Check states of PGs using "ceph pg dump" and for every PG that is not "active+clean", issue "ceph pg map <pg_num>" and get mapping OSDs. Check the state of those OSDs by looking at their logs under /var/log/ceph/.
Regards, Anand On Mon, May 23, 2016 at 6:53 AM, Ken Peng <k...@dnsbed.com> wrote: > Hi, > > # ceph -s > cluster 82c855ce-b450-4fba-bcdf-df2e0c958a41 > health HEALTH_ERR > 5 pgs inconsistent > 7 scrub errors > too many PGs per OSD (318 > max 300) > > > It's HEALTH_ERR above, how to fix up them? Thanks. > > > > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > > -- ---------------------------------------------------------------------------- Never say never.
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
