Ceph » RADOS

"activating" is not listed in the pg state table:

http://docs.ceph.com/docs/master/rados/operations/pg-states/

After upgrading to Luminous we ran into pg overdose protection that caused a large number of our pgs to get stuck in the "activating" state.

Here is sample output from ceph status:

root@idb-ceph4:~# ceph status
  cluster:
    id:     9de06f94-a841-4e17-ac7c-9a0d8d8791b8
    health: HEALTH_WARN
            738722/404087670 objects misplaced (0.183%)
            Reduced data availability: 460 pgs inactive
            Degraded data redundancy: 42719/404087670 objects degraded (0.011%), 513 pgs unclean, 110 pgs degraded, 110 pgs undersized
            too many PGs per OSD (221 > max 200)

  services:
    mon: 3 daemons, quorum idb-ceph4,idb-ceph5,idb-ceph6
    mgr: idb-ceph4(active), standbys: idb-ceph5, idb-ceph6
    osd: 128 osds: 128 up, 128 in; 513 remapped pgs

  data:
    pools:   16 pools, 9430 pgs
    objects: 128M objects, 126 TB
    usage:   409 TB used, 123 TB / 532 TB avail
    pgs:     4.878% pgs not active
             42719/404087670 objects degraded (0.011%)
             738722/404087670 objects misplaced (0.183%)
             8915 active+clean
             358  activating+remapped
             102  activating+undersized+degraded+remapped
             37   active+remapped+backfilling
             8    active+undersized+degraded+remapped+backfilling
             8    active+remapped+backfill_wait
             2    active+clean+scrubbing+deep

  io:
    recovery: 92647 kB/s, 97 objects/s