Advanced search

Forums : Technical Support : camb_boinc2docker erroring due to time?
Message board moderation

To post messages, you must log in.

AuthorMessage
Variable

Send message
Joined: 21 Oct 17
Posts: 12
Credit: 11,056,793
RAC: 12,634
Message 21552 - Posted: 22 Oct 2017, 19:08:23 UTC

Hi - I started trying out this project the last day or so. The legacy app seems to run fine but the VM app appears to be constantly erroring. Other VM projects like LHC have run fine. It's saying "exceeded elapsed time limit". Any ideas what's happening?

10/22/2017 2:05:32 PM | Cosmology@Home | Aborting task camb_boinc2docker_32028_1508609789.343757_0: exceeded elapsed time limit 1028.38 (86400.00G/85.93G)
10/22/2017 2:05:58 PM | Cosmology@Home | Computation for task camb_boinc2docker_32028_1508609789.343757_0 finished
10/22/2017 2:05:58 PM | Cosmology@Home | Output file camb_boinc2docker_32028_1508609789.343757_0_r618894800_0 for task camb_boinc2docker_32028_1508609789.343757_0 absent
10/22/2017 2:05:58 PM | Cosmology@Home | Starting task wu_102117_170958_0_0_0
ID: 21552 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 27 Sep 17
Posts: 62
Credit: 3,942,821
RAC: 445
Message 21553 - Posted: 22 Oct 2017, 22:28:12 UTC - in response to Message 21552.  

I wasn't able to look at work units on your computer since they're hidden.
ID: 21553 · Report as offensive     Reply Quote
Variable

Send message
Joined: 21 Oct 17
Posts: 12
Credit: 11,056,793
RAC: 12,634
Message 21554 - Posted: 23 Oct 2017, 14:15:52 UTC

Fixed - it should not be hidden now
ID: 21554 · Report as offensive     Reply Quote
Variable

Send message
Joined: 21 Oct 17
Posts: 12
Credit: 11,056,793
RAC: 12,634
Message 21555 - Posted: 23 Oct 2017, 14:24:17 UTC

From just sitting there and watching this machine for a while, it would run the camb_boinc2docker units for 15-20mins, enough to complete most of the work unit, but then fails with the error I pasted previously.
ID: 21555 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 27 Sep 17
Posts: 62
Credit: 3,942,821
RAC: 445
Message 21556 - Posted: 23 Oct 2017, 20:41:26 UTC - in response to Message 21555.  

Task 59459621, workunit 44736269, seems to have VirtualBox problems

You can try setting no new tasks and/or reset the Cosmology Project.

I don't know if you need to update your Boinc client from 7.6.33 to 7.8.3. I am on 7.8.3

Did you verify AMD SVM (virtualization) is turned on in the BIOS? I accidentally turned mine off after a BIOS update once.

Are you okay on RAM? Since you are only using one vCPU per camb_boinc2docker task, you could run out of RAM with a few tasks going.
ID: 21556 · Report as offensive     Reply Quote
Variable

Send message
Joined: 21 Oct 17
Posts: 12
Credit: 11,056,793
RAC: 12,634
Message 21557 - Posted: 23 Oct 2017, 22:05:25 UTC

Virtualization is turned on in the BIOS, I had to turn it on in order to run LHC@home which also uses VirtualBox. The machine has 16GB of RAM which was only maybe ~50% utilized while these work units were running. Are there any rules of thumb on boinc2docker tasks vs number of vCPU's assigned vs system RAM?
ID: 21557 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 27 Sep 17
Posts: 62
Credit: 3,942,821
RAC: 445
Message 21558 - Posted: 23 Oct 2017, 22:45:41 UTC - in response to Message 21557.  

Each boinc2docker task will use 2Gb per the virtual machine. The jobs use the same amount of memory even if you have it set to use more cores. I have mine set to run 8vCPU jobs and it runs two jobs at a time. I only have Cosmology running. As long as you aren't eating up all the ram in your computer, you should be fine.

You might need to update your BOINC client but I am not sure. Your logs look pretty much like mine until the virtualbox errors.

I use the following for my app_config.xml, placed in ProgramData\BOINC\projects\www.cosmologyathome.org
You can change the avg_ncpus to what ever you want.

<app_config>
<app_version>
<app_name>camb_boinc2docker</app_name>
<plan_class>vbox64_mt</plan_class>
<avg_ncpus>8</avg_ncpus>
</app_version>
</app_config>
ID: 21558 · Report as offensive     Reply Quote
Jonathan

Send message
Joined: 27 Sep 17
Posts: 62
Credit: 3,942,821
RAC: 445
Message 21566 - Posted: 27 Oct 2017, 2:22:00 UTC - in response to Message 21558.  

Did you sort out the errors? What was it, if so.
ID: 21566 · Report as offensive     Reply Quote
Variable

Send message
Joined: 21 Oct 17
Posts: 12
Credit: 11,056,793
RAC: 12,634
Message 21568 - Posted: 28 Oct 2017, 1:35:49 UTC

I tried running the multi-core app again today on this machine and now it seems to be working. Right now I'm running 2 tasks at 4 threads apiece, no errors. Weird. I didn't make any config changes to the machine at all.
ID: 21568 · Report as offensive     Reply Quote

Forums : Technical Support : camb_boinc2docker erroring due to time?