hosts fail cephadm check ceph分布式存告警

admin

health: HEALTH_WARN
            1 hosts fail cephadm check
# e) ?2 x( q6 V4 {+ q; KCEPHADM_HOST_CHECK_FAILED: 1 hosts fail cephadm check
# ceph health detail
, G$ T" Y. ^( r5 Z4 E3 tHEALTH_WARN 1 hosts fail cephadm check
[WRN] CEPHADM_HOST_CHECK_FAILED: 1 hosts fail cephadm check
, }4 w* R, Q9 o6 n: }    host gm268-1 failed check: Failed to connect to gm268-1 (gm268-1).
, [, X, J  m  z5 w3 e6 fPlease make sure that the host is reachable and accepts connections using the cephadm SSH key

; I, [! I* \0 m9 l. ?6 f0 BTo add the cephadm SSH key to the host:
+ Y+ H: v- N5 i/ z. _9 B( @1 b> ceph cephadm get-pub-key > ~/ceph.pub
/ [" I# B+ v3 W; x: ^> ssh-copy-id -f -i ~/ceph.pub root@gm268-1
2 Z. E+ A; f  y! t0 p% S
2 s3 t5 e3 ^9 S- A( S- Y2 ~+ |To check that the host is reachable:
% ]! s2 _. x. B> ceph cephadm get-ssh-config > ssh_config
> ceph config-key get mgr/cephadm/ssh_identity_key > ~/cephadm_private_key
> chmod 0600 ~/cephadm_private_key
' r  i' C+ p/ V0 t% t2 Y1 ^> ssh -F ssh_config -i ~/cephadm_private_key root@gm268-1



' L* @5 s( \3 ]8 h' [) N$ W

admin

[root@gm268-5 ~]# ceph cephadm check-host gm268-1
podman|docker (/usr/bin/podman) is present
6 E3 q. I/ m5 t0 [systemctl is present
lvcreate is present
Unit chronyd.service is enabled and running
8 u! ?! L  G* A8 BHostname "gm268-1" matches what is expected.
, A- V6 }# {) Q( F. DHost looks OK
gm268-1 (None) ok
[root@gm268-5 ~]# ceph cephadm check-host gm268-2
podman|docker (/usr/bin/podman) is present
systemctl is present
; g! H1 J  s6 P% r5 m0 ]lvcreate is present
) N! j$ S8 ?: W8 R2 k4 d4 cUnit chronyd.service is enabled and running
Hostname "gm268-2" matches what is expected.
) l# d% F5 z6 n8 t' t$ f( X3 e+ M5 J7 HHost looks OK
: V1 Q, \8 A% S' h$ Ggm268-2 (None) ok
4 I5 y- G4 Y& z' C  M) y9 H- h" D[root@gm268-5 ~]# ceph cephadm check-host gm268-3
8 U7 k2 ~  v  V( ]) tpodman|docker (/usr/bin/podman) is present
& v6 Y5 I9 g7 K; {7 q6 Msystemctl is present
8 ?2 n3 N: n7 X- q+ Hlvcreate is present
  l# e( I2 O( Y5 T( s( `" z, @Unit chronyd.service is enabled and running
Hostname "gm268-3" matches what is expected.
Host looks OK
5 }6 G9 V; Z/ Q' vgm268-3 (None) ok
0 w0 z; l& ?$ D+ {% }: {9 Y

admin

[root@gm268-1 lib64]# ceph -s
9 L' M$ d! F) X: `/ z5 h  cluster:
6 I9 W% j2 m5 Y, y9 }6 `$ N% a) k    id:     ce68aab8-8f46-11ed-88c0-ac1f6b3a30b9
    health: HEALTH_OK
7 b( @0 u/ T" {8 k) N) W* }
9 Q+ d2 T7 ~' D. u  services:
    mon: 3 daemons, quorum gm268-1,gm268-2,gm268-3 (age 11m)
+ a7 s8 Y' Z$ R  t" j& N* I; w' M1 j' H    mgr: gm268-2.zttohs(active, since 3w), standbys: gm268-3.sjagqo, gm268-1.jgdvxs
    mds: cephfs:1 {0=cephfs.gm268-2.xdsdoz=up:active} 2 up:standby
) x. r+ W3 p4 q, Y    osd: 41 osds: 41 up (since 10m), 41 in (since 2w); 1008 remapped pgs
& @% I% I+ c' x. ]+ M8 r+ [! t1 |
  data:
    pools:   3 pools, 9217 pgs
    objects: 10.14M objects, 14 TiB
    usage:   28 TiB used, 569 TiB / 597 TiB avail
$ t! X0 r9 m, W' q  }4 z$ T    pgs:     1026683/20286478 objects misplaced (5.061%)
, K" k: b$ t2 G$ s             8207 active+clean
             958  active+remapped+backfill_wait
             52   active+remapped+backfilling

4 S7 M! N- n- `4 ?. _# u1 i% b  io:
$ z1 _3 \8 n! b' t& t; `, e- U% C    recovery: 735 MiB/s, 14 keys/s, 484 objects/s
8 n  e/ u/ j- b, B+ V
2 @- l" ]8 G' e$ v3 V+ u2 o[root@gm268-1 lib64]# ceph cephadm check-host gm268-1
podman|docker (/usr/bin/podman) is present
systemctl is present
lvcreate is present
5 E6 S) f+ S# h! zUnit chronyd.service is enabled and running
Hostname "gm268-1" matches what is expected.
Host looks OK
% o! _. A* s# ]/ F2 t; P* f. ]gm268-1 (None) ok