climateprediction.net (CPDN) home page
Thread '4.45 client idle for no apparent reason'

Thread '4.45 client idle for no apparent reason'

Message boards : Number crunching : 4.45 client idle for no apparent reason
Message board moderation

To post messages, you must log in.

AuthorMessage
Profileold_user1179

Send message
Joined: 25 Aug 04
Posts: 11
Credit: 702,217
RAC: 0
Message 14237 - Posted: 9 Jul 2005, 16:17:01 UTC

I awoke this morning to find my client idle all night for no reason. My system idle process was at 99%. No relevant error or warning messages were in the log. I shutdown the client, reopened it and it started running again like normal. This might be a bug with the new client, so be warned.
ID: 14237 · Report as offensive     Reply Quote
ProfileThyme Lawn
Volunteer moderator

Send message
Joined: 5 Aug 04
Posts: 1283
Credit: 15,824,334
RAC: 0
Message 14241 - Posted: 9 Jul 2005, 17:02:16 UTC - in response to Message 14237.  
Last modified: 9 Jul 2005, 17:06:10 UTC

> I awoke this morning to find my client idle all night for no reason. My
> system idle process was at 99%. No relevant error or warning messages were in
> the log. I shutdown the client, reopened it and it started running again like
> normal. This might be a bug with the new client, so be warned.

Did BOINC run a periodic benchmark overnight? This has been seen on the Sulphur-cycle alpha and beta projects, with BOINC generating messages sucha as the following:

2005-06-21 07:48:22 [---] Suspending computation and network activity - running CPU benchmarks
2005-06-21 07:48:22 [climateprediction.net] Pausing result 2pvt_300148139_1 (removed from memory)
2005-06-21 07:48:22 [CPDN Alpha] Pausing result xagb_000290050_5 (removed from memory)
2005-06-21 07:48:24 [---] Running CPU benchmarks
2005-06-21 07:48:25 [---] request_reschedule_cpus: process exited
2005-06-21 07:48:32 [---] Aborting CPU benchmarks, one or more active tasks are still running.
2005-06-21 07:48:32 [---] Resuming computation and network activity
2005-06-21 07:48:32 [---] schedule_cpus: must schedule
2005-06-21 07:48:32 [climateprediction.net] Restarting result 2pvt_300148139_1 using hadsm3 version 4.12

<br><a href="http://www.teampicard.net/"><img src="http://www.teampicard.net/images/picardmini.gif"></a><a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/team_display.php?teamid=3">Join us here</a>
ID: 14241 · Report as offensive     Reply Quote
crandles
Volunteer moderator

Send message
Joined: 16 Oct 04
Posts: 692
Credit: 277,679
RAC: 0
Message 14242 - Posted: 9 Jul 2005, 17:05:35 UTC

Had a benchmark been run and got messages like this:

7/9/2005 6:39:51 AM|climateprediction.net|Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdn_cgi/cgi succeeded
7/9/2005 8:21:57 AM||Suspending computation and network activity - running CPU benchmarks
7/9/2005 8:21:57 AM|climateprediction.net|Pausing result 45r7_000294355_0 (removed from memory)
7/9/2005 8:21:59 AM||Running CPU benchmarks
7/9/2005 8:22:07 AM||Aborting CPU benchmarks, one or more active tasks are still running.
7/9/2005 8:22:07 AM||Resuming computation and network activity
7/9/2005 8:22:07 AM||request_reschedule_cpus: Resuming activities
7/9/2005 8:22:07 AM||request_reschedule_cpus: process exited

If so, Chris Sutton and Thyme Lawn have reported it to the BOINC development list.
_______________________________
Visit <a href="http://boinc-doc.net/boinc-wiki/index.php?title=Main_Page">BOINC WIKI</a> for help

And join <a href="http://www.boincsynergy.com/">BOINC Synergy</a> for all the news in one place.
ID: 14242 · Report as offensive     Reply Quote
Profileold_user1179

Send message
Joined: 25 Aug 04
Posts: 11
Credit: 702,217
RAC: 0
Message 14243 - Posted: 9 Jul 2005, 17:09:34 UTC - in response to Message 14242.  

Yes it did. Great, the problem has been reported. Thanks for the replys.

Regards,
Jon
ID: 14243 · Report as offensive     Reply Quote
Profileold_user1179

Send message
Joined: 25 Aug 04
Posts: 11
Credit: 702,217
RAC: 0
Message 14244 - Posted: 9 Jul 2005, 17:11:10 UTC - in response to Message 14243.  

Why is a benchmark even run for this type of project?
ID: 14244 · Report as offensive     Reply Quote
ProfileThyme Lawn
Volunteer moderator

Send message
Joined: 5 Aug 04
Posts: 1283
Credit: 15,824,334
RAC: 0
Message 14245 - Posted: 9 Jul 2005, 17:33:20 UTC - in response to Message 14244.  
Last modified: 9 Jul 2005, 18:02:45 UTC

&gt; Why is a benchmark even run for this type of project?

The benchmark is part of BOINC, and it's built in to run every 5 days.

It's needed for projects where credits are claimed by clients based on their benchmarks and granted by the server based on cross-comparison of what other computers claim for the same workunit. Which is not the way CPDN works.
<br><a href="http://www.teampicard.net/"><img src="http://www.teampicard.net/images/picardmini.gif"></a><a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/team_display.php?teamid=3">Join us here</a>
ID: 14245 · Report as offensive     Reply Quote
Profileold_user1179

Send message
Joined: 25 Aug 04
Posts: 11
Credit: 702,217
RAC: 0
Message 15319 - Posted: 24 Aug 2005, 0:20:32 UTC

Is there a work around yet for this problem? I've already lost days of CPU time as a result to this bug.
ID: 15319 · Report as offensive     Reply Quote
Profilegeophi
Volunteer moderator

Send message
Joined: 7 Aug 04
Posts: 2187
Credit: 64,822,615
RAC: 5,275
Message 15321 - Posted: 24 Aug 2005, 0:54:00 UTC - in response to Message 15319.  

&gt; Is there a work around yet for this problem? I've already lost days of CPU
&gt; time as a result to this bug.

You can check out <a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/forum_thread.php?id=2921">this thread</a> and determine if you want to use the 4.45b version that Ralic (Chris Sutton) has compiled. It does not suffer the benchmark problem. I've been running it the last week or so without problems.
ID: 15321 · Report as offensive     Reply Quote
ProfileHonza
Volunteer moderator

Send message
Joined: 5 Aug 04
Posts: 390
Credit: 2,475,242
RAC: 0
Message 15333 - Posted: 24 Aug 2005, 8:35:31 UTC - in response to Message 15319.  

Not much from UCB but you can use Chris Sutton's compilation of BOINC 4.45b that - among others - prlong benchmark timeout from 10 to 30 seconds. This was found as a working method to overcome BOINC-CPDN-benchmark bug/obstacle.
http://arnaudboinc.free.fr/

&gt; Is there a work around yet for this problem? I've already lost days of CPU time as a result to this bug.
_________________________
<i>Everyday i wished classic phpBB forum being back and after 60 prayers, all are </i><a href="http://www.climateprediction.net/board">invited</a>
ID: 15333 · Report as offensive     Reply Quote

Message boards : Number crunching : 4.45 client idle for no apparent reason

©2024 cpdn.org