Posts by MarkJ

1) Message boards : GPUs : second gpu not in uses (Message 100199)
Posted 4 days ago by MarkJ
Post:
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

His cards are listed as AMD Radeon so he won’t be doing any CUDA work.

It should say Config: Use all GPU’s in the log if it found it.
2) Message boards : GPUs : second gpu not in uses (Message 100197)
Posted 4 days ago by MarkJ
Post:
It doesn’t say it’s using all GPU’s so it didn’t like it. Recheck each point Jord has mentioned.

Also current version of BOINC for Windows is 7.16.7 I would upgrade to the latest version otherwise you might have issues connecting to certain projects.
3) Message boards : GPUs : second gpu not in uses (Message 100177)
Posted 5 days ago by MarkJ
Post:
Do you have a <use_all_gpus> tag in cc_config.xml?
4) Message boards : Questions and problems : Second manager window. (Message 99274)
Posted 13 Jun 2020 by MarkJ
Post:
When I try to connect a second manager window to my second computer, it asks for the machines name and password. Completing these results gives me "The password you have provided is incorrect, please try again.". I have just logged into that machine, and am sure, the name and password used are correct. Both machines are connected to ports on the same router. Both run Windows 8.1 x64. What is going wrong here?

It’s not your Windows (or Linux) password. It’s the one in gui_rpc_auth.cfg which is in your BOINC data directory.
5) Message boards : Questions and problems : Peer certificate cannot be authenticated with given CA certificates (Message 99187)
Posted 9 Jun 2020 by MarkJ
Post:
Yesterday (Mon 8th of June) Debian issued an updated ca-certificates. They added 13 and deleted 15, although it doesn't say which ones. I wouldn't be surprised if this has caught a number of organizations out, not just BOINC. I can still communicate with all my projects so they don't seem to have broken anything as far as I can tell.
6) Message boards : Questions and problems : BOINC Manager won't come up in Linux Mint (Message 99150)
Posted 6 Jun 2020 by MarkJ
Post:
Been running BOINC for years, been running it on Linux Mint for >1 year and now I can't get it to bring up the Manager. Tried restarting, tried a couple of simple command line things (sudo service boinc-client restart), quick searched the message boards.

I'm no pro at Linux; anyone have any advice for what to try next?

There was a similar message from David Pickett. He ran a strace on it and it would look like there is a desktop file that might be causing the issue. See this message

That suggests that in your home folder there will be a file called BOINC Manager-xxx that probably stores information about where to position it on your desktop.
7) Message boards : Questions and problems : remaining time to next checkpoint (Message 98956)
Posted 31 May 2020 by MarkJ
Post:
BoincTasks (a replacement for Boinc Manager) has this option. It also can show how long a go a task did a checkpoint and how many checkpoints it has made. But it won't show when next is due.

If you have checkpoints displayed in BT on the tasks tab, it highlights tasks that have exceeded the checkpoint interval in red.
8) Message boards : BOINC client : Why not use whole numbers instead of % for multiprocessor usage? (Message 98951)
Posted 31 May 2020 by MarkJ
Post:
Perhaps a better suggestion. It could have a slider with 0 to 100%. As you slide it display the percentage above the slider and the number of cores out of the total available underneath. You don’t need to change the core client to implement it, it’s a GUI change.

I believe there was mention of a developer looking at the GUI with a view to updating it.
9) Message boards : Questions and problems : Running the client on a different machine than the computation (Message 98883)
Posted 29 May 2020 by MarkJ
Post:
I am currently in the process of researching the possibility of running multiple machines that run BOINC computations and only having one central computer that is interfacing with BOINC servers. For this I assume I would have to modify the client so that the logic for execution is separate from the one managing tasks.

It does support use of a proxy server if you want a separate machine to communicate with the outside world, but it sounds like you were after more than that.
10) Message boards : Questions and problems : Unable to attach to Universe@home (Message 98864)
Posted 27 May 2020 by MarkJ
Post:
Moot point though. I found a post from myself to someone inquiring on how to get an ARM64 device to get work from the project. My own response said there are no applications for any active projects. The only applications that were available for ARM64 are for the deprecated Neutron Star and Black Hole formation project.

Rosetta have an ARM64 app if you want to use it over there.
11) Message boards : Questions and problems : Current Hardware Usage Column Removed? (Message 98776)
Posted 24 May 2020 by MarkJ
Post:
Or maybe the “Running 1 GPU + 1 CPU (device n)” message? That was shown under the Status column on the tasks tab.
12) Message boards : Questions and problems : WUs on external device (raspberrypi) (Message 98775)
Posted 24 May 2020 by MarkJ
Post:
I have redirected the entire root drive to external HDD in the past. I will see if i can find the blog post I did that explains how. Currently I have 2 x Pi4 running Rosetta off their SD cards.

It’s not as bad as it used to be when they unzipped a database of approx 1GB into the slot every time a task started. These days with the 4.20 app the database is in the project folder and remains there so a lot less writes. You can also increase the checkpoint interval time to reduce the writes. I have all of mine set to 600 sec.

Another option would be to symlink the /var/lib/boinc-client folder to some directory on your external drive.

Edit
Found the blog post from 2016. It’s here
13) Message boards : GPUs : I can't get Intel HD Graphics 4000 to work on Ubuntu 18.04 (Message 98680)
Posted 20 May 2020 by MarkJ
Post:
I recently have started using this computer for Boinc, running Einstein@home exclusively. It has an Intel i7-3612QM, which has integrated Intel HD Graphics 4000.
.
As far as I know Intel’s Neo drivers (aka intel-opencl-icd) support the 8th gen onwards so you can’t use them. On such an old CPU I would install beignet-opencl-icd.

Be aware that running OpenCL tasks on the iGPU will slow down the whole CPU so you lose productivity as a result. It’s usually faster running CPU only tasks.
14) Message boards : Questions and problems : Please, help - too many downloaded jobs/too many jobs in progress (Message 98608)
Posted 18 May 2020 by MarkJ
Post:
[
I do not understand what you stated: Rosetta have stated they start looking at results 48 hours after they send a batch out so setting the default run time as high as your is a waste of effort. Could you explain it more in details, please? I have read many pages on setting target CPU times and on all of them was written that it does not matter as all model needs to be tested anyway, so the longer time you set the more models you can calculate and it makes the process more efficent. However it can be wrong, so why do you think it is waste of time? I set 29h just because I want to reduce newtork use. My computers work 24/7 anyway, so I can use more of their time.

It was a post by one of the moderators in their Number Crunching forum. As such I would suggest a target run time of 12 hours. If you end up setting the cache back to one day then that is potentially work on your system for 36 hours (one day cache plus your target run time for work currently running). If you set the cache to 0.5 days it would be 24 hours.

At the moment I do not want to limit my cores, I want to use the full power of my computers. Maybe I will use it in the future, but not now.
Taking the opportunity I would like to ask if it is possible to configure BOINC that for instance 1 core is dedicated to one project, 2 cores are fixed to another project and the rest of cores for a third

You could use the app_config to limit how many cores each project can use at a time but if one project runs out of work you’ll end up with idle cores. Also I mentioned before that the 7.16 client has a fix for work fetch when used with max_concurrent, the older BOINC client you are running doesn’t have that fix so it will fetch more work than it can run.

BOINC is designed to share resources based upon your project weighting. It should balance things out over time.
15) Message boards : Questions and problems : Please, help - too many downloaded jobs/too many jobs in progress (Message 98574)
Posted 17 May 2020 by MarkJ
Post:
Better to be store 1 and an additional 0.1

Rosetta is a bit different. They have a target run time. It defaults to 8 hours if not set by the user but Mippi said he had it much higher. Work units will generally run for that length of time. They have allowed a +10 hour leeway where their watchdog timer will allow it to run over the target (it used to be 4 hours).

They are using 3 day deadlines. With a target run time of 12 hours, assuming he reduces it to that, it doesn’t leave much time to return them in. It would be better to cache a minimum of work in these circumstances. The project doesn’t seem to have a shortage of tasks at the moment.

Mippi also asked about app_config so assuming he sets that to half the available threads he gets half of them run at once. He is running an older version of BOINC which doesn’t have the work fetch/max concurrent logic that 7.16 has, so it gets tasks for all available threads even though they can’t all run at once. He may only need to reserve a couple of threads for the system in which case setting the % of processor available in BOINC would be better than using an app_config as it wouldn’t try and fetch more work than available threads.
16) Message boards : Questions and problems : Please, help - too many downloaded jobs/too many jobs in progress (Message 98568)
Posted 17 May 2020 by MarkJ
Post:
If you want to stick with the CLI you can use BOINCtui to visually see what they are doing, one machine at a time.

With multiple machines its best to use BOINCtasks on one machine that you trust (usually the one you use to remote into the others). It doesn't need to run BOINC at all. I'm using a Windows laptop to look after my cluster. If you don't have a windows machine you can run it under wine. You can download it from https://efmer.com/ and click on the BoincTasks option at the top of the screen. If you need help with configuring it just ask. It really makes managing a fleet of machines so much easier when you can see all of them on one screen.

If you can update your boinc-client to a later version.

To start with I would set all of the machines to no new tasks and let them run off their current work. Rosetta have stated they start looking at results 48 hours after they send a batch out so setting the default run time as high as your is a waste of effort. Set a zero cache. After it has finished off all work I I would set Rosetta closer to the default 8 hour target run time, maybe go for 12 hours. Adjust the target run time and tell the BOINC client to update so it picks up the latest setting from the project before requesting any new work.

As for the xml files BOINC doesn't use config_aux. It uses an app_config.xml file, It goes in the project specific folder. Under Debian and Ubuntu it would be:
/var/lib/boinc-client/projects/boinc.bakerlab.org_rosetta/app_config.xml

If you want to limit it to half the available cores then you would use something like this in it:
<app_config>
    <project_max_concurrent>6</project_max_concurrent>
</app_config>

That will limit it to running 6 at a time. Adjust the number as you want.

Once you've returned at least 11 and they've validated then you can adjust your cache setting. I run 0.1 days with no extra. That gives the project a fast turn around and you don't get overloaded with tasks. I'd suggest you have all of the machines with the same settings seeing as they are the same hardware config. The project don't recommend more than a 1 day cache due to the short deadlines (which are 3 days).
17) Message boards : BOINC Manager : Manager not starting since upgrade to ubuntu 20.04 (Message 98552)
Posted 16 May 2020 by MarkJ
Post:
I don’t find it much of a hinderance not being able to shut down the core client from the manager. You couldn’t start it up again anyway. I usually have a terminal window open at the same time as the manager and do the systemctl stop/start boinc-client from there.

Back to Dave’s problem. The reason why I suggested he use Gianfranco’s version is his manager presumably works unlike the one Dave has currently. I know Gianfranco spent some time finding a working wxWidgets.
18) Message boards : BOINC Manager : Manager not starting since upgrade to ubuntu 20.04 (Message 98549)
Posted 16 May 2020 by MarkJ
Post:
Maybe just nuke your version and use Gianfranco’s 7.16.6 version. He has an Ubuntu ppa here:
https://launchpad.net/~costamagnagianfranco/+archive/ubuntu/boinc
19) Message boards : Questions and problems : BOINC Manager will not connect with BOINC in Linux (Disconnected, Error -155 & -180) (Message 98548)
Posted 16 May 2020 by MarkJ
Post:
The standard Debian and Ubuntu install folders are:
/etc/boinc-client
/var/lib/boinc-client

The gui_rpc_auth.cfg file is in the /etc folder with a symlink in the /var/lib folder so you can edit it from either.

I see you have another thread going in the BOINC Manager forum so let’s use that instead of having two message threads for the same thing.
20) Message boards : Questions and problems : RPi 1 B+ Not Getting Work (RE: Science United Message 488) (Message 98547)
Posted 16 May 2020 by MarkJ
Post:
There are complaints regarding the Asteroids ARMv6 app not working in their message forums. They recently updated all their apps.

The Einstein BRP 1.06 app has been broken since Raspbian Stretch almost 2.5 years ago. There is a version that works for ARMv7 (Pi2 and later) but nothing for the ARMv6 (Zero and Pi1).


Next 20

Copyright © 2020 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.