Posts by l008com
|
1)
Message boards :
Questions and problems :
All Milkyway@Home GPU WU"s get Computation error
(Message 106167)
Posted 22 Nov 2021 by l008com Post: Here is a sample failed work unit: https://milkyway.cs.rpi.edu/milkyway/result.php?resultid=425929132 |
|
2)
Message boards :
Questions and problems :
All Milkyway@Home GPU WU"s get Computation error
(Message 106166)
Posted 22 Nov 2021 by l008com Post: I'm trying to get some GPU work going on my Mac Pro. It is running 10.13.6 and has a GeForce GT120. Not the best card in the world but it's all there is. I have the latest nVidia drivers installed and this card should be supported by the driver, opencl, and the project. And the project does indeed download work units. But almost instantly, these work units give a Computation Error. Every one. Over and over. What might be going on here? I'm not sure where to look to even try to troubleshoot this problem. |
|
3)
Message boards :
BOINC Manager :
7.16.19 Mac Invalid Package
(Message 105303)
Posted 2 Sep 2021 by l008com Post: I'm trying to update my client to 7.16.19 but I get an error each time. It says the digital signature is invalid. It suggest re-downloading which of course I've done. Seems like theres a problem with the posted file itself. 
|
|
4)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100934)
Posted 3 Oct 2020 by l008com Post: Actually, looking at the output from --get_tasks, I think it's been done already. 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. I'm on 7.6.11. Any idea why I'm missing this information? 10) ----------- name: 172425_Hs_T116268-ITGAV_wu-12_1601185078389_2 WU name: 172425_Hs_T116268-ITGAV_wu-12_1601185078389 project URL: http://gene.disi.unitn.it/test/ received: Sat Oct 3 07:48:43 2020 report deadline: Fri Oct 9 07:48:42 2020 ready to report: no state: downloaded scheduler state: scheduled active_task_state: EXECUTING app version num: 110 resources: 1 CPU estimated CPU time remaining: 13120.401551 slot: 0 PID: 11892 CPU time at last checkpoint: 10628.880000 current CPU time: 10633.100000 fraction done: 0.478333 swap size: 4303 MB working set size: 145 MB |
|
5)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100929)
Posted 3 Oct 2020 by l008com Post: So how do I get the runtime? |
|
6)
Message boards :
Questions and problems :
boinccmd values not matching up with GUI
(Message 100926)
Posted 3 Oct 2020 by l008com Post: I'm ultimately trying to run the boinccmd --get_taskscommand and parse that output into an sql table. It's going well except for one problem.... In the `boinccmd` results, the value for "Estimated CPU time remaining" corresponds exactly with the GUI's task column for "Remaining (estimated)". As you'd expect. However the `boinccmd` value for "current CPU time" does not correspond to the GUI's task column for "Elapsed". it is roughly about 75% smaller than that. What is the deal here? How can I get the elapsed time a work unit has been running via `boinccmd`? |
|
7)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99744)
Posted 8 Jul 2020 by l008com 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: I don't see on that page how you actually send and receive those commands? |
|
8)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99720)
Posted 8 Jul 2020 by l008com Post: That's up to you. 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? |
|
9)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99689)
Posted 7 Jul 2020 by l008com Post: The boinccmd utility contains the library routines for converting between XML and human formats as required. Well thats good, so how do you DO that? I can't find any mention in any of the online docs? |
|
10)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99683)
Posted 7 Jul 2020 by l008com Post: boinccmd still exists, and is documented here. Now that boinccmd is working, I'm confused by the difference here. Does boinccmd not use the RPC? If no, then how do I do RPC commands? The RPC page you linked to shows results coming back in XML yet boinccmd's results are all coming back as one long multi-line string. |
|
11)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99681)
Posted 7 Jul 2020 by l008com Post: If you can't find it any other way, it's certainly in the "(Unix command-line version)" on the download_all page. Yup apparently you have to download it as a separate piece on macos. |
|
12)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99680)
Posted 7 Jul 2020 by l008com Post: $ boinccmd -bash: boinccmd: command not found |
|
13)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99650)
Posted 6 Jul 2020 by l008com Post: The manager application goes right into /Applications as an app bundle. I don't see anything like boinccmd inside the app bundle anywhere. It's also not in the /Library/Application Support/BOINC Data folder. I don't know where else it would possibly be. I'm thinking the Mac client might not come with boinccmd? |
|
14)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99627)
Posted 5 Jul 2020 by l008com Post: Well if it's still around then where is it? Or where do I download it? I can't find it anywhere. |
|
15)
Message boards :
Questions and problems :
BOINC RPC Support?
(Message 99588)
Posted 2 Jul 2020 by l008com Post: I'm making a web based dashboard showing me info on all of my crypto wallets and distributed computing apps. The basic way this works is that timed scripts run and gather information from the apps, load it into an SQL database, and then the website pulls that info and shows it in a nice, neat, organized way. I had fairly good luck starting this, getting lots of info out of all of the crypto wallets very easily. Their RPC system is, while not very well documented, very easy to use. But I haven't had as much luck with BOINC. Lots of 10+ year old forum threads. Lots of conflicting, undated information. Pages about using boinccmd which doesn't appear to even exist any more? Can anyone point me in the right direction here? Whether its done via HTTP requests or by direct cli commands, either way as fine as long as I can use it to get info on all of the currently running work units. It seems that the BOINC Manager itself uses some form of RPC to communicate to the actual underlying BOINC processes so there must be some way to do it, no? |
|
16)
Message boards :
Questions and problems :
"Waiting for shared memory" on MacOS.
(Message 95821)
Posted 12 Feb 2020 by l008com Post: You need to change and increase the amount of memory BOINC can use overall. Read and follow instructions from link below. I'm already using 90%. And even with that setting, very little of my system's memory is being used. So that's definitely not it.[/quote] |
|
17)
Message boards :
Questions and problems :
"Waiting for shared memory" on MacOS.
(Message 95820)
Posted 12 Feb 2020 by l008com Post: It could be you need to configure more shared memory. Some years ago, when CPDN still worked on Macs, I came across a link to this page, which may help: Unfortunately that web page is written for Mac OS 10.3, circa 2005. The entire memory management system in current MacOS is completely different. Also the sysctl stuff is changed too. So that link doesn't really help. I found a bunch of links like that while searching for this problem, but they were all horribly outdated. Also that page still doesn't make it clear what the actual issue is. Why are yoyo@home work units running in to this problem but no other projects do? That said, I wonder if enabling performance mode will help this problem. I'll enable it and check back. Update: Turns out performance server mode doesn't exist on high sierra. But the sysctl.conf file does still work. SO the 2005 solution worked and now I have 12 yoyo work units running at once. I still don't understand what exactly this means, or why only this one project is affected by it. The only change I made was to the shared memory segments, going from 8 to 32 or something like that. |
|
18)
Message boards :
Questions and problems :
"Waiting for shared memory" on MacOS.
(Message 95815)
Posted 12 Feb 2020 by l008com Post: Well, for what it's worth, this machine appears to have 256KB of L2 cache per core, and 12 MB of total L3 cache. |
|
19)
Message boards :
Questions and problems :
"Waiting for shared memory" on MacOS.
(Message 95813)
Posted 12 Feb 2020 by l008com Post: I'm running a 12 virtual core Mac Pro running macOS High Sierra 10.13 I've been running BOINC for many years and never had this particular problem. But just recently, I started running yoyo@home. And with that project, I only seem to be able to run 8 yoyo work units at a time, with all other work units showing the "Waiting for shared memory" message. What does this mean? My system has 7 GB of RAM, which isn't a TON for sure, but also the memory pressure is green on this system and really low. Show maybe around 25%, which is very low. BOINC client itself is configured to use 90% of my RAM. So what exactly is going on here? It seems weird. I've been running projects like LHD, Einstein, Rosetta etc 12 at a time for years. What is it about yoyo@home that seems to limit it to 8 at a time. My suspicion is it's either some kind of bug, or some kind of unintentionally misleading error message, as there seems to be plenty of RAM available on this system. |
|
20)
Message boards :
BOINC client :
Edit Resource Share per Computer?
(Message 89148)
Posted 9 Dec 2018 by l008com Post: I see that the way you set Resource Share is not in the BOINC client but on each project's web site. But that means the settings are global. But say I have one computer I want to give high priority to Project A and low to Project B, so B essentially runs when there's no A work. But on a different computer I want to do normal use of Project B? Is this something you can do easily? |
Next 20
Copyright © 2023 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.