On Fri, 11 Nov 2011, Christian Brunner wrote: > 2011/11/11 Sage Weil <sage@xxxxxxxxxxxx>: > > Hi Christian, > > > > Do you have a core file? Can you dump the thread stack traces so we can > > see if it got hung up on a syscall or somewhere internally (thread apply > > all bt)? > > I'm not sure if it's from btrfs, but ther is no kernel warning at that time. > > I'm missing some debuginfo packages at the moment, but I managed to > install ceph-debuginfo. (If you need more I will ask a colleague to > subscribe the debuginfo channel on the RedHat Satellite Server, but > you will have to wait until Monday). > > Here is what I'm getting from the core file at the moment. Debug symbols for glibc would help, but so far I can tell that: - the filestore op_queue is blocked on the throttler (too much io queued) - the commit thread is also waiting for ops to finish - i see no actual thread processing the op_queue Usually that's because it hit a kernel bug and got killed. Not sure what else would make that thread disappear... sage
