Questions and Answers :
Windows :
BOINC downloaded new unit for wrong project after a fail
Message board moderation
Author | Message |
---|---|
Send message Joined: 20 Feb 06 Posts: 3 Credit: 109,516 RAC: 0 |
I run latest version of BOINC. I have been running WCG FAAH and Proteome folding for some time and recently added Climate prediction. All was going nicely. I was slightly surprised and dismayed by the duration of the Sulphur Cycle unit - but OK. It has now reached 8%, and obviously during this time many FAA and Proteome units have completed and reported. BOINC has behaved \"sensibly\" and regardes one work unit of Climate prediction to be enough! However, one of the Proteome units suffered an unrecoverable error. BOINC reported this, and obviously, should have just requested another FAA or Proteom workunit to replace the faulty one. BUT BOINC downloaded ANOTHER CLIMATE PRED UNIT INSTEAD! Obviously, I dont need another one at the moment, so I decided to abort it, assuming that I was freeing it, and that it could then be sent out to someone else. The aborted unit is still sitting there in BOINC, showing as \"Aborted by User\", and is still reported as in progress on my web results page. Why did BOINC download the wrong project? Apologies for asking 2 questions, but how do I get rid of aborted units and free them up for others? Robert |
Send message Joined: 5 Aug 04 Posts: 426 Credit: 2,426,069 RAC: 0 |
I would bet that you have a hyperthreaded or multi cpu system. When the other project errored out it needed another workunit and CPDN was the only project it was able to contact at that time. BOINC WIKI BOINCing since 2002/12/8 |
Send message Joined: 5 Aug 04 Posts: 1496 Credit: 95,522,203 RAC: 0 |
The aborted Work Unit should disappear after next contact with server, to \'report\' the result. You can \'force\' that with \'Update\' in the \'Projects\' tab. "We have met the enemy and he is us." -- Pogo Greetings from coastal Washington state, the scenic US Pacific Northwest. |
Send message Joined: 20 Feb 06 Posts: 3 Credit: 109,516 RAC: 0 |
I would bet that you have a hyperthreaded or multi cpu system. When the other project errored out it needed another workunit and CPDN was the only project it was able to contact at that time. Your right - Pentium D processor. Ideally, I want CPDN to use one processor and WCG to use the other. I have found out that it is possible to stop a project from downloading more work - and have done this for CPDN for the time being. This should stop it happening again. Thanks Robert |
Send message Joined: 20 Feb 06 Posts: 3 Credit: 109,516 RAC: 0 |
The aborted Work Unit should disappear after next contact with server, to \'report\' the result. You can \'force\' that with \'Update\' in the \'Projects\' tab. Aborted workunit did disappear. Hopefully it gets sent out to someone else for processing. Thanks for the reply Robert |
©2024 cpdn.org