hi there,

one of my nine machines is getting strange.

the server did not recive any trickle of this one since August 28th. here are some Logs;

stderrdae.txt:
<pre>
2005-08-25 18:37:13 [climateprediction.net] Project is down
2005-08-25 18:37:14 [climateprediction.net] Deferring communication with project for 3 hours, 21 minutes, and 48 seconds
2005-08-25 19:37:15 [climateprediction.net] Deferring communication with project for 2 hours, 21 minutes, and 48 seconds
2005-08-25 20:37:16 [climateprediction.net] Deferring communication with project for 1 hours, 21 minutes, and 47 seconds
2005-08-25 21:37:16 [climateprediction.net] Deferring communication with project for 21 minutes and 46 seconds
2005-08-28 13:16:36 [---] Aborting CPU benchmarks, one or more active tasks are still running.
</pre>

this is getting more and more often ... the computation of the project is noct restarting after the benchmark.

next morning the computer was booted up again:
stdoutdae.txt
<pre>
StartServiceCtrlDispatcher being called.
This may take several seconds. Please wait.
2005-08-29 11:01:31 [---] Starting BOINC client version 4.45 for windows_intelx86
2005-08-29 11:01:31 [---] Executing as a daemon
2005-08-29 11:01:31 [---] Data directory: C:\\Programme\\BOINC
2005-08-29 11:01:31 [---] BOINC is running as a service and as a non-system user.
2005-08-29 11:01:31 [---] No application graphics will be available.
2005-08-29 11:01:32 [climateprediction.net] Computer ID: 54302; location: work; project prefs: work
2005-08-29 11:01:32 [---] General prefs: from climateprediction.net (last modified 2005-07-27 21:08:48)
2005-08-29 11:01:32 [---] General prefs: using separate prefs for work
2005-08-29 11:01:32 [---] Remote control not allowed; using loopback address
2005-08-29 11:01:32 [climateprediction.net] Resuming computation for result 1535_200073804_1 using hadsm3 version 4.13
2005-08-29 11:01:45 [climateprediction.net] Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2005-08-29 11:01:45 [climateprediction.net] Requesting 0 seconds of work, returning 0 results
2005-08-29 11:01:57 [climateprediction.net] Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
2005-08-29 11:01:58 [---] Suspending work fetch because computer is overcommitted.
2005-08-29 11:01:58 [---] Using earliest-deadline-first scheduling because computer is overcommitted.
</pre>

what the heck means this?
i\'ve no idea what would cause this error. nothing is changed on this machine.

by the way: the computer computes and talks with the server all day long.
for example from stdoutdae.txt
<pre>
2005-08-30 10:59:22 [climateprediction.net] Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2005-08-30 10:59:22 [climateprediction.net] Requesting 0 seconds of work, returning 0 results
2005-08-30 10:59:23 [climateprediction.net] Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
[...]
2005-08-31 09:24:29 [climateprediction.net] Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2005-08-31 09:24:29 [climateprediction.net] Requesting 0 seconds of work, returning 0 results
2005-08-31 09:24:30 [climateprediction.net] Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
[... day by day ...]
2005-09-05 06:53:50 [climateprediction.net] Sending scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi
2005-09-05 06:53:50 [climateprediction.net] Requesting 0 seconds of work, returning 0 results
2005-09-05 06:53:51 [climateprediction.net] Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi succeeded
</pre>
but no tricklecheck was seen on my user-details.


the last message, that the compute would be overcomitted came at 2005-09-02 14:56:23.


what is to do, that the machine would be trickeling again?


greets, Micha

the problem is still there ...

please tell me anyone what to do!

Have you tried <a href="http://climateapps2.oucs.ox.ac.uk/cpdnboinc/forum_thread.php?id=2921">ver 4.45b of the BOINC client</a>? This seems to overcome the benchmarking problem.