Hi all, after our cluster problems with incomplete placementgroups, we've decided to remove our pools and create new ones. This was going fine in the beginning. After adding an additional OSD server, we now have 2 PGs that are stuck in the peering state: HEALTH_WARN 2 pgs peering; 2 pgs stuck inactive; 2 pgs stuck unclean pg 9.2e41 is stuck inactive for 52540.202628, current state peering, last acting [91,240,273] pg 9.bad5 is stuck inactive for 52540.077013, current state peering, last acting [335,64,273] pg 9.2e41 is stuck unclean for 65683.195508, current state peering, last acting [91,240,273] pg 9.bad5 is stuck unclean for 65683.218581, current state peering, last acting [335,64,273] pg 9.bad5 is peering, acting [335,64,273] pg 9.2e41 is peering, acting [91,240,273] I was checking the placementgroups with ceph pg query, but I found no reasons why the peering can not be completed. The out of ceph pg 9.2e41 query: http://pastebin.com/fyC8Qepv Any ideas? Regards, Christian _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |