| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: Multiple WU's in 'error' RESOLVED (but not solved!) |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 4
|
|
| Author |
|
|
dustydave
Cruncher Joined: Mar 5, 2006 Post Count: 21 Status: Offline Project Badges: 
|
Early this morning (Aug 24th 2012) I have received 43 WU's shown in error. My browser had closed down and one other background program reported being unable to run, so this may have been due to problems on my PC.
----------------------------------------BOINC reported it could not communicate with the server, but when I cleared that message and allowed the agent to run, all now seems OK, without a reboot. I am running Win7 on a Celeron dual processor laptop. Is there any way to determine what the cause actually was, so I can hopefully avoid this happening again. One of the messages: Result Name: CMD2_ 2533-MYH6.clustersOccur-2ZGV_ A.clustersOccur_ 200_ 581301_ 582000_ 1-- <core_client_version>6.10.58</core_client_version> <![CDATA[ <message> too many exit(0)s </message> <stderr_txt> INFO: No state to restore. Start from the beginning. </stderr_txt> ]]> Many thanks Dave [Edit 1 times, last edit by dustydave at Aug 25, 2012 6:00:00 PM] |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Hi,
The most frequent cause of this "too many exit(0)s, is a "too busy system". On my Linux system I've set BOINC to pause when the non-BOINC load is greater than 35%. The default is actually 25% [plus the Activity menu set to "Run based on preferences". In addition I've set "Leave Application in Memory when suspended", so when BOINC is paused, the task stays in memory i.e. does not get unloaded, so it can continue as soon as the high non-BOINC load abates. This has resulted in me almost never seeing these entries in the message log or in result output. |
||
|
|
dustydave
Cruncher Joined: Mar 5, 2006 Post Count: 21 Status: Offline Project Badges:
|
Thanks for the info. Seems odd that it might be what happened in this case, since I wasn't using the machine myself at the time. I do have it set to suspend if the CPUs go over 50% though; also as you suggest, leave application in memory and run on preferences. I guess it was an open program that failed, which then tripped up BOINC.
I was hoping there would be a rather more descriptive error log to look at. I have been running on the version 6.10.58 so today I have installed the latest 64 bit version 7.0.31 and in future when running overnight I'll close all other open programs. |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Could have been a quirky AV signature update too [exclude the already sandboxed BOINC datadir from AV scanning]. A periodic nightly index that is not low priority competing with the app you left open. The OS or AV Event viewer could reveal something.
Too Many exit(0)s is actually when the task restarted more than 100 times, i.e. a condition that the message log [stdoutdae.txt file] would probably expose. Don't think it's client version related BTW. |
||
|
|
|