BOINC 5.7.2 is out now in BETA. If you happen across specific problems, but aren't registered to the Alpha email list, leave your trouble here and we'll relay it to the devs.

---

BOINC 5.7.2 is out now in BETA. If you happen across specific problems, but aren't registered to the Alpha email list, leave your trouble here and we'll relay it to the devs.

I was a member ages ago - but 100% usage burned out my lap top fan.
I'm back, with manager 5.7.1 and rosetta@home project, was able to set to 60% OK.
However, application only runs for a few minutes, then stops for several minutes, I give up waiting and have to suspend/resume, or sometimes exit/restart Boinc manager. Application then picks up where it was, and runs for a few minutes again. My preferences have it set to run all the time (I think) but it seems to keep stopping on it's own.

While it ususally picks up where it left off, on occasion the completion of the application has suddenly gone from 1-2% up to 100%. Which of course sends as completed, up to BOINC/Rosetta land. I had this with 5.6.4 when I started a few days ago, upgrading to 5.7.1 didn't help. I'd rather not downgrade and use Threadmaster (sp?), I've a long history in QA and want get this resolved.
thanks
Mike

---

BOINC 5.7.2 is out now in BETA. If you happen across specific problems, but aren't registered to the Alpha email list, leave your trouble here and we'll relay it to the devs.

I was a member ages ago - but 100% usage burned out my lap top fan.
I'm back, with manager 5.7.1 and rosetta@home project, was able to set to 60% OK.
However, application only runs for a few minutes, then stops for several minutes, I give up waiting and have to suspend/resume, or sometimes exit/restart Boinc manager. Application then picks up where it was, and runs for a few minutes again. My preferences have it set to run all the time (I think) but it seems to keep stopping on it's own.

While it ususally picks up where it left off, on occasion the completion of the application has suddenly gone from 1-2% up to 100%. Which of course sends as completed, up to BOINC/Rosetta land. I had this with 5.6.4 when I started a few days ago, upgrading to 5.7.1 didn't help. I'd rather not downgrade and use Threadmaster (sp?), I've a long history in QA and want get this resolved.
thanks
Mike

I checked for more clues - Task status is switching every secong from "Running" to "Suspended - CPU throttled"

Plus, following being written every second to the messages section.
"02/11/2006 6:43:00 PM|rosetta@home|Task BENCH_ABRELAX_SAVE_ALL_OUT_4ubpA_BARCODE_R97_filters_1328_906_0: no shared memory segment"
I'll stop BOINC for now.
Mike

---

BOINC 5.7.2 is out now in BETA. If you happen across specific problems, but aren't registered to the Alpha email list, leave your trouble here and we'll relay it to the devs.

I was a member ages ago - but 100% usage burned out my lap top fan.
I'm back, with manager 5.7.1 and rosetta@home project, was able to set to 60% OK.
However, application only runs for a few minutes, then stops for several minutes, I give up waiting and have to suspend/resume, or sometimes exit/restart Boinc manager. Application then picks up where it was, and runs for a few minutes again. My preferences have it set to run all the time (I think) but it seems to keep stopping on it's own.

While it ususally picks up where it left off, on occasion the completion of the application has suddenly gone from 1-2% up to 100%. Which of course sends as completed, up to BOINC/Rosetta land. I had this with 5.6.4 when I started a few days ago, upgrading to 5.7.1 didn't help. I'd rather not downgrade and use Threadmaster (sp?), I've a long history in QA and want get this resolved.
thanks
Mike

I checked for more clues - Task status is switching every secong from "Running" to "Suspended - CPU throttled"

Plus, following being written every second to the messages section.
"02/11/2006 6:43:00 PM|rosetta@home|Task BENCH_ABRELAX_SAVE_ALL_OUT_4ubpA_BARCODE_R97_filters_1328_906_0: no shared memory segment"
I'll stop BOINC for now.
Mike

The CPU limiting is designed to do exactly as you are seeing. 50% mean 30 sec active 50% suspended. The 75% means on for 45 sec. and suspended for 15 sec.

---

Boinc V 7.4.36
Win7 i5 3.33G 4GB NVidia 470

BOINC 5.7.2 is out now in BETA. If you happen across specific problems, but aren't registered to the Alpha email list, leave your trouble here and we'll relay it to the devs.

I was a member ages ago - but 100% usage burned out my lap top fan.
I'm back, with manager 5.7.1 and rosetta@home project, was able to set to 60% OK.
However, application only runs for a few minutes, then stops for several minutes, I give up waiting and have to suspend/resume, or sometimes exit/restart Boinc manager. Application then picks up where it was, and runs for a few minutes again. My preferences have it set to run all the time (I think) but it seems to keep stopping on it's own.

While it ususally picks up where it left off, on occasion the completion of the application has suddenly gone from 1-2% up to 100%. Which of course sends as completed, up to BOINC/Rosetta land. I had this with 5.6.4 when I started a few days ago, upgrading to 5.7.1 didn't help. I'd rather not downgrade and use Threadmaster (sp?), I've a long history in QA and want get this resolved.
thanks
Mike

I checked for more clues - Task status is switching every secong from "Running" to "Suspended - CPU throttled"

Plus, following being written every second to the messages section.
"02/11/2006 6:43:00 PM|rosetta@home|Task BENCH_ABRELAX_SAVE_ALL_OUT_4ubpA_BARCODE_R97_filters_1328_906_0: no shared memory segment"
I'll stop BOINC for now.
Mike

The CPU limiting is designed to do exactly as you are seeing. 50% mean 30 sec active 50% suspended. The 75% means on for 45 sec. and suspended for 15 sec.

At 50% I do not see this running active for 30 seconds, then suspended for 30 seconds. Is this what your system shows? Mine switches EVERY second, initially from 100% to 1%, as observed using the Windows Task Manager, but after a minute or so gradually narrows the swings to betwen 55% and 45%. But then after another few minutes CPU usuage drops to 1%, and stays there - I've waited up to an hour. The task status indicator was still switching as I mentioned, but the task wasn't really running - as my laptop fan stopped cycling on and off. Plus -there was that repeating message error. Don't always se error messages when this happens though. I tried the eimstein@home project, same problems. Seems to vary from 4-20 minutes as tohow long this seems to run OK, so I don't know what other variables are involved. Doesn't seem to relate towheter I do other things on the computer or just leave it be. If you are a developer (or anyone else reading this is) please send me a note as to what specific information/tests you need so that I can help resolve this problem.

---

My apologies, you are correct about the switching. No excuse for my brain cramp. I should have rechecked before posting. I need to find some time to play with this feature.

My apologies, you are correct about the switching. No excuse for my brain cramp. I should have rechecked before posting. I need to find some time to play with this feature.

No problem. I'm a little frustrated I guess by this problem - I ran Find-a-drug for ages on another PC, wanted to do the same with BOINC on my laptop but couldn't because of the 100% usage. I thought there was a fix - looks like not quite yet.
Mike in Ottawa

---

BOINC 5.7.2 is out now in BETA. If you happen across specific problems, but aren't registered to the Alpha email list, leave your trouble here and we'll relay it to the devs.

I was a member ages ago - but 100% usage burned out my lap top fan.
I'm back, with manager 5.7.1 and rosetta@home project, was able to set to 60% OK.
However, application only runs for a few minutes, then stops for several minutes, I give up waiting and have to suspend/resume, or sometimes exit/restart Boinc manager. Application then picks up where it was, and runs for a few minutes again. My preferences have it set to run all the time (I think) but it seems to keep stopping on it's own.

While it ususally picks up where it left off, on occasion the completion of the application has suddenly gone from 1-2% up to 100%. Which of course sends as completed, up to BOINC/Rosetta land. I had this with 5.6.4 when I started a few days ago, upgrading to 5.7.1 didn't help. I'd rather not downgrade and use Threadmaster (sp?), I've a long history in QA and want get this resolved.
thanks
Mike

I checked for more clues - Task status is switching every secong from "Running" to "Suspended - CPU throttled"

Plus, following being written every second to the messages section.
"02/11/2006 6:43:00 PM|rosetta@home|Task BENCH_ABRELAX_SAVE_ALL_OUT_4ubpA_BARCODE_R97_filters_1328_906_0: no shared memory segment"
I'll stop BOINC for now.
Mike

The CPU limiting is designed to do exactly as you are seeing. 50% mean 30 sec active 50% suspended. The 75% means on for 45 sec. and suspended for 15 sec.

At 50% I do not see this running active for 30 seconds, then suspended for 30 seconds. Is this what your system shows? Mine switches EVERY second, initially from 100% to 1%, as observed using the Windows Task Manager, but after a minute or so gradually narrows the swings to betwen 55% and 45%. But then after another few minutes CPU usuage drops to 1%, and stays there - I've waited up to an hour. The task status indicator was still switching as I mentioned, but the task wasn't really running - as my laptop fan stopped cycling on and off. Plus -there was that repeating message error. Don't always se error messages when this happens though. I tried the eimstein@home project, same problems. Seems to vary from 4-20 minutes as tohow long this seems to run OK, so I don't know what other variables are involved. Doesn't seem to relate towheter I do other things on the computer or just leave it be. If you are a developer (or anyone else reading this is) please send me a note as to what specific information/tests you need so that I can help resolve this problem.

I see a similar problem. Since it seems that the 1 sec on/off switching of the cpu throttling is "beating" with the (presumeably) 1 sec default update time of the task manager, I can get a rock steady reading by setting the update speed to low.
However, after an indeterminate period of time, seemingly not triggered by anything particular,the cpu use drops to almost zero, and the boinc app shows the projects cycling between "running" and "suspended - cpu throttling". Once in that state, the app does not recover. It is almost as if the boinc app is "beating" with itself, ie it decides to switch itself on, and when it does so says "hey, time to turn off". I know none of this sounds very technical, but it is a problem that occasionally crops up in control systems.

Mike P

I saw a similar problem with BOINC 5.7.2 running the Help Defeat Cancer application on World Community Grid. The application got to about 270K memory utilization and then stopped using CPU (showing 0% CPU utilization). I have the overrides file set to use 80% CPU. It turned out the problem was having the preferences set to write to disk every 60 seconds....increasing to 999 seconds fixed the problem and the application ran to completion without a problem.

---

Just saw this problem again with BOINC 5.72 running the FAAH application from WCG. Stopping and starting BOINC caused the application to start running again properly. Please let me know what diagnostic info would help to debug this.

---

Just saw this problem again with BOINC 5.72 running the FAAH application from WCG. Stopping and starting BOINC caused the application to start running again properly. Please let me know what diagnostic info would help to debug this.

I have 5.7.1
Tried changing write to disc to 999 sec, and to 600 sec, and Leave applications in memory to Yes - no discernable changes in symptoms. Usually suspend/resume gets goibg again for a few minutes. If I'm doing nothing else on pc lasts laniger, like 10-15 minutes, if I'm doing email/browsing seems to die in 3-4 minutes.

I too would like to know what specific parameter changes and data would help in debugging this - have had no interest back from developers so far.

---

Have installed 5.7.3 but it identifies itself as 5.7.2.

Both 5.6.x as well as 5.7.x worked fine so far on my laptop even when running the CPU at 100% nearly 7/24 and using it for daily work.

Have installed 5.7.3 but it identifies itself as 5.7.2.

Both 5.6.x as well as 5.7.x worked fine so far on my laptop even when running the CPU at 100% nearly 7/24 and using it for daily work.

100% is a piece of cake for my laptop, but it overheats. Try 50% and let me know how long it hangs in there, please.
Einstein is good for maybe 10-15 minutes, Rosetta only 5 mins maybe. Both seem to run at the longer times before freezing if I do nothing else on the PC.
Running the Rosetta graphics is entertaining, but after a few minutes application not only freezes, but when I restart, it is at 100% with nasty computation errors.

---

Have installed 5.7.3 and before I was using 5.6.5.

5.6.5 worked well since I installed it. Yesterday or two days ago it started to change the status of a task from active to suspended - cpu throttled every second. Then I installed 5.7.3 and the same happens.

Here's my log, hope it helps:

17.11.2006 07:35:06||Starting BOINC client version 5.7.2 for windows_intelx86
17.11.2006 07:35:06||log flags: task, file_xfer, sched_ops
17.11.2006 07:35:06||Libraries: libcurl/7.15.5 OpenSSL/0.9.8a zlib/1.2.3
17.11.2006 07:35:06||Data directory: C:\\Programme\\BOINC
17.11.2006 07:35:06|SETI@home|Found app_info.xml; using anonymous platform
17.11.2006 07:35:06||Processor: 2 GenuineIntel Intel(R) Pentium(R) 4 CPU 3.00GHz
17.11.2006 07:35:06||Memory: 502.48 MB physical, 1.20 GB virtual
17.11.2006 07:35:06||Disk: 74.53 GB total, 61.83 GB free
17.11.2006 07:35:07|climateprediction.net|URL: http://climateprediction.net/; Computer ID: 493393; location: (none); project prefs: default
17.11.2006 07:35:07|Einstein@Home|URL: http://einstein.phys.uwm.edu/; Computer ID: 770812; location: (none); project prefs: default
17.11.2006 07:35:07|SETI@home|URL: http://setiathome.berkeley.edu/; Computer ID: 2768620; location: home; project prefs: default
17.11.2006 07:35:07||General prefs: from SETI@home (last modified 2006-10-18 07:03:38)
17.11.2006 07:35:07||Host location: home
17.11.2006 07:35:07||General prefs: no separate prefs for home; using your defaults
17.11.2006 07:35:07|climateprediction.net|Restarting task hadcm3ohc_1dzc_05613135_0 using hadcm3 version 515
17.11.2006 07:35:07|SETI@home|Restarting task 14jn03ab.20420.23504.634652.3.236_4 using setiathome_enhanced version 515
17.11.2006 07:35:07||compute_work_requests(): start
17.11.2006 07:36:08||compute_work_requests(): start
17.11.2006 07:37:08||compute_work_requests(): start
17.11.2006 07:38:09||compute_work_requests(): start
17.11.2006 07:39:10||compute_work_requests(): start
17.11.2006 07:40:10||compute_work_requests(): start
17.11.2006 07:41:10||compute_work_requests(): start
17.11.2006 07:42:11||compute_work_requests(): start
17.11.2006 07:43:11||compute_work_requests(): start
17.11.2006 07:44:11||compute_work_requests(): start
17.11.2006 07:45:11||compute_work_requests(): start
17.11.2006 07:46:11||compute_work_requests(): start
17.11.2006 07:47:13||compute_work_requests(): start
17.11.2006 07:48:13||compute_work_requests(): start
17.11.2006 07:49:13||compute_work_requests(): start
17.11.2006 07:50:13||compute_work_requests(): start
17.11.2006 07:51:14||compute_work_requests(): start
17.11.2006 07:52:14||compute_work_requests(): start
17.11.2006 07:53:14||compute_work_requests(): start
17.11.2006 07:54:14||compute_work_requests(): start
17.11.2006 07:55:14||compute_work_requests(): start
17.11.2006 07:56:16||compute_work_requests(): start
17.11.2006 07:57:16||compute_work_requests(): start
17.11.2006 07:58:16||compute_work_requests(): start
17.11.2006 07:59:16||compute_work_requests(): start
17.11.2006 08:00:16||compute_work_requests(): start
17.11.2006 08:01:17||compute_work_requests(): start
17.11.2006 08:02:17||compute_work_requests(): start
17.11.2006 08:03:17||compute_work_requests(): start
17.11.2006 08:04:18||compute_work_requests(): start
17.11.2006 08:05:19||compute_work_requests(): start
17.11.2006 08:06:19||compute_work_requests(): start
17.11.2006 08:07:19||compute_work_requests(): start
17.11.2006 08:08:19||compute_work_requests(): start
17.11.2006 08:09:20||compute_work_requests(): start
17.11.2006 08:10:20||compute_work_requests(): start
17.11.2006 08:11:21||compute_work_requests(): start
17.11.2006 08:12:21||compute_work_requests(): start
17.11.2006 08:13:21||compute_work_requests(): start
17.11.2006 08:14:22||compute_work_requests(): start
17.11.2006 08:15:22||compute_work_requests(): start
17.11.2006 08:16:22||compute_work_requests(): start
17.11.2006 08:17:22||compute_work_requests(): start
17.11.2006 08:18:22||compute_work_requests(): start
17.11.2006 08:19:23||compute_work_requests(): start
17.11.2006 08:20:24||compute_work_requests(): start
17.11.2006 08:21:24||compute_work_requests(): start
17.11.2006 08:22:24||compute_work_requests(): start
17.11.2006 08:23:24||compute_work_requests(): start
17.11.2006 08:24:25||compute_work_requests(): start
17.11.2006 08:25:25||compute_work_requests(): start
17.11.2006 08:26:25||compute_work_requests(): start
17.11.2006 08:27:26||compute_work_requests(): start
17.11.2006 08:28:26||compute_work_requests(): start
