On 9/4/19 1:34 AM, Eric Choi wrote: > Hi there, > > We are trying to upgrade a cluster in our test environment to Nautilus, > and we ran into this hurdle after upgrading Mons and Mgrs. Restarting > the first OSD, we noticed that ceph -s is reporting many PG's state as > unknown: > > [root@cephosd0041 ceph]# ceph pg stat > 996 pgs: 263 active+clean, 733 unknown; 71 GiB data, 364 GiB used, 11 > TiB / 11 TiB avail; 340 B/s rd, 3.0 KiB/s wr, 36 op/s > > As far I can research this is basically MGR's reporting that it didn't > receive information from OSD's, what can we do to resolve this? or is > this just a red herring by any chance? > This is indeed a MGR issue. Have you tried to restart them? That sometimes helps. I never upgraded from L to N in one go, so I can't say for sure what is going on. Wido > Thank you, > > -- > > Eric Choi > Senior Software Engineer 2 | Core Platform > > > > _______________________________________________ > ceph-users mailing list > ceph-users@xxxxxxxxxxxxxx > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |