Message boards :
Number crunching :
Strange problem with dual-core machine
Message board moderation
Author | Message |
---|---|
w1hue Send message Joined: 4 Aug 00 Posts: 69 Credit: 5,492,898 RAC: 7 |
I have two HP computers with dual core AMD processors that have been running BOINC projects for the past couple of years or so. BOINC preferences are set to use 100% of the available CPUs and they each have been running two tasks simultaneously without problems. Now, all of a sudden, one of them has decided that it only wants to run 1 task from time to time! (And yes, there are other tasks waiting to run or waiting to start.) If I restart BOINC manager, it runs two tasks for awhile and then drops down to 1 (and the task manager shows CPU usage around 50% rather than 100%). Later it will revert back to two tasks for awhile. Both machines are running BOINC Manager 7.0.25 -- and have been for awhile. I updated the problem machine to 7.0.27 and it behaves in the same manner -- running two tasks for awhile and then dropping down to one. It is not a temperature issue; I am running TThrottle on both machines with the temp limit set to 62 deg. C and they rarely reach that limit (with two tasks running, CPU temp is typically between 54 an 58 C -- these are small case machines). Of course when the problem machine drops down to one task, the CPU temp drops by a few degrees. Could it be power supply related? Any ideas? Anyone?? I guess I should add that the machine is running 32-bit WinXP SP3 with all current updates and CPU-only BOINC tasks. |
Claggy Send message Joined: 5 Jul 99 Posts: 4654 Credit: 47,537,079 RAC: 4 |
Same as i said on the Boinc forums, not enough information, none of your hosts attached to Seti (that have communicated with seti in the last month) are running Boinc 7.0.27, and the one host at Einstein running Boinc 7.0.27 doesn't have any work and is on Vista, so which host is the problem host? and how many tasks does it have across all your projects? Claggy |
Karsten Vinding Send message Joined: 18 May 99 Posts: 239 Credit: 25,201,931 RAC: 11 |
I have had it on my main computer too. It turned out to be milkyway's nbody multithreaded app, that made boinc do this. Some times when nbody had been running and using all eight cores, a number of cores would not start to crunch again afterwords. Boinc would sometimes resume anything from 5 to 8 task (at seldom times even fever), leaving the rest of the cores idle. Most of the times all cores would start again, though But since the nbody WU's crunched very fast, and I had many in my cache, the system would end up in this state quite often anyhow. I ended up dropping milkyway/nbody for this reason, and havent tested it with never versions of the boinc client. If you are running milkyway, this could be what is happening on your system as well. |
w1hue Send message Joined: 4 Aug 00 Posts: 69 Credit: 5,492,898 RAC: 7 |
Same as i said on the Boinc forums, not enough information, none of your hosts attached to Seti (that have communicated with seti in the last month) are running Boinc 7.0.27, The machine in question does not happen to be running SETI tasks at the moment, but what difference does that make? It is connected to the following projects: Cosmology@Home, Einstein@Home and World Community Grid under my user name (w1hue) and to boincsimp, maleriacontrol.net and rossetta@home under the user name SFCC (a local computer club). I am running a total of eight machines, 7 XP and one Vista. The problem host name is 'sunflower-b1' and there are currently two Cosmology@Home, one Einstein@Home and eight Rosetta@Home tasks in the queue. It is currently happily running two tasks: one Einstein & one Rosetta. One interesting thing: When it started acting up (i.e., only running a single task) after I restarted BOINC manager yesterday, it read preferences from 'world community' several times in a row. If you would like to see the log file for the past couple of days, I can upload it as a text file to my website tomorrow and post the URL. Anything else that you would care to know? |
w1hue Send message Joined: 4 Aug 00 Posts: 69 Credit: 5,492,898 RAC: 7 |
If you are running milkyway, this could be what is happening on your system as well. I gave up on milkyway some time ago and switched to Cosomology@Home. I didn't have the problems you mentioned, but I was getting a lot of compute errors and no credit for jobs that did complete. |
Grant Nelson Send message Joined: 7 May 12 Posts: 8022 Credit: 4,237,757 RAC: 0 |
I am running a 4 core (AMD) at full tilt for a month now and so far no problems but at first my cooling fan would run a lot at high speed. So I have an extra small house fan on it blowing right in the case to help from over heating and it works well. I don't if it would help you but it's worth a try. Cheers everybody Life is short so don't sip Beer speaks, people mumble |
w1hue Send message Joined: 4 Aug 00 Posts: 69 Credit: 5,492,898 RAC: 7 |
I am running a 4 core (AMD) at full tilt for a month now and so far no problems but at first my cooling fan would run a lot at high speed. So I have an extra small house fan on it blowing right in the case to help from over heating and it works well. I don't if it would help you but it's worth a try. It's not an over heating problem as I'm using TThrottle to keep the CPU temp below 62C (spec max is 70C). I've finally decided that a task was hogging memory and not allowing any other to run (the machine only has 1GB). Once that task completed, the machine has been happily running two tasks. Live and learn... |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.