I first reported this on the BOINC site for BOINC 4.32. I'm reporting it here because it only seems to happen with ClimatePrediction - I'm also participanting in Einstein@Home, ProteinPredictor@Home and SETI@Home - BOINC tries but fails to suspend hadsm3um_4.12 when its alloted time is up. I'm going to have to suspend my participation in the ClimatePrediction project if I can't get progress on this - my machine is getting hammered!

After a short while, I find 2 projects running when BOINC manager claims that only 1 project is running. For example, right now at 4pm local, Windows XP Task Manager tells me mfoldB125_4.28 (ProteinPredictor) and hadsm3um_4.12 (ClimatePrediction) are running, with elapsed run times of ~2mins and ~5hrs40mins, respectively. BOINC Manager tells me only mfoldB125 is running. My configuration is to allow only 1 at a time and to switch every 60 mins.

According to the messages window, at 10:14:06 it had suspended climateprediction.net. Presumably it failed to do so. Other projects then had some CPU time. At 14:22:25 BOINC again reported that it had suspended climateprediction.net. Interestingly, there is no intervening message saying that it had restarted climateprediction.net.

There appears to be some sort of scheduling bug. I hope the log snippet below is readable.

12/05/2005 09:14:06|climateprediction.net|Restarting result 3cga_200177677_0 using hadsm3 version 4.12
12/05/2005 09:14:08|ProteinPredictorAtHome|Started upload of h0007A_1_40675_2_0
12/05/2005 09:14:08|ProteinPredictorAtHome|Started upload of h0007A_1_40675_2_1
12/05/2005 09:14:09|ProteinPredictorAtHome|Finished upload of h0007A_1_40675_2_0
12/05/2005 09:14:09|ProteinPredictorAtHome|Throughput 6905 bytes/sec
12/05/2005 09:14:09|ProteinPredictorAtHome|Finished upload of h0007A_1_40675_2_1
12/05/2005 09:14:09|ProteinPredictorAtHome|Throughput 188439 bytes/sec
12/05/2005 09:14:09|ProteinPredictorAtHome|Started upload of h0007A_1_40675_2_2
12/05/2005 09:14:11|ProteinPredictorAtHome|Finished upload of h0007A_1_40675_2_2
12/05/2005 09:14:11|ProteinPredictorAtHome|Throughput 31384 bytes/sec
12/05/2005 10:14:06|climateprediction.net|Pausing result 3cga_200177677_0 (removed from memory)
12/05/2005 10:14:06|ProteinPredictorAtHome|Starting result h0007A_1_44283_0 using mfoldB125 version 4.28
12/05/2005 10:14:07||May run out of work in 1.00 days; requesting more
12/05/2005 10:14:07|ProteinPredictorAtHome|Requesting 11672.14 seconds of work
12/05/2005 10:14:07|ProteinPredictorAtHome|Sending request to scheduler: http://predictor.scripps.edu/predictor_cgi/cgi
12/05/2005 10:14:11|ProteinPredictorAtHome|Scheduler RPC to http://predictor.scripps.edu/predictor_cgi/cgi succeeded
12/05/2005 10:14:12|ProteinPredictorAtHome|Started download of h0007A_1_52104.ini
12/05/2005 10:14:12|ProteinPredictorAtHome|Started download of h0007A_1_52104.inp
12/05/2005 10:14:13|ProteinPredictorAtHome|Finished download of h0007A_1_52104.ini
12/05/2005 10:14:13|ProteinPredictorAtHome|Throughput 3177 bytes/sec
12/05/2005 10:14:13|ProteinPredictorAtHome|Finished download of h0007A_1_52104.inp
12/05/2005 10:14:13|ProteinPredictorAtHome|Throughput 460 bytes/sec
12/05/2005 10:14:13|ProteinPredictorAtHome|Started download of h0007A_1_52104.seq
12/05/2005 10:14:13|ProteinPredictorAtHome|Started download of h0007A_1_52104.res
12/05/2005 10:14:14|ProteinPredictorAtHome|Finished download of h0007A_1_52104.seq
12/05/2005 10:14:14|ProteinPredictorAtHome|Throughput 4149 bytes/sec
12/05/2005 10:14:14|ProteinPredictorAtHome|Finished download of h0007A_1_52104.res
12/05/2005 10:14:14|ProteinPredictorAtHome|Throughput 10 bytes/sec
12/05/2005 11:14:14|SETI@home|Restarting result 28ja05ab.22588.258.903410.243_1 using setiathome version 4.09
12/05/2005 11:14:14|ProteinPredictorAtHome|Pausing result h0007A_1_44283_0 (removed from memory)
12/05/2005 11:22:08|SETI@home|Computation for result 28ja05ab.22588.258.903410.243_1 finished
12/05/2005 11:22:08|Einstein@Home|Restarting result H1_0844.5__0844.6_0.1_T02_Fin1_2 using einstein version 4.79
12/05/2005 11:22:09|SETI@home|Started upload of 28ja05ab.22588.258.903410.243_1_0
12/05/2005 11:22:11|SETI@home|Finished upload of 28ja05ab.22588.258.903410.243_1_0
12/05/2005 11:22:11|SETI@home|Throughput 70738 bytes/sec
12/05/2005 12:22:08|Einstein@Home|Pausing result H1_0844.5__0844.6_0.1_T02_Fin1_2 (removed from memory)
12/05/2005 12:22:08|SETI@home|Starting result 19dc04aa.1338.19072.840898.66_3 using setiathome version 4.09
12/05/2005 12:22:13||May run out of work in 1.00 days; requesting more
12/05/2005 12:22:13|SETI@home|Requesting 65613.52 seconds of work
12/05/2005 12:22:13|SETI@home|Sending request to scheduler: http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi
12/05/2005 12:22:14|SETI@home|Scheduler RPC to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded
12/05/2005 12:22:15|SETI@home|Started download of 28ja05aa.29476.9458.586070.93
12/05/2005 12:22:24|SETI@home|Finished download of 28ja05aa.29476.9458.586070.93
12/05/2005 12:22:24|SETI@home|Throughput 39599 bytes/sec
12/05/2005 13:22:24|SETI@home|Pausing result 19dc04aa.1338.19072.840898.66_3 (removed from memory)
12/05/2005 14:22:25|climateprediction.net|Pausing result 3cga_200177677_0 (removed from memory)
12/05/2005 14:22:25|ProteinPredictorAtHome|Restarting result h0007A_1_44283_0 using mfoldB125 version 4.28
12/05/2005 14:45:40|ProteinPredictorAtHome|Computation for result h0007A_1_44283_0 finished
12/05/2005 14:45:40|Einstein@Home|Restarting result H1_0844.5__0844.6_0.1_T02_Fin1_2 using einstein version 4.79

