Hi all, this post is related to "Help needed, ceph fs down due to large stray dir" (https://www.spinics.net/lists/ceph-users/msg85394.html). We finally got the MDS up and could record what it is doing before and after getting hung. We opened tracker https://tracker.ceph.com/issues/69547 . The uploads were too large to attach. For a quick overview of the key findings see the 3 screen shots posted here: https://imgur.com/a/RF7ExSP . Apparently, the MDS gets hung in CInode::purge_stale_snap_data after populating the cache. I hope that the data uploaded will give someone a clue what is happening and how we can get out of it. Putting my hopes on tomorrow after sleeping a bit (well, at least trying to). Thanks for all your help with this critical situation. Best regards, ================= Frank Schilder AIT Risø Campus Bygning 109, rum S14 _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |