Hi everyone, I've recently become aware of this bug [1] that affects Reef and Squid. When the bug is triggered, OSD will fail to start with "bluefs mount failed to replay log: (5) Input/output error" after crashing with "bluestore/AvlAllocator.cc: 173: FAILED ceph_assert(rs->start <= start)". The Red Hat KB solution [2] recommends redeploying the OSD as soon as possible to avoid having to deal with multiple crashed OSDs in the cluster. Reef v18.2.7 will fix this issue via PR #62840 [3]. Squid v19.2.3 should fix this issue via PR #62839 [4]. Thought it might be worth sharing. Regards, Frédéric. [1] https://tracker.ceph.com/issues/70747 [2] https://access.redhat.com/solutions/7113657 [3] https://github.com/ceph/ceph/pull/62840 [4] https://github.com/ceph/ceph/pull/62839 _______________________________________________ ceph-users mailing list -- ceph-users@xxxxxxx To unsubscribe send an email to ceph-users-leave@xxxxxxx
 |