many specific to Vista 64. Optimising BOINC won't help you in the slightest - BOINC doesn't do any actual computation, it just manages and marshalls the project science applications which do the real work. SETI has a number of optimised applications, but none yet for CUDA - the BOINC release is just too new, and even the project's own application is still full of bugs. Read the SETI message boards periodically - any CUDA-optimised application will surface there first. Einstein doesn't allow optimised applications, but they do periodically release Beta and 'Power Users' applications which incorporate the latest, and fastest, techniques. Again, you need to read the project message boards to find out about these. Edit - snap! |
|
4679)
Message boards :
Questions and problems :
v6.5.0 only works as admin
(Message 22003)
Posted 22 Dec 2008 by Richard Haselgrove Post: You can reach it by pressing f8 while booting. That would be "Safe mode". I'll try it when I get a chance, but I think it still doesn't give you access to the sort of user-level security options that XP Pro has. |
|
4680)
Message boards :
BOINC client :
CUDA Project makes 1 CPU idle most of the time
(Message 22002)
Posted 22 Dec 2008 by Richard Haselgrove Post: aaaaaaaaaaagh To which I can only reply: AAAAAAAAAAAAAAAAAAGH Stop releasing half-baked BOINC clients with new features that haven't been fully thought through, declaring them stable with next-to-no testing, introducing new bugs (DCF), failing to fix old bugs .... .... and which require the use of NCPUs (plus resetting the thread priorities at application level) to do that very thing which BOINC was designed to do right at the very beginning, namely to put idle CPU cycles to productive use. Not your fault personally, Nicolas, of course: but if you dare look at the SETI boards to see what a confusing mess has been generated by this BOINC release, plus the equallly rushed, untested, un-thought-through, unsupported (there are too few SETI project staff to handle a release of this magnitude, especially so soon after their Astropulse release, which still hasn't had the post-release fine-tuning attention it deserves) SETI CUDA release, then you'd see why volunteers resort to experiments like this. |
|
4681)
Message boards :
BOINC client :
6.4.5 unnecessarily assigning high proiority CUDA
(Message 21893)
Posted 17 Dec 2008 by Richard Haselgrove Post: Either way, it's going to be a major problem for projects like SETI where CUDA and CPU programs run on the same project and share a DCF. That's a bug, in my book. I usually like to wait for confirmation from at least two matching reports before bloating the trac database. Not having a CUDA card, I don't feel like upgrading any of my machines to 6.4.5 yet, 'recommended' 'stable' or not. Besides, my existing machines are already joined to projects, including CPDN: I don't want to risk those tasks having their estimates multiplied by 100! These posts are by way of a public (more public than trac) "heads up": perhaps other people installing v6.4.5 and joining new projects such as SETI Beta could watch the DCF values as they go through the process, and report. |
|
4682)
Message boards :
BOINC client :
6.4.5 unnecessarily assigning high proiority CUDA
(Message 21888)
Posted 17 Dec 2008 by Richard Haselgrove Post: Default value expressed as a percentage rather than a fraction, anyone? The host is a new join on a new project. Look at the Task list: no tasks have been reported yet (tasks are never purged at SETI Beta). I don't know whether BeemerBiker has actually completed any tasks yet - an upload server failure at the project has prevented any tasks reporting since approximately 17:30 UTC yesterday. But SETI tasks are usually over-estimated: I'm running the same tasks on the CPU-only version of the program, and seeing reasonable DCF values below 0.5000 So either it's the default value (if no CUDA tasks have exited), or an extreme mis-interpretation of the running time (if they have exited but not been reported). Either way, it's going to be a major problem for projects like SETI where CUDA and CPU programs run on the same project and share a DCF. That's a bug, in my book. Edit - looking back at BeemerBiker's opening posts in this thread, the immediate EDF mode reported reinforces my hunch that it's an inital DCF problem. |
|
4683)
Message boards :
BOINC client :
6.4.5 unnecessarily assigning high proiority CUDA
(Message 21884)
Posted 17 Dec 2008 by Richard Haselgrove Post: These images, supplied by BeemerBiker at SETI Beta, suggest that there is a serious DCF bug in BOINC v6.4.5 Default value expressed as a percentage rather than a fraction, anyone?   The Astropulse task (top image) will have a running time of approx 40-45 hours on his CPU. |
|
4684)
Message boards :
Web interfaces :
Possible tag issue in forums
(Message 21876)
Posted 17 Dec 2008 by Richard Haselgrove Post: XML tags (and any other use of < - >) are now getting through the PM system correctly here, at SETI Main, and at SETI Beta. Thanks everyone for a quick and efficient response to the initial bug report. |
|
4685)
Message boards :
BOINC client :
6.4.5 unnecessarily assigning high proiority CUDA
(Message 21869)
Posted 16 Dec 2008 by Richard Haselgrove Post: The system is at home and I am accessing via remote desktop from work. Hasn't it been reported that CUDA processing is incompatible with RDP? I've been seeing people advocating a switch to VNC if you need to fiddle with GPU drivers during working hours ;-) |
|
4686)
Message boards :
BOINC Manager :
BOINC Bench for GPU
(Message 21839)
Posted 15 Dec 2008 by Richard Haselgrove Post: Regarding credit part of Richard's post. That's an excellent start - I'll suggest it over at SETI Beta (the more information, the better). But all they've done is add an extra information line into <stderr_txt> - that's not the same as formally recording it in the BOINC client metrics for re-use in work fetch, DCF, and credit functions. |
|
4687)
Message boards :
Web interfaces :
Possible tag issue in forums
(Message 21838)
Posted 15 Dec 2008 by Richard Haselgrove Post: Do you think if I can get Matt or Eric to update the code for the PM-bugfix tomorrow, I'll become a 'special user' too, along with all the other volunteer testers? (who are also automatically tagged at main, if their accounts can be linked up) And frequently blue in the face, if not in the background. |
|
4688)
Message boards :
Web interfaces :
Possible tag issue in forums
(Message 21837)
Posted 15 Dec 2008 by Richard Haselgrove Post: Email or PM Eric Korpela or Matt Lebowski. That they change it during the maintenance outage tomorrow. Done. |
|
4689)
Message boards :
Web interfaces :
Possible tag issue in forums
(Message 21834)
Posted 15 Dec 2008 by Richard Haselgrove Post: And while we're all here: this fancy new coloured background that Jord has had for a few weeks. I suppose it's triggered by one, other or both of his status tags. Useful: I can see it picking out Moderator or Project Scientist posts in a discussion thread. It's really not worth pestering him - there are other things I would much prefer he spent his time on. Hadn't seen the 'Highlight special users' option before (don't often visit those pages). Strange thing is, the preference is present and checked in my SETI Main account, but there is no highlighting on the SETI Main forums: must be in some sort of half-updated state. Do you think if I can get Matt or Eric to update the code for the PM-bugfix tomorrow, I'll become a 'special user' too, along with all the other volunteer testers? (who are also automatically tagged at main, if their accounts can be linked up) |
|
4690)
Message boards :
Web interfaces :
Possible tag issue in forums
(Message 21831)
Posted 15 Dec 2008 by Richard Haselgrove Post: And while we're all here: this fancy new coloured background that Jord has had for a few weeks. I suppose it's triggered by one, other or both of his status tags. Useful: I can see it picking out Moderator or Project Scientist posts in a discussion thread. But it doesn't work very well at SETI Beta, where everybody, without exception, is given a <Volunteer tester> tag. Is there any way the background could be made a bit more selective? |
|
4691)
Message boards :
Web interfaces :
Possible tag issue in forums
(Message 21830)
Posted 15 Dec 2008 by Richard Haselgrove Post: And thanks from me too. What's the best way of alerting/requesting SETI to apply the new code? That's where we really need it. (And probably SETI Beta too, though they usually do them in tandem). |
|
4692)
Message boards :
Web interfaces :
Possible tag issue in forums
(Message 21820)
Posted 15 Dec 2008 by Richard Haselgrove Post: No, here it works. It doesn't work in the Seti forums. No, it's OK in every forum I've seen - if anything, we have too many of them at SETI (heavy use/quoting of app_info.xml). It's the PM system which is broken, here and at SETI. |
|
4693)
Message boards :
BOINC Manager :
BOINC Bench for GPU
(Message 21767)
Posted 13 Dec 2008 by Richard Haselgrove Post: Better add a lot of questions on the subject of how BOINC deals with CUDA that I can put toward the developers at one time. I'm not going to ask them one question at a time. Good idea, but I don't think it's the right way to go. I'm not a programmer (at least, not at this level): all I can do is monitor what's happening in different places, and sometimes act as a messenger if a topic crops up in one place and has relevence somewhere else. Be aware that SETI@Home has started experimenting with a CUDA GPU application - very much a Beta app, and only available at their Beta project. This has generated a new wave of interest in the CUDA platform. SETI is, of course, open-source, and there has been a tradition over the years of various independent developers producing optimised versions of the SETI apps. The current torch-bearers in that tradition are the 'Lunatics', based at http://lunatics.kwsn.net/index.php. They had already started on their own (unsupported by NVidia) CUDA transfer for both SETI and Astropulse (the current official Beta is for SETI only). They have just been sent an official copy of the SETI CUDA Beta source code (the live repository is apparently currently unavailable for some reason), and I suspect they will be poring over it and coming up with all sorts of questions and suggestions at a phenomenal rate - probably all through the holiday season. What would be an efficient way to synergise these two parallel development efforts? The official SETI project development team are showing signs of severe overstrech - their Astropulse re-release has not gone well, and simple niggles have gone unaddressed for months. So I don't think we can ask or expect them to act as co-ordinators for BOINC issues. Me watching from the sidelines (I have access to the Lunatics pre-release test forums where much of the discussion takes place), passing questions to you, passing questions to David/Rom, and vice-versa, isn't going to work either. I don't (yet) even have a CUDA-capable graphics card, so I can't even participate in the testing and reproduce issues in a way I can report here. I think we need to find a new way of joining Berkeley (both BOINC and SETI) up with the rest of the world, but I don't know what it is yet. With regard to specific questions: the Lunatics team have come up against the same 'incorrect priority' issue in BOINC v6.4.5 as BeemerBiker: their workround is to use Process Lasso to manipulate the priorities of the three individual threads within the SETI Beta application. Some people seem to be increasing priorities, others decreasing them: anything except the default works, it seems! Another issue which is going to come up again and again and again is the thorny old subject of credit. I think BOINC has a duty to the projects to take a firm grip on this, and ensure that the credit-granting behaviour of CUDA-BOINC, is clear and transparent to users before a new round of credit wars breaks out. Here are two genuine examples from SETI of credit confusion which have been posted in the 60 hours or so since the Beta release: both wrong, contradictory, but will take up a lot of project time if BOINC doesn't get the central core right. I'm expressing both these points in my own words - the original posts were a bit difficult to follow, but I think I've caught the spirit of the original authors. "This new application can do the work in 60 seconds - it used to take 3,000 seconds. My RAC is going to be astronomical. Then SETI is going to have to reduce the credit per WU for all participants, to maintain cross-project parity." FALSE. The new SETI app can't do the work in 60 seconds. BOINC only reports the CPU time - the GPU and elapsed times (both unreported) are much longer. His RAC won't rise by anything like the proportion he's expecting. "This new application can do the work in 60 seconds - I'll get far less credit than everybody else, because credit is based on time, right?" FALSE - credit at SETI is based on a crude approximation of the number of FLOPs the calculation takes, and it shouldn't make any difference whether those FLOPs are performed on a CPU or a GPU. But what would be the credit awarded if a project still using benchmark*time credit released a CUDA app? Especially - to return to the original topic - since BOINC supplies neither a GPU benchmark nor the GPU time. |
|
4694)
Message boards :
BOINC Manager :
BOINC Bench for GPU
(Message 21763)
Posted 13 Dec 2008 by Richard Haselgrove Post: I just had a chat with one of the developers. He told me that GPUs report back what speed they run at when a program asks for it. BOINC doesn't store this information anywhere as BOINC will only query the card(s) once when BOINC starts up. The science application then handles the rest. That's a cop-out by the developers. Exectly the same thing could be said of the CPU speeds: CPUs will reply (or allow it to be calculated by experiment) if BOINC asks. BOINC has chosen to ask for that information, and to store/use the data in work fetch calculations. How does BOINC calculate the initial work fetch for a GPU/CUDA project application? |
|
4695)
Message boards :
Questions and problems :
Editing Client_State and Vista
(Message 21742)
Posted 11 Dec 2008 by Richard Haselgrove Post: If you took the default BOINC v6.2.19 installation settings, it will be running "as a service": the actual BOINC client won't shut down when you close BOINC Manager (you can check with Task Manager). Try using 'Shut down connected client' (Advanced view, Advanced menu) before closing BOINC Manager - then see if your edit takes. If I need to make a change like that (running BOINC v5.10.13 installed as a service under Vista Business), I shut down BOINC using the Services Control Panel and have no problem editing client_state.xml |
|
4696)
Message boards :
Questions and problems :
Changing BOINC rpc port
(Message 21717)
Posted 10 Dec 2008 by Richard Haselgrove Post: Does your router have a port redirection table? If so, you could use it to match (external) single IP, multiple ports --> (internal) multiple IPs, same port. Otherwise, you'd neet a startup batch file which includes boinccmd --gui_rpc_port N (from the BOINC Client Configuration Wiki) |
|
4697)
Message boards :
Questions and problems :
dedicating one core to a specific project of a multicore pc
(Message 21630)
Posted 4 Dec 2008 by Richard Haselgrove Post: Set your CPDN resource share to just over 50% - say a numeric value of 501. Allow CPDN to download one task only, then set No New Tasks. Keep all you other projects below 50% - say a total numeric value of 500. Then the CPDN short term debt will steadily, but slowly, increase: it will always be positive, and CPDN will always be the first task to run: with one task only on board, it will occupy one core. The other projects can (and will) fight it out amongst themselves for the remaining core. You'll need to micro-manage a bit - allow CPDN to download one new task every few months - but that shouldn't be too hard ;-) |
|
4698)
Message boards :
Questions and problems :
Work units lost when installing 6.2.19
(Message 21622)
Posted 3 Dec 2008 by Richard Haselgrove Post: I've taken all the clutter out to show that the old tasks that were warned about weren't started. We must be talking about SETI host 4630841 - that's the only T5750 on Bill's account, even looking at machines which haven't been active recently. There are no tasks shown as active/in progress for that host - he should abort anything and everything which the new install has found in that old folder. |
|
4699)
Message boards :
Questions and problems :
Work units lost when installing 6.2.19
(Message 21620)
Posted 3 Dec 2008 by Richard Haselgrove Post: Here are the first messagtes I got with the new Manager: There's something very odd about this statement. The new version of BOINC has found two tasks on your machine, warned you about missing deadlines, and started crunching them: all without contacting the SETI scheduler or downloading new work. Ergo, the workunits must have been on your computer all along, just not in the previously-active BOINC data folder and client_state.xml That suggests that you did something curious to the machine about six weeks ago (assuming these are typical SETI tasks with a ~3 week deadline still to run when you did whatever you did, now showing ~3 weeks overdue). What the BOINC v6.2.19 installation has done, faced with a choice of two data folders, is possibly to have picked the wrong one from your point of view: but that doesn't explain why it was faced with the choice in the first place. BTW, you should certainly abort the two tasks which have started: they are so old that they have been purged from the SETI database, and they will certainly contribute nothing to the search - neither science nor credit. |
|
4700)
Message boards :
BOINC Manager :
Minor sorting problem
(Message 21494)
Posted 24 Nov 2008 by Richard Haselgrove Post: ... get in the wild as soon as possible to see what problems it might create on a larger assortment of systems ... Not to mention a larger assortment of projects, many of whom haven't yet recovered from the v6.2 graphics compatibility change. You've got the BOINC testing paradigm described to a T. |
|
4701)
Message boards :
BOINC Manager :
Minor sorting problem
(Message 21488)
Posted 24 Nov 2008 by Richard Haselgrove Post: Just why is v6.4.x release more important than bugfixes? Trac [trac]#166[/trac] is over 2 years old, and patch code was added to it in April 2007. Now we're told that it's going to wait until v6.6.x. Timeline? Will bugfixes (in general, not just this particular one) have to take a back seat once again when the next bright idea comes along? I personally have reports [trac]#366[/trac], [trac]#525[/trac], and [trac]#588[/trac] still open. I regard [trac]#136[/trac] as a still-present bug, even though it was closed (on the basis of a false understanding, the first time round, although the second closure matches reality more closely). |
|
4702)
Message boards :
BOINC Manager :
Minor sorting problem
(Message 21397)
Posted 19 Nov 2008 by Richard Haselgrove Post: It's been reported before - message 20137 - but Jord was waiting for confirmation before passing it on to the developers. Thank you for supplying the confirmation, and also confirming that the problem still exists on the latest v6.2.19 recommended release. |
|
4703)
Message boards :
Questions and problems :
boinc release notes
(Message 21262)
Posted 15 Nov 2008 by Richard Haselgrove Post: How about looking at this thread? That's exactly the reference he's edited into the Wiki! |
|
4704)
Message boards :
Questions and problems :
Screen saver not working
(Message 21067)
Posted 1 Nov 2008 by Richard Haselgrove Post: Your guess is correct Richard, i have an opp app. It sounds as if you're in an application limbo, neither efficient nor pretty. By upgrading to BOINC v6.2.18, you have effectively killed off the graphics capability of your v5-compatible application. But by not upgrading your science application, you are missing out on the substantial performance increase obtained in the later optimised applications, released over six months ago (30 April 2008). I'll leave you to decide which version(s) you want to live with. |
|
4705)
Message boards :
Questions and problems :
Screen saver not working
(Message 21065)
Posted 1 Nov 2008 by Richard Haselgrove Post: My BOINC version; 6.2.18 SETI v5.28 is not a current SETI application. It won't, wherever it came from, have a BOINC v6 compatible graphics sub-system. My guess is that you have installed one of the third-party optimised SETI applications. Most optimised applications don't contain any graphics at all (and that includes the screensaver) - the eye-candy rather defeats the purpose of the optimisation, which is to do the science as quickly as possible. There are some optimised applications with graphics capability, for those who are prepared to sacrifice an element of the performance, but again they have not been re-engineered to fit into the new-ish BOINC v6 graphics model: the demand just isn't there, and the volunteer programmers have better things to do with their time. |
|
4706)
Message boards :
Questions and problems :
boinc not responding, OSX
(Message 20956)
Posted 24 Oct 2008 by Richard Haselgrove Post: e) Which option is selected under 'Activity', advanced view - "Run always", or "Run based on preferences"? |
|
4707)
Message boards :
Questions and problems :
Windows Vista Compatibility - Blocked Programs
(Message 20812)
Posted 15 Oct 2008 by Richard Haselgrove Post: It depends which bit of BOINC you want to start automatically when Windows starts. If you want the BOINC daemon (the bit that actually directs the science project applications to start crunching) to start automatically, then you need to install in Protected Application Execution mode - see Jord's adjacent (sticky) thread. If you just want the BOINC Manager (the pretty bit that lets you see and control what's going on) to start automatically but silently - well, you probably can't. But you don't need to in service/protected mode. Just delete any left-over shortcuts from the 'Startup' folder after you've done the service install. You may be able to make a silent-running shortcut later in your personal startup folder, but I haven't bothered - I just keep a handy shortcut I can click manually if I feel the urge to look in on what's happening. |
|
4708)
Message boards :
BOINC client :
app_info.xml parsing problem
(Message 20773)
Posted 12 Oct 2008 by Richard Haselgrove Post: Jord, do you want to raise the ticket, or shall I? Will do. May be later tonite, or even tomorrow - I want to watch Stephen Fry's road-trip round America this evening. |
|
4709)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 20772)
Posted 12 Oct 2008 by Richard Haselgrove Post: Wow! |
|
4710)
Message boards :
BOINC client :
app_info.xml parsing problem
(Message 20768)
Posted 12 Oct 2008 by Richard Haselgrove Post: A user at SETI has helped us track down the source of these extraneous spaces (thanks, MarkJ). On a bog-standard default Windows installation, the XML file extension is associated with Internet Explorer. The temptation for anyone needing to 'tweak' an XML file, therefore, is to: 1) Double-click on the file 2) Select the text they require in the resulting IE window 3) Ctrl-C to copy 4) Open the editor of their choice, paste, and carry on from there. Whatever you think about Microsoft's definition of "standards", it's clear that Internet Explorer thinks that the standard for a self-closed tag like <executable /> and <main_program /> is to use a space - it inserts one automatically when you use this procedure. And, this being the real world, I think it's important that BOINC at least recognises, and correctly parses, Microsoft's version of the standard. Jord, do you want to raise the ticket, or shall I? |
|
4711)
Message boards :
BOINC client :
DNS caching in 6.2.18 and onwards
(Message 20727)
Posted 9 Oct 2008 by Richard Haselgrove Post: Found a second machine trying to contact the reverse IP address: 09/10/2008 23:59:41|SETI@home|[file_xfer] Started download of file 18au08ag.30754.18068.12.8.90 09/10/2008 23:59:41||[file_xfer_debug] PERS_FILE_XFER::start_xfer(): URL: http://boinc2.ssl.berkeley.edu/sah/download_fanout/2d1/18au08ag.30754.18068.12.8.90 09/10/2008 23:59:42||[http_debug] [ID#0] info: About to connect() to boinc2.ssl.berkeley.edu port 80 (#0) 09/10/2008 23:59:42||[http_debug] [ID#0] info: Trying [color=red]13.240.68.208[/color]... 10/10/2008 00:00:03||[http_debug] [ID#0] info: Timed out 10/10/2008 00:00:03||[http_debug] [ID#0] info: Failed connect to boinc2.ssl.berkeley.edu:80; No error 10/10/2008 00:00:03||[http_debug] [ID#0] info: Expire cleared 10/10/2008 00:00:03||[http_debug] [ID#0] info: Closing connection #0 10/10/2008 00:00:03||[http_debug] HTTP error: couldn't connect to server This one was Windows XP, previous one Windows Vista - but still BOINC v5.10.13 Could someone check this out with a v6.2.xx build, please, next time there's an extended download server outage? (I can't test against 208.68.240.13 again, because Matt has kicked it, and it's working now). Edit: And a Windows 2000 machine: 2008-10-10 00:18:45 [SETI@home] [file_xfer] Started download of file 19au08ac.10227.8252.6.8.64 2008-10-10 00:18:45 [---] [http_debug] [ID#0] info: About to connect() to boinc2.ssl.berkeley.edu port 80 (#0) 2008-10-10 00:18:45 [---] [http_debug] [ID#0] info: Trying [color=red]18.240.68.208[/color]... 2008-10-10 00:19:08 [---] [http_debug] [ID#0] info: Timed out 2008-10-10 00:19:08 [---] [http_debug] [ID#0] info: Failed connect to boinc2.ssl.berkeley.edu:80; No error 2008-10-10 00:19:08 [---] [http_debug] [ID#0] info: Expire cleared 2008-10-10 00:19:08 [---] [http_debug] [ID#0] info: Closing connection #0 2008-10-10 00:19:08 [---] [http_debug] HTTP error: couldn't connect to server Note that this was trying to contact the (reverse of the) server that didn't have problems. Maybe we don't need an outage to test, just some congestion? That should be easy enough to organise at SETI.... |
|
4712)
Message boards :
BOINC client :
DNS caching in 6.2.18 and onwards
(Message 20724)
Posted 9 Oct 2008 by Richard Haselgrove Post: Am I right in thinking that BOINC v6.2.18 still has the bug whereby once a file transfer (download in this case) has been attempted and failed, subsequent retries use the same cached IP address, instead of a fresh DNS lookup? SETI has one download server address (boinc2.ssl.berkeley.edu), but two servers (208.68.240.13 and 208.68.240.18). Round-robin DNS should give you a random but equally-probable chance of connecting to either server. Today, 208.68.240.13 has an HTTP service problem, so download requests have a 50% chance of hitting the bad server. But once the IP is cached, every retry stays bad: only restarting the BOINC service seems to get things started again. And on one of my retries, I got 09/10/2008 17:10:47|SETI@home|[file_xfer] Started download of file 18au08af.30717.24612.15.8.186 09/10/2008 17:10:48||[http_debug] [ID#4] info: About to connect() to boinc2.ssl.berkeley.edu port 80 (#0) 09/10/2008 17:10:48||[http_debug] [ID#4] info: Trying [color=red]13.240.68.208[/color]... 09/10/2008 17:11:09||[http_debug] [ID#4] info: Timed out 09/10/2008 17:11:09||[http_debug] [ID#4] info: Failed connect to boinc2.ssl.berkeley.edu:80; No error (note the reverse-order IP address) - but that was on BOINC v5.10.13, and I know you don't like to talk about old versions here..... |
|
4713)
Message boards :
Questions and problems :
6.2.19
(Message 20614)
Posted 1 Oct 2008 by Richard Haselgrove Post: Running Vista and when I install I get I don't think this is a problem with v6.2.19 per se (though I haven't tried it, so I might be wrong). More likely, it's referring to the installation package for the version of BOINC you're currently running: it needs access to that, because that's where the information about what needs to be uninstalled is kept. Try re-running the previous installer first, and choose 'repair': after that, the new installer should work. If that doesn't work, try using the Windows Installer CleanUp Utility |
|
4714)
Message boards :
Questions and problems :
BOINC required for a project to run?
(Message 20578)
Posted 29 Sep 2008 by Richard Haselgrove Post: Do I understand correctly that BOINC is not required once a project, in this case Einstein@home, is underway. No, that isn't correct. The client, or daemon, part of BOINC is absolutely needed, whether you have one project or many. Indeed, it must be running on your system(s) if you are accumulating credits. But as Dagorath has explained, the other part of BOINC, the Manager, is not needed at all once things are under way, except for occasional monitoring and, well, management. If you have a Windows machine available for a supervisor or teacher to use on the same network, you might like to explore an add-on tool like BoincView. This would enable you to check the state of play, and to a certain extent control, the BOINC installations on all the students' machines, without having to visit each one and log off/log on each time. |
|
4715)
Message boards :
Questions and problems :
No New Task
(Message 20569)
Posted 29 Sep 2008 by Richard Haselgrove Post: Another idea I had is having a way to asign a project or projects to a CPU Core. If you are using a duel core you could asign 4 projects to each core! You can guess how many Project's a PC could do... A duel core could run 8 projects! It would be like having 2 PC's! The more cores the more project's! That would be <ncpus> in the cc_config.xml file. |
|
4716)
Message boards :
Questions and problems :
Problems running boinc under Vista x64
(Message 20561)
Posted 28 Sep 2008 by Richard Haselgrove Post: I moved the data directory to "documents/BOINC/" when i first made an attempt to install boinc 6 as Vista doesnt like programs writing data in the programs directory. When you say "documents", where exactly do you mean? By default, that is one of the movable folders associated with each separate user logon: my documents is not the same place as your documents, The "documents" in my start menu is actually 'C:\Users\Richard Haselgrove\Documents'. If you're referring to a similar location, it might explain why BOINC Manager couldn't attach properly when someone else was logged on to the computer. Depending how Windows is set up, users can be prevented from seeing or modifying each other's documents, even on the same machine. Every time i try to re-install boinc 6 now it has the data directory in "documents". Yes, BOINC remembers user choices made during installation, and uses them as the default setting when you re-install. Saves a lot of typing (and remembering) when updating frequently, such as during testing. |
|
4717)
Message boards :
Questions and problems :
Screensaver
(Message 20489)
Posted 25 Sep 2008 by Richard Haselgrove Post: The following projects have version 6 graphics: Leiden Classical Likewise LHC. |
|
4718)
Message boards :
Questions and problems :
Problem with BOINC 6.2.18
(Message 20486)
Posted 25 Sep 2008 by Richard Haselgrove Post: Hi What sort of SETI task is running on your desktop? SETI has two at the moment - MB (short), and Astropulse (very long). There has been a reported bug in Astropulse "whereby the Astropulse screensaver would sometimes freeze": a new version designed to fix this is in the final stages of Beta testing. |
|
4719)
Message boards :
BOINC Manager :
How to stop Boinc from auto starting on system boot.?
(Message 20377)
Posted 20 Sep 2008 by Richard Haselgrove Post: A few additional thoughts: 1) I'm not sure why BOINC running at startup might get in the way of video, RAID, etc., but I guess it must be to do with high CPU usage. BOINC itself doesn't use much CPU - it's the science applications that it launches that do that. So to get control of the system (just as a temporary work-round, while you sort out a long-term solution), you could try suspending all projects before shutting down/restarting the computer. 2) It does sound as if BOINC is currently installed to run as a service. If you find your way to the Services applet (either by running 'services.msc', as previously suggested, or from 'Administrative Tools' in the Control Panel), there's a new start-up option in Vista: "Automatic (Delayed Start)". That might be a good compromise setting to test. 3) If you have indeed installed BOINC as a service (deliberately or otherwise), you don't need to have the Manager running at all times. But it does sound as if you've got a shortcut link in a 'startup' program group somewhere. I saw this myself on the one and only v6.2.18 upgrade I've done so far (a deliberate service install), but I haven't investigated in detail under what circumstances the startup link is created at installation. Older versions of BOINC had it as an option on the final installer page: I have a suspicion there's a bug in there somewhere. |
|
4720)
Message boards :
Questions and problems :
Astropulse Wu's !!!
(Message 20329)
Posted 17 Sep 2008 by Richard Haselgrove Post: Astropulse Wu = 433 hr's To completion with a 10/15/2008 Deadline 725 hours from the time of your first post opening this thread takes us to 14:59 on 16 October 2008. You don't give us the time of the deadline, but this estimate is definitely past it. You were in EDF ('panic' - high priority) mode, whether you knew it or not. Now that the 'to completion' estimate has reduced, there will come a transition point where BOINC can revert to round robin scheduling. That transition point will depend on your cache (specifically: connect interval) setting. YoYo should then start to run: it will have a lot of debt to pay off, so it will run continuously for a long time: and it is entirely possible and plausible that it will run back past the transition point so that Astropulse restarts in EDF again for another stint, even before the debt is paid off. Rinse and repeat. All good clean fun. |
|
4721)
Message boards :
BOINC Manager :
boinc_6.2.18_macOSX_universal on 10.5 ... DONT !!!!!!!!!!!!!!!
(Message 20318)
Posted 16 Sep 2008 by Richard Haselgrove Post: There were a couple of posts on Einstein this evening - before it went off the air - suggesting: (a) Problems were starting to appear with Mac OS X science applications (b) There was the hint of a possibility that a Mac OS patch update might be implicated. I don't have a Mac, so can't assist from personal experience, but it might be something to keep an eye on. |
|
4722)
Message boards :
Questions and problems :
Astropulse Wu's !!!
(Message 20286)
Posted 16 Sep 2008 by Richard Haselgrove Post: EDF? Or ask the author of the science application - so far as I know, Astropulse hasn't been released for PS3 by Berkeley. |
|
4723)
Message boards :
Questions and problems :
Is it possible to increase delay between result upload and reporting?
(Message 20282)
Posted 16 Sep 2008 by Richard Haselgrove Post: I thought so: 16/09/2008 02:44:22|SETI@home|Sending scheduler request: To report completed tasks 16/09/2008 02:44:22|SETI@home|Reporting 1 tasks 16/09/2008 02:44:27|SETI@home|Scheduler RPC succeeded [server version 603] 16/09/2008 02:44:27|SETI@home|Deferring communication for 11 sec 16/09/2008 02:44:27|SETI@home|Reason: requested by project [b][color=red]16/09/2008 03:03:11|climateprediction.net|Deferring communication for 1 min 0 sec[/color][/b] 16/09/2008 03:03:11|climateprediction.net|Reason: Unrecoverable error for result hadcm3istd_0f6b_1920_160_05937811_2 (The device does not recognize the command. (0x16) - exit code 22 (0x16)) 16/09/2008 03:03:11|climateprediction.net|Computation for task hadcm3istd_0f6b_1920_160_05937811_2 finished 16/09/2008 03:03:11|climateprediction.net|Output file hadcm3istd_0f6b_1920_160_05937811_2_15.zip for task hadcm3istd_0f6b_1920_160_05937811_2 absent 16/09/2008 03:03:11|climateprediction.net|Output file hadcm3istd_0f6b_1920_160_05937811_2_16.zip for task hadcm3istd_0f6b_1920_160_05937811_2 absent 16/09/2008 03:03:11|Einstein@Home|Starting h1_1083.20_S5R4__1006_S5R4a_0 16/09/2008 03:03:13|Einstein@Home|Starting task h1_1083.20_S5R4__1006_S5R4a_0 using einstein_S5R4 version 604 16/09/2008 03:03:13|SETI@home|Sending scheduler request: To fetch work and so on. Notice how the one-minute comms deferral is initiated by the client, long before we get into any project-requested backoffs. If we did that for success results, as well as errors, the server databases would run smaller and cleaner [task ends --> reduce DCF (sometimes) --> scheduler contact to fetch work (sometimes) --> could double with reporting finished work, if delay introduced]. EDIT And in case anyone doesn't believe my point about DCF, I've just watched this on another box: 16/09/2008 11:40:42|SETI@home|Computation for task 28au08ab.26001.11115.3.8.14_2 finished 16/09/2008 11:40:42|SETI@home|Starting 28au08ab.4533.2526.4.8.12_1 16/09/2008 11:40:42|SETI@home|Starting task 28au08ab.4533.2526.4.8.12_1 using setiathome_enhanced version 528 16/09/2008 11:40:44|SETI@home|[file_xfer] Started upload of file 28au08ab.26001.11115.3.8.14_2_0 16/09/2008 11:41:07|SETI@home|[file_xfer] Finished upload of file 28au08ab.26001.11115.3.8.14_2_0 16/09/2008 11:41:07|SETI@home|[file_xfer] Throughput 1917 bytes/sec 16/09/2008 11:42:22|SETI@home|Computation for task 20au08ae.31360.1299.9.8.45_1 finished 16/09/2008 11:42:22|SETI@home|Starting 28au08ab.4533.2526.4.8.24_0 16/09/2008 11:42:22|SETI@home|Starting task 28au08ab.4533.2526.4.8.24_0 using setiathome_enhanced version 528 16/09/2008 11:42:24|SETI@home|[file_xfer] Started upload of file 20au08ae.31360.1299.9.8.45_1_0 [b][color=red]16/09/2008 11:42:25|SETI@home|Sending scheduler request: To fetch work[/color][/b] 16/09/2008 11:42:25|SETI@home|Requesting 23285 seconds of new work, and reporting 1 completed tasks 16/09/2008 11:42:30|SETI@home|[file_xfer] Finished upload of file 20au08ae.31360.1299.9.8.45_1_0 16/09/2008 11:42:30|SETI@home|[file_xfer] Throughput 10478 bytes/sec 16/09/2008 11:42:30|SETI@home|Scheduler RPC succeeded [server version 603] 16/09/2008 11:42:30|SETI@home|Deferring communication for 11 sec 16/09/2008 11:42:30|SETI@home|Reason: requested by project That was directly attributable to 16/09/2008 05:25:12|SETI@home|Computation for task ap_14au08ac_B4_P1_00396_20080912_08575.wu_1 finished 16/09/2008 05:25:14|SETI@home|[file_xfer] Started upload of file ap_14au08ac_B4_P1_00396_20080912_08575.wu_1_0 16/09/2008 05:25:22|SETI@home|[file_xfer] Finished upload of file ap_14au08ac_B4_P1_00396_20080912_08575.wu_1_0 16/09/2008 05:25:22|SETI@home|[file_xfer] Throughput 9065 bytes/sec - an Astropulse task (much higher DCF) finishing six hours earlier. That sawtooth wave generator in action. |
|
4724)
Message boards :
BOINC Manager :
Moving BOINC project to another computer
(Message 20280)
Posted 16 Sep 2008 by Richard Haselgrove Post: I've done this plenty of times with BOINC v5. Your BOINC v6.2.18 will be more difficult, but should be possible. You will need to move the whole BOINC data folder tree. Have a look at the BOINC v6 FAQ for where to find it. You will need to perform a 'clean' installation of BOINC on the new machine - i.e. install it normally, but don't attach to any projects. (Or allow any work you're currently doing to finish and report - you can re-attach to other projects later). Then, empty out the contents of the data directory you've just created on the new machine (but see *), and replace it with the entire contents of the data directory from the old machine - all the .xml and suchlike files from the folder, plus the projects, slots and any other subfolders. * One possible exception is the password file. I don't know enough about the security model of BOINC v6 to know whether you'll need this to match the accounts BOINC will run under on the new machine. Perhaps best to keep a copy just in case, or maybe Jord or someone else can tell us. Provided you can get BOINC to actually run on the new machine, you should have no problem reporting the work as normal. @ Dagorath: if you transfer the whole folder, you're also transferring the machine's "identity" - its Host ID - so the work will still be appearing to be coming from the same host. But @ccarson - that means that you do have to 'retire' the old machine - you can't have two computers attached to the same project with the same host ID. If you really want to copy just the single long-running CPDN model from the old machine, and merge it into a BOINC installation that's already running other projects (but not CPDN, because of the host ID problem), then it should be possible - there's a very long and detailled CPDN ReadMe - but I wouldn't advise it. |
|
4725)
Message boards :
BOINC client :
How to monitoring Boinc 6.xx with Boinc View 1.42?
(Message 20206)
Posted 12 Sep 2008 by Richard Haselgrove Post: In summary: Boinc View doesn't work with the 6.2 clients. Update request to a v6.2.18 client (Windows XP), commanded by BoincView v1.4.2 (also on Windows XP) SETI@home 12/09/2008 23:58:30 Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 1 completed tasks SETI@home 12/09/2008 23:58:35 Scheduler request succeeded: got 0 new tasks Other commands I use regularly (NNT/allow fetch, suspend/resume, etc.) also work correctly. |
|
4726)
Message boards :
BOINC client :
How to monitoring Boinc 6.xx with Boinc View 1.42?
(Message 20198)
Posted 12 Sep 2008 by Richard Haselgrove Post: In summary: Boinc View doesn't work with the 6.2 clients. Which bits don't work? Mine works fine for everything I need, with both the BOINC v6.2.14 I was using for weeks, and the v6.2.18 I upgraded to a couple of days ago when v.14 crashed horribly and trashed 76 tasks. |
|
4727)
Message boards :
BOINC client :
BOINC 6.2.xx - crashes all over the place
(Message 20172)
Posted 12 Sep 2008 by Richard Haselgrove Post: None of those actually addresses what happens when we try to start more concurrent tasks than we have cores. Well, I don't want to - but it seems my CC v6.2.14 did (at 10-Sep-2008 14:46:06, see log below), and that's what provoked the first attack of the shmget() faileds. |
|
4728)
Message boards :
BOINC Manager :
My Wish List - 2
(Message 20157)
Posted 11 Sep 2008 by Richard Haselgrove Post: Just tested it on my system, I can't reproduce that with my 6.2.18, but that may be because I don't have that much work for all the same date. Could someone else with a larger cache and 6.2.18 please check? Probably fairest to exit and reload BOINC Manager after editing regional controls - give it a chance to fully acclimatise to the new settings. We're testing steady state, not transitions. Won't be any problem to testers with daemon/service installs - crunching can continue through Manager testing. |
|
4729)
Message boards :
BOINC client :
BOINC 6.2.xx - crashes all over the place
(Message 20155)
Posted 11 Sep 2008 by Richard Haselgrove Post: Is anyone still performing post-mortems on v6.2.14, to get to the bottom of what caused all those Can't get shared memory segment name: shmget() failed messages? None of those actually addresses what happens when we try to start more concurrent tasks than we have cores. |
|
4730)
Message boards :
BOINC Manager :
My Wish List - 2
(Message 20140)
Posted 11 Sep 2008 by Richard Haselgrove Post: is that one of yours? I ask as I thought you were still using 5.10.13 and swearing at the rest. ;-) I am - but I need to keep one sacrificial host so I know exactly what I'm swearing at, LOL. That's the host which was running v6.2.14 until this happened yesterday. |
|
4731)
Message boards :
BOINC Manager :
My Wish List - 2
(Message 20137)
Posted 11 Sep 2008 by Richard Haselgrove Post: I saw that guy's post. BOINC sorts on date before time. So in decreasing sorting a deadline of 11-11-08 12:56am will come after a deadline of 11-11-08 09.00am Holy smoke, you're right. But it still seems to get confused by AM and PM: 
|
|
4732)
Message boards :
BOINC Manager :
My Wish List - 2
(Message 20135)
Posted 11 Sep 2008 by Richard Haselgrove Post: Gridview, in BOINC 5.10 and upwards, allows column sorting. What was in that new code? Just had a query at SETI, pointing out that sorting on the 'deadline' column uses a text sort, and if the user's regional settings have a single-digit hour format (h:mm:ss tt), then 9:59:59 doesn't sort next to 10:00:00. It works OK if you use hh:mm:ss, so that 09:59:59 comes before 10:00:00 Surely on type-numeric data columns (including date/time), the sorting should be done on the underlying numeric data, and then the formatting applied to the result? This appears to be "format as text first, then sort the text strings": not appropriate. NB I have checked that the sorting behaviour is as described in BOINC Manager v6.2.18 |
|
4733)
Message boards :
Questions and problems :
Outdated tasks get DL'd - ???
(Message 20091)
Posted 11 Sep 2008 by Richard Haselgrove Post: SETI tasks aren't resent, so they must be new ones. What exactly was the deadline date/time shown against the tasks in question (just a sample will do!): how does it compare with the deadline shown for the same task on the SETI website for your host. And, simplest of all, but easy to overlook: is the system clock accurate on your Mac? If it's a G4, it maybe quite elderly - the clock back-up battery could have failed. |
|
4734)
Message boards :
BOINC client :
BOINC 6.2.xx - crashes all over the place
(Message 20078)
Posted 10 Sep 2008 by Richard Haselgrove Post: Is anyone still performing post-mortems on v6.2.14, to get to the bottom of what caused all those Can't get shared memory segment name: shmget() failed messages? Worth a look - but it seems stderrdae.txt hasn't been written to since 7 May 2008, and only contains (multiple iterations of): UNRECOGNIZED: suspend_if_no_recent_input |
|
4735)
Message boards :
BOINC client :
BOINC 6.2.xx - crashes all over the place
(Message 20075)
Posted 10 Sep 2008 by Richard Haselgrove Post: Is anyone still performing post-mortems on v6.2.14, to get to the bottom of what caused all those Can't get shared memory segment name: shmget() failed messages? I have one 6.2.14 client for testing purposes. It was running absolutely smoothly, with no errors at all, until this happened: 10-Sep-2008 11:10:54 [lhcathome] Sending scheduler request: To fetch work. Requesting 57827 seconds of work, reporting 0 completed tasks 10-Sep-2008 11:11:16 [---] Project communication failed: attempting access to reference site 10-Sep-2008 11:11:18 [---] Internet access OK - project servers may be temporarily down. 10-Sep-2008 11:11:19 [lhcathome] Scheduler request failed: Couldn't connect to server 10-Sep-2008 11:28:04 [SETI@home] Computation for task 14au08af.23089.18477.5.8.223_1 finished 10-Sep-2008 11:28:04 [SETI@home] Starting ap_14au08aa_B0_P1_00115_20080909_26999.wu_0 10-Sep-2008 11:28:04 [SETI@home] Starting task ap_14au08aa_B0_P1_00115_20080909_26999.wu_0 using astropulse version 435 10-Sep-2008 11:28:06 [SETI@home] Started upload of 14au08af.23089.18477.5.8.223_1_0 10-Sep-2008 11:28:14 [SETI@home] Finished upload of 14au08af.23089.18477.5.8.223_1_0 10-Sep-2008 12:02:55 [SETI@home] Computation for task 15au08aa.16292.17250.5.8.155_1 finished 10-Sep-2008 12:02:55 [SETI@home] Starting 14au08af.23089.24203.5.8.28_0 10-Sep-2008 12:02:55 [SETI@home] Starting task 14au08af.23089.24203.5.8.28_0 using setiathome_enhanced version 528 10-Sep-2008 12:02:57 [SETI@home] Started upload of 15au08aa.16292.17250.5.8.155_1_0 10-Sep-2008 12:03:04 [SETI@home] Finished upload of 15au08aa.16292.17250.5.8.155_1_0 10-Sep-2008 12:57:31 [SETI@home] Computation for task 14au08af.23089.24203.5.8.28_0 finished 10-Sep-2008 12:57:31 [SETI@home] Starting 15au08aa.16292.20931.5.8.4_0 10-Sep-2008 12:57:31 [SETI@home] Starting task 15au08aa.16292.20931.5.8.4_0 using setiathome_enhanced version 528 10-Sep-2008 12:57:33 [SETI@home] Started upload of 14au08af.23089.24203.5.8.28_0_0 10-Sep-2008 12:57:40 [SETI@home] Finished upload of 14au08af.23089.24203.5.8.28_0_0 10-Sep-2008 13:21:00 [lhcathome] Sending scheduler request: To fetch work. Requesting 57797 seconds of work, reporting 0 completed tasks 10-Sep-2008 13:21:05 [lhcathome] Scheduler request succeeded: got 0 new tasks 10-Sep-2008 13:52:08 [SETI@home] Computation for task 15au08aa.16292.20931.5.8.4_0 finished 10-Sep-2008 13:52:08 [SETI@home] Starting 14au08ae.28085.72.13.8.135_0 10-Sep-2008 13:52:08 [SETI@home] Starting task 14au08ae.28085.72.13.8.135_0 using setiathome_enhanced version 528 10-Sep-2008 13:52:10 [SETI@home] Started upload of 15au08aa.16292.20931.5.8.4_0_0 10-Sep-2008 13:52:18 [SETI@home] Finished upload of 15au08aa.16292.20931.5.8.4_0_0 10-Sep-2008 14:46:06 [SETI@home] Computation for task 14au08ae.28085.72.13.8.135_0 finished [b][color=red]10-Sep-2008 14:46:06 [SETI@home] Starting 14au08ae.28085.890.13.8.242_0 10-Sep-2008 14:46:06 [SETI@home] Starting 14au08ae.28085.3344.13.8.113_1[/color][/b] 10-Sep-2008 14:46:07 [SETI@home] Computation for task 14au08ae.28085.890.13.8.242_0 finished 10-Sep-2008 14:46:07 [SETI@home] Output file 14au08ae.28085.890.13.8.242_0_0 for task 14au08ae.28085.890.13.8.242_0 absent 10-Sep-2008 14:46:07 [SETI@home] Computation for task 14au08ae.28085.3344.13.8.113_1 finished 10-Sep-2008 14:46:07 [SETI@home] Output file 14au08ae.28085.3344.13.8.113_1_0 for task 14au08ae.28085.3344.13.8.113_1 absent 10-Sep-2008 14:46:07 [SETI@home] Starting 14au08ae.28085.3753.13.8.163_1 10-Sep-2008 14:46:08 [SETI@home] Started upload of 14au08ae.28085.72.13.8.135_0_0 10-Sep-2008 14:46:08 [SETI@home] Computation for task 14au08ae.28085.3753.13.8.163_1 finished 10-Sep-2008 14:46:08 [SETI@home] Output file 14au08ae.28085.3753.13.8.163_1_0 for task 14au08ae.28085.3753.13.8.163_1 absent 10-Sep-2008 14:46:08 [SETI@home] Starting 14au08af.1803.6207.6.8.251_0 10-Sep-2008 14:46:09 [SETI@home] Computation for task 14au08af.1803.6207.6.8.251_0 finished 10-Sep-2008 14:46:09 [SETI@home] Output file 14au08af.1803.6207.6.8.251_0_0 for task 14au08af.1803.6207.6.8.251_0 absent This is a quad core, and is attached to a variety of projects: however as of today, every project is set to NNT except SETI and LHC. LHC had no work at the time, so effectively the host had become a SETI-only cruncher. Further, it had (and still has) three Astropulse tasks running - you can see where the third AP task started, at 11:28:04. They are 40-hour plus tasks, so that means that only one core remained available for SETI MB work, and you can see how the tasks start one at a time - at 12:02:55, 12:57:31, 13:52:08 etc. I run with a conservative 1 day cache, so there is no question of tasks being pre-empted for EDF. Then, at 14:46:06 (highlighted), BOINC tried to start two tasks at once. They both crashed with the "shmget() failed" error, and BOINC then proceeded to trash the remaining 74 tasks in the cache, one per second. Fortunately, it didn't trash the running Astropulse tasks, and it did go into a 24-hour backoff on scheduler contact with SETI (no reason apparent in the logs - the only scheduler contacts are: 10-Sep-2008 09:36:57 [SETI@home] Sending scheduler request: To fetch work. Requesting 101 seconds of work, reporting 2 completed tasks 10-Sep-2008 09:37:02 [SETI@home] Scheduler request succeeded: got 1 new tasks and 10-Sep-2008 17:41:07 [SETI@home] Fetching scheduler list 10-Sep-2008 17:41:12 [SETI@home] Master file download succeeded 10-Sep-2008 17:41:17 [SETI@home] Sending scheduler request: Requested by user. Requesting 0 seconds of work, reporting 86 completed tasks 10-Sep-2008 17:41:22 [SETI@home] Scheduler request succeeded: got 0 new tasks 10-Sep-2008 17:47:41 [---] Exit requested by user when I got home). So the only oddity I can see is that double task start at 14:46:06, which would have meant five tasks running on a four-core CPU. Host ID 4292666 at SETI, now upgraded to BOINC v6.2.18 (service install, as before). |
|
4736)
Message boards :
Questions and problems :
Is it possible to increase delay between result upload and reporting?
(Message 19886)
Posted 31 Aug 2008 by Richard Haselgrove Post: Richard wrote:You're showing your age here. SETI used to be set to 10-minutes-and-a-bit deferral, but it's been 11 seconds for a long time now Sorry, sorry, ..... just trying to insert a bit of light-heartedness. My only point was that your recollection of a 10-minute delay was correct, but that it had been changed some time ago and was no longer current. Anyway, that delay is irrelevant to the point under discussion. The delay you're referring to specifies the minimum interval between scheduler requests. It determines the time of the second (and subsequent) scheduler request, given that the first one has already taken place. I'm concerned about the timing of the first scheduler request in a sequence, which is entirely at the determination of the local BOINC client. As I hope I've demonstrated, there's a significant (and explicable) correlation between 'task exit' events and 'work fetch' events. If the BOINC client (free-standing, not under server direction) could insert a delay here, the overall work-flow would be more efficient. |
|
4737)
Message boards :
Questions and problems :
Is it possible to increase delay between result upload and reporting?
(Message 19884)
Posted 31 Aug 2008 by Richard Haselgrove Post: In your case I would more like ask for a check if the data has uploaded already, before trying to report. Has anyone at CPDN ever reported this behaviour to the BOINC devs? There was discussion about this relating to a certain third-party BOINC client which sported RRI as one of its attractions, long after the code had been deprecated in the official BOINC clients. When first released, that third-party client did indeed return results immediately, with the unfortunate side-effect that many tasks at SETI were rejected with validate errors - as Les puts it, the 'report' arriving before the 'data'. The last I heard, a 60-second delay had been added, but not properly debugged - on multicores, the counter wasn't reset if a second task finished within the minute. So if core 0 finished a task at time T, and core 1 finished at T+55 seconds, the second result could be reported 5 seconds after completion. |
|
4738)
Message boards :
Questions and problems :
Is it possible to increase delay between result upload and reporting?
(Message 19882)
Posted 31 Aug 2008 by Richard Haselgrove Post: It repeats asking every minute. If some task would be completed in that interwal it would be reported almost immediately after uploading. You're showing your age here. SETI used to be set to 10-minutes-and-a-bit deferral, but it's been 11 seconds for a long time now - see message 453335. (Flicked past that one while I was looking for the RRI thread!). |
|
4739)
Message boards :
Questions and problems :
Is it possible to increase delay between result upload and reporting?
(Message 19872)
Posted 31 Aug 2008 by Richard Haselgrove Post: I'd like to support Raistmer's suggestion for an enforced delay following task completion and result upload - in fact more than that, a 'protected interval' following task completion when scheduler requests are actively inhibited. This story goes back almost two years, to October 2006. I was questioning whether 'Return Results Immediately' was really such a costly event in BOINC database terms. The discussion is in the SETI@home thread Optimized Clients??, and you can see from Rom Walton's input that it was this discussion that led to his blog entry The evils of 'Returning Results Immediately'. Can I repeat my observation from that research: that for 20 out of the 81 results finished during the test, BOINC initiated a "work fetch" scheduler contact as soon as a task finished, in parallel with the process of uploading the result file. I'm convinced that this is no coincidence. I think that it relates to the re-calculation of RDCF when a task exits: if RDCF is decreased, then the work queue seems to shrink. This very afternoon I've been watching a host where the length of the work buffer (as expressed by BoincView) has been decreasing by an hour each time a task finishes. This has become more significant recently, because within the last month SETI@home has introduced a highly-effective saw-tooth wave generator for RCDF. It's called Astropulse. Astropulse WUs are big, slow, and rare. They have been deliberately introduced with a duration estimate which will tend to an RDCF of around 0.4 on Core2 architecture CPUs. They are typically run interleaved with Multibeam WUs, which are short, fast and common. If the same CPU is running them with the stock SETI science app, they tend towards an RDCF of ~0.25: if, as is often the case, the stock SETI app has been replaced with an optimised one, RDCF can in extreme cases get as low as 0.1 So a SETI host doing mixed work will see a sudden jump in RDCF as an AP task finishes, followed by an extended gentle decline as the commoner MB tasks pass through. This is exactly the scenario where work fetch is triggered at the instant each (MB) task finishes. I was arguing two years ago that an enforced moratorium on scheduler requests would result in more efficient server operation, because many results would be reported "this time" (along with the delayed work request), rather than "next time". To that increased efficiency, we can add the (potential) elimination of these 'validate errors' and the redundant work re-issue that they entail. I have a slight feeling that, following the previous discussion, I started to see the sort of back-off I'm proposing, in the form of a one-minute comms delay, but only in the case of tasks that exit with a science app error. Unfortunately, I don't generate many errors, so I can't look up a log entry at short notice! But there's no comms inhibition for normal status 0/finished file exits: I think there should be. |
|
4740)
Message boards :
Web interfaces :
New forum style corrupts non-ASCII characters ...
(Message 19707)
Posted 23 Aug 2008 by Richard Haselgrove Post: ... at least, as seen on the SETI messages boards... Ah, that's it. I saw similar corruption on SETI Beta a while ago, now corrected. So, testing today: SETI Beta renders correctly in ISO, fails in UTF-8 SETI Main fails to render in ISO, shows correctly in UTF-8 BOINC Dev fails to render in ISO, shows correctly in UTF-8 (but switches IE7 automatically to UTF-8) Any other project switched to the new style-sheets yet? |
|
4741)
Message boards :
Web interfaces :
New forum style corrupts non-ASCII characters ...
(Message 19692)
Posted 22 Aug 2008 by Richard Haselgrove Post: ... at least, as seen on the SETI messages boards. See, for example, http://setiathome.berkeley.edu/forum_thread.php?id=48860 |
|
4742)
Message boards :
BOINC client :
Boinc 6.2.18, Vista64bit: Benchmark crashes
(Message 19556)
Posted 18 Aug 2008 by Richard Haselgrove Post: That sounds like two separate bugs. 1) Under what circunstances does, or should, the installer create a Manager shortcut in a startup group? I found one too many in a test machine I loaded with 6.2.14 last week. 2) Do some of these benchmark problems happen during machine startup on a service install? Can't blame the user if there are other CPU-intensive tasks during that process. Vista has a new Service option for startups: "Automatic (delayed start)". Might be better if BOINC used that when available. |
|
4743)
Message boards :
Web interfaces :
Undocumented change to stats export function?
(Message 19355)
Posted 8 Aug 2008 by Richard Haselgrove Post: Yes, the officially-sanctioned, source-coded stats scripts in the official BOINC SVN repository hard-code full URLs to the XML files on each project. Which hasn't been updated since 05/03/07 - I read that as 3rd. May 2007, or 15 months ago. Time for a review/recheck, if projects like LHC have updated to export code which says something different? And have there really been no new projects in all that time? |
|
4744)
Message boards :
Web interfaces :
Undocumented change to stats export function?
(Message 19354)
Posted 8 Aug 2008 by Richard Haselgrove Post: Well, there's a reasonable start to a documentation framework at http://boinc.berkeley.edu/trac/wiki/XmlStats, but as you say no reference to the db_dump.xml construct. The 'host' xml structure is also outdated - fields <rpc_time> and <credit_per_cpu_sec> are undocumented. I haven't needed to download a recent copy of the other files, so I haven't checked them. Judging by the angst at Einstein (see No xml stats update since upgrade), and your opening remarks here, the actual undocumented change took place a long time ago, and the decision by the various stats sites to use fixed file names, rather than using the db_dump.xml to extract names which one might reasonably expect to be slowly-changing, is probably just as ancient. Anyway, Willy's drive to bring BOINCstats fully up-to-date has found me an extra 0.11 credits I didn't know I had (awarded at LHC since 14 July), so it can't all be bad. But the "Projects in which you are participating" cross-project stat for Einstein still hasn't updated. |
|
4745)
Message boards :
Web interfaces :
Undocumented change to stats export function?
(Message 19347)
Posted 8 Aug 2008 by Richard Haselgrove Post: Well, you aren't supposed to hard-code the XML file name anyway. You're supposed to parse db_dump.xml to know what the files are. <sarcasm>Didn't you see that in the documentation??</sarcasm> It's interesting to note: 1) When Einstein changed the file names back to their previous ones, BOINC Synergy picked up the update OK, thus, I think, confirming my theory that the name-change caused the problem in the first place. (Bruce and Willy are chasing each other all over the file name-space at BOINCstats, so that may take a little longer to settle down). 2) The cross-project stats shown on the bottom of everyone's account pages at every BOINC project are also broken - the Einstein one was stuck at the Sat 2nd August figure until today. Are you suggesting that the officially-sanctioned, source-coded stats repository wasn't written according to the documentation, either? |
|
4746)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 19314)
Posted 7 Aug 2008 by Richard Haselgrove Post: OK, I left it running all day. This is the log: StartServiceCtrlDispatcher being called. This may take several seconds. Please wait. 2008-08-07 11:59:16 [---] Starting BOINC client version 5.10.13 for windows_intelx86 2008-08-07 11:59:16 [---] log flags: task, file_xfer, sched_ops 2008-08-07 11:59:16 [---] Libraries: libcurl/7.16.1 OpenSSL/0.9.8e zlib/1.2.3 2008-08-07 11:59:16 [---] Executing as a daemon 2008-08-07 11:59:16 [---] Data directory: C:\Program Files\BOINC 2008-08-07 11:59:16 [---] BOINC is running as a service and as a non-system user. 2008-08-07 11:59:16 [---] No application graphics will be available. 2008-08-07 11:59:16 [Einstein@Home] Found app_info.xml; using anonymous platform 2008-08-07 11:59:16 [SETI@home] Found app_info.xml; using anonymous platform 2008-08-07 11:59:16 [---] Processor: 2 GenuineIntel Intel(R) Xeon(R) CPU 3040 @ 1.86GHz [x86 Family 6 Model 15 Stepping 2] 2008-08-07 11:59:16 [---] Processor features: fpu tsc pae nx sse sse2 mmx 2008-08-07 11:59:16 [---] Memory: 2.00 GB physical, 3.85 GB virtual 2008-08-07 11:59:16 [---] Disk: 25.00 GB total, 7.83 GB free 2008-08-07 11:59:16 [Einstein@Home] URL: http://einstein.phys.uwm.edu/; Computer ID: 889490; location: school; project prefs: school 2008-08-07 11:59:16 [lhcathome] URL: http://lhcathome.cern.ch/lhcathome/; Computer ID: 9654285; location: home; project prefs: default 2008-08-07 11:59:16 [SETI@home] URL: http://setiathome.berkeley.edu/; Computer ID: 3150564; location: school; project prefs: school 2008-08-07 11:59:16 [---] General prefs: from SETI@home (last modified 2008-04-07 16:56:21) 2008-08-07 11:59:16 [---] Host location: school 2008-08-07 11:59:16 [---] General prefs: using separate prefs for school 2008-08-07 11:59:16 [---] Preferences limit memory usage when active to 2047.27MB 2008-08-07 11:59:16 [---] Preferences limit memory usage when idle to 2047.27MB 2008-08-07 11:59:16 [---] Preferences limit disk usage to 7.83GB 2008-08-07 11:59:16 [SETI@home] Restarting task 03jl08ab.31147.72.15.8.48_1 using setiathome_enhanced version 528 2008-08-07 11:59:16 [SETI@home] Starting 17jn08ac.25817.100967.7.8.137_0 2008-08-07 11:59:16 [SETI@home] Starting task 17jn08ac.25817.100967.7.8.137_0 using setiathome_enhanced version 528 2008-08-07 12:09:22 [SETI@home] Sending scheduler request: To report completed tasks 2008-08-07 12:09:22 [SETI@home] Reporting 1 tasks 2008-08-07 12:26:49 [SETI@home] Computation for task 03jl08ab.31147.72.15.8.48_1 finished 2008-08-07 12:26:50 [Einstein@Home] Restarting task h1_0293.40_S5R4__78_S5R4a_0 using einstein_S5R4 version 604 2008-08-07 12:26:52 [SETI@home] [file_xfer] Started upload of file 03jl08ab.31147.72.15.8.48_1_0 2008-08-07 12:27:42 [SETI@home] Computation for task 17jn08ac.25817.100967.7.8.137_0 finished 2008-08-07 12:27:42 [SETI@home] Starting 17jn08ac.25817.100967.7.8.68_1 2008-08-07 12:27:42 [SETI@home] Starting task 17jn08ac.25817.100967.7.8.68_1 using setiathome_enhanced version 528 2008-08-07 12:27:44 [SETI@home] [file_xfer] Started upload of file 17jn08ac.25817.100967.7.8.137_0_0 2008-08-07 12:44:59 [Einstein@Home] Task h1_0293.40_S5R4__78_S5R4a_0 exited with zero status but no 'finished' file 2008-08-07 12:44:59 [Einstein@Home] If this happens repeatedly you may need to reset the project. 2008-08-07 12:45:43 [Einstein@Home] Restarting task h1_0293.40_S5R4__78_S5R4a_0 using einstein_S5R4 version 604 2008-08-07 12:52:27 [SETI@home] Computation for task 17jn08ac.25817.100967.7.8.68_1 finished 2008-08-07 12:52:27 [SETI@home] Starting 04jl08ab.30358.2935.10.8.110_1 2008-08-07 12:52:27 [SETI@home] Starting task 04jl08ab.30358.2935.10.8.110_1 using setiathome_enhanced version 528 2008-08-07 13:43:56 [SETI@home] Task 04jl08ab.30358.2935.10.8.110_1 exited with zero status but no 'finished' file 2008-08-07 13:43:56 [SETI@home] If this happens repeatedly you may need to reset the project. 2008-08-07 13:43:56 [SETI@home] Restarting task 04jl08ab.30358.2935.10.8.110_1 using setiathome_enhanced version 528 2008-08-07 13:43:57 [Einstein@Home] Task h1_0293.40_S5R4__78_S5R4a_0 exited with zero status but no 'finished' file 2008-08-07 13:43:57 [Einstein@Home] If this happens repeatedly you may need to reset the project. 2008-08-07 13:43:57 [Einstein@Home] Restarting task h1_0293.40_S5R4__78_S5R4a_0 using einstein_S5R4 version 604 2008-08-07 14:18:55 [SETI@home] Computation for task 04jl08ab.30358.2935.10.8.110_1 finished 2008-08-07 14:18:55 [SETI@home] Starting 04jl08ab.30358.4980.10.8.66_0 2008-08-07 14:18:55 [SETI@home] Starting task 04jl08ab.30358.4980.10.8.66_0 using setiathome_enhanced version 528 2008-08-07 14:44:03 [SETI@home] Task 04jl08ab.30358.4980.10.8.66_0 exited with zero status but no 'finished' file 2008-08-07 14:44:03 [SETI@home] If this happens repeatedly you may need to reset the project. 2008-08-07 14:44:56 [SETI@home] Restarting task 04jl08ab.30358.4980.10.8.66_0 using setiathome_enhanced version 528 2008-08-07 14:49:18 [SETI@home] Starting 18jn08aa.27205.11115.7.8.168_0 2008-08-07 14:49:18 [SETI@home] Starting task 18jn08aa.27205.11115.7.8.168_0 using setiathome_enhanced version 528 2008-08-07 15:21:29 [SETI@home] Computation for task 18jn08aa.27205.11115.7.8.168_0 finished 2008-08-07 15:21:29 [SETI@home] Starting 18jn08aa.27205.11115.7.8.207_1 2008-08-07 15:21:29 [SETI@home] Starting task 18jn08aa.27205.11115.7.8.207_1 using setiathome_enhanced version 528 2008-08-07 15:44:42 [SETI@home] Task 18jn08aa.27205.11115.7.8.207_1 exited with zero status but no 'finished' file 2008-08-07 15:44:42 [SETI@home] If this happens repeatedly you may need to reset the project. 2008-08-07 15:44:43 [SETI@home] Restarting task 18jn08aa.27205.11115.7.8.207_1 using setiathome_enhanced version 528 2008-08-07 15:45:33 [SETI@home] Task 18jn08aa.27205.11115.7.8.207_1 exited with zero status but no 'finished' file 2008-08-07 15:45:33 [SETI@home] If this happens repeatedly you may need to reset the project. 2008-08-07 15:45:33 [SETI@home] Restarting task 18jn08aa.27205.11115.7.8.207_1 using setiathome_enhanced version 528 2008-08-07 15:50:35 [SETI@home] Computation for task 18jn08aa.27205.11115.7.8.207_1 finished 2008-08-07 15:50:35 [SETI@home] Starting 04jl08ab.2409.13160.15.8.168_0 2008-08-07 15:50:35 [SETI@home] Starting task 04jl08ab.2409.13160.15.8.168_0 using setiathome_enhanced version 528 2008-08-07 15:57:01 [SETI@home] Computation for task 04jl08ab.30358.4980.10.8.66_0 finished 2008-08-07 15:57:01 [SETI@home] Starting 17jn08ad.32708.13571.10.8.246_1 2008-08-07 15:57:01 [SETI@home] Starting task 17jn08ad.32708.13571.10.8.246_1 using setiathome_enhanced version 528 2008-08-07 16:57:09 [Einstein@Home] Resuming task h1_0293.40_S5R4__78_S5R4a_0 using einstein_S5R4 version 604 2008-08-07 17:01:30 [SETI@home] Computation for task 04jl08ab.2409.13160.15.8.168_0 finished 2008-08-07 17:01:30 [SETI@home] Resuming task 17jn08ad.32708.13571.10.8.246_1 using setiathome_enhanced version 528 2008-08-07 17:58:06 [SETI@home] Starting 17jn08ad.32708.21751.10.8.235_0 2008-08-07 17:58:06 [SETI@home] Starting task 17jn08ad.32708.21751.10.8.235_0 using setiathome_enhanced version 528 2008-08-07 18:17:37 [SETI@home] Computation for task 17jn08ad.32708.13571.10.8.246_1 finished 2008-08-07 18:17:37 [SETI@home] Starting 17jn08ae.32037.1708.10.8.7_0 2008-08-07 18:17:37 [SETI@home] Starting task 17jn08ae.32037.1708.10.8.7_0 using setiathome_enhanced version 528 2008-08-07 18:45:33 [Einstein@Home] Task h1_0293.40_S5R4__78_S5R4a_0 exited with zero status but no 'finished' file 2008-08-07 18:45:33 [Einstein@Home] If this happens repeatedly you may need to reset the project. 2008-08-07 19:18:02 [Einstein@Home] Restarting task h1_0293.40_S5R4__78_S5R4a_0 using einstein_S5R4 version 604 2008-08-07 20:06:01 [SETI@home] Computation for task 17jn08ad.32708.21751.10.8.235_0 finished 2008-08-07 20:06:01 [SETI@home] Resuming task 17jn08ae.32037.1708.10.8.7_0 using setiathome_enhanced version 528 2008-08-07 20:18:02 [SETI@home] Starting 17jn08ad.8158.12753.11.8.219_1 2008-08-07 20:18:02 [SETI@home] Starting task 17jn08ad.8158.12753.11.8.219_1 using setiathome_enhanced version 528 2008-08-07 20:45:13 [SETI@home] Task 17jn08ae.32037.1708.10.8.7_0 exited with zero status but no 'finished' file 2008-08-07 20:45:13 [SETI@home] If this happens repeatedly you may need to reset the project. 2008-08-07 20:45:13 [SETI@home] Task 17jn08ad.8158.12753.11.8.219_1 exited with zero status but no 'finished' file 2008-08-07 20:45:13 [SETI@home] If this happens repeatedly you may need to reset the project. 2008-08-07 20:45:13 [SETI@home] Restarting task 17jn08ae.32037.1708.10.8.7_0 using setiathome_enhanced version 528 2008-08-07 20:45:13 [SETI@home] Restarting task 17jn08ad.8158.12753.11.8.219_1 using setiathome_enhanced version 528 2008-08-07 21:09:11 [SETI@home] Computation for task 17jn08ae.32037.1708.10.8.7_0 finished 2008-08-07 21:09:11 [Einstein@Home] Resuming task h1_0293.40_S5R4__78_S5R4a_0 using einstein_S5R4 version 604 2008-08-07 21:45:13 [SETI@home] Task 17jn08ad.8158.12753.11.8.219_1 exited with zero status but no 'finished' file 2008-08-07 21:45:13 [SETI@home] If this happens repeatedly you may need to reset the project. 2008-08-07 21:45:13 [SETI@home] Restarting task 17jn08ad.8158.12753.11.8.219_1 using setiathome_enhanced version 528 2008-08-07 22:06:39 [---] Exit requested by user Notice: lots of tasks exiting with zero status but no 'finish' file. Lots of tasks finished too, but none of them uploaded. Then I upgraded to BOINC v5.10.45, to keep Jord happy. All the files uploaded while BOINC was running benchmarks (version change), but as soon as computation started, we were back to app 50%, app 35%, BOINC 15% and the manager froze. Stopped the service with services.msc, started the manager as an application: 50%, 50%, 0% and the manager is responsive. Did the security patches and the reboot. Service now running 50, 50, 0 and manager is responsive. Re-installed v5.10.13. Service runs properly, manager fails to load: BOINC.DLL is missing. Use /a switch to extract, move to folder, loads OK - all seems good. Has already reported tasks uploaded by v5.10.45. Decide not to meddle any further. -------------------------------- It <feels> as if something got strangled in the comms layer of the machine - which seems odd, since I had no problem getting into it with remote desktop, over the same network interface. But the question which is intriguing me is: why did BOINC --daemon behave so differently from BOINCmgr|BOINC? Both v5.10.13 and v5.10.45 behaved the same (not at all to my surprise), so it's a current problem - as I said, it's a domain controller, so we're not going into the v6.2.x sequence. Ideas? |
|
4747)
Message boards :
Web interfaces :
Undocumented change to stats export function?
(Message 19310)
Posted 7 Aug 2008 by Richard Haselgrove Post: Several projects have been updating their server code recently - major updates at LHC and Einstein, more incremental ones at CPDN and SETI. Each of those four has suffered from delays in transferring statistics to external stats sites. As I type, the 'last update' is showing as: LHC - Mon 14 July Einstein - Sat 02 Aug at both BOINCstats and BOINC Synergy. Einstein exported around 14:00 UTC today - Bernd posted, and I checked that the folder which was previously empty is now populated. There's an interesting post by Toby (KSWN) at LHC: Looks like the names of the XML files changed (from team_id.gz to team.gz and same for host and user). I just updated my database to reflect this and it seems to be working again. They couldn't have ................. ............, could they? |
|
4748)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 19284)
Posted 7 Aug 2008 by Richard Haselgrove Post: Why not try 5.10.45 and see if it does the same thing? If it doesn't, there was a fix somewhere between .13 and .45 (it's only 32 iterations, something must've been added, changed, prodded, fixed). I may try that after working hours this evening. But it's going back to v5.10.13 afterwards. (Did you hear the one about the validate errors at SETI if you use RRI? That's why it's set at a nice safe 15 minutes) (or 14:24, to be pedantic). It's definitely something about the state the machine's in today, so I'll definitely be rebooting it remotely when I've finished - got some security patches to install while I'm there, I see. So any further tests I should run need to be notified in the next 6 hours or so. |
|
4749)
Message boards :
BOINC client :
BOINC cc using excess CPU as service
(Message 19275)
Posted 7 Aug 2008 by Richard Haselgrove Post: Could anyone explain the difference between these two screenshots, please? BOINC running as a service BOINC running as an application As a service, BOINC.EXE was using between 6% and 13% of CPU cycles (fluctuating on each refresh of task manager). BOINC Manager wouldn't open properly - not even the menu is visible, and it was unresponsive. I had to crash-close it. If I stopped the service, the manager opened normally, started the cc, and gave all CPU to the project applications, exactly as it should. This behaviour was fully reproducible - daemon wasted cycles, client-manager worked as normal. Normally, there's no problem - the service daemon uses next-to-no cycles, and the (dual core) apps run at 50% each. This is out-of-envelope behaviour for the host, today only: but like I said, reproducible. Test environment is awkward - it's a remote server which I have authority to manage using Remote Desktop. It's only a small test and training network, which means I'll be able to reboot and test again overnight - but not in the middle of the working day. Also, it's a Windows 2003 Small Business Server, which means it absolutely has to run as a Domain Controller - so no upgrade to BOINC v6.2.x. BOINC also has to run as a service, because 99.9% of the time nobody is logged into the console. You'll see I'm running v5.10.13 (my favourite), but I don't remember there being many relevant changes between that and v5.10.45. I'll stick with .13, thank you very much, because the 15 minute connect/reporting interval I've set makes it easier to monitor without the overhead of setting up the whole VPN/RDP management thing. |
|
4750)
Message boards :
BOINC client :
app_info.xml parsing problem
(Message 19085)
Posted 2 Aug 2008 by Richard Haselgrove Post: Bump. This problem is now showing up on SETI Main as well. The application developers are rather busy on social engineering issues right now, so assistance from the rest of the BOINC community would be appreciated. |
|
4751)
Message boards :
Questions and problems :
Boing 6.2.14 on PentiumD WinXP
(Message 19025)
Posted 1 Aug 2008 by Richard Haselgrove Post: I've installed the new version but now run only one project with cpu idle at 50%. If you follow the link, you'll find that the user's only computer is a P4 @ 2.80 GHz, with 1GB of RAM and running Windows XP. It's running the least demanding of the CPDN models, an SM3 for which the recommended memory statement is: 57MB used by the model, 256MB RAM recommended. Astropulse has a similar memory footprint: Minimum CPU: 1.6 GHz Minimum RAM: 256 MB Miniumum disk space: 128 MB and it occupies about 48 MB while running. Doesn't sound like memory starvation to me. |
|
4752)
Message boards :
Questions and problems :
Boing 6.2.14 on PentiumD WinXP
(Message 19009)
Posted 31 Jul 2008 by Richard Haselgrove Post: I've installed the new version but now run only one project with cpu idle at 50%. That depends on the projects in question. I think you're remembering old readings with SETI on one HT 'core', and Einstein on the other. There has been a recent report (2 days ago) that with CPDN and Astropulse sharing a HT processor, the Astrolpulse task runs seven times slower than when it runs on its own (no figures for what happened to the CPDN task). |
|
4753)
Message boards :
BOINC client :
app_info.xml parsing problem
(Message 18820)
Posted 25 Jul 2008 by Richard Haselgrove Post: We're experimenting with app_info.xml files at SETI Beta, in anticipation of the release of Astropulse. SETI's third party optimised applications make heavy use of the anonymous platform mechanism, even on standard operating systems. We've had a case where a user was using an app_info including the constructs <executable /> and <main_program /> (with spaces). He then combined this with a second block specifying the Astropulse application, using the constructs <executable/> and <main_program/> (without spaces). All tasks for the original application - the one specified with spaces - then failed with "No main program specified". I have reproduced the failure on my own machine (by adding spaces), and advised that the spaces should be removed from the reporting problem machine. Further reports as I receive them. The question is, why did it (apparently) work before editing? I have checked the optimised application download package, and the spaces are present in the pre-supplied app_info. It's been available for download for almost three months (since 1st. May), so I would have expected problems to have shown up and been corrected by now. All of this is happening in BOINC 5 (mine v5.10.13, his v5.10.45), so the debug output introduced in [trac]#703[/trac] isn't automatically available. I'm not sure whether <unparsed_xml> will help, but I'll try it if no-one can tell me exactly what's wrong with the spacing. |
|
4754)
Message boards :
Questions and problems :
icon in notification tray in system install
(Message 18398)
Posted 13 Jul 2008 by Richard Haselgrove Post: I performed a system install since this computer is often on but not logged in to Windows, so BOINC could run during all that unused time. When I logged in I set the BOINC Manager icon in the notification area by the clock to "always hide" because I don't want to see it there all the time. However, when I log off or reboot the icon reappears in the notification area, and when I check the settings for it, it's back to "hide when inactive". As I read it, your 'system' install is the same thing as the 'service' install the other posters have suggested. 'Service' install is the more standard terminology, so it will avoid the slight confusion if we use that from now on. With a service install, there is no need for any BOINC icon to appear in the system tray, hidden or not. However, if you have previously had a 'user' BOINC installation on the computer, you may find that there is a leftover BOINC Manager shortcut in a startup group somewhere - either your personal one, or in the 'all users' area. Look for a BOINC entry under 'startup' in Start button-->All Programs, and delete it if present. |
|
4755)
Message boards :
Web interfaces :
BOINC's own website - translations missing
(Message 17292)
Posted 10 May 2008 by Richard Haselgrove Post: Everything is fixed again. Looks like someone woke up. :-) Not that it's any consolation, but my pre-existing trac login ('remember me on this computer', or whatever it is) is still working - so they can't have hosed the entire user database. |
|
4756)
Message boards :
Web interfaces :
BOINC's own website - translations missing
(Message 17286)
Posted 10 May 2008 by Richard Haselgrove Post: There's a serious problem with this website, especially the downloads page. I'll try a few PMs - anyone one got any quicker routes? (I've already sent a PM to DaveA from my SETI account - that should also trigger an email notification with the full text, unless he's got that facility turned off.) |
|
4757)
Message boards :
BOINC client :
Message from server: platform 'hppa64-hp-hpux' not found
(Message 17263)
Posted 9 May 2008 by Richard Haselgrove Post: Most projects have an 'applications' page - reached from their main BOINC page, or by typing apps.php after their main url. The apps page usually lists the names of the current executable files, and the platforms they're intended for. For some reason, SETI's apps page isn't working at the moment. I very much doubt that you will get a compatible application downloaded automatically just by subscribing to SETI, but the other benefit of that project is that the 'science' application is open source, and variants have been made available by other volunteers for all sorts of platforms and operating systems. Several websites offering applications for HPUX are included in this post by Dotsch on the SETI message board. |
|
4758)
Message boards :
BOINC client :
BOINC 6.2.1 outstanding bugs roll-up thread
(Message 17231)
Posted 7 May 2008 by Richard Haselgrove Post: I'll open the batting. Trac [trac]#588[/trac] (indefinite suspension of computing when changing system clock) is still present in v6.2.1, as is the other problem discussed in the related Benchmarking bug - indefinite suspension of computing thread (<active_frac> set very low and doesn't self-correct). In fact, the 'indefinite suspension' bug is even worse than I described it at the time: BOINC Manager shows the tasks as still 'running, high priority', but no computing is being done (CPU time not increasing), and a remote viewer like BOINCview shows the 'suspended' status. |
|
4759)
Message boards :
BOINC client :
BOINC 6.2.1 outstanding bugs roll-up thread
(Message 17230)
Posted 7 May 2008 by Richard Haselgrove Post: From Rom's email quoted in the changelog thread: We are feature complete for 6.2, from here on out 6.2 will just be bug fixes. Any of the remaining new features that didn’t make it are going to be punted to the next release. Windows 9x should work again. Why don't we take him at his word, and gather together here any bugs which have been reported, but haven't (yet) made it into the v6.2.1 pre-release? |
|
4760)
Message boards :
Web interfaces :
Venues and CPU types
(Message 17017)
Posted 1 May 2008 by Richard Haselgrove Post: The lack of the physical count is really aggravating for me..... Have you been following the two parallel threads at SETI? Your computers page in Number Crunching, and Lesson Today: Databases (Apr 29 2008) in Technical News. Nicolas (posting as PovAddict) found a quote from David Anderson: Getting the result count for a host is a potentially expensive query, and doing this for each host in a long list would be bad. (in a mailing list, not a code changeset or trac ticket). Yet we have Matt Lebofsky, actually managing probably the most highly-stressed servers in the BOINC business, saying ... we can handle a ton more web page queries ... The jury is still out. |
|
4761)
Message boards :
BOINC client :
IRQL_NOT_LESS_OR_EQUAL
(Message 16914)
Posted 27 Apr 2008 by Richard Haselgrove Post: After BOINC have run for about 1 - 2 hours I usually get this error: http://www.microsoft.com/technet/prodtechnol/windows2000serv/reskit/w2000Msgs/6077.mspx?mfr=true I got a rash of these when I built a Q9300 onto a cheap mobo recently. The stop error (in Win XP) also listed the file name where the problem ocurred, and in my case it was a NVidia graphics card device driver. A concentrated session tracking down all the latest drivers, and an updated installation of DirectX 9.0c, seems finally to have reached stability - touch wood. |
|
4762)
Message boards :
BOINC Manager :
Task computation error gone on reboot
(Message 16888)
Posted 26 Apr 2008 by Richard Haselgrove Post: I think the way the WU outcome is reported on the Tasks page of a project's website also changes the same way: I'm still not certain, but I can supply some corroborative evidence. SETI Beta WU 1105766 is a normal 'surplus to requirements' cancellation. SETI Beta WU 1106057 is the same thing, but with a local BOINC restart between the server issuing the cancellation instruction, and the client reporting that it has received it. We get the "Done" client state, instead of the proper one. |
|
4763)
Message boards :
BOINC client :
Ubuntu 8.04 throttling
(Message 16885)
Posted 26 Apr 2008 by Richard Haselgrove Post: I have a Dell E1505 with a Core Duo T2500 running on Ubuntu 8.04. Its full clock speed is 2 GHz, but when the computer is idle, it runs at 1 GHz. It can also run at 1.33 GHz and 1.67 GHz, depending on the load. I've noticed that BOINC seems to be a lower priority process, so the CPU stays throttled at 1 GHz. When I have a user process also running at the same time, the CPU ramps up to 2 GHz, and BOINC actually goes faster (I can see the estimated time to completion decrease.) Is there any way to get BOINC to unthrottle the CPU? (I know temperature is not an issue, as the Windows client causes the CPU to step up to 2 GHz. The fans eventually kick in, plus I have a laptop cooler, and the CPU temp will settle in at 60 degrees centigrade.) This is becoming an increasingly common question. There has been a long discussion about it at SETI (with suggestions for how to deal with it), and two recent threads at Einstein here and here. But I agree that it's a general BOINC issue, not specific to any one project, and the best place to discuss it is here at BOINC - though possibly in the 'BOINC core client' area rather than the 'BOINC Manager' area. |
|
4764)
Message boards :
Web interfaces :
Venues and CPU types
(Message 16877)
Posted 26 Apr 2008 by Richard Haselgrove Post: Now #631. So is the OS sub-type and a link to (but not a count of) tasks assigned. The link to the actual computer details is currently broken, |
|
4765)
Message boards :
Web interfaces :
Venues and CPU types
(Message 16871)
Posted 25 Apr 2008 by Richard Haselgrove Post: "Be careful what you wish for. You might just get it." Well done Les - it turns out that it was a wish, by one SETI user, turned into a trac ticket [trac]#569[/trac] by our own dear Ageless, and closed two months ago so it doesn't show on the default trac filter. And no, it didn't ask for the change to be made this way. Looking at [trac]changeset:14897[/trac], $host->p_model should have been kept in preference to $host->p_vendor (I think we know which vendor makes Pentiums, and which vendor makes Athlons). And there's a lot of affection for $ltr. |
|
4766)
Message boards :
Web interfaces :
Venues and CPU types
(Message 16868)
Posted 25 Apr 2008 by Richard Haselgrove Post: The natives are restless, again. Strong pressure (there and in Technical News) to bring back a direct link to 'Results for computer'. Any chance we could get David Anderson to make his sixth visit to these public fora in 30 months, and explain what the **** he's on about? |
|
4767)
Message boards :
Web interfaces :
Venues and CPU types
(Message 16865)
Posted 25 Apr 2008 by Richard Haselgrove Post: The 'Your computers' listing at SETI has been changed (breaking various things, and causing yet more stress for the part-time project staff, as usual). The owner's display has a new 'Location' column (displaying the information elsewhere known as 'venue'), which is marginally useful, but to make space, they've shrunk the 'CPU Type' column to the point of uselessness. IMO, anyway. Anyone? |
|
4768)
Message boards :
BOINC Manager :
Unable To Connect To Any Projects
(Message 16817)
Posted 24 Apr 2008 by Richard Haselgrove Post: 4/23/2008 was Wednesday. Actually, SETI's maintenance was switched to Tuesdays a while back, so that's not it. |
|
4769)
Message boards :
BOINC client :
can't download new work units
(Message 16789)
Posted 22 Apr 2008 by Richard Haselgrove Post: The way I understand the problem (and maybe I don't understand it fully), CPDN accumulating debt while running solo for a year won't cause the user any real grief during that year so there is no need for any warning while CPDN is running solo. If I understand correctly, the user runs into trouble when s/he turns on other projects. At that point BOINC could advise the user that corrective action needs to be taken. If corrective action means reseting the debts and/or adjusting shares then so be it. The 'problem' often manifests itself the other way round: the user has been running a mix of projects, got used to the task-switching behaviour, and then wonders why it stops when they add CPDN to the mix. @ JM7 I think you're getting this the wrong way round. BOINC has defined that resource shares are a long-term committment, and written code to match. Now you are trying to force users to accept that paradigm. Why should users not choose which projects to support with their CPU cycles on a short-term basis too? That is just as valid a contribution to the science. |
|
4770)
Message boards :
BOINC client :
can't download new work units
(Message 16702)
Posted 17 Apr 2008 by Richard Haselgrove Post: What should I be looking at? One thing would be the time statistics for the computer. On any project website, towards the bottom of your computer details, you should see a block like: % of time BOINC client is running While BOINC running, % of time work is allowed Average CPU efficiency Task duration correction factor What figures do you see there? |
|
4771)
Message boards :
BOINC Manager :
Task computation error gone on reboot
(Message 16689)
Posted 16 Apr 2008 by Richard Haselgrove Post: I think you may be right but I'm not about to induce a computation error in any of the tasks I have running at the moment to verify (too much CPU time invested). Are you certain? Can anyone else verify? I have a slightly unstable new build which has thrown up a few errors recently. If I see another one, I'll do the restart and see what changes. Of course, by typing that I'm trying to invoke Murphy - when you want an error, everything runs smoothly. But does this count as a double-Murphy? |
|
4772)
Message boards :
BOINC Manager :
Task computation error gone on reboot
(Message 16686)
Posted 16 Apr 2008 by Richard Haselgrove Post: Anway, I submitted Trac ticket #621 for this issue. I think the way the WU outcome is reported on the Tasks page of a project's website also changes the same way: Client error Compute error if the task is reported before a BOINC restart, and Client error Done if it's reported afterwards. That could be fixed at the same time. |
|
4773)
Message boards :
Web interfaces :
Cached project forum pages?
(Message 16623)
Posted 12 Apr 2008 by Richard Haselgrove Post: CPDN Beta has: CPDN Beta now has <!-- $Id: translation.inc 14916 2008-03-14 21:40:18Z davea $ --> <!-- $Id: pm.inc 14019 2007-11-01 23:04:39Z davea $ --> <!-- $Id: team.inc 14394 2007-12-18 20:28:08Z davea $ --> <!-- $Id: text_transform.inc 14866 2008-03-07 04:38:06Z boincadm $ --> <!-- $Id: stats_sites.inc 15020 2008-04-06 03:15:36Z davea $ --> <!-- $Id: user.inc 14626 2008-01-28 16:12:28Z boincadm $ --> <!-- $Id: profile.inc 14866 2008-03-07 04:38:06Z boincadm $ --> <!-- $Id: util.inc 14854 2008-03-06 00:40:48Z davea $ --> and it's become as bad as the others. Any chance we could move forwards, instead of backwards? And why do threads become 'unread' after ?31 days? |
|
4774)
Message boards :
BOINC client :
Problems with "in use" detection in Ubuntu Linux
(Message 16600)
Posted 11 Apr 2008 by Richard Haselgrove Post: Hi, Erlend. Welcome to BOINC forums. Read the saga at trac ticket [trac]#463[/trac]! |
|
4775)
Message boards :
BOINC client :
Boinc using cpu on Remote Computer
(Message 16589)
Posted 10 Apr 2008 by Richard Haselgrove Post: Hi all, Does the remote server run on the same domain as the user's ordinary workstation, and does she use the same login credentials on both machines? If so, her domain account could have been set up with a profile stored on the domain server, and standard settings (including desktop and startup items) could migrate between machines that way. It wouldn't necessarily mean that BOINC, or any science applications, were actually running on the server, just that some of the icons were visible. Solutions - don't use server profiles unless strictly necessary: use different logins for daily workstation / server administration roles: uninstall BOINC from the workstation (it should leave the CPDN model intact, but make a backup to be on the safe side) and re-install it as a service without an automatic startup for the manager. |
|
4776)
Message boards :
BOINC Manager :
My Wish List - 2
(Message 16581)
Posted 10 Apr 2008 by Richard Haselgrove Post: I've added ticket #620 for this. Fair enough - thanks for the extended explanation, and I have no quarrel at all with your proposed underlying functionality. It all goes to re-affirm that core coding, and user interface design, are two seperate disciplines and needn't (perhaps shouldn't) be entrusted to the same person. But it is important that both parties understand their common goal, and have good communication between them. |
|
4777)
Message boards :
BOINC Manager :
My Wish List - 2
(Message 16579)
Posted 10 Apr 2008 by Richard Haselgrove Post: Also, if another project with very long WUs comes on stream .... Like CPDN Beta. |
|
4778)
Message boards :
BOINC client :
CPU Benchmarks
(Message 16576)
Posted 10 Apr 2008 by Richard Haselgrove Post: Fedora 7 with KDE has a GUI to do this. But I don't think I ever changed what it was set at (dynamic FWIW, choices are Performance, Dynamic and Powersave). On second glance, it's probably specific to KDE because it's in KPowersave. I'd guess that Gnome has something similar though. I'll need to boot back to Windows this weekend, so before I go back to Fedora I'll poke around in Ubuntu/Gnome and see what I can find. It probably also matters what SpeedStep / Cool'n'Quiet controls are available in your BIOS, and what they're currently set at. I think this is a difficult class of problem for volunteer helpers to deal with. It's most likely to be reported at a project level (and with a complaint of RAC falling, rather than benchmarks as here): and I would actually expect BOINC benchmarks to run at full speed (normal priority), and the CPU speed to drop only when the computer is otherwise unused and the project science apps are the only significant thing running. But the issue wouldn't be specific to any one particular project: it's genuinely a BOINC issue, because it arises from BOINC assigning low CPU priority ("nice") to project science apps. That raises the question of the relationship between BOINC, and project, helpdesks and discussion forums. I think I'll just leave that one hanging. |
|
4779)
Message boards :
BOINC client :
5.10.45 Fedora 7 64 bit always suspended
(Message 16573)
Posted 10 Apr 2008 by Richard Haselgrove Post: I have it set to always run and the network is always available on this client. What do the message logs say? BOINC always runs benchmarks after a version change, and there are bugs in this area...... |
|
4780)
Message boards :
BOINC client :
CPU Benchmarks
(Message 16569)
Posted 10 Apr 2008 by Richard Haselgrove Post: Hi guys. The CPU performance measures for my machine show much higner figures for when it had windows vista on it than at the minute, using ubuntu linux. Any ideas why? Do BOINC projects run slower on linux? Not necessarily, but possibly. There are two quite separate issues. a) The benchmark figure reported by BOINC. This is, frankly, meaningless for resolving questions like this. Ignore it. b) The actual speed your CPU is running. If you've been running Vista, you presumably have a reasonably modern and powerful CPU in your computer. These usually have a power-saving or cool-running slow mode, and it's possible that Linux has selected this mode for the low-priority science applications launched by BOINC. We had a long discussion about this at SETI recently: that link drops you in the middle, at the best answer: Toby says to use cat /proc/cpuinfo to find the current CPU speed, and sudo cpufreq-selector -g performance to fix it if it turns out to be slow. Edit - I did in fact put in a request for this information to be added to the BOINC FAQ, so that all help-desk advisors would be aware of it. Not unreasonably I got a reply that someone with personal experience, and understanding of the CPUs/Linux versions affected, would be best placed to do the write-up. I don't have the experience to be that person: any takers? |
|
4781)
Message boards :
BOINC Manager :
My Wish List - 2
(Message 16564)
Posted 10 Apr 2008 by Richard Haselgrove Post: 1) vacation manager: I'd like to tell BOINC that my computer will be off during a certain time span. e.g. ~7 days from 2008-01-01 12:00 till 2008-01-08 06:00. Sorry, Peter, I really don't like this idea at all. I've added the following comment to your ticket. I think this is going to be really unfriendly and confusing for ordinary users - "I'm going to be away for a week, but I can fit in a couple of days work before then. So what figure do I type in this box? Minus five? Huh?" How many other programs do you use that work like that? |
|
4782)
Message boards :
Web interfaces :
Cached project forum pages?
(Message 16546)
Posted 9 Apr 2008 by Richard Haselgrove Post: Just noticed an odd sub-feature of the original 'cached pages' problem (which is still here). Using IE7 on Windows (both Vista SP1 and XP SP2), if I save up a reference from a forum page, by right-clicking a link and choosing 'Open in New Tab', and then go back to the page I originally arrived from, the page refreshes and the yellow pages are cleared. Just an observation. But it reminds me of an aggravation with the recent design changes on these boards. "Send message" (as in PM) is wrong. It doesn't send a message, it's a link to a message composition page. What I want to be able to do is to open the message form in a new tab by right-clicking, so I can keep my place on the forum page and refer back to it while I'm composing the private message. You can't do that with a button in IE7. (Workround - right-click to the user's account page, and PM from there - but I don't like unnecessary two-step workrounds). |
|
4783)
Message boards :
BOINC Manager :
BOINC network/cpu times
(Message 16532)
Posted 8 Apr 2008 by Richard Haselgrove Post: Not a whole lot that can be done about it though. There are numbers that cannot be represented exactly in binary. This is almost certainly pure rounding problems. Store it in seconds, like every other BOINC time? |
|
4784)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16477)
Posted 5 Apr 2008 by Richard Haselgrove Post: Found it! (Subject to checking and validation - please confirm) http://boinc.berkeley.edu/trac/browser/trunk/boinc/client/cs_benchmark.C?rev=12128 307 bool CLIENT_STATE::cpu_benchmarks_poll() {
308 int i;
309 static double last_time = 0;
310 if (!benchmarks_running) return false;
311
312 if (now < last_time + 1) return false;
313 last_time = now;
314
315 active_tasks.send_heartbeats(); If benchmarks have been run in the current BOINC session, at some time in the future (as a result of the clock fumbling we've been talking about), the static variable last_time will have been initialised and will have a value of, for example, Wed, 30 Apr 2008 13:25:40 UTC.So the test at line 312 will be satisfied, and the application will loop until the cows come home (or Wed, 30 Apr 2008 13:25:40 UTC, whichever comes sooner). That explains why exiting BOINC and re-starting allows benchmarks to run properly: the variable will be undefined and correctly initialised to zero. Solution: explicitly set the value of last_time to zero on all possible exit routes out of the benchmarking loop, so that it's properly initialised for next time. NB that's OK: this is a timing variable for the benchmark duration, nothing to do with the 5-day interval between benchmarks. That's tested at 250 double diff = now - host_info.p_calculated; 251 if (diff < 0) return true; 252 253 return ((run_cpu_benchmarks || diff > BENCHMARK_PERIOD)); |
|
4785)
Message boards :
BOINC client :
active_frac Question (BUG?)
(Message 16475)
Posted 5 Apr 2008 by Richard Haselgrove Post: I think I've now tracked down this bug to line 69 in time_stats.C: details and reasoning in Benchmarking bug - indefinite suspension of computing. The problem is initiated if the computer user inadvertently sets the computer's clock to some time/date in the future, and then resets it to a more accurate time/date. All updating of time_stats metrics is then inhibited until ten (?)seconds(?) after the maximum time/date ever set on that computer's clock while BOINC is running. |
|
4786)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16474)
Posted 5 Apr 2008 by Richard Haselgrove Post: OK, this is a new post - the previous poster wasn't worth replying to, so I won't reply. He was right, however, right to say that I have all the data available, and he was also right to say that I should have posted the whole story. The key datum is <last_update>1209561940.387707</last_update> in the <time_stats> in my post of 11:29 UTC. Using http://www.onlineconversion.com/unix_time.htm, that equates to Wed, 30 Apr 2008 13:25:40 UTC - still 25 days in the future. Here are lines 127-151 of time_stats.C: // Update time statistics based on current activities
// NOTE: we don't set the state-file dirty flag here,
// so these get written to disk only when other activities
// cause this to happen. Maybe should change this.
//
void TIME_STATS::update(int suspend_reason) {
double dt, w1, w2;
bool is_active = !(suspend_reason & ~SUSPEND_REASON_CPU_USAGE_LIMIT);
if (last_update == 0) {
// this is the first time this client has executed.
// Assume that everything is active
on_frac = 1;
connected_frac = 1;
active_frac = 1;
first = false;
last_update = gstate.now;
log_append("power_on", gstate.now);
} else {
dt = gstate.now - last_update;
[color=red]if (dt <= 10) return;[/color]
w1 = 1 - exp(-dt/ALPHA); // weight for recent period
w2 = 1 - w1; // weight for everything before that
// (close to zero if long gap)(sorry, I can't use [code] for code, because of the indent bug on these boards) I call BUG at line 148 (highlighted). This contains an implied assumption that time is always monotonic (i.e. the clock hasn't been fiddled with - which is where we came in). The intention of the test is clearly to reduce workload by only re-calculating active_frac at intervals of 10 time_units or more: the effect is to inhibit updating following a clock-fiddle until (MAX(clock) + 10) is reached. The test should be [color=red]if (ABS(dt) <= 10) return;[/color](or whatever the C construct is - sorry, I'm a VB programmer) Now, I suppose it's up to me to find the line number of the original benchmarking bug. Correction: the 10 time_unit test is at line number 69 of http://boinc.berkeley.edu/trac/browser/trunk/boinc/client/time_stats.C?rev=4610 - I got the first number from my Visual Studio editor, working on the text version of the file which the BOINC/Wiki search function found first. |
|
4787)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16471)
Posted 5 Apr 2008 by Richard Haselgrove Post: For now, suggest to exit BOINC, open client_state.xml with ASCII text-editor and set that value to the the march 26 one. Also set the DCF's to 1.000000 so at least crunching and work fetching return to normality. The DCF's are indicative of the situation slowly returning to normality.... right now BOINC figures things complete much faster than the other parms indicate. That's exactly the reply I was trying to avoid. In my original post (message 16467), I put a footnote in small print. If you had followed recommended Forum practice, and used 'Reply to Post' (for threading purposes), instead of 'Post to Thread', you would have seen it. Since you clearly missed it, here it is for the visually-impaired: Please, no sticking-plaster replies: I know what to change and how to change it, but I'm researching whether there's a need to put in another bug report |
|
4788)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16469)
Posted 5 Apr 2008 by Richard Haselgrove Post: Why dont you post the whole top section of the client_state.xml + some project DCF's so we can have a integral view rather than this step by step guessing game. From current client_state.xml: <host_info>
<timezone>3600</timezone>
<domain_name>ANONYMOUS</domain_name>
<ip_addr>192.168.173.13</ip_addr>
<host_cpid>e90761a879d5bf174f2e7e32671872db</host_cpid>
<p_ncpus>1</p_ncpus>
<p_vendor>GenuineIntel</p_vendor>
<p_model> Intel(R) Pentium(R) 4 CPU 2.00GHz [x86 Family 15 Model 2 Stepping 4]</p_model>
<p_features>fpu tsc sse sse2 mmx</p_features>
<p_fpops>1050903119.868637</p_fpops>
<p_iops>1698914891.321735</p_iops>
<p_membw>1000000000.000000</p_membw>
<p_calculated>1207315872.154749</p_calculated>
<m_nbytes>536133632.000000</m_nbytes>
<m_cache>1000000.000000</m_cache>
<m_swap>1310920704.000000</m_swap>
<d_total>39990591488.000000</d_total>
<d_free>5179965440.000000</d_free>
<os_name>Microsoft Windows XP</os_name>
<os_version>Home Edition, Service Pack 2, (05.01.2600.00)</os_version>
<accelerators>NVIDIA GeForce3 Ti 200</accelerators>
</host_info>
<time_stats>
<on_frac>0.805489</on_frac>
<connected_frac>-1.000000</connected_frac>
<active_frac>0.045279</active_frac>
<cpu_efficiency>0.937373</cpu_efficiency>
<last_update>1209561940.387707</last_update>
</time_stats>
<net_stats>
<bwup>6416.736822</bwup>
<avg_up>29442415.841511</avg_up>
<avg_time_up>1207383404.717249</avg_time_up>
<bwdown>61101.566988</bwdown>
<avg_down>1165019550.171246</avg_down>
<avg_time_down>1207380038.842249</avg_time_down>
</net_stats>Einstein: <duration_correction_factor>0.342887</duration_correction_factor> SETI: <duration_correction_factor>0.260244</duration_correction_factor> (both with Power/Optimised apps, respectively). No other Project entries. From a client_state.xml.bak file dated 26 May 2006 - must have been the last time I used BoincDV to reset debts: <time_stats>
<on_frac>0.998095</on_frac>
<connected_frac>1.000000</connected_frac>
<active_frac>0.999851</active_frac>
<cpu_efficiency>0.949493</cpu_efficiency>
<last_update>1148667485.875000</last_update>
</time_stats>- I would judge that to be pretty normal for this machine: _efficiency @ ~95% is partly because it's the BoincView monitoring host for my LAN. Since there is no easy way to monitor changes in debt values over time (the subject of a different bug report), I wrote myself a small utility to record and graph project debt values. I'll adapt it to log time_stats over time, and report next weekend. Any other tags you would like a time series for? Edit - here are the Einstein tasks for this host. The report/fetch contact at 5 Apr 2008 11:45:46 UTC today was triggered as the LTD from the last high priority run rose above -3600. Initial metrics for the new task are: Computation time to completion: 16 hours 34 minutes 'Work buffer' from BoincView: 16 days 12 hours Project shares are equal (50% - 100::100) Task immediately went into high priority: 05/04/2008 12:43:48|Einstein@Home|Sending scheduler request: To fetch work. Requesting 1490 seconds of work, reporting 1 completed tasks 05/04/2008 12:43:58|Einstein@Home|Scheduler request succeeded: got 1 new tasks 05/04/2008 12:44:00|Einstein@Home|Starting h1_0907.30_S5R3__78_S5R3b_0 05/04/2008 12:44:08|Einstein@Home|Starting task h1_0907.30_S5R3__78_S5R3b_0 using einstein_S5R3 version 436 2nd. edit: Another observation - I run a 0.01 day CI, and a 1 day AC. That work fetch would normally be for (87264 - ε) seconds, where ε = ~4,363 seconds for a 95% CPU efficiency. That demonstrates how the active_frac corruption impinges on work fetch. |
|
4789)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16467)
Posted 5 Apr 2008 by Richard Haselgrove Post: Could anyone reading this thread comment on how/when BOINC updates its time stats, please? When I performed the clock forward / clock back experiment that started this thread, one of the side effects that I noticed was a drop in time metrics: <active_frac>0.045279</active_frac> Six days later, with 24/7 BOINC running (v5.10.45 service install under Windows XP - neither BOINC nor the computer have been restarted), the active_frac remains exactly the same to six decimal places - compare the code above, which is a current paste from client_state, with the figure in my message 16142. ??? I would have expected a similar sort of trap-door function as TDCF - in this case, quick to fall and slow to rise, but no recovery at all? The frac is so low that even on this medium-speed machine (2.0GHz P4), new Einstein tasks at 50% share go immediately into high priority. If, as I suspect, there's no automatic recovery mechanism (or a broken mechanism) for active_frac, that might explain why so many people report problems with high priority and cache sizes on the various project message boards. Please, no sticking-plaster replies: I know what to change and how to change it, but I'm researching whether there's a need to put in another bug report] |
|
4790)
Message boards :
BOINC Manager :
My Wish List - 2
(Message 16435)
Posted 4 Apr 2008 by Richard Haselgrove Post: Wishlist request: "Forward" button for private message in inbox, as in all good email programs. |
|
4791)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16433)
Posted 4 Apr 2008 by Richard Haselgrove Post: OK, there's 2 ways to check a date: There's a third way: 3) a) Double-click on clock in system tray. b) Fiddle with clock. c) Click 'cancel'. *************** I've just seen your lastest post. Andy's friend wouldn't have been running BOINC on the POS, because it has no internet connection. For the same reason, he/she wouldn't have been able to Google for any of the proper tools. I agree with everything that's been said about perpetual calendars. However, everyone who posts here is by definition a Nerd or a Geek, and we understand about things like system integrity. Microsoft, on the other hand, has spent 12 years (since the release of Windows 95) providing end users with a little facility which looks and feels like a perpetual calendar, and which is always guaranteed to be visible onscreen and one doubleclick away from use (unless you're one of those people who hide the taskbar). Think back twelve years: would you have Googled for a perpetual calendar then? With all the overhead of establishing the dial-up internet connection first? It's no wonder that people have got into an engrained habit of (ab)using the system clock for date look-ups. And because it's there, and because it's a habit, people will go on using it: and it will go on being a problem until the last copy of Windows XP is consigned to the great bit-bucket in the sky. BOINC just has to live in the real world. Having a robust, independent, self-validating, self-correcting internal time reference for BOINC is obviously the way forward. But my betting is that that isn't going to be in place this year, for all the reasons that people who've got knowledge of the internal code of BOINC have explained already. In the meantime, can I remind you yet again that there is something that causes an indefinite hang between Suspending computation - running CPU benchmarks and [benchmark_debug] Starting floating-point benchmark Isn't that worth solving? |
|
4792)
Message boards :
BOINC Manager :
Possible Boinc Conflict with Skype (Windows XP)
(Message 16388)
Posted 3 Apr 2008 by Richard Haselgrove Post: Recently I've been experiencing computer crashes with Skype (blue screen of death). I'm running Windows XP professional on a 3 GHz machine with 2GB main memory and lots of disk space. I'd be surprised if this is the case, since the BOINC Online Help system uses Skype - I would expect that one of the BOINC volunteer helpers would have noticed and reported this by now if it was a generic problem. Also, I don't know of a BOINC module which would commit 'almost all' of a 2GB main memory. You couldn't mean on of the BOINC project scientific applications, could you? Did you happen to make a note of the name - it would help identify which project might be involved, and their message boards might be a good place to start looking for fellow-sufferers. In the meantime, as a temporary work-round, you could lower the percentage of the computer's memory BOINC is allowed to use while it thinks the computer is not in use - 75% (1.5GB for BOINC, 512MB for Windows and Skype) might be a starting point. |
|
4793)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16381)
Posted 3 Apr 2008 by Richard Haselgrove Post: Another selling point for BOINC! Keeps your computer running smoothly and safely. (because, IIRC, Windows update fails if the local computer clock is sufficiently different from the Microsoft servers' estimation of time). |
|
4794)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16378)
Posted 3 Apr 2008 by Richard Haselgrove Post: .... and because NTP servers are way more accurate than project servers. Which raises another question: why would anyone run a BOINC server which isn't set up for automatic synchronisation with an NTP server? Matt Lebofsky at SETI occasionally forgets when he's setting up a new web server made up of cannibalised/donated parts, and it soon becomes obvious - the latest post on a message board is some time in the future (SETI has separate servers for the database and the web front end). But he always corrects it as soon as he notices or I point it out ;-) |
|
4795)
Message boards :
Web interfaces :
Error on Forum search, if not logged in
(Message 16374)
Posted 3 Apr 2008 by Richard Haselgrove Post: If anyone has a look at this, could they have a go at trac [trac]#557[/trac] while they're at it, please? That bug also relates to setHighlightTerms(), and hasn't been acknowledged by the project developers in over a month (or fixed, the last time I checked). |
|
4796)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16282)
Posted 1 Apr 2008 by Richard Haselgrove Post: Guys, guys, .... There are going to be problems with time recording for as long as we have clocks and different timezones. We can argue till the cows come home (and what time is that? Do cows know about DST?) whether it is the duty of BOINC and other software to ignore, correct, notify or work round such errors as it finds. But while - or preferably before - we discuss all these arcana, there's a bug to be fixed. It seems that, under certain circumstances as documented in my trac ticket, BOINC just STOPS. Period. That isn't on my list of acceptable responses to a time glitch. It appeares that there's a code sequence along the lines of IF <weird time detected> SUSPEND computing for benchmarking WAIT [color=red]<for something that isn't going to happen>[/color] START <floating point benchmark> I'm wondering if this is something that was introduced round about v5.8.16: - core client: if benchmark time is in the future (due to user tweak) always run benchmarks - if it came in with a fix like that, it might explain why it was missed in pre-release testing. Anyone got a better idea? |
|
4797)
Message boards :
BOINC client :
(temporarily) Solving the LHC/BOINC crashing problem.
(Message 16281)
Posted 1 Apr 2008 by Richard Haselgrove Post: I removed the sections in client_state.xml that Ageless suggests (<project>...</project> and <active_task>...</active_task>), so there were no references to LHC in the file. BOINC started just fine: I have other projects on the box, so I wanted to keep networking enabled. However, I hadn't read JM7's comment about the 'account_' file, so BOINC tried to reconnect to LHC with a 'project initialisation' request. I suspended the LHC Project which had reappeared in the projects list, but BOINC kept sending the initialistation requests (BUG? v5.10.13, as usual) and evenually crashed. I closed it down, removed the (largely empty) <project>...</project> which had reappeared in client_state, and parked the account_ file in a handy folder out of the way. This time, when I restarted BOINC, all seemed to work normally. I found two files in the BOINC folder, "master_lhcathome.cern.ch_lhcathome.xml" and "sched_reply_lhcathome.cern.ch_lhcathome.xml", both datestamped at the time BOINC was trying to re-initialise the project. Both of them appear to be copies of a recent front page of the LHC website. I would expect that for 'master_', but 'sched_reply_'?????? |
|
4798)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16142)
Posted 30 Mar 2008 by Richard Haselgrove Post:
Also noted. Since performing self-sacrifice in the name of science, I'm getting <active_frac>0.045279</active_frac> This machine runs 24/7/365. |
|
4799)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16139)
Posted 30 Mar 2008 by Richard Haselgrove Post: Just offered a temporary solution. You can do what you please with it. That's fine. We solved the problem brought to us by the original SETI user. The logs I posted from my own machine were self-inflcted in the interests of research. As an amateur programmer myself, I know all the mantras which programmers utter when a bug report comes in, and I've used most of them myself. "Is it reproducible?" "What did the message actually say?" "Are you using the latest version?" "What were you doing at the time?" "Is your antivirus up-to-date?" "Have you applied the latest service pack?" "Oh no, our program could never do that." (often false) "Is your video/printer driver up-to-date?" "Why on earth do you want to do that?" etc. etc. I just think that it's grown-up and responsible to try to work through as many as possible of them before I open a trac ticket. |
|
4800)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16137)
Posted 30 Mar 2008 by Richard Haselgrove Post: Use a utility like LClock (LonghornClock) and you have a clock in place of the regular with a calendar function and sans risk of changing the real clock inadvertantly. It's an additional menu option to get thru or 1 Click, Calendar, double clock, the time setting applet. Been using it now on XP for a few years. Yes, yes, yes, but....... Can you force every Windows XP user to install a read-only clock/calendar? (At least this problem will go away when Vista is universal on the desktop). And can you train every Windows XP user to cancel out of the existing clock, every time they check a date? (When I hit this problem in the real world, with a telesales database, the problem went away when we upgraded from Windows 98 to domain-controlled Windows 2000 for the sales floor, and gave the users restricted rights. But the boss's orders were still all over the place, because his logon had to have administrative rights and could change the clock. But I digress). The BUG is that benchmarking starts, but doesn't complete. There should be no way that that can happen, period. It's called fault-tolerance, and it should apply even when the 'fault' is a naïve user. |
|
4801)
Message boards :
BOINC client :
Benchmarking bug - indefinite suspension of computing
(Message 16133)
Posted 30 Mar 2008 by Richard Haselgrove Post: Just come across this as a result of a problem-solving session at SETI - seems to be reproducible in current (v5.10.45) version for Windows. Scenario: BOINC running as a service on Windows XP. Do that typical end-user thing of using the system clock as a holiday planner (checking a date next month). Inadvertently click 'OK' instead of 'cancel' - sets the clock a month ahead. Some time later, you (or Internet Time) notice that the clock is wrong, and move it back to the correct month. BOINC computation stops with an endless benchmark shortly after the second time change. {Edit - opened trac ticket [trac]#588[/trac]}. You'll get a message log something like this: 2008-03-30 12:28:49 [Einstein@Home] Resuming task h1_0907.30_S5R3__166_S5R3b_1 using einstein_S5R3 version 436 2008-04-30 14:09:40 [---] Running CPU benchmarks 2008-04-30 14:09:40 [---] Suspending computation - running CPU benchmarks 2008-04-30 14:09:42 [---] [benchmark_debug] Starting floating-point benchmark 2008-04-30 14:09:52 [---] [benchmark_debug] Ended floating-point benchmark 2008-04-30 14:09:57 [---] [benchmark_debug] Starting integer benchmark 2008-04-30 14:10:07 [---] [benchmark_debug] Ended integer benchmark 2008-04-30 14:10:10 [---] [benchmark_debug] Ended benchmark 2008-04-30 14:10:11 [---] [benchmark_debug] CPU 0 has finished 2008-04-30 14:10:11 [---] [benchmark_debug] 1 out of 1 CPUs done 2008-04-30 14:10:11 [---] [benchmark_debug] CPU 0: fp 1038127090.301003 int 1675825412.162456 intloops 27696000.000000 inttime 9.406250 2008-04-30 14:10:11 [---] Benchmark results: 2008-04-30 14:10:11 [---] Number of CPUs: 1 2008-04-30 14:10:11 [---] 1038 floating point MIPS (Whetstone) per CPU 2008-04-30 14:10:11 [---] 1676 integer MIPS (Dhrystone) per CPU 2008-04-30 14:10:12 [---] Resuming computation 2008-03-30 14:13:41 [---] Running CPU benchmarks 2008-03-30 14:13:41 [---] Suspending computation - running CPU benchmarks 2008-03-30 14:17:24 [---] Exit requested by user To pause/resume tasks hit CTRL-C, to exit hit CTRL-BREAK StartServiceCtrlDispatcher being called. This may take several seconds. Please wait. 2008-03-30 14:17:26 [---] Starting BOINC client version 5.10.13 for windows_intelx86 2008-03-30 14:17:26 [---] log flags: task, file_xfer, sched_ops, benchmark_debug 2008-03-30 14:17:26 [---] Libraries: libcurl/7.16.1 OpenSSL/0.9.8e zlib/1.2.3 2008-03-30 14:17:26 [---] Executing as a daemon 2008-03-30 14:17:26 [---] Data directory: C:Program FilesBOINC 2008-03-30 14:17:26 [---] BOINC is running as a service and as a non-system user. 2008-03-30 14:17:26 [---] No application graphics will be available. 2008-03-30 14:17:27 [Einstein@Home] Found app_info.xml; using anonymous platform 2008-03-30 14:17:27 [SETI@home] Found app_info.xml; using anonymous platform 2008-03-30 14:17:27 [---] Processor: 1 GenuineIntel Intel(R) Pentium(R) 4 CPU 2.00GHz [x86 Family 15 Model 2 Stepping 4] 2008-03-30 14:17:27 [---] Processor features: fpu tsc sse sse2 mmx 2008-03-30 14:17:27 [---] Memory: 511.30 MB physical, 1.22 GB virtual 2008-03-30 14:17:27 [---] Disk: 37.24 GB total, 4.92 GB free 2008-03-30 14:17:27 [Einstein@Home] URL: http://einstein.phys.uwm.edu/; Computer ID: 1036916; location: home; project prefs: default 2008-03-30 14:17:27 [SETI@home] URL: http://setiathome.berkeley.edu/; Computer ID: 1791152; location: work; project prefs: work 2008-03-30 14:17:27 [---] General prefs: from Einstein@Home (last modified 2007-12-07 10:01:47) 2008-03-30 14:17:27 [---] Host location: home 2008-03-30 14:17:27 [---] General prefs: using separate prefs for home 2008-03-30 14:17:27 [---] Preferences limit memory usage when active to 511.30MB 2008-03-30 14:17:27 [---] Preferences limit memory usage when idle to 511.30MB 2008-03-30 14:17:27 [---] Preferences limit disk usage to 4.92GB 2008-03-30 14:17:27 [---] Running CPU benchmarks 2008-03-30 14:17:30 [---] [benchmark_debug] Starting floating-point benchmark 2008-03-30 14:17:40 [---] [benchmark_debug] Ended floating-point benchmark 2008-03-30 14:17:46 [---] [benchmark_debug] Starting integer benchmark 2008-03-30 14:17:55 [---] [benchmark_debug] Ended integer benchmark 2008-03-30 14:17:59 [---] [benchmark_debug] Ended benchmark 2008-03-30 14:18:01 [---] [benchmark_debug] CPU 0 has finished 2008-03-30 14:18:01 [---] [benchmark_debug] 1 out of 1 CPUs done 2008-03-30 14:18:01 [---] [benchmark_debug] CPU 0: fp 1044264943.457189 int 1983678791.328194 intloops 29952000.000000 inttime 8.593750 2008-03-30 14:18:01 [---] Benchmark results: 2008-03-30 14:18:01 [---] Number of CPUs: 1 2008-03-30 14:18:01 [---] 1044 floating point MIPS (Whetstone) per CPU 2008-03-30 14:18:01 [---] 1984 integer MIPS (Dhrystone) per CPU 2008-03-30 14:18:09 [Einstein@Home] Restarting task h1_0907.30_S5R3__166_S5R3b_1 using einstein_S5R3 version 436 2008-03-30 14:21:45 [---] Exit requested by user StartServiceCtrlDispatcher being called. This may take several seconds. Please wait. 30-Mar-2008 14:22:49 [---] Starting BOINC client version 5.10.45 for windows_intelx86 30-Mar-2008 14:22:49 [---] log flags: task, file_xfer, sched_ops, benchmark_debug 30-Mar-2008 14:22:49 [---] Libraries: libcurl/7.18.0 OpenSSL/0.9.8e zlib/1.2.3 30-Mar-2008 14:22:49 [---] Executing as a daemon 30-Mar-2008 14:22:49 [---] Data directory: C:Program FilesBOINC 30-Mar-2008 14:22:49 [---] BOINC is running as a service and as a non-system user. 30-Mar-2008 14:22:49 [---] No application graphics will be available. 30-Mar-2008 14:22:49 [Einstein@Home] Found app_info.xml; using anonymous platform 30-Mar-2008 14:22:49 [SETI@home] Found app_info.xml; using anonymous platform 30-Mar-2008 14:22:49 [---] Processor: 1 GenuineIntel Intel(R) Pentium(R) 4 CPU 2.00GHz [x86 Family 15 Model 2 Stepping 4] 30-Mar-2008 14:22:49 [---] Processor features: fpu tsc sse sse2 mmx 30-Mar-2008 14:22:49 [---] OS: Microsoft Windows XP: Home Edition, Service Pack 2, (05.01.2600.00) 30-Mar-2008 14:22:49 [---] Memory: 511.30 MB physical, 1.22 GB virtual 30-Mar-2008 14:22:49 [---] Disk: 37.24 GB total, 4.83 GB free 30-Mar-2008 14:22:49 [---] Local time is UTC +1 hours 30-Mar-2008 14:22:49 [---] Version change (5.10.13 -> 5.10.45) 30-Mar-2008 14:22:49 [Einstein@Home] URL: http://einstein.phys.uwm.edu/; Computer ID: 1036916; location: home; project prefs: default 30-Mar-2008 14:22:49 [SETI@home] URL: http://setiathome.berkeley.edu/; Computer ID: 1791152; location: work; project prefs: work 30-Mar-2008 14:22:49 [---] General prefs: from Einstein@Home (last modified 07-Dec-2007 10:01:47) 30-Mar-2008 14:22:50 [---] Host location: home 30-Mar-2008 14:22:50 [---] General prefs: using separate prefs for home 30-Mar-2008 14:22:50 [---] Preferences limit memory usage when active to 511.30MB 30-Mar-2008 14:22:50 [---] Preferences limit memory usage when idle to 511.30MB 30-Mar-2008 14:22:50 [---] Preferences limit disk usage to 4.83GB 30-Mar-2008 14:22:50 [---] Running CPU benchmarks 30-Mar-2008 14:22:53 [---] [benchmark_debug] Starting floating-point benchmark 30-Mar-2008 14:23:02 [---] [benchmark_debug] Ended floating-point benchmark 30-Mar-2008 14:23:07 [---] [benchmark_debug] Starting integer benchmark 30-Mar-2008 14:23:17 [---] [benchmark_debug] Ended integer benchmark 30-Mar-2008 14:23:21 [---] [benchmark_debug] Ended benchmark 30-Mar-2008 14:23:23 [---] [benchmark_debug] CPU 0 has finished 30-Mar-2008 14:23:23 [---] [benchmark_debug] 1 out of 1 CPUs done 30-Mar-2008 14:23:23 [---] [benchmark_debug] CPU 0: fp 1030769230.769231 int 1932793606.913454 intloops 31200000.000000 inttime 9.187500 30-Mar-2008 14:23:23 [---] Benchmark results: 30-Mar-2008 14:23:23 [---] Number of CPUs: 1 30-Mar-2008 14:23:23 [---] 1031 floating point MIPS (Whetstone) per CPU 30-Mar-2008 14:23:23 [---] 1933 integer MIPS (Dhrystone) per CPU 30-Mar-2008 14:23:32 [Einstein@Home] Restarting task h1_0907.30_S5R3__166_S5R3b_1 using einstein_S5R3 version 436 30-Apr-2008 14:24:32 [---] Running CPU benchmarks 30-Apr-2008 14:24:32 [---] Suspending computation - running CPU benchmarks 30-Apr-2008 14:24:35 [---] [benchmark_debug] Starting floating-point benchmark 30-Apr-2008 14:25:08 [---] [benchmark_debug] Ended floating-point benchmark 30-Apr-2008 14:25:10 [---] [benchmark_debug] Starting integer benchmark 30-Apr-2008 14:25:12 [---] [benchmark_debug] Ended integer benchmark 30-Apr-2008 14:25:14 [---] [benchmark_debug] Ended benchmark 30-Apr-2008 14:25:16 [---] [benchmark_debug] CPU 0 has finished 30-Apr-2008 14:25:16 [---] [benchmark_debug] 1 out of 1 CPUs done 30-Apr-2008 14:25:16 [---] [benchmark_debug] CPU 0: fp 1032357177.148344 int 1912681740.570187 intloops 5776000.000000 inttime 1.718750 30-Apr-2008 14:25:16 [---] Benchmark results: 30-Apr-2008 14:25:16 [---] Number of CPUs: 1 30-Apr-2008 14:25:16 [---] 1032 floating point MIPS (Whetstone) per CPU 30-Apr-2008 14:25:16 [---] 1913 integer MIPS (Dhrystone) per CPU 30-Apr-2008 14:25:17 [---] Resuming computation 30-Mar-2008 14:27:00 [---] Running CPU benchmarks 30-Mar-2008 14:27:00 [---] Suspending computation - running CPU benchmarks 30-Mar-2008 14:30:04 [---] Exit requested by user StartServiceCtrlDispatcher being called. This may take several seconds. Please wait. 30-Mar-2008 14:30:06 [---] Starting BOINC client version 5.10.45 for windows_intelx86 30-Mar-2008 14:30:06 [---] log flags: task, file_xfer, sched_ops, benchmark_debug 30-Mar-2008 14:30:06 [---] Libraries: libcurl/7.18.0 OpenSSL/0.9.8e zlib/1.2.3 30-Mar-2008 14:30:06 [---] Executing as a daemon 30-Mar-2008 14:30:06 [---] Data directory: C:Program FilesBOINC 30-Mar-2008 14:30:06 [---] BOINC is running as a service and as a non-system user. 30-Mar-2008 14:30:06 [---] No application graphics will be available. 30-Mar-2008 14:30:06 [Einstein@Home] Found app_info.xml; using anonymous platform 30-Mar-2008 14:30:06 [SETI@home] Found app_info.xml; using anonymous platform 30-Mar-2008 14:30:07 [---] Processor: 1 GenuineIntel Intel(R) Pentium(R) 4 CPU 2.00GHz [x86 Family 15 Model 2 Stepping 4] 30-Mar-2008 14:30:07 [---] Processor features: fpu tsc sse sse2 mmx 30-Mar-2008 14:30:07 [---] OS: Microsoft Windows XP: Home Edition, Service Pack 2, (05.01.2600.00) 30-Mar-2008 14:30:07 [---] Memory: 511.30 MB physical, 1.22 GB virtual 30-Mar-2008 14:30:07 [---] Disk: 37.24 GB total, 4.84 GB free 30-Mar-2008 14:30:07 [---] Local time is UTC +1 hours 30-Mar-2008 14:30:07 [Einstein@Home] URL: http://einstein.phys.uwm.edu/; Computer ID: 1036916; location: home; project prefs: default 30-Mar-2008 14:30:07 [SETI@home] URL: http://setiathome.berkeley.edu/; Computer ID: 1791152; location: work; project prefs: work 30-Mar-2008 14:30:07 [---] General prefs: from Einstein@Home (last modified 07-Dec-2007 10:01:47) 30-Mar-2008 14:30:07 [---] Host location: home 30-Mar-2008 14:30:07 [---] General prefs: using separate prefs for home 30-Mar-2008 14:30:07 [---] Preferences limit memory usage when active to 511.30MB 30-Mar-2008 14:30:07 [---] Preferences limit memory usage when idle to 511.30MB 30-Mar-2008 14:30:07 [---] Preferences limit disk usage to 4.84GB 30-Mar-2008 14:30:07 [---] Running CPU benchmarks 30-Mar-2008 14:30:10 [---] [benchmark_debug] Starting floating-point benchmark 30-Mar-2008 14:30:20 [---] [benchmark_debug] Ended floating-point benchmark 30-Mar-2008 14:30:26 [---] [benchmark_debug] Starting integer benchmark 30-Mar-2008 14:30:36 [---] [benchmark_debug] Ended integer benchmark 30-Mar-2008 14:30:38 [---] [benchmark_debug] Ended benchmark 30-Mar-2008 14:30:40 [---] [benchmark_debug] CPU 0 has finished 30-Mar-2008 14:30:40 [---] [benchmark_debug] 1 out of 1 CPUs done 30-Mar-2008 14:30:40 [---] [benchmark_debug] CPU 0: fp 1029145728.643216 int 1641931187.505236 intloops 26640000.000000 inttime 9.234375 30-Mar-2008 14:30:40 [---] Benchmark results: 30-Mar-2008 14:30:40 [---] Number of CPUs: 1 30-Mar-2008 14:30:40 [---] 1029 floating point MIPS (Whetstone) per CPU 30-Mar-2008 14:30:40 [---] 1642 integer MIPS (Dhrystone) per CPU 30-Mar-2008 14:30:43 [Einstein@Home] Restarting task h1_0907.30_S5R3__166_S5R3b_1 using einstein_S5R3 version 436 |
|
4802)
Message boards :
BOINC client :
Dual core
(Message 16003)
Posted 19 Mar 2008 by Richard Haselgrove Post: Cross-referencing to SETI thread boinc resource share, where I've posted some data-logged graphs of BOINC STD and (indirectly) task switching. Doesn't immediately answer the opening question about Work Fetch (although my data logger tracks LTD as well), but it certainly demonstrates that the current BOINC isn't as efficient as it could be on a multicore. |
|
4803)
Message boards :
Web interfaces :
Cached project forum pages?
(Message 15891)
Posted 14 Mar 2008 by Richard Haselgrove Post: Additional information from the SETI thread: SETI has: <!-- SVN VERSIONS --> <!-- $Id: translation.inc 14562 2008-01-15 22:22:08Z davea $ --> <!-- $Id: pm.inc 14019 2007-11-01 23:04:39Z davea $ --> <!-- $Id: team.inc 14394 2007-12-18 20:28:08Z davea $ --> <!-- $Id: text_transform.inc 14264 2007-11-20 00:55:35Z davea $ --> <!-- $Id: stats_sites.inc 14562 2008-01-15 22:22:08Z davea $ --> <!-- $Id: user.inc 14626 2008-01-28 16:12:28Z boincadm $ --> <!-- $Id: profile.inc 14562 2008-01-15 22:22:08Z davea $ --> <!-- $Id: util.inc 14747 2008-02-15 15:25:44Z davea $ --> sending HTTP/1.x 200 OK Date: Fri, 14 Mar 2008 07:34:45 GMT Server: Apache/2.2.6 (Fedora) X-Powered-By: PHP/5.2.4 Connection: close Transfer-Encoding: chunked Content-Type: text/html; charset=UTF-8 CPDN Beta has: <!-- SVN VERSIONS --> <!-- $Id: translation.inc 14562 2008-01-15 22:22:08Z davea $ --> <!-- $Id: pm.inc 14019 2007-11-01 23:04:39Z davea $ --> <!-- $Id: team.inc 14394 2007-12-18 20:28:08Z davea $ --> <!-- $Id: text_transform.inc 14264 2007-11-20 00:55:35Z davea $ --> <!-- $Id: stats_sites.inc 14562 2008-01-15 22:22:08Z davea $ --> <!-- $Id: user.inc 14546 2008-01-14 16:32:34Z davea $ --> <!-- $Id: profile.inc 14562 2008-01-15 22:22:08Z davea $ --> <!-- $Id: util.inc 14474 2008-01-06 18:09:23Z boincadm $ --> sending HTTP/1.x 200 OK Date: Fri, 14 Mar 2008 12:51:31 GMT Server: Apache/2.2.6 (Fedora) X-Powered-By: PHP/5.2.4 Expires: Thu, 19 Nov 1981 08:52:00 GMT Cache-Control: no-store, no-cache, must-revalidate, post-check=0, pre-check=0 Pragma: no-cache Connection: close Transfer-Encoding: chunked Content-Type: text/html; charset=UTF-8 The only SVN changes are user.inc (SETI's is 14 days newer) and util.inc (40 days newer). |
|
4804)
Message boards :
Web interfaces :
Cached project forum pages?
(Message 15865)
Posted 13 Mar 2008 by Richard Haselgrove Post: Forum pages are never cached. I have noticed this problem with unread messages, but it can't be a cache problem... Well, anything which can be cleared by pressing F5 feels like a cached page to me. Are you thinking server-side cache, or client-side (browser) cache? |
|
4805)
Message boards :
BOINC client :
Dual core
(Message 15864)
Posted 13 Mar 2008 by Richard Haselgrove Post: Are u using the 'Connect every X days' to buffer or the 'Additional Buffer X Days' aka 'Cache' option? Settings are 0.01 'Connect every', 1.0 days 'Additional Cache'. |
|
4806)
Message boards :
BOINC client :
Dual core
(Message 15862)
Posted 13 Mar 2008 by Richard Haselgrove Post: Well, much to my surprise, I woke up this morning to find I had a cache of SETI Beta work to keep my second allocated core working. Well, not quite a second allocated core. Apart from the 'upgrade' to BOINC v5.10.45, there has been one other change in the experimental conditions - LHC has started sending out lots of work, so the allocation is now 2/9 (22.22%), instead of 2/8 (25%). So this is what I saw this morning:   (direct link) ................... (direct link) That's SEVEN SETI Beta tasks running at the same time, including the Astropulse one at the top. Then they task-switched, and now I'm on SIX SETI main, and two CPDN. I didn't get the debt values when they downloaded (I was asleep at the time), but when they switched out after an hour's run, BETA was at STD -11,650 LTD +645,133: next highest LTD on the system is LHC at +299,351 The log is curious. I ran dry on the second core at 05:33:13, but it didn't fetch any work for three hours - in fact, it actually reported six tasks (at 4.5 hours each, that's a days work) and requested nothing. Then it started requesting 1.5 hours work, and receiving 4.5 hours work, again and again and again. Here's the filtered log: SETI@home Beta Test 13/03/2008 05:33:13 Computation for task 27mr07ag.15695.11524.3.10.182_2 finished SETI@home Beta Test 13/03/2008 05:33:15 Started upload of 27mr07ag.15695.11524.3.10.182_2_0 SETI@home Beta Test 13/03/2008 05:33:22 Finished upload of 27mr07ag.15695.11524.3.10.182_2_0 SETI@home Beta Test 13/03/2008 05:51:31 Sending scheduler request: To report completed tasks. Requesting 0 seconds of work, reporting 6 completed tasks SETI@home Beta Test 13/03/2008 05:51:37 Scheduler request succeeded: got 0 new tasks SETI@home Beta Test 13/03/2008 08:36:21 Sending scheduler request: To fetch work. Requesting 7502 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:36:26 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:36:28 Started download of 27mr07ag.15695.15205.3.10.180 SETI@home Beta Test 13/03/2008 08:36:34 Finished download of 27mr07ag.15695.15205.3.10.180 SETI@home Beta Test 13/03/2008 08:41:05 Sending scheduler request: To fetch work. Requesting 4401 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:41:10 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:41:12 Started download of 27mr07ag.15695.15205.3.10.168 SETI@home Beta Test 13/03/2008 08:41:18 Finished download of 27mr07ag.15695.15205.3.10.168 SETI@home Beta Test 13/03/2008 08:41:26 Sending scheduler request: To fetch work. Requesting 5338 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:41:31 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:41:33 Started download of 27mr07ag.15695.15205.3.10.182 SETI@home Beta Test 13/03/2008 08:41:39 Finished download of 27mr07ag.15695.15205.3.10.182 SETI@home Beta Test 13/03/2008 08:41:46 Sending scheduler request: To fetch work. Requesting 5394 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:41:51 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:41:53 Started download of 27mr07ag.15695.15205.3.10.137 SETI@home Beta Test 13/03/2008 08:41:59 Finished download of 27mr07ag.15695.15205.3.10.137 SETI@home Beta Test 13/03/2008 08:42:07 Sending scheduler request: To fetch work. Requesting 5450 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:42:12 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:42:14 Started download of 27mr07ag.15695.15205.3.10.193 SETI@home Beta Test 13/03/2008 08:42:20 Finished download of 27mr07ag.15695.15205.3.10.193 SETI@home Beta Test 13/03/2008 08:42:27 Sending scheduler request: To fetch work. Requesting 5510 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:42:32 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:42:34 Started download of 27mr07ag.15695.15205.3.10.96 SETI@home Beta Test 13/03/2008 08:42:40 Finished download of 27mr07ag.15695.15205.3.10.96 SETI@home Beta Test 13/03/2008 08:42:47 Sending scheduler request: To fetch work. Requesting 5565 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:42:52 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:42:55 Started download of 27mr07ag.15695.15205.3.10.176 SETI@home Beta Test 13/03/2008 08:43:01 Finished download of 27mr07ag.15695.15205.3.10.176 SETI@home Beta Test 13/03/2008 08:43:08 Sending scheduler request: To fetch work. Requesting 5619 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:43:13 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:43:15 Started download of 27mr07ag.15695.15205.3.10.147 SETI@home Beta Test 13/03/2008 08:43:21 Finished download of 27mr07ag.15695.15205.3.10.147 SETI@home Beta Test 13/03/2008 08:43:28 Sending scheduler request: To fetch work. Requesting 2093 seconds of work, reporting 0 completed tasks SETI@home Beta Test 13/03/2008 08:43:33 Scheduler request succeeded: got 1 new tasks SETI@home Beta Test 13/03/2008 08:43:35 Started download of 27mr07ag.15695.15205.3.10.190 SETI@home Beta Test 13/03/2008 08:43:41 Finished download of 27mr07ag.15695.15205.3.10.190 SETI@home Beta Test 13/03/2008 08:45:14 Starting 27mr07ag.15695.15205.3.10.180_2 SETI@home Beta Test 13/03/2008 08:45:14 Starting task 27mr07ag.15695.15205.3.10.180_2 using setiathome_enhanced version 600 SETI@home Beta Test 13/03/2008 09:24:37 Starting 27mr07ag.15695.15205.3.10.168_2 SETI@home Beta Test 13/03/2008 09:24:37 Starting task 27mr07ag.15695.15205.3.10.168_2 using setiathome_enhanced version 600 SETI@home Beta Test 13/03/2008 09:25:27 Starting 27mr07ag.15695.15205.3.10.182_1 SETI@home Beta Test 13/03/2008 09:25:27 Starting task 27mr07ag.15695.15205.3.10.182_1 using setiathome_enhanced version 600 SETI@home Beta Test 13/03/2008 09:25:33 Starting 27mr07ag.15695.15205.3.10.137_2 SETI@home Beta Test 13/03/2008 09:25:33 Starting task 27mr07ag.15695.15205.3.10.137_2 using setiathome_enhanced version 600 SETI@home Beta Test 13/03/2008 09:26:23 Starting 27mr07ag.15695.15205.3.10.193_2 SETI@home Beta Test 13/03/2008 09:26:23 Starting task 27mr07ag.15695.15205.3.10.193_2 using setiathome_enhanced version 600 SETI@home Beta Test 13/03/2008 09:28:12 Starting 27mr07ag.15695.15205.3.10.96_2 SETI@home Beta Test 13/03/2008 09:28:12 Starting task 27mr07ag.15695.15205.3.10.96_2 using setiathome_enhanced version 600 |
|
4807)
Message boards :
Web interfaces :
Cached project forum pages?
(Message 15842)
Posted 12 Mar 2008 by Richard Haselgrove Post: Since maybe a month or so ago, as some projects open up or others upgrade their server SW, I'm very often noticing one annoying problem - I see a list of threads, some of them marked unread, then open few one after another, read through, go back and forth to another ones... After reading 5-6 threads (and loosing an overview, which ones), I can still see them marked as unread and when opening them, the same (previously unread) posts are still marked as such. Just explicitly refreshing any page 'repairs' the particular page and the unread marks disappear. Quite an active discussion about this at SETI: Yellow pages still here! (someone coined the term "Yellow Pages" for the yellow 'unread' marker). The general consensus seems to be that people would prefer the old (uncached) version, where the page is re-loaded (and hence a new, updated set of yellow pages shown) when the browser 'back' button is clicked. It's cross-platform, and cross-browser, so it's clearly a change in the way the pages are being sent out by the new versions of the forum software. Edit - the change has been tracked down to <!-- $Id: util.inc 14747 2008-02-15 15:25:44Z davea $ --> |
|
4808)
Message boards :
BOINC client :
Dual core
(Message 15806)
Posted 11 Mar 2008 by Richard Haselgrove Post: I'd been wondering whether CPDN had any sort of 'special case' treatment in the work fetch algorithm. I have a second machine, a quad core, which had got itself into a very similar situation. I ran it 100% on Einstein for a while, as a timing test for a new Beta application: other projects were flushed down to zero tasks remaining, but CPDN had a single part-crunched task in suspended animation so LTD grew to a massive value. Then, when I opened up the machine to other projects at the end of the test, Einstein was inhibited from fetching work, the single CPDN task (controlled by NNT) wasn't enough to use up the resource share, and STD grew as well. I ended up with CPDN in a very similar state to the SETI Beta saga I described earlier in this thread: STD up against the stop at 86,400: LTD somewhere in the 400,000s: one task in progress: additional cores available. But this time, when I released NNT, BOINC initiated a CPDN work fetch within a few seconds. I'm wondering whether there's anything in the work fetch calculation which considers the size of the running task against the time remaining before deadline. In CPDN's case, those figures are about 96 days to complete a full task, against 30 months deadline. So there is plenty of time available to complete two tasks, even if for some reason they have follow each other on a single core. But for the SETI Beta/Astropulse example, the figures are 268 hours estimate/134 hours actual computation time (RDCF permanently messed up by the other SETI application), against 2 weeks deadline. So in the initial stages at least, there would not be enough time to slot in a second task on the same core as the first (after it completes), and still meet the deadline for that second task. This may be the point at which BOINC fails to take the availability of a second core into account: if the calculation was done on the basis that any hypothetical new task wouldn't start until the current one has finished, that would account for the behaviour I've observed. The machine is in a state of transition at the moment. The long-running Beta task has just finished, and I have about a day's worth of short-running tasks to work through, as per my cache preferences. I also have one brand-new long-running task, with initial timings so tight that it spent a couple of brief periods running at high priority. When the short tasks are exhausted, I'll watch to see if it fetches any more, or whether I'm in for several more days running under share allocation, watching STD bump against the stop and LTD climbing inexorably. |
|
4809)
Message boards :
BOINC Manager :
Column sorting, anyone?
(Message 15770)
Posted 10 Mar 2008 by Richard Haselgrove Post: Actually there is a trick without registry meddling. It's by hitting 1 particular column and restores the order to the sequence in which the work arrived.... and just tested it on a 5.10.42 client and it still does it. You end up with the running jobs on top, most progressed highest, but that's where I like to have them, so think to have hinted at the one to sort on. Do you mean screenshot 3 or screenshot 4? They can't both be right - was lhcathome the first or the last of the unstarted tasks to have arrived? Anyway, the challenge was to return to the original screenshot - no prizes so far, I'm afraid. And for screenshot 2 - see the edit to my previous post. Bit of a problem there (but seems only to apply if you're in fresh-out-of-the-box, unsorted mode. |
|
4810)
Message boards :
BOINC Manager :
Column sorting, anyone?
(Message 15765)
Posted 10 Mar 2008 by Richard Haselgrove Post: Now that column sorting is an established feature in the v5.10.xx range, is there any official way to unsort a grid - e.g. to return the tasks pane to 'as issued' order, to see which task is going to run next (all other things being equal). When you have a bright, shining, brand new BOINC installation, and you switch it to Advanced Grid view, Tasks pane, there'll be nothing to see - the list will be empty. When you attach to one or more projects, tasks will be issued, and they will start to fill up the pane in the order they are issued. And as you work through various tasks, from various projects, they'll get into a glorious muddle - see  screenshot But because they're in the order they were issued, you can see, for any project, which is the next task to run, or how many tasks you've got to get through before you get to the one from the faulty SETI splitter that you're interested in - whatever. Now play around with column sorting for a while. The challenge is to return to screenshot order, using only BOINC-provided tools, and without losing sight of the pretty progress %age graphics bar. Accessible view is cheating - you lose the graphics. Deadline sorting doesn't do it - I've had everything from 5 years deadline (previous CPDN models) to 15 minutes deadline (Orbit rng test program) on this box - and projects like SETI usually have variable deadlines within the project (this is an unusually boring set). Like I say, I can do it with regedit - see if you can do it without. Edit - look carefully at that screenshot - it reveals a big bug in the v5.10.45 GUI. Notice how the column headers are lined up with the data columns - not. I was prevailed upon to upgrade from v5.10.13 to v5.10.45 over the weekend, to demonstrate that a work fetch bug hadn't been fixed in the interim. It looks as if previously saved values for the column widths have been applied to the data area - that looks much like I had it before - but the column headers are completely different. Could somebody check if they look like an out-of-the-box configuration? OK - so that screenshot didn't show it. This is what I'm seeing now:  screenshot 2 Now how did I do that? Switching from Grid to Accessible view, and back again, I think. Yes, reproducible. Forget all that whinge about upgrading. |
|
4811)
Message boards :
BOINC Manager :
Column sorting, anyone?
(Message 15756)
Posted 10 Mar 2008 by Richard Haselgrove Post: Now that column sorting is an established feature in the v5.10.xx range, is there any official way to unsort a grid - e.g. to return the tasks pane to 'as issued' order, to see which task is going to run next (all other things being equal). In Windows, I can do it by using regedit and deleting the sort keys in [HKEY_CURRENT_USER\Software\Space Sciences Laboratory, U.C. Berkeley\BOINC Manager\TasksGrid] "SortColumn"=dword:00000004 "SortAscending"=dword:00000000 but that's not very good advice to give a n00b on a help desk. I know you can temporarily unsort the data by switching to accessible view, but some people like to have both native ordering and progress bars. Edit - I keep losing backslashes in the registry key |
|
4812)
Message boards :
BOINC client :
Download Problems-Help Please.
(Message 15726)
Posted 9 Mar 2008 by Richard Haselgrove Post: Hi. Have been a member for some years but the last week all downloads have failed. Also a program called KWSN_2.4_SSE3-Core2_MB.exe has been trying to download and fails constantly. Can anyonen tell me what this program is and how I can get back to downloading WU's. I've tried resetting the project and downlading the BOINC prog again. Thanks Ian KWSN_2.4_SSE3-Core2_MB.exe is a third-party optimised application designed for use on the SETI@home project. If it is SETI that is trying to download the file, I suggest that you post this question again on the SETI message boards - it's not really a BOINC problem. It sounds as if an attempt to optimise the SETI computations has gone wrong. The guys at SETI will help you either to get the optimisation working properly, or to return things to the status quo ante. When you post at SETI, it will be helpful if you include some more details about your computer set-up - operating system version, processor type etc. |
|
4813)
Message boards :
BOINC client :
Dual core
(Message 15720)
Posted 8 Mar 2008 by Richard Haselgrove Post: You seem to have hit the highlights. Ack, no short-term guarantees. But this failure to fetch work distorts (unnecessarily, imv) the work-flow pattern for BOINC as a whole, and could in the end be detrimental to production science: doesn't matter on a Beta project, of course. For example: STD is limit-stopped, but LTD continues to grow. So now SETI general is at -10,856 LTD, and fetch-inhibited. Because I was inveigled into 'upgrading' to v5.10.45 seven hours ago, that means that I now have ten completed SETI tasks which are reporting-inhibited as well. When, eventually, SETI Beta starts requesting new work, the debt mountain means that many cores (probably at least six) will start working on the same project at the same time. I don't know if you've been following threads like Peter Stoll's SETI and Einstein Cooperation on a Q6600, but that's a very inefficient way of utilising a multicore. ---------------------- Bed-time in the UK now. I'll leave both debug flags active overnight, and see if we can catch another of the elusive Beta requests in action: but unless you have any specific further requests, I'll turn them off in the morning so I can keep a longer, less detailed log history (and save some CPU cycles for science). |
|
4814)
Message boards :
BOINC client :
Dual core
(Message 15718)
Posted 7 Mar 2008 by Richard Haselgrove Post: Please read this page and create a cc_config.xml file with the following flags turned on: Please look further into the thread: I have a cc_config.xml file, I know how to use it, and I had <work_fetch_debug> turned on most of last night. With <cpu_sched_debug> turned on as well, I'm getting a huge amount of output - what would be most useful to you? Here's a sample - this is repeated every 5 seconds. 07/03/2008 23:00:11||[cpu_sched_debug] Request enforce CPU schedule: Checkpoint reached 07/03/2008 23:00:11||[cpu_sched_debug] enforce_schedule(): start 07/03/2008 23:00:11|SETI@home Beta Test|[cpu_sched_debug] want to run: ap_22mr07ac_B7_P1_00398_20080208_13385.wu_3 07/03/2008 23:00:11|climateprediction.net|[cpu_sched_debug] want to run: hadcm3istd_026w_1920_160_15925025_0 07/03/2008 23:00:11|climateprediction.net|[cpu_sched_debug] want to run: hadcm3istd_0bob_1920_160_05936405_5 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] want to run: 30mr07ac.30604.3079.12.7.174_0 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] want to run: 30mr07aa.11888.2526.12.7.173_1 07/03/2008 23:00:11|Einstein@Home|[cpu_sched_debug] want to run: h1_0836.30_S5R3__316_S5R3b_0 07/03/2008 23:00:11|Einstein@Home|[cpu_sched_debug] want to run: h1_0836.30_S5R3__315_S5R3b_1 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] want to run: 30mr07ac.20541.1085042.6.7.171_3 07/03/2008 23:00:11|SETI@home Beta Test|[cpu_sched_debug] processing ap_22mr07ac_B7_P1_00398_20080208_13385.wu_3 07/03/2008 23:00:11|SETI@home Beta Test|[cpu_sched_debug] ap_22mr07ac_B7_P1_00398_20080208_13385.wu_3 is already running 07/03/2008 23:00:11|climateprediction.net|[cpu_sched_debug] processing hadcm3istd_026w_1920_160_15925025_0 07/03/2008 23:00:11|climateprediction.net|[cpu_sched_debug] hadcm3istd_026w_1920_160_15925025_0 is already running 07/03/2008 23:00:11|climateprediction.net|[cpu_sched_debug] processing hadcm3istd_0bob_1920_160_05936405_5 07/03/2008 23:00:11|climateprediction.net|[cpu_sched_debug] hadcm3istd_0bob_1920_160_05936405_5 is already running 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] processing 30mr07ac.30604.3079.12.7.174_0 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] 30mr07ac.30604.3079.12.7.174_0 is already running 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] processing 30mr07aa.11888.2526.12.7.173_1 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] 30mr07aa.11888.2526.12.7.173_1 is already running 07/03/2008 23:00:11|Einstein@Home|[cpu_sched_debug] processing h1_0836.30_S5R3__316_S5R3b_0 07/03/2008 23:00:11|Einstein@Home|[cpu_sched_debug] h1_0836.30_S5R3__316_S5R3b_0 is already running 07/03/2008 23:00:11|Einstein@Home|[cpu_sched_debug] processing h1_0836.30_S5R3__315_S5R3b_1 07/03/2008 23:00:11|Einstein@Home|[cpu_sched_debug] didn't preempt 30mr07ac.30604.1081361.12.7.33_0: tr 2137.500601 tsc 6.099602 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] processing 30mr07ac.20541.1085042.6.7.171_3 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] 30mr07ac.20541.1085042.6.7.171_3 is already running 07/03/2008 23:00:11||[cpu_sched_debug] finished preempt loop, nrunning 8 07/03/2008 23:00:11|climateprediction.net|[cpu_sched_debug] hadcm3istd_026w_1920_160_15925025_0 sched state 2 next 2 task state 1 07/03/2008 23:00:11|climateprediction.net|[cpu_sched_debug] hadcm3istd_0bob_1920_160_05936405_5 sched state 2 next 2 task state 1 07/03/2008 23:00:11|SETI@home Beta Test|[cpu_sched_debug] ap_22mr07ac_B7_P1_00398_20080208_13385.wu_3 sched state 2 next 2 task state 1 07/03/2008 23:00:11|Einstein@Home|[cpu_sched_debug] h1_0836.30_S5R3__316_S5R3b_0 sched state 2 next 2 task state 1 07/03/2008 23:00:11|Einstein@Home|[cpu_sched_debug] h1_0836.30_S5R3__315_S5R3b_1 sched state 1 next 1 task state 9 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] 30mr07ac.30604.3079.12.7.174_0 sched state 2 next 2 task state 1 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] 30mr07aa.11888.2526.12.7.173_1 sched state 2 next 2 task state 1 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] 30mr07ac.20541.1085042.6.7.171_3 sched state 2 next 2 task state 1 07/03/2008 23:00:11|SETI@home|[cpu_sched_debug] 30mr07ac.30604.1081361.12.7.33_0 sched state 2 next 2 task state 1 07/03/2008 23:00:11||[cpu_sched_debug] enforce_schedule: end Periodically, I get this block as well: 07/03/2008 23:02:34||[work_fetch_debug] Request work fetch: timer 07/03/2008 23:02:34||[work_fetch_debug] compute_work_requests(): start 07/03/2008 23:02:34||[cpu_sched_debug] CPU efficiency old 0.991033 new 0.991026 wall 487.668808 CPU 482.714105 w 0.994372 e 0.989840 07/03/2008 23:02:34||[work_fetch_debug] compute_work_requests(): cpu_shortfall 0.000000, overall urgency OK 07/03/2008 23:02:34|climateprediction.net|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:34|CPDN Beta|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:34|Einstein@Home|[work_fetch_debug] project has no shortfall; skipping 07/03/2008 23:02:34|lhcathome|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:34|orbit@home|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:34|SETI@home|[work_fetch_debug] project is overworked; skipping 07/03/2008 23:02:34|SETI@home Beta Test|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:40||[work_fetch_debug] Request work fetch: Project backoff ended 07/03/2008 23:02:40||[work_fetch_debug] compute_work_requests(): start 07/03/2008 23:02:40||[cpu_sched_debug] CPU efficiency old 0.991026 new 0.991013 wall 48.671997 CPU 47.065467 w 0.999437 e 0.966993 07/03/2008 23:02:40||[work_fetch_debug] compute_work_requests(): cpu_shortfall 0.000000, overall urgency OK 07/03/2008 23:02:40|climateprediction.net|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:40|CPDN Beta|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:40|Einstein@Home|[work_fetch_debug] project has no shortfall; skipping 07/03/2008 23:02:40|lhcathome|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:40|orbit@home|[work_fetch_debug] work fetch: project not contactable; skipping 07/03/2008 23:02:40|SETI@home|[work_fetch_debug] project is overworked; skipping 07/03/2008 23:02:40|SETI@home Beta Test|[work_fetch_debug] project has no shortfall; skipping Edit: Here's a section from last night, on one of the rare occasions when it decided to fetch work from SETI Beta: 2008-03-06 20:35:18 [---] [work_fetch_debug] Request work fetch: Project backoff ended 2008-03-06 20:35:18 [---] [work_fetch_debug] compute_work_requests(): start 2008-03-06 20:35:18 [---] [work_fetch_debug] compute_work_requests(): cpu_shortfall 0.000000, overall urgency OK 2008-03-06 20:35:18 [climateprediction.net] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:35:18 [CPDN Beta] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:35:18 [Einstein@Home] [work_fetch_debug] project has no shortfall; skipping 2008-03-06 20:35:18 [lhcathome] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:35:18 [SETI@home] [work_fetch_debug] project has no shortfall; skipping 2008-03-06 20:35:18 [SETI@home Beta Test] [work_fetch_debug] project has no shortfall; skipping 2008-03-06 20:35:18 [orbit@home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:19 [---] [work_fetch_debug] Request work fetch: timer 2008-03-06 20:36:19 [---] [work_fetch_debug] compute_work_requests(): start 2008-03-06 20:36:19 [---] [work_fetch_debug] compute_work_requests(): cpu_shortfall 13.950803, overall urgency Need 2008-03-06 20:36:19 [climateprediction.net] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:19 [CPDN Beta] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:19 [Einstein@Home] [work_fetch_debug] best project so far 2008-03-06 20:36:19 [lhcathome] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:19 [SETI@home] [work_fetch_debug] project has less LTD than Einstein@Home 2008-03-06 20:36:19 [SETI@home Beta Test] [work_fetch_debug] best project so far 2008-03-06 20:36:19 [orbit@home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:19 [SETI@home Beta Test] [work_fetch_debug] compute_work_requests(): work req 4.650268, shortfall 0.000000, urgency OK 2008-03-06 20:36:23 [---] [work_fetch_debug] time_until_work_done(): est 0.000000 ssr 800.000000 apr 7.931308 prs 200.000000 2008-03-06 20:36:23 [SETI@home Beta Test] Sending scheduler request: To fetch work 2008-03-06 20:36:23 [SETI@home Beta Test] Requesting 5 seconds of new work 2008-03-06 20:36:48 [SETI@home Beta Test] Scheduler RPC succeeded [server version 601] 2008-03-06 20:36:48 [SETI@home Beta Test] Deferring communication for 7 sec 2008-03-06 20:36:48 [SETI@home Beta Test] Reason: requested by project 2008-03-06 20:36:48 [---] [work_fetch_debug] Request work fetch: RPC complete 2008-03-06 20:36:49 [---] [work_fetch_debug] compute_work_requests(): start 2008-03-06 20:36:49 [---] [work_fetch_debug] compute_work_requests(): cpu_shortfall 87.374924, overall urgency Need 2008-03-06 20:36:49 [climateprediction.net] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:49 [CPDN Beta] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:49 [Einstein@Home] [work_fetch_debug] best project so far 2008-03-06 20:36:49 [lhcathome] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:49 [SETI@home] [work_fetch_debug] project has less LTD than Einstein@Home 2008-03-06 20:36:49 [SETI@home Beta Test] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:49 [orbit@home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:49 [Einstein@Home] [work_fetch_debug] compute_work_requests(): work req 14.562487, shortfall 0.000000, urgency OK 2008-03-06 20:36:51 [SETI@home Beta Test] [file_xfer] Started download of file 11oc06aa.18571.545437.14.10.88 2008-03-06 20:36:54 [---] [work_fetch_debug] time_until_work_done(): est 80451.142322 ssr 800.000000 apr 7.931366 prs 100.000000 2008-03-06 20:36:54 [Einstein@Home] Sending scheduler request: To fetch work 2008-03-06 20:36:54 [Einstein@Home] Requesting 15 seconds of new work 2008-03-06 20:36:56 [---] [work_fetch_debug] Request work fetch: Project backoff ended 2008-03-06 20:36:56 [---] [work_fetch_debug] compute_work_requests(): start 2008-03-06 20:36:56 [---] [work_fetch_debug] compute_work_requests(): cpu_shortfall 237.182610, overall urgency Need 2008-03-06 20:36:56 [climateprediction.net] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:56 [CPDN Beta] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:56 [Einstein@Home] [work_fetch_debug] best project so far 2008-03-06 20:36:56 [lhcathome] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:56 [SETI@home] [work_fetch_debug] project has less LTD than Einstein@Home 2008-03-06 20:36:56 [SETI@home Beta Test] [work_fetch_debug] best project so far 2008-03-06 20:36:56 [orbit@home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:56 [SETI@home Beta Test] [work_fetch_debug] compute_work_requests(): work req 79.060870, shortfall 0.000000, urgency OK 2008-03-06 20:36:57 [SETI@home Beta Test] [file_xfer] Finished download of file 11oc06aa.18571.545437.14.10.88 2008-03-06 20:36:57 [SETI@home Beta Test] [file_xfer] Throughput 82665 bytes/sec 2008-03-06 20:36:59 [Einstein@Home] Scheduler RPC succeeded [server version 601] 2008-03-06 20:36:59 [Einstein@Home] Deferring communication for 1 min 0 sec 2008-03-06 20:36:59 [Einstein@Home] Reason: requested by project 2008-03-06 20:36:59 [---] [work_fetch_debug] Request work fetch: RPC complete 2008-03-06 20:36:59 [---] [work_fetch_debug] compute_work_requests(): start 2008-03-06 20:36:59 [---] [work_fetch_debug] compute_work_requests(): cpu_shortfall 0.000000, overall urgency OK 2008-03-06 20:36:59 [climateprediction.net] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:59 [CPDN Beta] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:59 [Einstein@Home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:59 [lhcathome] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:59 [SETI@home] [work_fetch_debug] best project so far 2008-03-06 20:36:59 [SETI@home Beta Test] [work_fetch_debug] project has no shortfall; skipping 2008-03-06 20:36:59 [orbit@home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:36:59 [SETI@home] [work_fetch_debug] compute_work_requests(): work req 1173.277255, shortfall 1173.277255, urgency Need 2008-03-06 20:37:04 [---] [work_fetch_debug] time_until_work_done(): est 211151.817560 ssr 800.000000 apr 7.931474 prs 300.000000 2008-03-06 20:37:04 [SETI@home] Sending scheduler request: To fetch work 2008-03-06 20:37:04 [SETI@home] Requesting 1173 seconds of new work 2008-03-06 20:37:09 [SETI@home] Scheduler RPC succeeded [server version 601] 2008-03-06 20:37:09 [SETI@home] Deferring communication for 11 sec 2008-03-06 20:37:09 [SETI@home] Reason: requested by project 2008-03-06 20:37:09 [---] [work_fetch_debug] Request work fetch: RPC complete 2008-03-06 20:37:09 [---] [work_fetch_debug] compute_work_requests(): start 2008-03-06 20:37:09 [---] [work_fetch_debug] compute_work_requests(): cpu_shortfall 0.000000, overall urgency OK 2008-03-06 20:37:09 [climateprediction.net] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:09 [CPDN Beta] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:09 [Einstein@Home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:09 [lhcathome] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:09 [SETI@home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:09 [SETI@home Beta Test] [work_fetch_debug] project has no shortfall; skipping 2008-03-06 20:37:09 [orbit@home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:11 [SETI@home] [file_xfer] Started download of file 30mr07aa.8116.23794.10.7.128 2008-03-06 20:37:18 [SETI@home] [file_xfer] Finished download of file 30mr07aa.8116.23794.10.7.128 2008-03-06 20:37:18 [SETI@home] [file_xfer] Throughput 69525 bytes/sec 2008-03-06 20:37:20 [---] [work_fetch_debug] Request work fetch: Project backoff ended 2008-03-06 20:37:20 [---] [work_fetch_debug] compute_work_requests(): start 2008-03-06 20:37:20 [---] [work_fetch_debug] compute_work_requests(): cpu_shortfall 0.000000, overall urgency OK 2008-03-06 20:37:20 [climateprediction.net] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:20 [CPDN Beta] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:20 [Einstein@Home] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:20 [lhcathome] [work_fetch_debug] work fetch: project not contactable; skipping 2008-03-06 20:37:20 [SETI@home] [work_fetch_debug] project has no shortfall; skipping 2008-03-06 20:37:20 [SETI@home Beta Test] [work_fetch_debug] project has no shortfall; skipping 2008-03-06 20:37:20 [orbit@home] [work_fetch_debug] work fetch: project not contactable; skipping |
|
4815)
Message boards :
BOINC client :
Dual core
(Message 15716)
Posted 7 Mar 2008 by Richard Haselgrove Post: You wrote before that the SETI.Beta job takes 202 hours. The remaining run time on that job will be, is my guess, the cause for that project not being entitled to backfill the buffer. 86400 STD less remaining run time on job and the relevant weight broadly fills the bill. SETI.Beta jobs actually take ~144 hours. 202 hours was BOINC's estimate, at that time, of time remaining to completion (estimate incorrect as RDCF is messed up - it's a Beta project). Time remaining is currently estimated at 131 hours 15 mins. It's about 2 hours since I upgraded. STD has remained limit-stopped at 86,400: LTD has increased to 591,333. LHC has requested work six times, Orbit has requested work 11 times. SETI Beta has not requested work at all. It is unable to fulfil the resource share I requested, because it isn't clever enough to recognise that the work in its buffer is indivisible, and can't be split across two cores. I (and I think Mike Gibson) think that it should recognise that an available resource (a second core) is available to honour the resource share, that it can't be used by the task already downloaded because it can't be split across two cores, and download more work as permitted by debt etc. |
|
4816)
Message boards :
BOINC client :
Dual core
(Message 15714)
Posted 7 Mar 2008 by Richard Haselgrove Post: OK, I've upgraded. v5.10.45 it is. Current debt report: Share___Project__________STD____________LTD 200 ... CPDN .......... 1280 ... -1,132,438 100 ... CPDN Beta ........ 0 ...... 200,000 (inactive) 100 ... Einstein ... -47,285 ....... 27,571 100 ... LHC .............. 0 ...... 307,296 100 ... Orbit ............ 0 ........ 9,009 300 ... SETI ....... -40,471 ........ 4,503 200 ... SETI Beta ... 86,400 ...... 584,057 Now please tell me why since the upgrade, BOINC has requested new work (in sequence) from SETI, Orbit (none available) and LHC (none available): but it has NOT requested any work from SETI Beta. SETI Beta project is active (running one task), and NOT in comms deferral. BUG. |
|
4817)
Message boards :
BOINC client :
Dual core
(Message 15713)
Posted 7 Mar 2008 by Richard Haselgrove Post: Well, there is the more radical solution to go to the project tab, select each project that is allowed to fetch work, but has none in progress or buffer and reset those projects. That should right them for the most part but some residual LTD. OR ditch 5.10.13 and get on a current recommended version.... old trusty is up for long overdue retirement.... why else are we now on like 5.10.45. As somebody just wrote in a message board far, far away .... this is a bug report. Sure, I can work round it. Or I can ignore it. (Which I've been doing for weeks). But what I'm trying to do is to document the bug, so that somebody can fix it. Apart from the (irrelevent) bugfix at v5.10.17, can you point to one bit of evidence that this bug has been corrected between v5.10.13 and v5.10.45? Sigh. OK, I know: "Upgrade to the latest version, re-test, report again if the problem is still apparent." |
|
4818)
Message boards :
BOINC client :
Dual core
(Message 15710)
Posted 7 Mar 2008 by Richard Haselgrove Post: Thanks for dropping by. I don't have all of the information, but, it is entirely possible that BOINC is blocking downloads from one of the projects because of time pressure. If BOINC cannot finish a task within 90% of the time from now to the computation deadline based on round robin scheduling, it is in time trouble. The computation deadline is the report deadline - (Connect ever X + task switch interval + safety margin). The safety margin is either 24 hours or 0 depending on the version of BOINC. This is to avoid making an already bad situation worse. A project that has a task under time pressure is not allowed to download more work. Plugging figures into the formula, I get: Computation deadline = [19 March 2008 07:06:50] minus (0.01 day + 1 hour + 24 hours) or about 05:52 on 18 March 2008. Current completion estimate is 12:28:16 on 13 March 2008 (and dropping), so I think we can rule out time pressure. If there is sufficient work to cover all of the cores of a machine for the queue time, and there is any task with deadline pressure, no project will be asked for work. Again, no sense in making a bad situation worse. Work, in general, is being fetched as expected. Requests are regularly going out for typically 70 seconds of SETI general work (cache currently at 2 days 17 hours, against 1 day x 3 cores requested), and around 62,500 seconds of each of LHC and Orbit (caches currently empty). If a task has enough work to cover its share of the work queue, it will not be asked for work. It already has its share downloaded. Of course, if there is no time task with time pressure, enough work from someplace will be downloaded to keep the queue full for all cores. It just may not be the exact mix that you thought you specified. The question is, why is no work being requested for SETI Beta, which has by far the highest debt values (both STD and LTD), no unstarted work, and only one active task against a requested resource share of 2/8? I can only explain that by assuming that BOINC is looking at the "6 days to completion" figure for the current task, converting that into 3 days x 2 cores resource share as the measure of work available, and concluding that there is sufficient to meet my request of 1 day additional cache. What BOINC seems not to be taking into account is that the 6 day completion figure is a single monolithic task, and cannot be processed across two cores as resource share would specify. I think it should. |
|
4819)
Message boards :
BOINC client :
Dual core
(Message 15708)
Posted 7 Mar 2008 by Richard Haselgrove Post: Mike, I've been round the blocks a few times with BOINC, and I think I've used most of the tricks you describe. My debt values are deliberately skewed: I run with a baseline of -1,000,000 for CPDN (and No New Tasks set as well) so I will never download a new CPDN task without it being a deliberate choice on my part: and a baseline of +200,000 for all other active projects, so they should never be inhibited from fetching work by the "LTD less than negative of task switch interval" rule. And as you say, there are debt values for CPDN Beta, LHC and Orbit (all of which have been inactive for the last few hours) which I left out for clarity - that's why the figures don't balance to zero as they should. LHC is a special case. In order to distribute the limited amount of available work fairly among the crunchers, they have unusually strict rules about issuing work (maximum two tasks per request, no repeat request for 15 minutes, quota 10 per CPU per day, quota limited to 4 CPUs). I think any problem with LHC is more likely to be supply restriction, rather than BOINC request calculations. You asked about checking debt figures. I know of two tools: BoincDV, which displays a snapshot of the current values on request (and can also reset them to zero if you wish): and BoincView, which can monitor/control/log multiple BOINC hosts across a network. BoincView doesn't display debt by default, but you can select columns for both STD and LTD on the projects tab - they update in real time, every 60 seconds. (Incidentally, you can find a bug report from me - [trac]#136[/trac] - on these pages saying that debt isn't updated when networking is suspended, and a contradiction from David Anderson saying that debt is only updated once an hour anyway. He was wrong). Anyway, back to the problem at hand. The current values from BV are: SETI Beta - STD 86,400 LTD 559,335 BOINC believes that the single SETI Beta task running on the host will finish in 6 days' time, on 13 March - deadline is 19 March. It has downloaded, completed and reported one additional short (1 hour) task overnight, and requested 40 seconds or so of work three more times: each time, it then asked for Einstein work instead, and stopped asking for SETI Beta work (SETI Beta is out of work at the moment, which confuses things - but BOINC didn't even continue trying). Einstein is at STD -44,969 LTD +43,810 cache 30 hours. So I think you're right: there is a bug in the work fetch algorithm for dual-cores and upwards, but it's obscure and I don't think I can put it into clear enough language to make it an issue for the main BOINC development team. However, if we could attract the attention of John McLeod VII (who wrote this particular bit of code) to this thread, he might be more interested. |
|
4820)
Message boards :
BOINC client :
Dual core
(Message 15699)
Posted 6 Mar 2008 by Richard Haselgrove Post: Came across this thread while pondering a work-fetch problem. I think it's a related issue. Bear with me, this is going to get complicated. I have 8 cores. Neglecting projects which currently have no work or are disabled, they have been allocated: Share___Project__________STD____________LTD 200 ... CPDN ........ -6,253 ... -1,129,611 100 ... Einstein ... -21,185 ....... 74,984 300 ... SETI ....... -14,491 ....... 52,856 200 ... SETI Beta ... 41,931 ...... 495,423 Those shares add up to 800, so the cores should be working 2, 1, 3, 2 on the four projects. By the debt figures, SETI Beta should be the next to both fetch work and schedule a task. Yet I have only one SETI Beta task on board, and BOINC shows no sign of fetching another one. The single SETI Beta task I have is an Astropulse WU, currently estimating 202 hours remaining for a 19 March deadline. Tight, but with a 0.9888 efficiency for the machine as a whole, and a low CI/cache, there's no deadline pressure. In fact, it will finish in plenty of time, but BOINC doesn't know that (SETI Beta RDCFs are confused by having two separate applications under test). From previous experience - this has been nagging me for some time - BOINC will only start to fetch Beta work for a second core when the time remaining on the AP task has dropped below my cache setting (1 day). By then, STD will have bumped up against the limit stop of 86,400, and LTD will be up around 850,000 - both figures have gone up by over 1,000 since I started typing this. As Mike Gibson says, BOINC doesn't seem to be correctly allowing for the availability of extra cores when making work fetch calculations/decisions. My BOINC - as Ageless well knows - is my old faithful v5.10.13 for Windows: but I've checked the change log, and there's no report of any adjustment to this part of the scheduler since then. Edit - I've turned on <work_fetch_debug> in cc_config.xml - let's see what that throws up. At the moment, it's saying: 06/03/2008 13:52:03||[work_fetch_debug] compute_work_requests(): cpu_shortfall 0.000000, overall urgency OK 06/03/2008 13:52:03|SETI@home Beta Test|[work_fetch_debug] project has no shortfall; skipping |
|
4821)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 15480)
Posted 18 Feb 2008 by Richard Haselgrove Post: Found yet another bug in Forum Advanced Search. This time, it happens whether or not you restrict the search to a single message board. Trying to deal with a bug at SETI which causes major loss of completed work with the words "[Error] Error on file upload: no command" in the BOINC message log. So I did an advanced search on "Error on file upload". This found several message log excerpts, including some which displayed like this: ont color="blue">1/25/2008 11:54:14 PM|SETI@home|Computation for task 28no06ah.19864.24612.15.6.131_0 finishedont> 1/25/2008 11:54:14 PM|SETI@home|Starting 28no06ah.30219.5798.16.6.159_0 1/25/2008 11:54:14 PM|SETI@home|Starting task 28no06ah.30219.5798.16.6.159_0 using setiathome_enhanced version 528 ont color="red">1/25/2008 11:54:14 PM|SETI@home|Sending scheduler request: To fetch workont> 1/25/2008 11:54:14 PM|SETI@home|Requesting 14764 seconds of new work ont color="green">1/25/2008 11:54:16 PM|SETI@home|[file_xfer] Started upload of file 28no06ah.19864.24612.15.6.131_0_0 1/25/2008 11:54:18 PM|SETI@home|[file_xfer] Finished upload of file 28no06ah.19864.24612.15.6.131_0_0ont> The original message was: [color=blue]1/25/2008 11:54:14 PM|SETI@home|Computation for task 28no06ah.19864.24612.15.6.131_0 finished[/color] 1/25/2008 11:54:14 PM|SETI@home|Starting 28no06ah.30219.5798.16.6.159_0 1/25/2008 11:54:14 PM|SETI@home|Starting task 28no06ah.30219.5798.16.6.159_0 using setiathome_enhanced version 528 [color=red]1/25/2008 11:54:14 PM|SETI@home|Sending scheduler request: To fetch work[/color] 1/25/2008 11:54:14 PM|SETI@home|Requesting 14764 seconds of new work [color=green]1/25/2008 11:54:16 PM|SETI@home|[file_xfer] Started upload of file 28no06ah.19864.24612.15.6.131_0_0 1/25/2008 11:54:18 PM|SETI@home|[file_xfer] Finished upload of file 28no06ah.19864.24612.15.6.131_0_0[/color] Looking at the page source for the browser display for the Forum Search Results, the first line of the examples above looks like this: <f<span class="highlight">on</span>t color="blue">1/25/2008 11:54:14 PM|SETI@home|Computati<span class="highlight">on</span> for task 28no06ah.19864.24612.15.6.131_0 finished</f<span class="highlight">on</span>t><br /> - in other words, the "funky search result formatting" in [trac]changeset:14267[/trac] has been applied to absolutely everything, including search terms embedded in HTML tags arising from the conversion of BBCode 'color' tags. Not good. Edit: Trac [trac]#557[/trac] |
|
4822)
Message boards :
BOINC client :
Error on file upload: no command
(Message 15477)
Posted 18 Feb 2008 by Richard Haselgrove Post: I'm trying to gather further information on this problem in a new SETI thread: Rash of Validate errors So far, we've had one report of it still happening under BOINC v5.10.30, but the wording has changed: [error] Error on file upload: no command Giving up on upload of xxxxx: fatal upload error |
|
4823)
Message boards :
BOINC client :
Error on file upload: no command
(Message 15460)
Posted 16 Feb 2008 by Richard Haselgrove Post: Could anyone with BOINC code access have a look at this one for me, please? Had a SETI server outage earlier today. But this problem started much earlier: Validate errors II (warning - long thread). One machine - only - got itself into a state with SETI - only - where every upload ended with: Error on file upload: no command [file_xfer] Permanently failed upload of xxxxx Giving up on upload of xxxxx: server rejected file None of my other machines was affected (though other users have reported it), and none of the other projects on the affected machine had any problems with uploads. I restarted BOINC on the affected machine during today's total SETI outage, and it has uploaded/reported/been credited since then, so it was a transient problem. What troubles me is that while the upload problems were happening, about 34 hours, every single upload was terminated as a permanent error at the first upload attempt - no retries, no backoff, just a straight slam dunk. Is that how it's supposed to work? BOINC is v5.10.13, Windows 32bit, running on a Windows 2003 server as a service (but I have other W2K3 servers which didn't suffer the problem). |
|
4824)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 15051)
Posted 20 Jan 2008 by Richard Haselgrove Post: Fixed again: [trac]changeset:14603[/trac]. Frankly, the busier I can keep him with 'extremely minor bug fixes', the less time he'll have to b****r up the rest of BOINC. Anyone got a libcurl proxy we can bother him with? |
|
4825)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 14996)
Posted 17 Jan 2008 by Richard Haselgrove Post: The fatal error has gone away, so I've been able to reproduce the bug. Again, it only shows if you restrict the search to a single forum thread. If you search on 'bug' in this forum (Web interfaces), it says that Jord's message #14835 was posted 2 days ago. Seems to show all messages found in a thread as if they were posted at the time of the most recent post in that thread. Edit - and to confirm the point, it's now telling me that Jord's message was posted 17 seconds ago. |
|
4826)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 14910)
Posted 15 Jan 2008 by Richard Haselgrove Post: More jiggery-pokery with advanced search: Fatal error: Call to a member function setHighlightTerms() on a non-object in /disks/setifiler1/home/boincadm/projects/beta/html/user/forum_search_action.php on line 169 I'd used searching successfully a few minutes earlier, but noticed another bug and was trying to recreate the circumstances! The problem I saw was with the message "Posted [x length of time] ago by...." in the message headers: under some circumstances (tbc), the [x length of time] seems to be the user's most recent post in the thread, rather than the age of the actual message. |
|
4827)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 14855)
Posted 12 Jan 2008 by Richard Haselgrove Post: Ticket has been closed, and the fix seems to have been deployed to the forums I originally reported (here, SETI, SETI Beta). Thanks, David. |
|
4828)
Message boards :
BOINC Manager :
Bug in 5.10.30--"Use At Most" CPU Option missing
(Message 14848)
Posted 11 Jan 2008 by Richard Haselgrove Post: .... What sort of machine are you running? There's a similar report of DDE failure in this thread at SETI. Turned out not to be BOINC at all, but a specific piece of OEM preinstall software on an IBM Thinkpad. Details in the thread. |
|
4829)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 14842)
Posted 11 Jan 2008 by Richard Haselgrove Post: Just tested CPDN, which has the older-style search results format ("[Read the rest of this post]" after each message, rather than an ID number in the message header). No problem there, so the issue may have arisen when they re-designed the results page. |
|
4830)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 14840)
Posted 11 Jan 2008 by Richard Haselgrove Post: Ah - just tried it here, seems to be dependent on whether you restrict the search to a single forum. Try it again with "Forum Only display posts from this forum" set to something other than 'All'. |
|
4831)
Message boards :
Web interfaces :
Bug in Forum Advanced Search
(Message 14834)
Posted 11 Jan 2008 by Richard Haselgrove Post: Just spotted a bug in forum_search_action.php Doing an advanced search (here, at SETI, or at SETI Beta), you get a list of messages back. At the end of the header row for each message, there's a forum link to the Message ID - except it isn't. The number shown, and linked as the #message, is actually the thread ID number instead (repeated): a search for 'bug' in this forum returns (inter alia) http://boinc.berkeley.edu/dev/forum_thread.php?id=2369&nowrap=true#2369 - my opening post in that thread was #14479. Should be easy just to switch the field displayed and referenced in the link. |
|
4832)
Message boards :
Web interfaces :
User profile images on Forum index pages
(Message 14596)
Posted 1 Jan 2008 by Richard Haselgrove Post: The images have gone from the forum index pages, but they've popped up again for 'Friends'. If you're not on dialup, have a look at Misfit's account page. Good thing SETI has lots of bandwidth to spare, eh? ;-) |
|
4833)
Message boards :
Web interfaces :
User profile images on Forum index pages
(Message 14509)
Posted 24 Dec 2007 by Richard Haselgrove Post: I think it's related... show_user.php is now requiring login. So you can't see who their friends are? Actually, I'm finding I can see the abbreviated form of the account pages without being logged in. It may have changed again. Where exactly did you have the problem? |
|
4834)
Message boards :
Web interfaces :
User profile images on Forum index pages
(Message 14496)
Posted 24 Dec 2007 by Richard Haselgrove Post: The images have now been removed from the main SETI project. However, you can still see then on the SETI Beta forums for the time being, if you want a preview of what DA has in store for you in the future. |
|
4835)
Message boards :
Web interfaces :
User profile images on Forum index pages
(Message 14490)
Posted 24 Dec 2007 by Richard Haselgrove Post: David Anderson, the BDFL of BOINC. See [trac]changeset:14438[/trac]. Trac [trac]#525[/trac] |
|
4836)
Message boards :
Web interfaces :
User profile images on Forum index pages
(Message 14480)
Posted 24 Dec 2007 by Richard Haselgrove Post: As well as the thumbnail images, they've also implemented a 'Friends' scheme. Once you've been offered friendship by another user, and accepted, that fact is shown on both the respective account pages, for all and sundry to see. For ever, as far as we've been able to determine in brief testing. Where was that on the core client v6.00 roadmap? |
|
4837)
Message boards :
Web interfaces :
User profile images on Forum index pages
(Message 14479)
Posted 24 Dec 2007 by Richard Haselgrove Post: Have a look at the SETI Forum Index. Those aren't avatars - they're thumbnails of the user profile image. They've appeared, like a rash, without warning, sometime in the last hour. People are reporting them on the statistics pages too. As an absolute, unarguable BUG: They don't respect the Community preference for 'Hide Avatars' or 'Show images as links'. They need an 'off' switch, for dial-up users if nobody else: either one of the existing ones, or a new one of their own. But who the **** implemented this on the Sunday leading into an extended holiday, when the SETI servers happen to be under a particularly severe stress load? I'm sure it's the last thing the project staff would have wanted or chosen to implement. Has it been foisted on them by a remote control BOINC stealth attack? |
|
4838)
Message boards :
BOINC client :
Testing BOINC v6.00
(Message 14414)
Posted 18 Dec 2007 by Richard Haselgrove Post: BOINC 6.1 has temporarily gone back to the drawing board. A couple of us have tested 6.1.5 but we came across too many problems to even throw it onto the Alpha testers group. Mainly problems with the installer and the service, though. I don't mind running a test version by hand, but "No graphics" rather scuppers any idea of testing - the new standalone graphics component in SETI! I don't mind running through manual installation steps either, if anyone has a pre-installer version with Vista graphics - if it's pre-release, you may prefer to PM me at SETI or SETI Beta. |
|
4839)
Message boards :
BOINC client :
Testing BOINC v6.00
(Message 14412)
Posted 18 Dec 2007 by Richard Haselgrove Post: Does anybody have a viable BOINC v6.00 (32bit Windows) available for testing yet? I know Crunch3r has one called Windows 32bit 6.1.0.32, but I'd prefer an official(ish) one for testing. The reason I ask is that SETI Beta have just put up a v6.00 science app with Vista compatible graphics for testing - I'd like to test it properly. Many thanks in advance. |
|
4840)
Message boards :
Web interfaces :
BOINC dev Forum broken
(Message 13477)
Posted 2 Nov 2007 by Richard Haselgrove Post: Seems to be fixed now - although I'd reported it as trac [trac]#458[/trac] in the meantime. The SETI message boards now have this in the page source: <!-- $Id: forum_thread.php 14023 2007-11-02 18:48:48Z davea $ --> <!-- $Id: forum.inc 14023 2007-11-02 18:48:48Z davea $ --> where this page has <!-- $Id: forum_thread.php 14019 2007-11-01 23:04:39Z davea $ --> <!-- $Id: forum.inc 14019 2007-11-01 23:04:39Z davea $ --> 1) The appearance of the message button is even worse with the SETI version than this version. 2) Would I be right in reading 14019 and 14023 as SVN changeset revision numbers? In which case, why is the latest update on the Server Updates page still stuck at 13964? |
|
4841)
Message boards :
Web interfaces :
BOINC dev Forum broken
(Message 13443)
Posted 2 Nov 2007 by Richard Haselgrove Post: Seems to be fixed now - although I'd reported it as trac [trac]#458[/trac] in the meantime. While we're on the subject, the new "Send [private] message" button under the user name or avatar in the left-hand-column of the forum tables is either oversized or padded too far right: the edge of the button is not visible. |
|
4842)
Message boards :
Web interfaces :
Citation index for forum posts
(Message 12212)
Posted 23 Aug 2007 by Richard Haselgrove Post: Just an off-the wall idea that occurred during a 'feedback to project' discussion at SETI. How easy would it be to maintain a count of how many times a message board post was referred to as a hyperlink target in another message? In other words, if enough people posted "Read this message", then 'this message' would rise to the top of the rankings as the most useful post - a bit like a simple version of the Google page rankings. Then, with a suitable decay factor, there could be page displays for 'Most useful tip' in Q&A, 'Current problem workrounds' in Number Crunching, 'Funniest Joke' in the Cafe, etc. etc. [I'm interested in #2 in that list, but hey, let's be inclusive]. I've no idea how easy it would be to implement this - just thought I'd throw it open for discussion. |
|
4843)
Message boards :
BOINC client :
Work fetch bug in BOINC v5.10.13
(Message 11969)
Posted 8 Aug 2007 by Richard Haselgrove Post: During a recent SETI outage recovery, I noticed a problem related to 'Redundant result - Cancelled by server' (server-side abort of work which has reached quorum and been validated before the host has even started computation - a new feature which I think only SETI has implemented so far). When the servers came back up, my fast box was allocated a lot of WUs quite quickly, but struggled to download them through the congestion. However, other people obviously did download them, because some became redundant before I'd even managed to complete the download. Once I'd reported the (aborted) result, my trusty BoincView was showing green across the board, but no new work was being fetched. I discovered that the machine was still re-trying five downloads, but getting 'file not found' (obviously, because the WU had been assimilated and the datapak deleted from the server). I just aborted the meaningless downloads, and work fetch started back up as normal. BOINC needs to check for, and cancel, any associated transfers when doing a server-mandated abort. I'll log a Trac report. Edit - trac 366. |
|
4844)
Message boards :
Server programs :
Server scheduler: problems with anonymous platform
(Message 10360)
Posted 19 May 2007 by Richard Haselgrove Post: As part of the recovery from their recent server breakdown, SETI@home have updated their servers to the latest BOINC source tree - partly, I believe, because they also moved from a Solaris server to Linux. Users using any form of the anonymous platform mechanism are finding that they get "HTTP internal server error" whenever they request new work. Users on project-supported platforms can get work by disabling anonymous platform: users on non-project-supported platforms (OS/2 Warp, 64-bit) are left high and dry. I have logged a trac ticket (#194), but any project admin thinking about upgrading server software in the near future might like to check this problem first. |
|
4845)
Message boards :
BOINC Manager :
Transferring to a new operating system
(Message 10357)
Posted 19 May 2007 by Richard Haselgrove Post: I've been through a very similar process, but doing Vista as an 'in situ' upgrade over a working XP install. I managed to keep everything - project host IDs, stats, work in progress (including 5 CPDN models - this is a multicore!) etc. I would suggest:
|
|
4846)
Message boards :
BOINC client :
What Happened to the Edit Post on this board -
(Message 10250)
Posted 14 May 2007 by Richard Haselgrove Post: Yes, looking OK here. Edit - including edits! |
|
4847)
Message boards :
BOINC client :
What Happened to the Edit Post on this board -
(Message 10237)
Posted 13 May 2007 by Richard Haselgrove Post: ... 'might' ..., that would be. Still no edit function here. |
|
4848)
Message boards :
BOINC client :
What Happened to the Edit Post on this board -
(Message 10236)
Posted 13 May 2007 by Richard Haselgrove Post: In an earlier thread on this subject, Jord suggested that the problem night be limited to IE. I'm using IE7 on Win XP. |
|
4849)
Message boards :
BOINC client :
What Happened to the Edit Post on this board -
(Message 10230)
Posted 13 May 2007 by Richard Haselgrove Post: SETI and SETI Beta both have the new PM function, but are showing headers/footers/red-x/edit. Either there's a big difference between the SETI board and this board (which I doubt), or the developers have got access to a fixed version but haven't deployed it here yet. PS 1: no, I'm not a moderator at SETI PS 2: posting with no edit facility should do wonders for our typing skills! |
|
4850)
Message boards :
BOINC Manager :
Forum bugs and other shananigans
(Message 10206)
Posted 11 May 2007 by Richard Haselgrove Post: Cool. The error message was cleared within a minute of posting here. |
|
4851)
Message boards :
BOINC Manager :
Forum bugs and other shananigans
(Message 10205)
Posted 11 May 2007 by Richard Haselgrove Post: Headers and footers still not visible here. The new version (with private messageing) has been deployed at SETI, but we're getting errors: Warning: mysql_fetch_object(): supplied argument is not a valid MySQL result resource in /disks/setifiler1/home/boincadm/projects/sah/html/inc/forum.inc on line 578 so it looks like an 'include' file hasn't been copied. |
|
4852)
Message boards :
BOINC Manager :
Forum bugs and other shananigans
(Message 10198)
Posted 11 May 2007 by Richard Haselgrove Post: (and the link to edit the typo in my last post....) |
|
4853)
Message boards :
BOINC Manager :
Forum bugs and other shananigans
(Message 10197)
Posted 11 May 2007 by Richard Haselgrove Post: For the time being, the afore-mention red-x seems to have disappeared: as have the message numbers, date and time stamps. (though strangely, all of the above are visible while I'm typing this on the 'post a reply' screen - it's just when browsing they all disappear). |
|
4854)
Message boards :
BOINC client :
Not updating debt - network disabled
(Message 9934)
Posted 27 Apr 2007 by Richard Haselgrove Post: I've added my version to trac as ticket # 136. Mike - I think your version has been mentioned before, but I couldn't lay my hands on a reference easily. I think it's a good idea, but I'll leave you to add it to trac if you want to. |
|
4855)
Message boards :
BOINC client :
Not updating debt - network disabled
(Message 9929)
Posted 27 Apr 2007 by Richard Haselgrove Post: Could there be a problem with BOINC failing to update the debt values when networking is disabled? I'm running BOINC v5.8.16 under Windows Vista, on an 8-core dual Xeon. I'm monitoring it remotely with BOINCView 1.4.2, which is capable of displaying STD and LTD in real time. [The columuns - on the projects tab - are not visbile by default, but can be selected] When BOINC networking is enabled (normal), the debt values update every 60 seconds or so, and I can predict pretty accurately when applications are going to switch and which project is going to be 'called forward' next. For the last couple of days, I've been running with networking disabled (because the CPDN servers are full, and there's an alpha test going on at SETI). While networking is off, BV doesn't update the displayed debt figures regularly, only at major events like tasks finishing. Worse, I don't thing BOINC is updating the values internally, either. One project got up to +9274 STD, and has gradually taken over all 8 cores (it's on 3/8ths resource share). As the first task finished, it jumped immediately to -2987 STD. The most recent task switch happened less than 30 minutes before the recalculation, so if the debts had been up-to-date, there's no way it should have taken an extra core and pre-empted a running project. The project which got the core released by the task end was already at +7627 STD. |
|
4856)
Message boards :
BOINC client :
Octo Mac Pro - only 4 jobs running
(Message 9828)
Posted 22 Apr 2007 by Richard Haselgrove Post: Simple first thing to check - what's the 'maximum number of CPUs' preference on the project you've most recently joined it to/updated from? |
|
4857)
Message boards :
BOINC Manager :
Boinc isn't using my CPUs it seems.
(Message 9727)
Posted 20 Apr 2007 by Richard Haselgrove Post: I've set Boinc to use up to 100% and I'm running the climateprediction.net project. Each of the four instances that the BOINC manager has created show time remaining in the thousands of hours. 75 days per work unit can't be right I thought so I checked my CPU usage and boinc is using 0.20% and BOINCManager is using 2%. 75 days per workunit is absolutely right. I'm running 4 CPDN workunits on a Dell Windows workstation (along with SETI and Einstein). My quads only run at 1.86Ghz, so my 110 day estimates (at 50+% done) are in line with yours. Again, BOINC using 0.20% and BOINCManager using 2% is absolutely right. BOINC doesn't actually do anything at all (like all bosses, it directs and controls other workers, but contributes nothing at all to productivity). The real question is, how much CPU %age do the CPDN worker applications get when you're not clockwatching? As others have said, BOINC might be telling the workers to down tools and make the place look tidy while the owner (you!) is around. |
|
4858)
Message boards :
BOINC client :
Linux Installation Problem
(Message 9514)
Posted 11 Apr 2007 by Richard Haselgrove Post: It's beginning to sound as if you need to do some detective work, to find out whether it's a corrupt download or a broken (un)zipper. Do you have any chance of downloading the package on another machine - preferably using a different ISP - and transferring via CD or memory stick? That would help to track down whether it's getting mangled in transit. And for the other half of the test, have you tried to install any gzipped packages from other sources recently, to test that all your local tools are working properly? |
|
4859)
Message boards :
BOINC Manager :
dualboot machines: same work units in both partitions????
(Message 9502)
Posted 11 Apr 2007 by Richard Haselgrove Post: Do you have (or could you make) a third 'data' partition visible to both OSs? Or are you saying that the same C:\BOINC folder is visible all the time, whichever OS is in control? Provided you do have a single folder which is visible all the time, and both OSs are flavours of Windows, you should be able to install BOINC to the common repository from both OSs, and crunch the same WUs with the same apps in either mode. Wouldn't work for a Linux/Windows dual boot, though. |
|
4860)
Message boards :
BOINC client :
Using BBCode
(Message 9258)
Posted 31 Mar 2007 by Richard Haselgrove Post: And arguably, tho I didn't mention it before, compliance with the law in England & Wales (Disability Discrimination Act). It opens up liability to litigation if we (*) don't take reasonable steps to include disabled people, and as the standards have been pushing ALT text for the convenience of blind users ever since the IMG tag came in, nobody can claim it is an unreasonable request. Now that is a huge pending litigation mountain... And while we're on the subject of the DDA, whocrazy's post on the BOINC Manager board alerts us to at least as serious - IMHO, possibly more serious - breach of the same Act. I took Microsoft to task under the Act when they started asking users to re-key the code from the Certificate of Authenticity, as part of an early version of the Genuine Windows Advantage anti-piracy scheme. I'm pleased to say they took it seriously, and I think they changed the validation system - at least, I haven't been caught out by it again. I think the Open Source community should strive to achieve the same standards as we would expect of Microsoft, Apple et al. |
|
4861)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8985)
Posted 22 Mar 2007 by Richard Haselgrove Post: Batman's report of the error turns out to be in a Crunch3r optimisation (BOINC v5.9.0.32 dated 08-Jan-2007 15:08), so it rather pre-dates this discussion. Anyone seen any signs of it happening in BOINC v5.8.15 (19-Feb-2007 18:19) or BOINC v5.8.16 (01-Mar-2007 08:30)? (or on any platform other than Windows?) |
|
4862)
Message boards :
BOINC client :
Client re-enables work requests
(Message 8948)
Posted 21 Mar 2007 by Richard Haselgrove Post: Could a similar approach be used for this problem (simple GUI discarding custom settings in global_prefs_override.xml), discussed a few days ago in the Number of CPUs changing thread? David did say his new technique might be useful elsewhere.... |
|
4863)
Message boards :
Web interfaces :
search box on these boards
(Message 8828)
Posted 17 Mar 2007 by Richard Haselgrove Post: Can we have a search box on these message boards please? SETI@Home have, quietly and without fanfare, recently introduced a search box at the top level of the 'Message boards' menu - previously it was only shown on 'Questions and Answers'. Let's hope that change makes it into the standard template, and gets copied to all other projects - who knows, even to this "technology demonstrator" site for the whole BOINC system /-) ? |
|
4864)
Message boards :
BOINC Manager :
Trojan boinc installation by rogue member
(Message 8775)
Posted 15 Mar 2007 by Richard Haselgrove Post: Is this where you saw W's name? (I'm not even prepared to use his full name in a post because he's probably also a Google-rating junkie.) Yes, that where we saw it - but it's just a random pick from http://boinc.berkeley.edu/chart_list.php, where's he's static at number 11. |
|
4865)
Message boards :
BOINC Manager :
Trojan boinc installation by rogue member
(Message 8771)
Posted 15 Mar 2007 by Richard Haselgrove Post: What a shame! Actually, it's not "volunteer of the day", but "random volunteer of the moment" - just try hitting 'refresh' when viewing the front page. But I'm not surprised you jumped to that conclusion - it gave me quite a turn the first time I saw his/her name on the page, just after the story had broken. But on a more serious note - shouldn't the account have been purged by now? It's almost a month since the problem was first spotted. [Edit - especially since the wording visitors see is "Wate is contributing 283 billion floating-point operations per second (GFLOPS)". Does the fact that his/her project percentages add up to 10.1% mean that some projects have purged the data, or just that there's a margin-width bug in the pie-chart generator? /edit] |
|
4866)
Message boards :
BOINC client :
Number of CPUs changing
(Message 8769)
Posted 15 Mar 2007 by Richard Haselgrove Post: I hear what you say, but that doesn't mean I like the underlying implications :-( |
|
4867)
Message boards :
BOINC client :
Number of CPUs changing
(Message 8765)
Posted 15 Mar 2007 by Richard Haselgrove Post: But yes, any preferences you set in a hand written global_preferences_Override.xml file are overwritten with the Simple GUI's override file. The Simple GUI isn't for the professionals. It's for the people starting with BOINC and expecting to find some basic preferences setting ability. Whatever the target audience, that's just plain bad programming - a bug, in my book. If the simple GUI wants to 'own' a settings file (i.e. have permission to overwrite/delete things put there by some other process), then it should have its own file - it shouldn't use a file designated for professionals to use in a different way. OTOH, when I started writing this sort of code (about Windows 3.1 !), we used .INI configuration files for this sort of thing. The Windows API had - and still has, up to XP SP2, not tried it on Vista yet - ProfileFile calls which allow you to read and set individual items in the file, without disturbing the rest of the contents. Surely by now someone's got a library which can do the same thing with .XML files? It'll make things much more maintainable in the future if they would put the underlying file handling on a firmer footing first, before adding the extra bells and whistles. |
|
4868)
Message boards :
BOINC client :
Dedicated Bonic computer question
(Message 8619)
Posted 8 Mar 2007 by Richard Haselgrove Post: And on SETI. Approx 25% improvement for Quad channel over Dual channel. |
|
4869)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8553)
Posted 5 Mar 2007 by Richard Haselgrove Post: (best take the code out... read it yourself on file_upload_scheduler.c Yep, that's where it happens - I think we hit the return error at around line 243. But we have to debug on from that point: why does the user's PC upload a smaller number of bytes than it originally declared that it wanted to upload? Answer: because the file itself (on the user's PC) has been modified between the CGI request, and the actual upload: because something - a still unknown something, in my book - has triggered the science application to re-start computation on a WU that it should know it's finished with. |
|
4870)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8537)
Posted 5 Mar 2007 by Richard Haselgrove Post: Jord, I think the difference between us is that you're in 'explanation' mode, while I'm in 'debug' mode, trying to use the explanations to get back to the orginial cause of the problem. [file_xfer] Temporarily failed upload of 11no03aa.4732.8626.304816.3.23_0_0: transient upload error Obviously, I accept that the words 'transient upload error' are in the source code, and will be output to the scheduler response and hence the local message log under failure conditions. So what was the failure? [error] Error on file upload: socket read incomplete: asked for 12226, got 8963 Ah, the server asked for 12226 bytes, but the upload stopped uploading at 8963 bytes - some sort of premature end-of-file signal. Where did those figures come from? How did the server know to expect 12226 byres? I'm assuming that the initial contact comes from the client PC: something along the lines of "Hey, I've finished some work and I've got something for you - it'll be xxx bytes, and called yyy - let me know when you're ready for it." (May be a single message, or may be broken into sub-messages - doesn't really matter at this level). The server then sets up the storage space, and requests the upload when it's ready. BOINC then uploads the file, but it's a different size to the one it originally signalled. Why? [The next clue comes from Josef Segur in a BOINC Alpha email. Thanks to Byron for quoting it on SETI Beta at message 14952] I've seen the problem most often with SETI WUs. The SETI upload file has the structure {Copy of WU header}{one or more, or no, data of interest}{final summary with 'best of' information}. The undersize uploads (I've preserved some) have the first two sections, but the final summary is missing: this is the bit that Joe couldn't understand. My logs earlier in this thread - I won't bother quoting them again - provide the answer to Joe's question. The WUs finish computation (and presumably BOINC stores the result file size at this point): then the same WU starts computation all over again, overwriting the result file with a new copy of the header and, possibly, some of the data - but crucially, no final summary, because the second computation hasn't finished yet. Meanwhile, BOINC is continuing to process the first 'exit - finished' event, and offers the file for upload. So my question is "Why - and under what circumstances - would BOINC re-start computation on a WU which has just exited with a normal 'finished' status?". My observation is that two results from the same project finishing in quick succession (no more than 1 second apart) is a necessary but not sufficient event trigger. Nicolas's reply to Pepo and myself earlier in this thread was 'David Anderson says it's fixed in BOINC 5.8.11', but he couldn't or wouldn't point to a specific changelog entry which addressed these symptoms: the two entries cited in this thread refer to different abend situations (though the 2 Feb entry might cover this one too). In the meantime, I've posted two counter-examples which show that that the problem was not completely fixed in BOINC 5.8.11 - which is why I keep banging on about it! (sorry!) I've not seen any problems in .15 (or 16, or whatever we're up to now), but I'll keep monitoring and post anything I see. In the meantime, if you come across anything while wou're reading the new code for documentation purposes, please let us know. Maybe we can meet in the middle and nail this thing once and for all. |
|
4871)
Message boards :
BOINC Manager :
i BET NONE OF YOU CAN FIX THIS PROBLEM!!
(Message 8534)
Posted 5 Mar 2007 by Richard Haselgrove Post: If you look in your BOINC folder, you should see an XML file called something like 'master_malaria....' (I don't know the exact name, as I don't run that project). If you open that file, somewhere inside it you should find a pair of <scheduler></scheduler> tags. For SETI, it's the only line in the file: for Einstein, it's almost at the bottom of a copy of the home page, just above the footers. For Malaria, ??? The scheduler url between the tags would be a good place to start your research. |
|
4872)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8518)
Posted 4 Mar 2007 by Richard Haselgrove Post: Ageless, Thanks for dropping by to visit this thread, and for your helpful, as always, reminder about the FAQs. But I don't think it can be the explanation for this class of events. 1) The general sequence of events that I'm seeing is: WU finishes computation. Same WU starts computation again. BOINC attempts upload of results file. - so the initial cause of the problem happens on the local PC, before any communication takes place with the project servers. Of course, all subsequent attempts at uploading fail, but that seems to be the result of the initial problem, not its cause. 2) Whenever I've had this problem, I've found that aborting the upload transfer clears the problem - the WU is put into a 'ready to report' state, and is cleared at the next scheduler contact. Of course, there's no credit for the WU, but the local BOINC can continue crunching error-free. No reset, not even a restart, and certainly no need of attention by project admins. I think we must be talking about different problems. FWIW, I have not yet experienced this problem since upgrading to BOINC 5.8.15, nor have I picked up on any other reports on BOINC later than 5.8.11. We did have another one reported by Ulrich Metzner on SETI Main this morning, but he's come back to say he's using 5.7.5, so it doesn't add much to the diagnosis. |
|
4873)
Message boards :
BOINC Manager :
"Exit 0 status no finished file"
(Message 8443)
Posted 26 Feb 2007 by Richard Haselgrove Post: I think David's comment only related to the rash of spurious zero status messages every time a project was paused for any reason (round-robin switch or whatever). The original messages are still there: I've just had one, tracked down to an old-fashioned clock synchronisation. |
|
4874)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8430)
Posted 25 Feb 2007 by Richard Haselgrove Post: We've now got two full logs of this happening under 5.8.11, and they both show the same pattern as before: two WUs from the same project finish within a second: the second one is re-started by BOINC. Andy's (WinterKnight) is slightly different, in that the upload started just before the computation restart, but the end result was the same. The second case was logged by 'The Eternal' on the SETI Beta board at message 15232, but I'll copy the log here to keep them together. Nicolas, could you give us a little bit more on the context in which David said it was fixed in 5.8.11? Could the fix possibly have been delayed even beyond that release? WinterKnight's log: 2007-02-23 00:08:38 [SETI@home] Starting 11no03aa.4732.20290.1022148.3.153_0 2007-02-23 00:08:38 [SETI@home] Starting task 11no03aa.4732.20290.1022148.3.153_0 using setiathome_enhanced version 515 2007-02-23 00:08:45 [SETI@home] Computation for task 11no03aa.4732.20290.1022148.3.152_3 finished 2007-02-23 00:08:45 [SETI@home] Starting 11no03aa.4732.20290.1022148.3.160_0 2007-02-23 00:08:45 [SETI@home] Starting task 11no03aa.4732.20290.1022148.3.160_0 using setiathome_enhanced version 515 2007-02-23 00:08:46 [SETI@home] Computation for task 11no03aa.4732.20290.1022148.3.153_0 finished 2007-02-23 00:08:47 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.20290.1022148.3.152_3_0 2007-02-23 00:08:48 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.20290.1022148.3.153_0_0 2007-02-23 00:08:50 [SETI@home] Starting 11no03aa.4732.20290.1022148.3.153_0 2007-02-23 00:08:50 [SETI@home] Starting task 11no03aa.4732.20290.1022148.3.153_0 using setiathome_enhanced version 515 2007-02-23 00:08:52 [SETI@home] Computation for task 11no03aa.4732.20290.1022148.3.160_0 finished 2007-02-23 00:08:52 [SETI@home] Starting 11no03aa.4732.20290.1022148.3.193_2 2007-02-23 00:08:52 [SETI@home] Starting task 11no03aa.4732.20290.1022148.3.193_2 using setiathome_enhanced version 515 2007-02-23 00:08:52 [SETI@home] Starting 11no03aa.4732.20290.1022148.3.183_0 2007-02-23 00:08:52 [SETI@home] Starting task 11no03aa.4732.20290.1022148.3.183_0 using setiathome_enhanced version 515 2007-02-23 00:08:54 [SETI@home] [file_xfer] Finished upload of file 11no03aa.4732.20290.1022148.3.152_3_0 2007-02-23 00:08:54 [SETI@home] [file_xfer] Throughput 7082 bytes/sec 2007-02-23 00:08:54 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.20290.1022148.3.160_0_0 2007-02-23 00:08:58 [SETI@home] [file_xfer] Finished upload of file 11no03aa.4732.20290.1022148.3.160_0_0 2007-02-23 00:08:58 [SETI@home] [file_xfer] Throughput 23188 bytes/sec 2007-02-23 00:08:59 [SETI@home] Computation for task 11no03aa.4732.20290.1022148.3.193_2 finished 2007-02-23 00:08:59 [SETI@home] Starting 11no03aa.4732.20290.1022148.3.226_1 2007-02-23 00:08:59 [SETI@home] Starting task 11no03aa.4732.20290.1022148.3.226_1 using setiathome_enhanced version 515 2007-02-23 00:09:00 [SETI@home] Computation for task 11no03aa.4732.20290.1022148.3.183_0 finished 2007-02-23 00:09:01 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.20290.1022148.3.193_2_0 2007-02-23 00:09:04 [SETI@home] [file_xfer] Finished upload of file 11no03aa.4732.20290.1022148.3.193_2_0 2007-02-23 00:09:04 [SETI@home] [file_xfer] Throughput 17441 bytes/sec 2007-02-23 00:09:04 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.20290.1022148.3.183_0_0 2007-02-23 00:09:05 [SETI@home] Starting 11no03aa.4732.20290.1022148.3.183_0 2007-02-23 00:09:05 [SETI@home] Starting task 11no03aa.4732.20290.1022148.3.183_0 using setiathome_enhanced version 515 2007-02-23 00:09:06 [SETI@home] Computation for task 11no03aa.4732.20290.1022148.3.226_1 finished 2007-02-23 00:09:06 [SETI@home] Resuming task 18se03aa.2009.5280.553406.3.26_2 using setiathome_enhanced version 515 2007-02-23 00:09:06 [SETI@home] Starting 09ap00aa.21686.30674.54816.3.32_1 2007-02-23 00:09:06 [SETI@home] Starting task 09ap00aa.21686.30674.54816.3.32_1 using setiathome_enhanced version 515 2007-02-23 00:09:09 [SETI@home] Sending scheduler request: To fetch work 2007-02-23 00:09:09 [SETI@home] Requesting 12843 seconds of new work, and reporting 4 completed tasks 2007-02-23 00:09:15 [SETI@home] Scheduler RPC succeeded [server version 507] 2007-02-23 00:09:15 [SETI@home] Deferring communication for 11 sec 2007-02-23 00:09:15 [SETI@home] Reason: requested by project 2007-02-23 00:09:30 [SETI@home] Sending scheduler request: To fetch work 2007-02-23 00:09:30 [SETI@home] Requesting 1147 seconds of new work 2007-02-23 00:09:35 [SETI@home] Scheduler RPC succeeded [server version 507] 2007-02-23 00:09:35 [SETI@home] Deferring communication for 11 sec 2007-02-23 00:09:35 [SETI@home] Reason: requested by project 2007-02-23 00:18:50 [SETI@home] Sending scheduler request: To fetch work 2007-02-23 00:18:50 [SETI@home] Requesting 28 seconds of new work 2007-02-23 00:18:55 [---] Project communication failed: attempting access to reference site 2007-02-23 00:18:55 [SETI@home] [file_xfer] Temporarily failed upload of 11no03aa.4732.20290.1022148.3.153_0_0: http error 2007-02-23 00:18:55 [SETI@home] Backing off 1 min 0 sec on upload of file 11no03aa.4732.20290.1022148.3.153_0_0 The Eternal's log: 2007-02-21 06:03:59 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.8626.304816.3.18_0_0 2007-02-21 06:03:59 [SETI@home] Computation for task 11no03aa.4732.8626.304816.3.22_0 finished . 2007-02-21 06:03:59 [SETI@home] Starting 11no03aa.4732.8626.304816.3.23_0 . 2007-02-21 06:03:59 [SETI@home] Starting task 11no03aa.4732.8626.304816.3.23_0 using setiathome_enhanced version 519 2007-02-21 06:04:01 [SETI@home] [file_xfer] Finished upload of file 11no03aa.4732.8626.304816.3.18_0_0 2007-02-21 06:04:01 [SETI@home] [file_xfer] Throughput 67350 bytes/sec 2007-02-21 06:04:01 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.8626.304816.3.22_0_0 2007-02-21 06:04:04 [SETI@home] [file_xfer] Finished upload of file 11no03aa.4732.8626.304816.3.22_0_0 2007-02-21 06:04:04 [SETI@home] [file_xfer] Throughput 214138 bytes/sec 2007-02-21 06:13:28 [SETI@home Beta Test] Sending scheduler request: To fetch work 2007-02-21 06:13:28 [SETI@home Beta Test] Requesting 107916 seconds of new work 2007-02-21 06:13:33 [SETI@home Beta Test] Scheduler RPC succeeded [server version 509] 2007-02-21 06:13:33 [SETI@home Beta Test] Deferring communication for 7 sec 2007-02-21 06:13:33 [SETI@home Beta Test] Reason: requested by project 2007-02-21 06:13:33 [SETI@home Beta Test] Deferring communication for 42 min 44 sec 2007-02-21 06:13:33 [SETI@home Beta Test] Reason: no work from project . 2007-02-21 06:23:23 [SETI@home] Computation for task 11no03aa.4732.8626.304816.3.29_0 finished 2007-02-21 06:23:23 [SETI@home] Resuming task 19au03aa.16709.20786.704834.3.98_1 using setiathome_enhanced version 519 . 2007-02-21 06:23:24 [SETI@home] Computation for task 11no03aa.4732.8626.304816.3.23_0 finished 2007-02-21 06:23:25 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.8626.304816.3.29_0_0 . 2007-02-21 06:23:25 [SETI@home] Starting 11no03aa.4732.8626.304816.3.23_0 . 2007-02-21 06:23:25 [SETI@home] Starting task 11no03aa.4732.8626.304816.3.23_0 using setiathome_enhanced version 519 . 2007-02-21 06:23:26 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.8626.304816.3.23_0_0 2007-02-21 06:23:27 [SETI@home] [file_xfer] Finished upload of file 11no03aa.4732.8626.304816.3.29_0_0 2007-02-21 06:23:27 [SETI@home] [file_xfer] Throughput 51216 bytes/sec 2007-02-21 06:23:28 [SETI@home] [error] Error on file upload: socket read incomplete: asked for 12226, got 7426: No such file or directory . 2007-02-21 06:23:28 [SETI@home] [file_xfer] Temporarily failed upload of 11no03aa.4732.8626.304816.3.23_0_0: transient upload error . 2007-02-21 06:23:28 [SETI@home] Backing off 1 min 0 sec on upload of file 11no03aa.4732.8626.304816.3.23_0_0 . 2007-02-21 06:24:28 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.8626.304816.3.23_0_0 2007-02-21 06:24:30 [SETI@home] [error] Error on file upload: socket read incomplete: asked for 12226, got 8963: No such file or directory . 2007-02-21 06:24:30 [SETI@home] [file_xfer] Temporarily failed upload of 11no03aa.4732.8626.304816.3.23_0_0: transient upload error . 2007-02-21 06:24:30 [SETI@home] Backing off 1 min 0 sec on upload of file 11no03aa.4732.8626.304816.3.23_0_0 2007-02-21 06:25:24 [SETI@home] Sending scheduler request: To fetch work 2007-02-21 06:25:24 [SETI@home] Requesting 13653 seconds of new work, and reporting 7 completed tasks 2007-02-21 06:25:29 [SETI@home] Scheduler RPC succeeded [server version 507] 2007-02-21 06:25:29 [SETI@home] Deferring communication for 11 sec 2007-02-21 06:25:29 [SETI@home] Reason: requested by project 2007-02-21 06:25:30 [SETI@home] [error] garbage_collect(); still have active task for acked result 11no03aa.4732.8626.304816.3.19_3; state 9 . 2007-02-21 06:25:30 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.8626.304816.3.23_0_0 2007-02-21 06:25:30 [SETI@home] Computation for task 11no03aa.4732.8626.304816.3.19_3 finished 2007-02-21 06:25:36 [SETI@home] [file_xfer] Started download of file 15au03aa.20522.4642.779818.3.216 2007-02-21 06:25:38 [SETI@home] [error] Error on file upload: socket read incomplete: asked for 12226, got 8963: No such file or directory . 2007-02-21 06:25:38 [SETI@home] [file_xfer] Temporarily failed upload of 11no03aa.4732.8626.304816.3.23_0_0: transient upload error . 2007-02-21 06:25:38 [SETI@home] Backing off 1 min 0 sec on upload of file 11no03aa.4732.8626.304816.3.23_0_0 |
|
4875)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8406)
Posted 24 Feb 2007 by Richard Haselgrove Post: Here are the relevant lines from my C2D's stdoutdae.txt file, Email received OK, thanks, and the data is all there. At a first glance, it's a slightly different pattern, but it's still a finish-restart-upload sequence. And as SETI result 481839220 confirms, this was under BOINC v5.8.11 I'll post more of the log when I've gone through it and highlighted the relevant points. |
|
4876)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8403)
Posted 24 Feb 2007 by Richard Haselgrove Post: Here are the relevant lines from my C2D's stdoutdae.txt file, I'd like to go a little further back, if that's possible, to see the actual computation stages for 11no03aa.4732.20290.1022148.3.153 Or if you could forward the whole thing to initial dot surname at btinternet dot com, I'll have a look through it here. |
|
4877)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8371)
Posted 23 Feb 2007 by Richard Haselgrove Post: Pepo - Great minds think alike (and cross-post), LOL |
|
4878)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8370)
Posted 23 Feb 2007 by Richard Haselgrove Post: Still reading! The closest would seem to be: David 2 Feb 2007 But that change should have been in 5.8.9 - 05 Feb. And the bug is still in that version. Can I re-report it, please? |
|
4879)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8368)
Posted 23 Feb 2007 by Richard Haselgrove Post: Many thanks for that link - I'll be hanging on to it for future reference. If the fix was first in 5.8.11, and I saw the problem in 5.8.9, then the checkin would be between 05 Feb and 09 Feb, right? (There was no 5.8.10 for Windows). David wrote 8 checkin notes during that timeframe. I can't reconcile any of them with the symptoms I've been describing. I'll read a few more entries before and after tomorrow, when I've had some sleep (and when I'm sober!) |
|
4880)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8366)
Posted 23 Feb 2007 by Richard Haselgrove Post: According to David Anderson, this problem was fixed on 5.8.11. Looking at the datestamps on /dl, 5.8.11 was released on 09 Feb. So how does that marry with the checkin quoted by Pepo, DA on 15 Feb? Could somebody, anybody, please post a public link to the full checkin changelog? And if there isn't one, make one? [/rant]Sorry, posts at this time of day/night tend to be made under the affluence of incohol. |
|
4881)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8357)
Posted 22 Feb 2007 by Richard Haselgrove Post: OK, I've upgraded to 5.8.15, which is the latest at the time of writing. It's difficult to reproduce it to order, because it's so critical on timing (2 WUs finish within 1 second), but I'll certainly report whatever I see. Would any extra logging help? |
|
4882)
Message boards :
BOINC client :
Error on file upload: Socket Read incomplete
(Message 8348)
Posted 22 Feb 2007 by Richard Haselgrove Post: Is anyone making any progress on this (increasingly common) error? Every instance I've investigated in detail shares the same characteristsics: 2 WUs from the same project finish 1 second apart. The second WU then re-starts before uploading, when one would expect a different WU to start. Here's a standard log showing the full sequence of events: I'm happy to do any enhanced logging which might help to track it down. Edit: this particular log is from BOINC 5.8.9, running as a service under Windows XP. 2007-02-21 20:12:11 [SETI@home] Sending scheduler request: To fetch work 2007-02-21 20:12:11 [SETI@home] Requesting 13201 seconds of new work, and reporting 1 completed tasks 2007-02-21 20:12:17 [SETI@home] Scheduler RPC succeeded [server version 507] 2007-02-21 20:12:17 [SETI@home] Deferring communication for 11 sec 2007-02-21 20:12:17 [SETI@home] Reason: requested by project 2007-02-21 20:12:19 [SETI@home] [file_xfer] Started download of file 11no03aa.4732.12274.92334.3.38 2007-02-21 20:12:19 [SETI@home] [file_xfer] Started download of file 11no03aa.4732.12274.92334.3.43 2007-02-21 20:12:26 [SETI@home] [file_xfer] Finished download of file 11no03aa.4732.12274.92334.3.38 2007-02-21 20:12:26 [SETI@home] [file_xfer] Throughput 68354 bytes/sec 2007-02-21 20:12:26 [SETI@home] [file_xfer] Finished download of file 11no03aa.4732.12274.92334.3.43 2007-02-21 20:12:26 [SETI@home] [file_xfer] Throughput 68756 bytes/sec 2007-02-21 20:12:26 [SETI@home] [file_xfer] Started download of file 11no03aa.4732.12274.92334.3.44 2007-02-21 20:12:26 [SETI@home] [file_xfer] Started download of file 11no03aa.4732.12274.92334.3.45 2007-02-21 20:12:27 [SETI@home] Starting 11no03aa.4732.12274.92334.3.38_3 2007-02-21 20:12:27 [SETI@home] Starting task 11no03aa.4732.12274.92334.3.38_3 using setiathome_enhanced version 517 2007-02-21 20:12:27 [SETI@home] Starting 11no03aa.4732.12274.92334.3.43_3 2007-02-21 20:12:27 [SETI@home] Starting task 11no03aa.4732.12274.92334.3.43_3 using setiathome_enhanced version 517 2007-02-21 20:12:27 [SETI@home] Starting 01no03aa.5768.1328.765888.3.102_0 2007-02-21 20:12:27 [SETI@home] Starting task 01no03aa.5768.1328.765888.3.102_0 using setiathome_enhanced version 517 2007-02-21 20:12:27 [SETI@home] Starting 01no03aa.5768.1233.373588.3.77_0 2007-02-21 20:12:27 [SETI@home] Starting task 01no03aa.5768.1233.373588.3.77_0 using setiathome_enhanced version 517 2007-02-21 20:12:36 [SETI@home] [file_xfer] Finished download of file 11no03aa.4732.12274.92334.3.44 2007-02-21 20:12:36 [SETI@home] [file_xfer] Throughput 44306 bytes/sec 2007-02-21 20:12:36 [SETI@home] [file_xfer] Finished download of file 11no03aa.4732.12274.92334.3.45 2007-02-21 20:12:36 [SETI@home] [file_xfer] Throughput 44219 bytes/sec 2007-02-21 20:12:36 [SETI@home] [file_xfer] Started download of file 15au03aa.20522.7056.765914.3.4 2007-02-21 20:12:38 [SETI@home] Starting 11no03aa.4732.12274.92334.3.44_3 2007-02-21 20:12:38 [SETI@home] Starting task 11no03aa.4732.12274.92334.3.44_3 using setiathome_enhanced version 517 2007-02-21 20:12:38 [SETI@home] Starting 11no03aa.4732.12274.92334.3.45_0 2007-02-21 20:12:38 [SETI@home] Starting task 11no03aa.4732.12274.92334.3.45_0 using setiathome_enhanced version 517 2007-02-21 20:12:46 [SETI@home] [file_xfer] Finished download of file 15au03aa.20522.7056.765914.3.4 2007-02-21 20:12:46 [SETI@home] [file_xfer] Throughput 47095 bytes/sec 2007-02-21 20:12:53 [SETI@home] Computation for task 11no03aa.4732.12274.92334.3.38_3 finished 2007-02-21 20:12:54 [SETI@home] Computation for task 11no03aa.4732.12274.92334.3.43_3 finished 2007-02-21 20:12:55 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.12274.92334.3.38_3_0 2007-02-21 20:12:55 [SETI@home] Starting 11no03aa.4732.12274.92334.3.43_3 2007-02-21 20:12:55 [SETI@home] Starting task 11no03aa.4732.12274.92334.3.43_3 using setiathome_enhanced version 517 2007-02-21 20:12:56 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.12274.92334.3.43_3_0 2007-02-21 20:12:58 [SETI@home] Computation for task 11no03aa.4732.12274.92334.3.44_3 finished 2007-02-21 20:12:58 [SETI@home] Resuming task 19au03aa.16709.19889.648582.3.172_0 using setiathome_enhanced version 517 2007-02-21 20:12:58 [SETI@home] Resuming task 19au03aa.16709.20530.392322.3.171_3 using setiathome_enhanced version 517 2007-02-21 20:12:59 [SETI@home] Computation for task 11no03aa.4732.12274.92334.3.45_0 finished 2007-02-21 20:13:03 [SETI@home] [file_xfer] Finished upload of file 11no03aa.4732.12274.92334.3.38_3_0 2007-02-21 20:13:03 [SETI@home] [file_xfer] Throughput 4505 bytes/sec 2007-02-21 20:13:03 [SETI@home] [file_xfer] Started upload of file 11no03aa.4732.12274.92334.3.44_3_0 2007-02-21 20:13:09 [SETI@home] [error] Error on file upload: socket read incomplete: asked for 16382, got 7421: No such file or directory 2007-02-21 20:13:09 [SETI@home] [file_xfer] Temporarily failed upload of 11no03aa.4732.12274.92334.3.43_3_0: transient upload error |
|
4883)
Message boards :
BOINC client :
Alpha 5.7.5 - upload error
(Message 6684)
Posted 28 Nov 2006 by Richard Haselgrove Post: Alpha email list is here. Thanks for the link, but I think I'll keep in on the boards for now. I hope I may be able to contribute more to testing now I've got more cores, but I'll let it settle down a bit first. I'm going to post it on SETI later, and possibly on lunatics.at (since it was one of Simon's apps that went belly-up). But I do read those (SETI) boards quite closely, and I haven't seen any similar report that I can recall. The unusual thing about the event was two results finishing within a second of each other, and I wondered if BOINC (or SETI, or Windows) got the two result files muddled up. But that c**p error message is certainly all down to BOINC! |
|
4884)
Message boards :
BOINC client :
Alpha 5.7.5 - upload error
(Message 6682)
Posted 28 Nov 2006 by Richard Haselgrove Post: (Yes I know it's alpha, but the alpha boards don't seem to be an active location for bug reports!) Installed 5.7.5 as a service under Win XP SP2 on a brand-new dual Xeon workstation - 8 cores. Attached to SETI (only) for the time being. First two WUs finished almost at the same time, and I got: 28/11/2006 12:48:47|SETI@home|Started upload of file 07jn03aa.5768.25602.767320.3.93_3_0 28/11/2006 12:48:47||Starting 07jn03aa.5768.25602.767320.3.92_0 28/11/2006 12:48:47|SETI@home|Starting task 07jn03aa.5768.25602.767320.3.92_0 using setiathome_enhanced version 517 28/11/2006 12:48:48|SETI@home|Started upload of file 07jn03aa.5768.25602.767320.3.92_0_0 28/11/2006 12:48:52|SETI@home|Error on file upload: socket read incomplete: asked for 11649, got 7425: No such file or directory 28/11/2006 12:48:52|SETI@home|Finished upload of file 07jn03aa.5768.25602.767320.3.93_3_0 28/11/2006 12:48:52|SETI@home|Throughput 9801 bytes/sec 28/11/2006 12:48:52|SETI@home|Temporarily failed upload of 07jn03aa.5768.25602.767320.3.92_0_0: transient upload error in the message log. It's a badly-written error message - from other checking, I think it means the uploaded file was shorter than expected (7425 bytes instead of 11649) I also got some very odd information in the 'Tasks' tab of BOINC Manager: BOINC Manager - (localhost) CPU time: 00:01:25 Progress: 1.897% To completion: --- Report deadline: 02/12/2006 19:14:43 Status: Uploading After a BOINC restart, I got: 28/11/2006 13:21:30|SETI@home|State file error: result 07jn03aa.5768.25602.767320.3.92_0 is in wrong state and CPU time: 01:43:28 Progress: 100.000% To completion: --- Report deadline: 02/12/2006 19:14:43 Status: Uploading - but the upload kept on failing at about 75%, with the same error message in the log. In the end, I aborted it, and reverted down to BOINC 5.6.5 for the time being. I have fuller notes, and I'm willing to do more testing, if it would help. |
|
4885)
Message boards :
BOINC Manager :
BOINC distribution under the GNU Public License (GPL)
(Message 5953)
Posted 8 Oct 2006 by Richard Haselgrove Post: Eric - many thanks for your prompt and helpful response. I'm sorry you had to register as a new user in order to be able to post! And a warm welcome to my sparring partner from Einstein. But apart from that, my attempt to start a broader debate seems to have failed spectacularly. Let's bump the thread a little. IANAL, but... Here in the UK, it's clear that a club or association is not recognised as a legal entity, unless it chooses to go through one of the formal processes of incorporation - usually as a company, but there are others. Disregarding the confrontational example of sueing/being sued, the commonest practical implication relates to property (real estate). An unincorporated club cannot enter into a contract to buy or lease land or buildings. The workround is for the members to appoint "Trustees" to hold the property on their behalf. These are not usually the executive or 'point' officers (Chair, Secretary, Webmaster) - partly because officerships often change by election each year, and it becomes expensive to transfer property every time. Also, the trustees act as legal long-stop to pick up the pieces if it all goes pear-shaped: members failing to elect a committee that a working majority have confidence in, for example. And it would be the trustees who sue and are sued on behalf of the members. So I suppose it would be possible for individuals to hold a GPL licence "in trust" for the other members of the organisation, but it all sounds a bit tortuous, and presupposes the existence of a formal trust deed. The club/team example is not the only place where the "You" in the GPL is ambiguous. Towards the end of David and Eric's recent interview with Lookers, David talks about plans to expand the BOINC user-base by promulgating it more widely within UCB. I'm sure that in practice he'll want to do that under GPL, but for argument's sake, how widely could you circulate a 'closed' UCB version under GPL?
Computer Studies students on a distributed computing course? Almost certainly Full-time enrolled students on campus? Probably Part-time students on distance-learning courses, studying at home? er... Day-release or sandwich course students, studying at work in their employers' time? Non-tenured research associates? Alumni?
|
|
4886)
Message boards :
BOINC Manager :
BOINC distribution under the GNU Public License (GPL)
(Message 5913)
Posted 5 Oct 2006 by Richard Haselgrove Post: I have been having a spirited debate in the Einstein forum (this thread), sparked by the OP's throwaway remark about the development of a 'closed' version of BOINC, with extra facilities only available to members of one team, and restricted from distribution to the community at large. The question arises whether this is permitted under the GPL. However, I find that I'm on rather shaky ground, because it seems to me that there are serious defects in the "official" distribution of BOINC from this website under GPL. For instance, the 'BOINC software development' page claims that BOINC software is distributed under the lesser GPL, which turns out to be false - read the 'copying' file in the root of the source tarball. Also, the distributed executable (Windows package only inspected) contains a copyright notice, but none of the other information resources required under GPL - limited warranty, notification of redistribution rights, and the text of the GPL itself. Could anyone from the project please
Comment on and review actual practice in relation to that policy
|
Previous 20
Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document
under the terms of the GNU Free Documentation License,
Version 1.2 or any later version published by the Free Software Foundation.