17.11.2006 08:29:27||compute_work_requests(): start
17.11.2006 08:30:27||compute_work_requests(): start
17.11.2006 08:31:27||compute_work_requests(): start
17.11.2006 08:32:27||compute_work_requests(): start
17.11.2006 08:33:28||compute_work_requests(): start
17.11.2006 08:34:28||compute_work_requests(): start
17.11.2006 08:35:29||compute_work_requests(): start
17.11.2006 08:36:29||compute_work_requests(): start
17.11.2006 08:37:30||compute_work_requests(): start
17.11.2006 08:38:30||compute_work_requests(): start
17.11.2006 08:39:30||compute_work_requests(): start
17.11.2006 08:40:30||compute_work_requests(): start
17.11.2006 08:41:30||compute_work_requests(): start
17.11.2006 08:42:31||compute_work_requests(): start
17.11.2006 08:43:32||compute_work_requests(): start
17.11.2006 08:44:32||compute_work_requests(): start
17.11.2006 08:45:32||compute_work_requests(): start
17.11.2006 08:46:33||compute_work_requests(): start
17.11.2006 08:47:33||compute_work_requests(): start
17.11.2006 08:48:33||compute_work_requests(): start
17.11.2006 08:49:33||compute_work_requests(): start
17.11.2006 08:50:34||compute_work_requests(): start
17.11.2006 08:51:35||compute_work_requests(): start
17.11.2006 08:52:35||compute_work_requests(): start
17.11.2006 08:53:35||compute_work_requests(): start
17.11.2006 08:54:35||compute_work_requests(): start
17.11.2006 08:55:35||compute_work_requests(): start
17.11.2006 08:56:36||compute_work_requests(): start
17.11.2006 08:57:36||compute_work_requests(): start
17.11.2006 08:58:36||compute_work_requests(): start
17.11.2006 08:59:37||compute_work_requests(): start
17.11.2006 09:00:37||compute_work_requests(): start
17.11.2006 09:01:38||compute_work_requests(): start
17.11.2006 09:02:38||compute_work_requests(): start
17.11.2006 09:03:38||compute_work_requests(): start
17.11.2006 09:04:38||compute_work_requests(): start
17.11.2006 09:05:38||compute_work_requests(): start
17.11.2006 09:06:40||compute_work_requests(): start
17.11.2006 09:07:40||compute_work_requests(): start
17.11.2006 09:08:40||compute_work_requests(): start
17.11.2006 09:09:40||compute_work_requests(): start
17.11.2006 09:10:40||compute_work_requests(): start
17.11.2006 09:11:41||compute_work_requests(): start
17.11.2006 09:12:41||compute_work_requests(): start
17.11.2006 09:13:41||compute_work_requests(): start
17.11.2006 09:14:42||compute_work_requests(): start
17.11.2006 09:15:42||compute_work_requests(): start
17.11.2006 09:16:43||compute_work_requests(): start
17.11.2006 09:17:43||compute_work_requests(): start
17.11.2006 09:18:43||compute_work_requests(): start
17.11.2006 09:19:43||compute_work_requests(): start
17.11.2006 09:20:43||compute_work_requests(): start
17.11.2006 09:21:44||compute_work_requests(): start
17.11.2006 09:22:45||compute_work_requests(): start
17.11.2006 09:23:45||compute_work_requests(): start
17.11.2006 09:24:45||compute_work_requests(): start
17.11.2006 09:25:45||compute_work_requests(): start
17.11.2006 09:26:46||compute_work_requests(): start
17.11.2006 09:27:46||compute_work_requests(): start
17.11.2006 09:28:46||compute_work_requests(): start
17.11.2006 09:29:46||compute_work_requests(): start
17.11.2006 09:30:47||compute_work_requests(): start
17.11.2006 09:31:48||compute_work_requests(): start
17.11.2006 09:32:48||compute_work_requests(): start
17.11.2006 09:33:48||compute_work_requests(): start
17.11.2006 09:34:48||compute_work_requests(): start
17.11.2006 09:35:49||compute_work_requests(): start
17.11.2006 09:36:49||compute_work_requests(): start
17.11.2006 09:37:49||compute_work_requests(): start
17.11.2006 09:38:50||compute_work_requests(): start
17.11.2006 09:39:50||compute_work_requests(): start
17.11.2006 09:40:51||compute_work_requests(): start
17.11.2006 09:41:51||compute_work_requests(): start
17.11.2006 09:42:51||compute_work_requests(): start
17.11.2006 09:43:51||compute_work_requests(): start
17.11.2006 09:44:51||compute_work_requests(): start
17.11.2006 09:45:53||compute_work_requests(): start
17.11.2006 09:46:53||compute_work_requests(): start
17.11.2006 09:47:53||compute_work_requests(): start
17.11.2006 09:48:53||compute_work_requests(): start
