代码改变世界

ceph 扩容后healt-error

2021-10-26 20:11  dribs  阅读(789)  评论(0)    收藏  举报 
[ceph@ceph0001 ceph-deploy]$ ceph -s
  cluster:
    id:     965a2a5c-b7b4-4d67-a7af-ce8e244b76b2
    health: HEALTH_ERR
            Degraded data redundancy (low space): 1 pg backfill_toofull
            too few PGs per OSD (20 < min 30)

  services:
    mon: 3 daemons, quorum ceph0001,ceph0002,ceph0003 (age 2h)
    mgr: ceph0003(active, since 2h)
    mds: cephfs:1 {0=ceph0002=up:active} 1 up:standby
    osd: 20 osds: 20 up (since 37m), 20 in (since 37m); 72 remapped pgs
    rgw: 4 daemons active (ceph0001, ceph0002, ceph0003, ceph0004)

  data:
    pools:   11 pools, 136 pgs
    objects: 139.50k objects, 242 GiB
    usage:   692 GiB used, 25 TiB / 26 TiB avail
    pgs:     182877/396251 objects misplaced (46.152%)
             66 active+remapped+backfill_wait
             64 active+clean
             5  active+remapped+backfilling
             1  active+remapped+backfill_wait+backfill_toofull

  io:
    client:   1.3 MiB/s rd, 1.9 MiB/s wr, 308 op/s rd, 323 op/s wr
    recovery: 47 MiB/s, 40 objects/s

[ceph@ceph0001 ceph-deploy]$ ceph health detail
HEALTH_ERR Degraded data redundancy (low space): 1 pg backfill_toofull; too few PGs per OSD (20 < min 30)
PG_DEGRADED_FULL Degraded data redundancy (low space): 1 pg backfill_toofull
    pg 2.18 is active+remapped+backfill_wait+backfill_toofull, acting [3,1,2]
TOO_FEW_PGS too few PGs per OSD (20 < min 30)

　　

 

 

慢慢等回填完，逐渐降下去