---

Scheduler problems are a known 'feature' of the dev versions of BOINC. Also, a problem with 'trickles', which are unique to CPDN.

See Carl's post at the bottom of <a href="http://www.climateprediction.net/board/viewtopic.php?t=2887&amp;postdays=0&amp;postorder=asc&amp;start=90"> this</a> thread. He also posted something similar in another area of this forum.

I think that, basicly, until this gets sorted, CPDN doesn't work with dev BOINC.

Les

'fraid I don't know about the scheduler problems (haven't experienced them), but I've been keeping up with the latest BIONC dev versions and CPDN trickles which weren't working for me for a while now seem to be fine since I upgraded to 4.42 yesterday.

---

&gt; I think that, basicly, until this gets sorted, CPDN doesn't work with dev
&gt; BOINC.

I would not go quite that far - but it is true that there are problems still to be ironed out. I have noticed myself occasional problems in the handling of the CPDN application by the new BOINC manager which may be apparent when running multi project but not I think caused by it, as it can happen when attached only to a single project. So I'm not so sure it is the scheduler. Even closing the BOINC manager can leave the CPDN programmes open, which should not happen. I've only observed the odd instance myself and have not been able to pin down any common element.

I recommend updating to the current development version if you want to persevere with that. Alternatively, you can drop back to 4.25 if you want to stay with BOINC manager, or revert to 4.19 as the last proved stable version, but I don't know if that is advisable for the other projects you are running.

---

I've been having problems with the latest devs too... But if you can get hold of the installer for CC v4.35, I don't think you'll be disappointed. It seems to me to be the most usable of what's been coming out of the BOINC stable of late.

---

<img src="http://boinc.mundayweb.com/one/stats.php?userID=444&amp;trans=off">

Sorry Andrew
Should have said: Doesn't work reliably with dev versions.
I'm OK on 4.25, the last production version.

Les

This still happens with 4.43, which is officially now the production version.

---