Computation errors without known cause on cuda tasks

Message boards : Number crunching : Computation errors without known cause on cuda tasks
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Akio
Avatar

Send message
Joined: 18 May 11
Posts: 375
Credit: 32,129,242
RAC: 0
United States
Message 1907785 - Posted: 17 Dec 2017, 23:38:27 UTC

Well guys, I don't know why this started happening on this one rig after a full shutdown and reboot, but alas, there is something seriously long causing the errors with cuda tasks. I've run as many diagnostics as I can and even tore the rig apart and put it back together. I've reported all of the tasks and am just going to cycle through what I have left buffered. I'll try another approach to fix whatever has gone wrong after the holidays.
ID: 1907785 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34744
Credit: 261,360,520
RAC: 489
Australia
Message 1907787 - Posted: 17 Dec 2017, 23:47:47 UTC

I'd do a few passes of Memtest86+ to check the system memory and run the manufacturer's diagnostics on the hard drive to check both 1st, but then I'd grab the latest Lunatics installer (0.45 Beta6) and change from that outdated CUDA app over to the SoG app. ;-)

Cheers.
ID: 1907787 · Report as offensive
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1907788 - Posted: 17 Dec 2017, 23:50:00 UTC - in response to Message 1907785.  

Were you running any games on that system?
ID: 1907788 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 34744
Credit: 261,360,520
RAC: 489
Australia
Message 1907790 - Posted: 18 Dec 2017, 0:07:37 UTC

Have you checked to make sure that the system clock is correct?

Cheers.
ID: 1907790 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13722
Credit: 208,696,464
RAC: 304
Australia
Message 1907822 - Posted: 18 Dec 2017, 6:36:57 UTC
Last modified: 18 Dec 2017, 6:45:39 UTC

I'm with WIggo- check your date & time are set correctly- particularly the date.
And how many WUs are you trying to run at once? those run times are ridiculously long, even for CUDA50. My GTX750Tis have better run times while running 2WU at a time.

I'd re-run the Lunatics Beta6 installer & select SoG. And make sure you're running 1 WU at a time on the system that is working OK.


EDIT- and i'd check your CPU temperatures as most of your CPU WUs are going to be invalid- almost all of them are Inconclusinve & appear to be finding too many pulses and overflowing.

That system is very sick.
PSU, CPU, memory all need thorough checking.
Grant
Darwin NT
ID: 1907822 · Report as offensive

Message boards : Number crunching : Computation errors without known cause on cuda tasks


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.