Mark Nelson <mnelson@...> writes: > PGs are pool specific, so the other pool may be totally healthy while > the first is not. If it turns out it's a hardware problem, it's also > possible that the 2nd pool may not hit all of the same OSDs as the first > pool, especially if it has a low PG count. > Just to be clear: I have a cluster with three servers and three osds. The replica count is three so it is impossible that I am not touching all osds. How can I tell ceph to discard those pgs? Thanks again for help, Mario _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |