ceph config help mon_max_pg_per_osd查看默认值

admin

[root@compute01 ~]# ceph config help mon_max_pg_per_osd
, q2 J. f  b& j+ u, I# d) i! Tmon_max_pg_per_osd - Max number of PGs per OSD the cluster will allow
  (uint, advanced)
  Default: 250
  Minimum: 1
. E( D6 M1 O+ p' i- T& A  Maximum:
  Can update at runtime: true
  Services: [mgr]
& t# P: o" k$ C* pIf the number of PGs per OSD exceeds this, a health warning will be visible in `ceph status`.  This is also used in automated PG management, as the threshold at which some pools' pg_num may be shrunk in order to enable increasing the pg_num of others.
" L( R, ?# T+ A& V

& ?/ K& U, H$ I查看ceph的mon_max_pg_per_osd默认值：
ceph daemon mgr.`hostname -s` config get mon_max_pg_per_osd
" R* w  u$ v$ M: R2 i1 `. _- J) {{
: _" O) }  \9 J& A$ Y    "mon_max_pg_per_osd": "1000"
" e+ v. P7 Z" U7 n! S0 W3 N9 N}

6 e+ e' p& M- ^. ]/ e1 e& l! Z在线实时调整配置参数(重启后失效)
' o( m) k' x/ q) d; a
```
[root@ceph ~]# ceph daemon mgr.ceph1 config set mon_max_pg_per_osd 300
0 D3 i. R$ L( @) f( O{
# Y% @# D# R* c" M9 r! I% G/ J- M; O    "success": "mon_max_pg_per_osd = '300' (not observed, change may require restart) "
6 P2 m7 t; `' m% K" Y}
0 H7 o/ F* l/ O& _! I: F* Y```
& k" S4 V" y3 }* f0 |* c
: z4 R7 u  K# f3 @9 K0 Q0 a: R% ^; c
& W7 N# o7 E5 a$ ]* `
启动保存配置
需要在/etc/ceph/ceph.conf加上以下内容
& I) ~# m5 \: X9 ?9 R
```
6 F9 s- G- K+ [: ~$ f[mgr.ceph]
mon_max_pg_per_osd = 300
```

3 x; G$ X4 Y/ {8 I" h6 u
- u+ `5 d2 s% S1 j* ?! s

admin

查看ceph mon_max_pg_per_osd  最大连接数
, S* ~. k# H7 o. s* U( ?ceph daemon mgr.`hostname -s` config get mon_max_pg_per_osd
{
- \- y/ _) q3 m% L: e. \    "mon_max_pg_per_osd": "1000"
}

admin

[root@compute01 ~]#  ceph daemon mgr.compute01 config set mon_max_pg_per_osd 1000
2 n% o: d5 a% E- E  p. V- L+ L{
* y& t) s" _- E* k) S    "success": "mon_max_pg_per_osd = '1000' (not observed, change may require restart) "
}