No tasks for one pc but lots for other pc
Message boards :
Number crunching :
No tasks for one pc but lots for other pc
Message board moderation
Author | Message |
---|---|
Send message Joined: 31 Aug 15 Posts: 10 Credit: 6,017,448 RAC: 0 |
I have 2 machines running the project, the first: http://denis.usj.es/denisathome/results.php?userid=9188 has loads of tasks but the second: http://denis.usj.es/denisathome/results.php?hostid=13486 is no longer receiving any. Any advice would be appreciated, thanks. |
Send message Joined: 28 Apr 15 Posts: 29 Credit: 1,426,883 RAC: 0 |
Are you running other BOINC projects on the second machine? If you have done a lot of DENIS tasks recently, then the BOINC scheduler may decide it is time to run the other ones. That is especially so if you have added a new project to the second machine; BOINC will prefer it for awhile (about a week on my machines) before returning to the first project. |
Send message Joined: 31 Aug 15 Posts: 10 Credit: 6,017,448 RAC: 0 |
Hi, no other boinc projects on those machines now (they were running climate prediction.net until yesterday), the one with no tasks has boinc with just this project, the one getting lots of tasks is running boinc with just this project on the cpu and folding at home on the gpu. |
Send message Joined: 28 Apr 15 Posts: 29 Credit: 1,426,883 RAC: 0 |
I used to run Climate Prediction Net with World Community Grid/CEP2, and they did not get along very well. I can remember times when there was no CPDN work available, but still the CEP2 would not download and at least one core was idle. Maybe that is fixed with later BOINC versions, but now I just avoid that combination. It might be something similar with DENIS. You could try removing DENIS from the BOINC manager (and CPDN if it is still attached) and then adding the DENIS project back again. |
Send message Joined: 31 Aug 15 Posts: 10 Credit: 6,017,448 RAC: 0 |
Thanks Jim1348 ! Removed the project and added again as you said and the tasks are flowing in +++ Thanks again :) |
Send message Joined: 19 Apr 15 Posts: 9 Credit: 10,017,468 RAC: 0 |
Hi currob, It might just be a co-incidence but they seem to have removed the limit on how many jobs you can download per core (or made it much, much larger than it was before). Whereas an 8 core machine was restricted to 80 jobs I now have over a thousand jobs which I'm guessing has been limited by my cache setting. When they released the new jobs after the server went down lots of machines might have sucked all the jobs up and left nothing for your second machine until more got released. Just a possibility ! Welcome to the team :-) |
Send message Joined: 31 Aug 15 Posts: 10 Credit: 6,017,448 RAC: 0 |
Welcome to the team :-) Thanks for the warm welcome, it's nice to be contributing to the project. :) |
Send message Joined: 11 Apr 15 Posts: 24 Credit: 4,366,045 RAC: 0 |
The job limit is still here : 02/09/2015 10:21:08 | DENIS@Home | This computer has reached a limit on tasks in progress 4 threads, 40 jobs queued ... |
Send message Joined: 19 Apr 15 Posts: 9 Credit: 10,017,468 RAC: 0 |
I have one with 4 cores with the 10 times core limit but another 4 core machine with 150+ tasks and an 8 core machine with over 900 tasks. They are all set up the same, it only happened after the server crash yesterday. |
Send message Joined: 16 Mar 15 Posts: 219 Credit: 14,859 RAC: 0 |
Hi! After the crash we missed to set up again the limit. Now it's set again. Best regards, Joel. |
Send message Joined: 19 Apr 15 Posts: 9 Credit: 10,017,468 RAC: 0 |
Hi! :-( |