On Sun, May 8, 2016 at 11:57 PM, Robert Sander <r.sander@xxxxxxxxxxxxxxxxxxx> wrote: > Am 29.04.2016 um 17:11 schrieb Robert Sander: > >> As the backfilling with the full weight of the new OSDs would have run >> for more than 28h and no VM was usable we re-weighted the new OSDs to >> 0.1. The backfilling finished after about 2h and we planned to increase >> the weight slowly when suddenly the RBD was corrupted. > > After going through the logfiles I may have found the culprit. > > Is it possible to lose data if "ceph osd crush add" is called on the > same OSD multiple times? Whoa, that would sound like some seriously bad behavior. Any more insights on this? Can you confirm that this was indeed the root cause? Cheers, Florian _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |