Bugzilla – Bug 4111
locking / synch problems
Last modified: 2006-04-07 16:43:52
You need to
before you can comment on or make changes to this bug.
I think that something doesn't get awaken properly in the gram code in trunk
(or might be a deadlock problem) causing time outs:
The container.log contains periodic jvm stack traces. Notice that only one job
can be created at a time.
Also, the reason might be that .notify() is called before something enters .wait
(). So using Semaphore (from core) will fix it, if that's the problem.
I can't reproduce this with throughput tests running simple job submissions.
Staging jobs are hanging, but according to my own thread dumps it looks like RFT
is hanging. I'll try this again once Ravi's related bugs (#3840 and #3959) are
All the thoroughput tests have been succeeding for a while now. This problem
seems to have been resolved.