Hi, I've recently had a "glusterfs event", where the gluster cluster stopped responding.
>From what I can determine two combined events occurred: - one of the brick node had been restarted, but failed to actually come up - of the two remaining brick nodes, one appeared to develop xfs related errors. I did a 'gluster volume heal gv0 full' but I'm 100% sure if things are back to normal now. I have some messages in the log files, do these indicate further issues / problems to be looked at? glusterFS[2189]: [2014-05-23 03:20:53.931282] C [glusterd-op-sm.c:3648:glusterd_op_txn_complete] 0-management: Unable to clear local lock, ret: -1 Also, I see Gluster 3.5 has been released (but no version for i386 on Debian?), having never done an upgrade is it safe to upgrade each node one-by-one or should then be done in a particular order or ... ? Thanks, Anand -- “Don’t be sad because it’s over. Smile because it happened.” – Dr. Seuss
_______________________________________________ Gluster-users mailing list [email protected] http://supercolony.gluster.org/mailman/listinfo/gluster-users
