mds crashes in EMetaBlob::replay
while testing cephfs using the debian wheezy packages on a fairly large volume (2TB) i ran into random unreproduceable client-stalls. now mds starts but crashes shortly after.
i am willing to debug this further, as ceph was promising before.
- Status changed from New to Need More Info
can you put 'debug mds = 20' in the ceph.conf, restart ceph-mds, and then attach the resulting log (assuming it crashes again)?
- Project changed from Ceph to CephFS
aww. i had debug ms = 20 in my ceph.conf. sorry.
the new one is attached
This looks like a problem with what's in the journal, but soo much MDS code has changed since then that I don't think we can make sense of this report. Are you in a position to retest against latest master?
should i test the same btrfs volume with a new ceph? if so i might get to it in the next month. please close with insufficient data. i will reopen when i found the time.
unfortunately i don't have the hardware around anymore to replicate the whole test.
- Status changed from Need More Info to Rejected
Understood. I'm sorry we weren't able to dig in when it happened. When do you get around to retesting we should be in a better position to follow up.
Also available in: Atom