All tasks fail on one computer

Fardringle
Fardringle
Joined: 11 Mar 06
Posts: 7
Credit: 111,958,997
RAC: 831
Topic 223482

https://einsteinathome.org/host/12768720

This is an older computer and GPU, but it still works fine for gaming and hasn't had any problems on every other project it has run. And in fact it received quite a few credits while running Einstein last year. But I recently let it start getting work again and noticed that ALL tasks it has tried to run are failing after running for about 80 seconds.

Any ideas what is causing this and how I can fix it?

 

Thanks!

Richie
Richie
Joined: 7 Mar 14
Posts: 656
Credit: 1,702,989,778
RAC: 0

You may have been running

You may have been running tasks from "Gamma-ray pulsar binary search #1 on GPUs" last year. Your GPU has tried to run only the "Gravitational Wave search O2 Multi-Directional" tasks now. These GW GPU tasks have been more demanding for the GPU (memory amount). Is your GPU running only one task at a time ?

If even one GW task at a time doesn't work then you could change the settings and select only the "Gamma-ray pulsar binary search #1 on GPUs". Your card should be able to run those tasks fine. In this case, you could also uncheck the "Run test applications" option.

https://einsteinathome.org/account/prefs/project

Fardringle
Fardringle
Joined: 11 Mar 06
Posts: 7
Credit: 111,958,997
RAC: 831

Yes, only one task at a time.

Yes, only one task at a time. I'll try disabling the Gravitational Wave application to see what happens. Thanks for the suggestion!

archae86
archae86
Joined: 6 Dec 05
Posts: 3,156
Credit: 7,174,644,931
RAC: 732,609

Fardringle wrote:I'll try

Fardringle wrote:
I'll try disabling the Gravitational Wave application to see what happens. 

As Richie mentioned, you would be wise also to set "run test applications?" to No.  I think a Yes in that setting has at least sometimes over-ridden other restrictions.  And the stderr for the one of your failures I glanced at had "beta" in the application name.

Fardringle
Fardringle
Joined: 11 Mar 06
Posts: 7
Credit: 111,958,997
RAC: 831

I did that, too.

I did that, too. Thanks.

Probably due to the large number of errors, the project put that computer on "hold" and deferred communication for several hours and immediately resets the timer if I manually request an update. I'll check on it once it is allowed to get more work in about 3 hours.

archae86
archae86
Joined: 6 Dec 05
Posts: 3,156
Credit: 7,174,644,931
RAC: 732,609

Fardringle wrote:... deferred

Fardringle wrote:
... deferred communication for several hours and immediately resets the timer if I manually request an update. I'll check on it once it is allowed to get more work in about 3 hours.

The clock that tells you how long you must wait and the clock that actually decides to honor your request or not are sometimes mismatched by hours.

In my case (it may be time zone dependent) the one that tells you how long remains references a random time between midnight UTC and midnight + 1 hour UTC.  But quite often the request made then is again refused, the automatic timer puts off my request by another day.  But if I try now and again manually, generally it accepts in no more than nine additional hours, sometimes a lot less.

Good luck.  Keep trying.

 

 

Fardringle
Fardringle
Joined: 11 Mar 06
Posts: 7
Credit: 111,958,997
RAC: 831

Seems to be running OK with

Seems to be running OK with those changes now. Thanks.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.