| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Support
Forum: BOINC Agent Support
Thread: WU"s sit idle after a day of running |
| No member browsing this thread |
|
Thread Status: Active Total posts in this thread: 8
|
|
| Author |
|
|
hunterkasy
Senior Cruncher USA Joined: Dec 8, 2008 Post Count: 300 Status: Offline Project Badges: 
|
Boinc version: 6.12.34 (64)
Dell Inspiron Intel dual core 3 ghtz 3 gigs memory win 7 home (64) The problem I am having after around a day of running the WU's they end up sitting idle until I restart the computer. after I restart the computer the wu's will run like they are set at (60% of cpu) then after around a day the wu's will sit idle. if I open up the task manager it will show the cpu running a 0-1% If I don't restart the machine I end up with wu's erroring out do to max time exceeded what can I do to fix this Thanks for any help |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Which specific sciences are doing this?
--//-- |
||
|
|
hunterkasy
Senior Cruncher USA Joined: Dec 8, 2008 Post Count: 300 Status: Offline Project Badges:
|
GO Fight Against Malaria
Drug Search for Leishmaniasis it does it on both of them |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
Yes, exactly [see prior threads in DSFL/GFAM forums on stuck work units]. The VINA engine using science apps don't like the constant start stop way of BOINC when using the CPU Time percent [u mention you got it set to 60%]. Set CPU time to 100% and use alternate tools such as TThrottle or ThreadMasterGUI to slow down crunching to reduce CPU temperature [mostly for laptops], the only reason why BOINC has this control. As of yet the techs have not found a fix for this issue and seemingly only existing on Windows [where else].
The quickest way to immediately reduce system temps if an issue, when not done for a long time or never, is to open the desktop case and clean out all fans and cooling elements [carefully]. For laptops use compressed air. The Chat Room forum has a dust-bunnies thread if you like to visit prior discussion. --//-- |
||
|
|
BSD
Senior Cruncher Joined: Apr 27, 2011 Post Count: 224 Status: Offline |
|
||
|
|
hunterkasy
Senior Cruncher USA Joined: Dec 8, 2008 Post Count: 300 Status: Offline Project Badges:
|
Yes, exactly [see prior threads in DSFL/GFAM forums on stuck work units]. The VINA engine using science apps don't like the constant start stop way of BOINC when using the CPU Time percent [u mention you got it set to 60%]. Set CPU time to 100% and use alternate tools such as TThrottle or ThreadMasterGUI to slow down crunching to reduce CPU temperature [mostly for laptops], the only reason why BOINC has this control. As of yet the techs have not found a fix for this issue and seemingly only existing on Windows [where else]. The quickest way to immediately reduce system temps if an issue, when not done for a long time or never, is to open the desktop case and clean out all fans and cooling elements [carefully]. For laptops use compressed air. The Chat Room forum has a dust-bunnies thread if you like to visit prior discussion. --//-- thanks, do you know if it's having problem with all windows or just 7? I have a xp machine that I just installed boinc on, I have not had any problems with my Ubuntu machines |
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
I have two P4s, one with WinXP set at 80% of CPU time (non-HT) and one with Debian set at 33% (HT on), crunching DSFL and GFAM with no problems at all. Just noisy fans :)
|
||
|
|
Former Member
Cruncher Joined: May 22, 2018 Post Count: 0 Status: Offline |
There's no rhyme or reason other than the structure of the control + worker app way of working with VINA on Windows [cant say the issue focuses on any particular Windows version, else the techs would see that in their detail fail stats]. CEP2 which has a similar structure using Q-Chem and a control + worker app has no reports, but then these units cut off at 12 hours, so may just escape attention. Seen a few CEP2 wingman that have an endless cycle of restarts at the first checkpoint, but these 'would' cut off after 100 sequential restarts IIRC.
--//-- |
||
|
|
|