17.11.2006 09:49:53||compute_work_requests(): start
17.11.2006 09:50:54||compute_work_requests(): start
17.11.2006 09:51:54||compute_work_requests(): start
17.11.2006 09:52:54||compute_work_requests(): start
17.11.2006 09:53:55||compute_work_requests(): start
17.11.2006 09:54:55||compute_work_requests(): start
17.11.2006 09:55:56||compute_work_requests(): start
17.11.2006 09:56:56||compute_work_requests(): start
17.11.2006 09:57:56||compute_work_requests(): start
17.11.2006 09:58:56||compute_work_requests(): start
17.11.2006 09:59:56||compute_work_requests(): start
17.11.2006 10:00:58||compute_work_requests(): start
17.11.2006 10:01:58||compute_work_requests(): start
17.11.2006 10:02:58||compute_work_requests(): start
17.11.2006 10:03:58||compute_work_requests(): start
17.11.2006 10:04:58||compute_work_requests(): start
17.11.2006 10:05:58||compute_work_requests(): start
17.11.2006 10:06:59||compute_work_requests(): start
17.11.2006 10:07:59||compute_work_requests(): start
17.11.2006 10:08:59||compute_work_requests(): start
17.11.2006 10:10:00||compute_work_requests(): start
17.11.2006 10:11:00||compute_work_requests(): start
17.11.2006 10:12:01||compute_work_requests(): start
17.11.2006 10:13:01||compute_work_requests(): start
17.11.2006 10:14:01||compute_work_requests(): start
17.11.2006 10:15:02||compute_work_requests(): start
17.11.2006 10:16:02||compute_work_requests(): start
17.11.2006 10:17:02||compute_work_requests(): start
17.11.2006 10:18:03||compute_work_requests(): start
17.11.2006 10:19:04||compute_work_requests(): start
17.11.2006 10:20:04||compute_work_requests(): start
17.11.2006 10:21:04||compute_work_requests(): start
17.11.2006 10:22:04||compute_work_requests(): start
17.11.2006 10:23:04||compute_work_requests(): start
17.11.2006 10:24:04||compute_work_requests(): start
17.11.2006 10:25:06||compute_work_requests(): start
17.11.2006 10:26:06||compute_work_requests(): start
17.11.2006 10:27:06||compute_work_requests(): start
17.11.2006 10:28:06||compute_work_requests(): start
17.11.2006 10:29:07||compute_work_requests(): start
17.11.2006 10:30:07||compute_work_requests(): start
17.11.2006 10:31:07||compute_work_requests(): start
17.11.2006 10:32:08||compute_work_requests(): start
17.11.2006 10:33:09||compute_work_requests(): start
17.11.2006 10:34:09||compute_work_requests(): start
17.11.2006 10:35:09||compute_work_requests(): start
17.11.2006 10:36:09||compute_work_requests(): start
17.11.2006 10:37:10||compute_work_requests(): start
17.11.2006 10:38:11||compute_work_requests(): start
17.11.2006 10:39:11||compute_work_requests(): start
17.11.2006 10:40:11||compute_work_requests(): start
17.11.2006 10:41:11||compute_work_requests(): start
17.11.2006 10:42:12||compute_work_requests(): start
17.11.2006 10:43:12||compute_work_requests(): start
17.11.2006 10:44:12||compute_work_requests(): start
17.11.2006 10:45:12||compute_work_requests(): start
17.11.2006 10:46:13||compute_work_requests(): start
17.11.2006 10:47:14||compute_work_requests(): start
17.11.2006 10:48:14||compute_work_requests(): start
17.11.2006 10:49:14||compute_work_requests(): start
17.11.2006 10:50:14||compute_work_requests(): start
17.11.2006 10:51:14||compute_work_requests(): start
17.11.2006 10:51:45|SETI@home|Computation for task 14jn03ab.20420.23504.634652.3.236_4 finished
17.11.2006 10:51:45|SETI@home|Restarting task 10jn03aa.8062.16608.403388.3.92_5 using setiathome_enhanced version 515
17.11.2006 10:51:45||compute_work_requests(): start
17.11.2006 10:51:47|SETI@home|Started upload of file 14jn03ab.20420.23504.634652.3.236_4_0
17.11.2006 10:51:55|SETI@home|Finished upload of file 14jn03ab.20420.23504.634652.3.236_4_0
17.11.2006 10:51:55|SETI@home|Throughput 5503 bytes/sec
17.11.2006 10:52:33||compute_work_requests(): start
17.11.2006 10:52:35|SETI@home|Sending scheduler request: Requested by user
17.11.2006 10:52:35|SETI@home|Reporting 1 tasks
17.11.2006 10:52:40|SETI@home|Scheduler RPC succeeded [server version 507]
17.11.2006 10:52:40|SETI@home|Deferring scheduler requests for 11 seconds
17.11.2006 10:52:40||compute_work_requests(): start
17.11.2006 10:52:52||compute_work_requests(): start

