| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: OpenPandemics - COVID-19 Project
Thread: process exited with code 10 |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 4
|
|
| Author |
|
|
dondee
Advanced Cruncher Joined: Jan 16, 2006 Post Count: 100 Status: Offline Project Badges: 
|
All of my opng dated 11-4 or later has error "process exited with code 10".
----------------------------------------Error: boinc_get_opencl_ids() failed with error -1 Any one else seen this or something like it? [Edit 1 times, last edit by dondee at Nov 5, 2023 1:54:26 PM] |
||
|
|
Bryn Mawr
Senior Cruncher Joined: Dec 26, 2018 Post Count: 337 Status: Offline Project Badges:
|
All of my opng dated 11-4 or later has error "process exited with code 10". Error: boinc_get_opencl_ids() failed with error -1 Any one else seen this or something like it? Yes, every OPNG WU I’ve received since 12:00 03-11-23 has failed after 00:00:01. |
||
|
|
alanb1951
Veteran Cruncher Joined: Jan 20, 2006 Post Count: 873 Status: Offline Project Badges: 
|
[Operating systems and client versions unspecified in the above -- my response is Linux-based, using client 7.20.5, with NVIDIA GPUs]
I'm seeing this from the opposite side (and was wondering if anyone would post a query about it!...) During the 48 hours of 3rd and 4th November I ran 218 OPNG tasks, all of which ran to conclusion and most of which have validated. However, 86 wingmen had the error you are both describing and a further 8 wingmen had a slightly different error during the first AutoDock job - "process exited with code 255 (0xff, -1)" All the code 255 cases seem to have been from the same host (Gentoo, and client 7.24.1); most of the code 10 cases seem to be associated with about half a dozen systems on Ubuntu with various 5.x and 6.x kernels and client versions from 7.16.x to 7.22.x (the rest seem to be from a GNOME 44 system...) So there doesn't seem to be anything wrong with the actual tasks, and it's not down to client version either(!) but something appears to have confused the BOINC client. One thought is that there has been an [automatic] driver update that lost the OpenCL component[*1], another is that it might self-repair if the client is restarted[*2]. The same may apply on Windows, but I have no access to relevant information... For what it's worth, I haven't had an OPNG error or invalid result under normal running circumstances[*3] in a very long time... (hope I haven't jinxed it!) Cheers - Al. *1 If the OpenCL driver was present at client start-up but the driver updates without a system shutdown, tasks will start but fail in some way when the GPU is actually sought. *2 If restarting, disable GPU tasks before doing so, then check that OpenCL appears in the start-up stuff in the BOINC log on restart; if it doesn't, you have an answer(!) and if it does appear enable one GPU task and see if it fails the same way... *3 I've "been there, done that" (forgetting to stop BOINC before starting an update, for instance!) in the [distant] past, and am now very cautious about driver updates! |
||
|
|
Bryn Mawr
Senior Cruncher Joined: Dec 26, 2018 Post Count: 337 Status: Offline Project Badges:
|
OK, it’s affecting both of my systems, identical setup, Ryzen 9/3900 running Ubuntu 22.04.3 LTS and Boinc 7.20.5 with a very slow GT710 video card.
----------------------------------------I’ve been away since Friday and I’m not aware of any updates to either system. I do, however, allow updates without shutting Boinc down so I’ll reboot and see what effect that has. [Edit 1 times, last edit by Bryn Mawr at Nov 5, 2023 6:53:36 PM] |
||
|
|
|