Aborted by User??

Message boards : Number crunching : Aborted by User??
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Bill G Special Project $75 donor
Avatar

Send message
Joined: 1 Jun 01
Posts: 1282
Credit: 187,688,550
RAC: 182
United States
Message 1740161 - Posted: 6 Nov 2015, 15:01:59 UTC
Last modified: 6 Nov 2015, 15:03:23 UTC

I seem to be having a bunch of Aborted by User WUs which of course show up as Errors. When I look at the Task Results it seems to say there was an error. Am I doing something wrong as this computer was just having the current Invalids until recentl?. The computer is: http://setiathome.berkeley.edu/show_host_detail.php?hostid=7226971

SETI@home classic workunits 4,019
SETI@home classic CPU time 34,348 hours
ID: 1740161 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1740164 - Posted: 6 Nov 2015, 15:15:09 UTC - in response to Message 1740161.  

The actual error message is

Exit status 201 (0xc9) EXIT_MISSING_COPROC

Although BOINC thinks you have a GPU, it isn't the right type (CUDA) for the tasks you've been issued. Quite possibly Windows 10 up to its driver updating tricks again (there was a new NVidia driver release this morning - within the last couple of days, anyway).
ID: 1740164 · Report as offensive
Profile AyalaZero
Volunteer tester

Send message
Joined: 14 Aug 05
Posts: 21
Credit: 10,910,119
RAC: 0
United States
Message 1740166 - Posted: 6 Nov 2015, 15:23:13 UTC - in response to Message 1740161.  

It's likely still due to the Server going down. The new Nvidia driver seems to be working fine for me. I'd wait about a week, see if it goes back to normal.
ID: 1740166 · Report as offensive
Profile William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 2037
Credit: 17,689,662
RAC: 0
Message 1740169 - Posted: 6 Nov 2015, 15:26:40 UTC

I traced that error here.

If the problem persists, start from there and my following posts.
Else, from Richard's comment I'd say you are looking at an update artifact.
[and we need to see if we can get a better error message coded - one that doesn't aggrevate the user]

I'm out for the weekend, I hope somebody else can assist further if necessary.
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1740169 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1740172 - Posted: 6 Nov 2015, 15:27:51 UTC - in response to Message 1740166.  

It's likely still due to the Server going down. The new Nvidia driver seems to be working fine for me. I'd wait about a week, see if it goes back to normal.

Not the server problem. These are new workunits created today, and Bill's wingmates don't have any problems with them.
ID: 1740172 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1740173 - Posted: 6 Nov 2015, 15:35:39 UTC - in response to Message 1740164.  

... it isn't the right type (CUDA) for the tasks you've been issued.

Correction - they're Astropulse tasks, like in the thread William linked, so it's probably the OpenCL support which has gone AWOL again.

Similar troubleshooting process: restart BOINC (heck, why not reboot the computer to be safe), and look at BOINC's startup messages.
ID: 1740173 · Report as offensive
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14649
Credit: 200,643,578
RAC: 874
United Kingdom
Message 1740177 - Posted: 6 Nov 2015, 16:05:30 UTC - in response to Message 1740173.  

The most recent AP task which ran detected

 OpenCL Platform Name:					 NVIDIA CUDA
  Name:						 GeForce GTX 750 Ti
  Vendor:					 NVIDIA Corporation
  Driver version:				 358.50
  Version:					 OpenCL 1.2 CUDA

That was yesterday morning - task 4498598077

Then there are a bunch supposedly still in progress - are they still displayed in BOINC Manager? If so, what 'Status' is shown? And all AP tasks issued after 04:00 today have failed immediately.

I can't see any sign of the host contacting the servers between midnight and 4am (UTC) - that sounds like the sort of timetable for Microsoft to be messing with drivers.
ID: 1740177 · Report as offensive
Profile Bill G Special Project $75 donor
Avatar

Send message
Joined: 1 Jun 01
Posts: 1282
Credit: 187,688,550
RAC: 182
United States
Message 1740178 - Posted: 6 Nov 2015, 16:05:58 UTC - in response to Message 1740173.  
Last modified: 6 Nov 2015, 16:07:32 UTC

... it isn't the right type (CUDA) for the tasks you've been issued.

Correction - they're Astropulse tasks, like in the thread William linked, so it's probably the OpenCL support which has gone AWOL again.

Similar troubleshooting process: restart BOINC (heck, why not reboot the computer to be safe), and look at BOINC's startup messages.


I have been off fixing the problem. I noticed that there was no OpenCL for that computer and realized that it had upgraded the OS to a newer release of the Insider Win10. That is actually a reinstall of windows so it changed to drivers and after I reloaded/upgraded the drivers for the NVIDIA card I now have OpenCL again and I am sure the problem is gone. Sorry to have bothered all of you and thanks for your help. Someday I may actually remember what is going on.

SETI@home classic workunits 4,019
SETI@home classic CPU time 34,348 hours
ID: 1740178 · Report as offensive
Profile Bill G Special Project $75 donor
Avatar

Send message
Joined: 1 Jun 01
Posts: 1282
Credit: 187,688,550
RAC: 182
United States
Message 1740179 - Posted: 6 Nov 2015, 16:18:48 UTC

And you were correct about the timing Richard. I forgot that I told this computer to update to the new OS last night when I went to bed and the install must have been going on most of that time and the system was waiting for me to logon this morning, so BOINC does not start running until I log on.

I was easily confused this morning as TeamViewer upgraded yesterday and there a few things going on with that as well. (I monitor the computers remotely from my laptop).

Again, Thank You

SETI@home classic workunits 4,019
SETI@home classic CPU time 34,348 hours
ID: 1740179 · Report as offensive

Message boards : Number crunching : Aborted by User??


 
©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.