Using beta9, under configure repository options->job settings->failure detection I set the ‘mark task as failed after it has generated this many errors’. but the actual number of times a job is run and errors out is not always capped at this vaules. For example, if I set the value to 1 and repeatedly resubmit a 5 frame job where all the frames will fail, the frames generate this many errors:
7,4,3,1,2
4,1,1,1,1
1,1,1,1,3
2,1,1,1,1
2,4,2,2,1
2,4,3,2,1
2,1,3,1,1
1,1,2,2,1
Did you restart the slaves after setting the error limit to 1? The slaves only update their cached repository settings every 10 minutes, so if you didn’t restart the slaves and performed this test immediately after saving the repository settings, that could explain the behavior you’re seeing.
Cheers,
I made the change to the limit several hours ago so the slaves would have updated by now but I restarted all the slaves just to be certain. It appears there are still problems consistently respecting the number of errors. Here are the values:
2,4,1,2,9
1,1,1,2,1
1,2,1,1,4
1,1,1,1,1
1,1,1,1,1
1,2,1,1,1
1,1,3,1,1
2,1,2,1,1
1,1,1,1,1
Cool, thanks for checking that. We’ve logged this as a bug.