ceph 状态 health: HEALTH_WARN no active mgr

admin

health: HEALTH_WARN
health: HEALTH_WARN
            no active mgr
/ @( X0 v, L$ W* n# H3 x, W( n
解决：Ceph Manager Daemon，简称ceph-mgr。 该组件的主要作用是分担和扩展monitor的部分功能，减轻monitor的负担，让更好地管理ceph存储系统ceph
8 U, _5 x3 D# j
! w0 r/ t; y) {" E6 cceph在 luminous中新加入了mgr功能模块，手动安装mgr即可
0 Q- [- Z9 e  y7 M( a* L% d8 n
ceph-deploy  mgr  create  ceph1  ceph2 ceph3
9 G; @; X9 m% p6 U3 U3 g3 q( {

admin

crush rule 常用命令记录

" T6 ]$ t$ K4 N; p
; f# G) `- J% _' n
: t. ~0 D2 ?, m+ Y8 Q: o4 f) Tceph osd getcrushmap -o /tmp/crush
2 |& |2 [5 `# m% m8 G- H5 |/ Y

! I- M' w8 w, d- m3 @crushtool -d /tmp/crush -o /tmp/crush.txt
3 b2 t% L5 |& x; b9 Q$ zcrushtool -c /tmp/crush.txt -o /tmp/crush.bin
ceph osd setcrushmap -i /tmp/crush.bin

admin

ceph -s
6 r" U. p  O  L3 \. n5 i2 X" F3 l; Fceph -s
  cluster:
  [3 c' E. ^& |9 r$ C! x9 x% E    id:     04f83ed1-096e-41c9-a214-5a1fa4724bc3
: k, P0 c" v0 ~9 v, h) S7 S    health: HEALTH_OK
$ A6 m. n% s! h6 v* ~9 V8 i0 Y5 W
  services:
# Z# N  ?7 c6 N. d    mon: 3 daemons, quorum ceph1,ceph2,ceph3
    mgr: ceph1(active), standbys: ceph2, ceph3
. P7 B9 `1 W: m5 W( `    osd: 3 osds: 3 up, 3 in

- u( E+ z4 j! g; b7 J  data:
    pools:   0 pools, 0 pgs
( W" i$ V1 c; E    objects: 0  objects, 0 B
0 O# {+ t9 u# ^" F2 W+ Y: E    usage:   3.0 GiB used, 297 GiB / 300 GiB avail
/ k) `6 E* T1 D4 J4 ]7 x    pgs: