WU errors

Message boards : Number crunching : WU errors

To post messages, you must log in.

AuthorMessage
Variable

Send message
Joined: 5 Jan 18
Posts: 4
Credit: 926,701,248
RAC: 2,457,873
Message 13592 - Posted: 15 Jan 2018, 15:51:48 UTC

Over the weekend I started experiencing Enigma GPU calculation errors on my Linux machine which uses GTX 1050Ti cards. Is this happening to anyone else? The config of the machine has not changed. Last week it ran these units fine, then suddenly they all started failing a couple days ago, despite Boinc & machine restarts.

Stderr output
<core_client_version>7.6.31</core_client_version>
<![CDATA[
<message>
process exited with code 195 (0xc3, -61)
</message>
<stderr_txt>
08:15:44 (435): wrapper (7.7.26016): starting (812 +dev#, cuda_fermi)
Device number read from command line: -1
08:15:44 (435): wrapper (7.7.26016): starting (812 +dev#, cuda_fermi)
08:15:44 (435): wrapper: running enigma_gpu (-d 0 -R -o results.txt 04trigr.cur 04unigr.cur 04ciphertext)
execv() failed: : Permission denied
08:15:45 (435): enigma_gpu exited; CPU time 0.000671
08:15:45 (435): app exit status: 0x6c
08:15:45 (435): called boinc_finish(195)

</stderr_txt>
]]>
ID: 13592 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Easton West

Send message
Joined: 2 Oct 17
Posts: 2
Credit: 10,014,270
RAC: 64,406
Message 13593 - Posted: 15 Jan 2018, 22:41:39 UTC - in response to Message 13592.  

Seems to be affecting everyone running Enigma GPU v1.12 (cuda_fermi) on Nvidia + Linux. Old WUs on 1.10 are fine. 1.12's probably bugged.

me (Intel CPU) - http://www.enigmaathome.net/results.php?hostid=231569
nes (AMD CPU) - http://www.enigmaathome.net/results.php?hostid=232327
anonymous (Intel CPU) - http://www.enigmaathome.net/results.php?hostid=220084
anonymous (Intel CPU) - http://www.enigmaathome.net/results.php?hostid=216569
ID: 13593 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Variable

Send message
Joined: 5 Jan 18
Posts: 4
Credit: 926,701,248
RAC: 2,457,873
Message 13594 - Posted: 16 Jan 2018, 3:13:33 UTC

It's not just Linux & 1.12 units though. I have a Win10 machine that is now failing as well with all 1.10 units.

http://www.enigmaathome.net/result.php?resultid=366129616
ID: 13594 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Variable

Send message
Joined: 5 Jan 18
Posts: 4
Credit: 926,701,248
RAC: 2,457,873
Message 13595 - Posted: 16 Jan 2018, 18:05:51 UTC

My Win7 machine is running 1.10 units just fine. Can someone from the project comment? I could put a lot more compute resources on this project, except most of it is failing when I try.
ID: 13595 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Easton West

Send message
Joined: 2 Oct 17
Posts: 2
Credit: 10,014,270
RAC: 64,406
Message 13599 - Posted: 18 Jan 2018, 4:28:45 UTC

Just got some new 1.13 WUs, so far so good! Thanks for the quick update.
ID: 13599 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Werinbert

Send message
Joined: 22 Jul 13
Posts: 7
Credit: 8,558,470
RAC: 21,678
Message 13614 - Posted: 6 Apr 2018, 15:30:06 UTC

I am still getting errors with 1.10 WUs. As my computer previously crunched the WUs just fine and I have experience intermittent WU problems in the past I thought it was just buggy files. However, my inability to crunch has been going on for a couple months now with no good WUs seen. I have tried the various fixes described a while ago and they don't seem to help.

Any help on finding a solution would be greatly appreciated so I can get back to crunching for this project.
ID: 13614 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Crack

Send message
Joined: 4 May 09
Posts: 1
Credit: 58,781,139
RAC: 101,135
Message 13617 - Posted: 25 Apr 2018, 17:38:49 UTC - in response to Message 13614.  

I think it's a driver problem:
<core_client_version>7.8.3</core_client_version>
<![CDATA[
<message>
(unknown error) - exit code 195 (0xc3)</message>
<stderr_txt>
wrapper [26118 765 +dev#]: starting (GPU #0)
18:55:48 (10660): wrapper: running enigma_cuda.exe (-d 0 -R -o results.txt 04trigr.cur 04unigr.cur 04ciphertext)
CUDA error: C:/Proj/Other/Enigma/enigma-cuda/enigma-cuda-lib/cuda_code.cu:784:  CUDA driver version is insufficient for CUDA runtime versionInitialization failed: CUDA error: C:/Proj/Other/Enigma/enigma-cuda/enigma-cuda-lib/cuda_code.cu:784:  CUDA driver version is insufficient for CUDA runtime version
app exit status: 0x2
18:55:49 (10660): called boinc_finish

</stderr_txt>
]]>
ID: 13617 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Werinbert

Send message
Joined: 22 Jul 13
Posts: 7
Credit: 8,558,470
RAC: 21,678
Message 13620 - Posted: 30 Apr 2018, 6:46:50 UTC

I am running the most recent driver (397.31) and I am still getting errors.
<core_client_version>7.8.3</core_client_version>
<![CDATA[
<message>
(unknown error) - exit code 195 (0xc3)</message>
<stderr_txt>
wrapper [26118 765 +dev#]: starting (GPU #0)
23:40:52 (2692): wrapper: running enigma_cuda.exe (-d 0 -R -o results.txt 04trigr.cur 04unigr.cur 04ciphertext)
app exit status: 0x1
23:40:53 (2692): called boinc_finish

</stderr_txt>
]]>
ID: 13620 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : Number crunching : WU errors




Copyright © 2018 TJM