---

Have installed 5.7.3 and before I was using 5.6.5.

5.6.5 worked well since I installed it. Yesterday or two days ago it started to change the status of a task from active to suspended - cpu throttled every second. Then I installed 5.7.3 and the same happens.

(edited)

Are you running with the CPU at 100% or have you selected less?

---

I can tell you all with the throttling problem that I'm having the same problem.

Those "Computer work request" messages are spurious and weren't supposed to be turned on in the build.

5.7.4 should fix that small problem (I've downloaded it, just need to install it).


I've started a thread on the alpha test discussion list about this problem.


The list can be found here.

---

Kathryn :o)

I can tell you all with the throttling problem that I'm having the same problem.

Those "Computer work request" messages are spurious and weren't supposed to be turned on in the build.

5.7.4 should fix that small problem (I've downloaded it, just need to install it).


I've started a thread on the alpha test discussion list about this problem.


The list can be found here.

5.7.4 gives me the same problems as before -when throttling (with Einstein), runs a few minutes and then processing stops.

And I'm relatively new here, but if this is a Beta version, why would the issues go to an alpha version discussion area. Do the developers not look at this beta thread also?

---

Make sure you report this behaviour on the Einstein forums as well.

As for the software, technically it's Alpha software. But rest assured, I have pointed out some of the details to the developers. They know this thread exists. ;)

---

I have tested what the CPU percentage usage does in seconds, starting from 100% when it runs flat out:

90%: 9 second run, 1 second idle.
80%: 4 second run, 1 second idle.
70%: 3 second run, 1 second idle, 2 second run, 1 second idle, 3 second run, etc.
60%: 1 second run, 1 second idle, 2 second run, 1 second idle, 1 second run, etc.
50%: 1 second run, 1 second idle, 1 second run, 1 second idle, etc.
40%: 1 second run, 1 second idle, 1 second run, 2 seconds idle, 1 second run, etc.
30%: 1 second run, 1 second idle, 1 second run, 3 seconds idle, 1 second run, 2 seconds idle, etc.
20%: 1 second run, 4 seconds idle, 1 second run, 4 seconds idle, etc.
10%: 1 second run, 9 seconds idle.

It doesn't work with Einstein@Home. EAH will just sit idle in your task manager, never getting CPU.

---

Answer by John McLeod on the Alpha email list:

There are also the cases where the % is not a multiple of 10.

99% would be 99 seconds on and 1 second off.

75% would be 3 on one off.

25% would be 1 on 3 off.

33% would be 1 on 2 off.

66% would be 2 on one off.

The code checks once per second to see if it has exceeded the CPU fraction on. If so, it then turns off, and starts checking for being under CPU fraction on, if so, it turns on.

Example:
75%
Start on.
At one second on, it is at 100% CPU usage. Turn off.
At two seconds (one on, one off) it is at 50% CPU usage. Turn on.
At three seconds (two on, one off) it is at 66% CPU usage. Stay on.
At 4 seconds (three on, one off) it is at 75% usage. (**uncertain how an exact match is handled - assumption) Stay on.
At 5 seconds (4 on, one off) it is at 80% usage. Turn off.
At 6 seconds (4 on, 2 off) it is at 66% usage. Turn on.

...

Note that the polling loop for everything runs once per second.

jm7