#289 closed Defect (invalid)
Aborted by project
Reported by: | mozeskriebels | Owned by: | davea |
---|---|---|---|
Priority: | Minor | Milestone: | 5.10 |
Component: | Client - Scheduler Policy | Version: | |
Keywords: | Cc: |
Description
Since I have installed boinc application 5.10.2 and 5.10.6 I've got a lot of "aborted by project" status on results. On the seti website it states client error. The results are aborted before any cpu time/calculation is done on the results. They are just standing in wait and go to aborted by project on the next "project update".
On clients running the stable version this is not happening. Is this a small bug? Maybe 10 days connection time + 10 additional days is to much for the project to handle???
Change History (6)
comment:1 Changed 17 years ago by
comment:2 Changed 17 years ago by
"Aborted by project" is a feature, not a bug. It's to avoid crunching work that it knows you won't finish by the deadline.
comment:3 Changed 17 years ago by
Resolution: | → invalid |
---|---|
Status: | new → closed |
comment:4 Changed 17 years ago by
Resolution: | invalid |
---|---|
Status: | closed → reopened |
One of the aborted results has a deadline of 10-7-2007. I've work that has a shorter deadline (6-7-2007)and work with a longer deadline (13-7-2007). The result in the middle is being aborted.
Is this work being aborted a side effect from the additional 10 days? Is the scheduler working properly if the work that gets scheduled also gets aborted due to resource shortage? I get a lot of aborted results. I think it is not good for the project if the results that are being scheduled are also being aborted on large scale.
I will try to lower the additional days of work to find out of the aborted work get less.
comment:5 Changed 17 years ago by
Resolution: | → invalid |
---|---|
Status: | reopened → closed |
Projects may also abort unstarted work that already has quorum in server. This is to stop you from wasting computing power on unneeded work.
comment:6 Changed 17 years ago by
you're right. The problem is that the quorum was reached. The error was a bit misleading. Sorry for the problems.
Good job with all the new versions.
What was the deadline on the tasks which were automatically aborted? Longer or shorter than 10 days from the moment they were aborted?
(The 10 days connection time parameter tells Boinc that the task is not guaranteed to report until 10 days are up).