I've had a 4 node ceph cluster working well for month.
This weekend I added a 5th node to the cluster and after many hours of
rebalancing I have the following warning:
HEALTH_WARN 1 pgs incomplete; 1 pgs stuck inactive; 1 pgs stuck
unclean
But, my big problem is that the cluster is almost useless, the
throughput went from great to literally just a few blocks/second.
And even on the cluster itself, an "rbd ls" command is nearly
instantaneous, while "rbd ls -l" hasn't completed after 15 minutes....
Any ideas what to look at? All of the ceph bits are up to date as of
yesterday (ceph version 0.61.7 (8f010aff684e820ecc837c25ac77c7a05d7191ff).
Thanks for any help/suggestions!!
Jeff
--
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
