记一次ceph pg unfound处理过程

今天检查ceph集群，发现有pg丢失，于是就有了本文~~~

1.查看集群状态

[root@k8snode001 ~]# ceph health detail
HEALTH_ERR 1/973013 objects unfound (0.000%); 17 scrub errors; Possible data damage: 1 pg recovery_unfound, 8 pgs inconsistent, 1 pg repair; Degraded data redundancy: 1/2919039 objects degraded (0.000%), 1 pg degraded
OBJECT_UNFOUND 1/973013 objects unfound (0.000%)
    pg 2.2b has 1 unfound objects
OSD_SCRUB_ERRORS 17 scrub errors
PG_DAMAGED Possible data damage: 1 pg recovery_unfound, 8 pgs inconsistent, 1 pg repair
    pg 2.2b is active+recovery_unfound+degraded, acting [14,22,4], 1 unfound
    pg 2.44 is active+clean+inconsistent, acting [14,8,21]
    pg 2.73 is active+clean+inconsistent, acting [25,14,8]
    pg 2.80 is active+clean+scrubbing+deep+inconsistent+repair, acting [4,8,14]
    pg 2.83 is active+clean+inconsistent, acting [14,13,6]
    pg 2.ae is active+clean+inconsistent, acting [14,3,2]
    pg 2.c4 is active+clean+inconsistent, acting [8,21,14]
    pg 2.da is active+clean+inconsistent, acting [23,14,15]
    pg 2.fa is active+clean+inconsistent, acting [14,23,25]
PG_DEGRADED Degraded data redundancy: 1/2919039 objects degraded (0.000%), 1 pg degraded
    pg 2.2b is active+recovery_unfound+degraded, acting [14,22,4], 1 unfound

从输出发现pg 2.2b is active+recovery_unfound+degraded, acting [14,22,4], 1 unfound 现在我们来查看pg 2.2b，看看这个pg得想想信息。

[root@k8snode001 ~]# ceph pg dump_json pools    |grep 2.2b
dumped all
2.2b       2487                  1        1         0       1  9533198403 3048     3048                active+recovery_unfound+degraded 2020-07-23 08:56:07.669903  10373'5448370  10373:7312614  [14,22,4]         14  [14,22,4]             14  10371'5437258 2020-07-23 08:56:06.637012   10371'5437258 2020-07-23 08:56:06.637012             0

可以看到它现在只有一个副本

2.查看pg map

[root@k8snode001 ~]# ceph pg map 2.2b
osdmap e10373 pg 2.2b (2.2b) -> up [14,22,4] acting [14,22,4]

从pg map可以看出，pg 2.2b分布到osd [14,22,4]上

3.查看存储池状态

[root@k8snode001 ~]# ceph osd pool stats k8s-1
pool k8s-1 id 2
  1/1955664 objects degraded (0.000%)
  1/651888 objects unfound (0.000%)
  client io 271 KiB/s wr, 0 op/s rd, 52 op/s wr

[root@k8snode001 ~]# ceph osd pool ls detail|grep k8s-1
pool 2 'k8s-1' replicated size 3 min_size 1 crush_rule 0 object_hash rjenkins pg_num 256 pgp_num 256 last_change 88 flags hashpspool,selfmanaged_snaps stripe_width 0 application rbd

4.尝试恢复pg 2.2b丢失的块

[root@k8snode001 ~]# ceph pg repair 2.2b

如果一直修复不成功，可以查看卡住PG的具体信息，主要关注recovery_state，命令如下

[root@k8snode001 ~]# ceph pg 2.2b  query
{
    "......
    "recovery_state": [
        {
            "name": "Started/Primary/Active",
            "enter_time": "2020-07-21 14:17:05.855923",
            "might_have_unfound": [],
            "recovery_progress": {
                "backfill_targets": [],
                "waiting_on_backfill": [],
                "last_backfill_started": "MIN",
                "backfill_info": {
                    "begin": "MIN",
                    "end": "MIN",
                    "objects": []
                },
                "peer_backfill_info": [],
                "backfills_in_flight": [],
                "recovering": [],
                "pg_backend": {
                    "pull_from_peer": [],
                    "pushing": []
                }
            },
            "scrub": {
                "scrubber.epoch_start": "10370",
                "scrubber.active": false,
                "scrubber.state": "INACTIVE",
                "scrubber.start": "MIN",
                "scrubber.end": "MIN",
                "scrubber.max_end": "MIN",
                "scrubber.subset_last_update": "0'0",
                "scrubber.deep": false,
                "scrubber.waiting_on_whom": []
            }
        },
        {
            "name": "Started",
            "enter_time": "2020-07-21 14:17:04.814061"
        }
    ],
    "agent_state": {}
}

如果repair修复不了；两种解决方案，回退旧版或者直接删除

5.解决方案

回退旧版
[root@k8snode001 ~]# ceph pg  2.2b  mark_unfound_lost revert
直接删除
[root@k8snode001 ~]# ceph pg  2.2b  mark_unfound_lost delete

6.验证

我这里直接删除了，然后ceph集群重建pg,稍等会再看，pg状态变为active+clean

[root@k8snode001 ~]#  ceph pg  2.2b query
{
    "state": "active+clean",
    "snap_trimq": "[]",
    "snap_trimq_len": 0,
    "epoch": 11069,
    "up": [
        12,
        22,
        4
    ],

再次查看集群状态

[root@k8snode001 ~]# ceph health detail
HEALTH_OK