Advanced search

Message boards : Graphics cards (GPUs) : acemdbeta erroring out

Author Message
ETQuestor
Send message
Joined: 11 Jul 09
Posts: 27
Credit: 1,000,618,568
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12500 - Posted: 14 Sep 2009 | 17:23:46 UTC

It looks like recently that my machine started to use the "acemdbeta" instead of the usual "acemd", which had been running fine. The computation seems to fail out each time; errors below.

I've made no recent changes to my GPU, GPU driver, BOINC, or OS:

OS = Fedora 11, 2.6.30.6_x86_64 kernel
GPU driver = x86_64 Kernel Module 185.18.36
GPU = GeForce 9600 GSO
BOINC = 6.6.37 for x86_64-pc-linux-gnu




13-Sep-2009 15:03:34 [GPUGRID] Starting p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3
13-Sep-2009 15:03:34 [GPUGRID] Starting task p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3 using acemdbeta version 602
13-Sep-2009 15:03:45 [GPUGRID] [error] Can't rename output file p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_5 to projects/www.gpugrid.net/p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_5: Error -1
13-Sep-2009 15:03:51 [GPUGRID] [error] Can't rename output file p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_7 to projects/www.gpugrid.net/p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_7: Error -1
13-Sep-2009 15:03:51 [GPUGRID] Computation for task p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3 finished
13-Sep-2009 15:03:51 [GPUGRID] Output file p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_1 for task p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3 absent
13-Sep-2009 15:03:51 [GPUGRID] Output file p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_2 for task p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3 absent
13-Sep-2009 15:03:51 [GPUGRID] Output file p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_3 for task p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3 absent
13-Sep-2009 15:03:52 [GPUGRID] Started upload of p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_0
13-Sep-2009 15:03:52 [GPUGRID] Started upload of p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_6
13-Sep-2009 15:03:58 [GPUGRID] Finished upload of p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_0
13-Sep-2009 15:03:58 [GPUGRID] Finished upload of p40000-IBUCH_random_pYEEI_1009-0-5-RND3455_3_6

Profile K1atOdessa
Send message
Joined: 25 Feb 08
Posts: 249
Credit: 370,186,977
RAC: 0
Level
Asp
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12503 - Posted: 14 Sep 2009 | 18:46:46 UTC - in response to Message 12500.

It looks like recently that my machine started to use the "acemdbeta" instead of the usual "acemd", which had been running fine. The computation seems to fail out each time; errors below.

I've made no recent changes to my GPU, GPU driver, BOINC, or OS:

OS = Fedora 11, 2.6.30.6_x86_64 kernel
GPU driver = x86_64 Kernel Module 185.18.36
GPU = GeForce 9600 GSO
BOINC = 6.6.37 for x86_64-pc-linux-gnu


It's a Linux beta task. Given these are beta WU's, you shouldn't expect them to work 100%. The error rate in these should be greater than the non-beta WU's. There is an option in the GPUGrid preferences to elect not to receive beta WU's if you don't want some failed WU's.

ETQuestor
Send message
Joined: 11 Jul 09
Posts: 27
Credit: 1,000,618,568
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12504 - Posted: 14 Sep 2009 | 19:10:28 UTC - in response to Message 12503.

It's a Linux beta task. Given these are beta WU's, you shouldn't expect them to work 100%. The error rate in these should be greater than the non-beta WU's. There is an option in the GPUGrid preferences to elect not to receive beta WU's if you don't want some failed WU's.



Understood. I intentionally selected the option to run "test" applications and I don't have a problem with a few blowing up. However, every WU has errored out almost immediately in this fashion since the new acemdbeta was downloaded, so I wanted to make sure it was reported. I wasn't sure where else to report this beside the forum.

Profile GDF
Volunteer moderator
Project administrator
Project developer
Project tester
Volunteer developer
Volunteer tester
Project scientist
Send message
Joined: 14 Mar 07
Posts: 1947
Credit: 629,356
RAC: 0
Level
Gly
Scientific publications
watwatwatwatwat
Message 12512 - Posted: 15 Sep 2009 | 14:20:44 UTC - in response to Message 12500.
Last modified: 15 Sep 2009 | 14:24:13 UTC

hi,
we are looking into it.

thanks.
gdf


[EDIT the last ones seem to work. I don't know if it was a particular batch or it is your machine]

ETQuestor
Send message
Joined: 11 Jul 09
Posts: 27
Credit: 1,000,618,568
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12514 - Posted: 15 Sep 2009 | 17:35:41 UTC - in response to Message 12512.

Hmm, OK. I'm waiting to get my new daily quota to see if it continues to blow up. If there is any additional information or troubleshooting you want me to try, please let me know...happy to do it.

fractal
Send message
Joined: 16 Aug 08
Posts: 87
Credit: 1,248,879,715
RAC: 0
Level
Met
Scientific publications
watwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwatwat
Message 12519 - Posted: 16 Sep 2009 | 0:48:10 UTC

My 9600gso / linux box started getting mostly errors this week as well. This machine was getting an error every month or two before this weeks problems. The wu's don't appear to be beta ( 325-GIANNI_DOPd-2-25-RND0623 ) and are finished by someone else.

I updated to the latest greatest cuda driver today so will see if that helps.

Post to thread

Message boards : Graphics cards (GPUs) : acemdbeta erroring out

//