一、故障现象：
查看ceph的集群状态：too many PGs per OSD (698 > max 300)
|
1
2
3
4
5
6
7
8
9
10
11
12
13
|
# ceph -s
cluster e2ca994a-00c4-477f-9390-ea3f931c5062
health HEALTH_WARN
too many PGs per OSD (698 > max 300)
monmap e1: 3 mons at {hz-01-ops-tc-ceph-02=172.16.2.231:6789
/0
,hz-01-ops-tc-ceph-03=172.16.2.172:6789
/0
,hz-01-ops-tc-ceph-04=172.16.2.181:6789
/0
}
election epoch 14, quorum 0,1,2 hz-01-ops-tc-ceph-03,hz-01-ops-tc-ceph-04,hz-01-ops-tc-ceph-02
osdmap e54: 5 osds: 5 up, 5
in
flags sortbitwise,require_jewel_osds
pgmap v1670: 1164 pgs, 3 pools, 14640 kB data, 22 objects
240 MB used, 224 GB / 224 GB avail
1164 active+clean
# ceph --show-config | grep mon_pg_warn_max_per_osd
mon_pg_warn_max_per_osd = 300
|
二、调整ceph配置信息
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
|
# cd /my-cluster
# vim ceph.conf
添加如下参数：
mon_pg_warn_max_per_osd = 1024
# ceph-deploy --overwrite-conf config push hz-01-ops-tc-ceph-04 hz-01-ops-tc-ceph-02 hz-01-ops-tc-ceph-03
[ceph_deploy.conf][DEBUG ] found configuration
file
at:
/root/
.cephdeploy.conf
[ceph_deploy.cli][INFO ] Invoked (1.5.39):
/usr/bin/ceph-deploy
--overwrite-conf config push hz-01-ops-tc-ceph-04 hz-01-ops-tc-ceph-02 hz-01-ops-tc-ceph-03
[ceph_deploy.cli][INFO ] ceph-deploy options:
[ceph_deploy.cli][INFO ] username : None
[ceph_deploy.cli][INFO ] verbose : False
[ceph_deploy.cli][INFO ] overwrite_conf : True
[ceph_deploy.cli][INFO ] subcommand : push
[ceph_deploy.cli][INFO ] quiet : False
[ceph_deploy.cli][INFO ] cd_conf : <ceph_deploy.conf.cephdeploy.Conf instance at 0x7fb51b241320>
[ceph_deploy.cli][INFO ] cluster : ceph
[ceph_deploy.cli][INFO ] client : [
'hz-01-ops-tc-ceph-04'
,
'hz-01-ops-tc-ceph-02'
,
'hz-01-ops-tc-ceph-03'
]
[ceph_deploy.cli][INFO ] func : <
function
config at 0x7fb51bac8320>
[ceph_deploy.cli][INFO ] ceph_conf : None
[ceph_deploy.cli][INFO ] default_release : False
[ceph_deploy.config][DEBUG ] Pushing config to hz-01-ops-tc-ceph-04
[hz-01-ops-tc-ceph-04][DEBUG ] connected to host: hz-01-ops-tc-ceph-04
[hz-01-ops-tc-ceph-04][DEBUG ] detect platform information from remote host
[hz-01-ops-tc-ceph-04][DEBUG ] detect machine
type
[hz-01-ops-tc-ceph-04][DEBUG ] write cluster configuration to
/etc/ceph/
{cluster}.conf
[ceph_deploy.config][DEBUG ] Pushing config to hz-01-ops-tc-ceph-02
[hz-01-ops-tc-ceph-02][DEBUG ] connected to host: hz-01-ops-tc-ceph-02
[hz-01-ops-tc-ceph-02][DEBUG ] detect platform information from remote host
[hz-01-ops-tc-ceph-02][DEBUG ] detect machine
type
[hz-01-ops-tc-ceph-02][DEBUG ] write cluster configuration to
/etc/ceph/
{cluster}.conf
[ceph_deploy.config][DEBUG ] Pushing config to hz-01-ops-tc-ceph-03
[hz-01-ops-tc-ceph-03][DEBUG ] connected to host: hz-01-ops-tc-ceph-03
[hz-01-ops-tc-ceph-03][DEBUG ] detect platform information from remote host
[hz-01-ops-tc-ceph-03][DEBUG ] detect machine
type
[hz-01-ops-tc-ceph-03][DEBUG ] write cluster configuration to
/etc/ceph/
{cluster}.conf
|
在mon节点上重启服务：
# systemctl restart ceph-mon.target
三、然后在管理节点上再次查看集群
|
1
2
3
4
5
6
7
8
9
10
11
12
|
# ceph -s
cluster e2ca994a-00c4-477f-9390-ea3f931c5062
health HEALTH_OK
monmap e1: 3 mons at {hz-01-ops-tc-ceph-02=172.16.2.231:6789
/0
,hz-01-ops-tc-ceph-03=172.16.2.172:6789
/0
,hz-01-ops-tc-ceph-04=172.16.2.181:6789
/0
}
election epoch 20, quorum 0,1,2 hz-01-ops-tc-ceph-03,hz-01-ops-tc-ceph-04,hz-01-ops-tc-ceph-02
osdmap e54: 5 osds: 5 up, 5
in
flags sortbitwise,require_jewel_osds
pgmap v1779: 1164 pgs, 3 pools, 14640 kB data, 22 objects
240 MB used, 224 GB / 224 GB avail
1164 active+clean
# ceph --show-config | grep mon_pg_warn_max_per_osd
mon_pg_warn_max_per_osd = 1024
|
本文转自 冰冻vs西瓜 51CTO博客，原文链接：http://blog.51cto.com/molewan/2061192，如需转载请自行联系原作者
