将设为首页浏览此站
开启辅助访问 天气与日历 收藏本站联系我们切换到窄版

易陆发现论坛

 找回密码
 开始注册
查看: 35|回复: 0
收起左侧

octopus版本ceph 分布式问题 ceph status HEALTH_WARN

[复制链接]
发表于 2021-6-10 17:01:51 | 显示全部楼层 |阅读模式

马上注册,结交更多好友,享用更多功能,让你轻松玩转社区。

您需要 登录 才可以下载或查看,没有帐号?开始注册

x
ceph status
, O) I$ H8 a( J/ @% Q8 U( j  L  cluster:
* E5 Q* C& @/ W9 r    id:     ed60a380-c9bd-11eb-ad78-000c29ad02de, U* l6 t8 J* ?9 j
    health: HEALTH_WARN5 W0 u% R: l' k6 g" h2 H1 a7 D" x
            Degraded data redundancy: 1 pg undersized9 N& u6 p" x- ]- P3 q6 @6 X$ {
% z6 S/ h9 l; C. _' p# u1 l
  services:
3 `4 N4 r7 h( ^" W( M    mon: 1 daemons, quorum controller (age 18m)7 h8 ^  t7 q, J
    mgr: controller.ifotuc(active, since 7m)) }% _2 [$ _7 t; S' H: ^" g
    osd: 6 osds: 6 up (since 12m), 6 in (since 12m); 1 remapped pgs1 e. t9 v  I, C9 g$ I1 H
: g( ]- j1 S; x( \* Q+ ?
  data:- U2 e! @- O! t, |2 M
    pools:   1 pools, 1 pgs
3 {2 F) \5 l" ~* \/ G    objects: 0 objects, 0 B# B1 r) ]% P$ g
    usage:   6.0 GiB used, 114 GiB / 120 GiB avail7 b3 h4 ^* y$ C$ F( Y
    pgs:     1 active+undersized+remapped ! y( A4 P! v7 a  B( e
Degraded data redundancy: 1 pg undersized
& J- E5 Q; f7 _9 P) K
8 Y# t9 }0 j/ Q! C) N8 @9 ^/ f8 x. R0 s' r+ J8 A# H7 d! ?. v
( u9 K8 r$ U- V* K. V( ?& y& ~
将调小size和min_size  r5 H; D; N, W- R- w: r
ceph osd pool ls detail
5 n; Y! d9 I( H# `1 K0 j3 P/ Wpool 1 'device_health_metrics' replicated size 3 min_size 1 crush_rule 0 object_hash rjenkins pg_num 1 pgp_num 1 autoscale_mode on last_change 51 flags hashpspool stripe_width 0 pg_num_min 1 application mgr_devicehealth
2 f8 d: t4 m' x' p! S
: X% o  G. U' ]' d4 [) e. b. G- B+ Eceph osd pool set device_health_metrics  size 2
/ B( ]8 I$ o+ R& ?
' E; T" l# B: pceph osd pool  set device_health_metrics  min_size 1# M8 O9 E/ `) }! [. ]) b" Z
set pool 1 min_size to 1
1 E" O5 T! q& O) }; I2 R' I. G! d9 v
[root@controller ~]# ceph -s3 X3 j) l8 M$ X6 j; {5 h: i* p  j
  cluster:
$ V; {5 P0 j, a% k    id:     ed60a380-c9bd-11eb-ad78-000c29ad02de: x# S+ M$ Z+ [2 {! X) _3 C& H1 r
    health: HEALTH_OK9 L7 j8 i: j, q6 L# z. U# c: [3 }
  {4 `. L7 V! H$ L1 \+ I' T; [) R
  services:
7 N1 C1 {3 x! N/ C# H$ F6 Y# s    mon: 1 daemons, quorum controller (age 17m)1 L7 ^: W$ {. l
    mgr: controller.ifotuc(active, since 5m): L, N( ]) @0 K  C3 [
    osd: 6 osds: 6 up (since 11m), 6 in (since 11m); 1 remapped pgs
5 ^1 ^' L& N3 m  o; a% {* ] . H. s8 H' U: e: d( h1 a
  data:7 ]( t2 k9 |7 {
    pools:   1 pools, 1 pgs4 x; Y0 B3 r& ]/ P( P/ {
    objects: 0 objects, 0 B$ A1 }+ o  @% J  n5 Z0 A9 R
    usage:   6.0 GiB used, 114 GiB / 120 GiB avail
" U" O/ k. ]4 e3 Z+ O6 |    pgs:     1 active+undersized+remapped
; _" B3 k1 ]- s# v2 k0 R+ r
" {! l/ J, {: Z8 }1 _& W$ @
, c2 f0 g: X4 M; e; s[root@controller ~]# ceph -s# s& ~+ v0 T  C8 x3 L
  cluster:" x) z7 I$ [8 Z3 r8 f& I
    id:     ed60a380-c9bd-11eb-ad78-000c29ad02de
- x, b. ]6 S! c' }! T9 G; p. M% ^    health: HEALTH_OK
8 \6 d7 ?& z8 ^5 L+ c* v
! L- V4 D  R$ w" v4 ^- _  services:
& @5 `& {4 L8 v( v/ E5 f    mon: 1 daemons, quorum controller (age 21m)
$ q6 f* ?8 |* D/ y2 B    mgr: controller.ifotuc(active, since 9m)! G( ^) D# q& R
    osd: 6 osds: 6 up (since 15m), 6 in (since 15m); 1 remapped pgs( i: ?9 g# b7 a6 d% N
$ [8 X8 M2 a4 y- B# s
  data:
& ~7 S. o: J: ^; d) r    pools:   1 pools, 1 pgs
8 z' ~5 h6 M" @    objects: 0 objects, 0 B
5 `, d$ R+ v# r. ~+ G, c    usage:   6.0 GiB used, 114 GiB / 120 GiB avail6 f' ?3 t" N+ F" G* U# L
    pgs:     1 active+clean+remapped( n$ D8 V6 [; Y1 n: o

4 d: T5 P. K' x$ ~设置低于默认值就正常了。奇怪的问题。这是octopus版本的问题。# J3 E# J0 Y, z6 }4 p
device_health_metrics  是由cephadm安装后默认生成的。
8 J3 y' U6 u0 N5 u1 x. \9 u2 N
您需要登录后才可以回帖 登录 | 开始注册

本版积分规则

关闭

站长推荐上一条 /4 下一条

如有购买积分卡请联系497906712

QQ|返回首页|Archiver|手机版|小黑屋|易陆发现 点击这里给我发消息

GMT+8, 2021-6-20 21:33 , Processed in 0.050407 second(s), 24 queries .

Powered by 龙睿 bbs168x X3.2

© 2001-2020 Comsenz Inc.

快速回复 返回顶部 返回列表