ceph config help mon_max_pg_per_osd查看默认值

admin

[root@compute01 ~]# ceph config help mon_max_pg_per_osd
mon_max_pg_per_osd - Max number of PGs per OSD the cluster will allow
' Q* ^' c# s+ M; `7 x1 L; `  (uint, advanced)
  Default: 250
' G3 }2 V* [' r$ h$ G7 K' Z7 d  Minimum: 1
5 a# s% H  L# E$ ^- d9 ?  Maximum:
9 Z9 o3 E9 N3 C! r2 t, m  Can update at runtime: true
* ]/ F3 b* N/ D6 f# U4 W  Services: [mgr]
If 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.

4 s+ @7 z- s) ^( p) T" M
查看ceph的mon_max_pg_per_osd默认值：
! M: ?. c4 ?# n/ ^* hceph daemon mgr.`hostname -s` config get mon_max_pg_per_osd
{
    "mon_max_pg_per_osd": "1000"
7 J5 K8 _+ n( l9 D}

( M+ u; i0 h1 [在线实时调整配置参数(重启后失效)
+ m+ E2 r" T) }8 x1 A+ v
```
7 q- i" C2 J8 l6 z[root@ceph ~]# ceph daemon mgr.ceph1 config set mon_max_pg_per_osd 300
: y! ^1 s. L& [0 [6 ?! H{
    "success": "mon_max_pg_per_osd = '300' (not observed, change may require restart) "
$ g: ^# V& L2 N# j  R5 w1 s}
3 W/ }  Z0 c6 M- V8 n```
7 ~. m, y( X6 Y- v2 t5 I5 X2 M
3 A! O3 m) A3 @* I5 k- c* j
  V7 w+ N9 ?6 M7 O& i- h- a$ v* [
启动保存配置
需要在/etc/ceph/ceph.conf加上以下内容
) v3 N  U+ i1 c1 x0 f; y' Q! g
```
% ]/ C) _9 ~0 {8 O& {7 a6 l0 h[mgr.ceph]
- V% W2 T8 ?- L7 W$ W1 ^. Ymon_max_pg_per_osd = 300
& R  B: O2 p+ ?) ^+ S0 \7 U7 H```
% w8 \# z; V8 R% ?
! |: p7 B# h& L( e# k

. V" f, s6 K' J; r6 c. y2 w

admin

查看ceph mon_max_pg_per_osd  最大连接数
- F( k# e/ g0 t! Eceph daemon mgr.`hostname -s` config get mon_max_pg_per_osd
# H  ]/ ^2 \6 i) g0 O2 u: Y& h{
& e+ u9 u% O! R( \' B' X8 E    "mon_max_pg_per_osd": "1000"
8 i  Q8 C& ?. U3 E  T3 o8 `}

admin

[root@compute01 ~]#  ceph daemon mgr.compute01 config set mon_max_pg_per_osd 1000
{
: X+ J. Z: Q7 M# Q2 _    "success": "mon_max_pg_per_osd = '1000' (not observed, change may require restart) "
}