hosts fail cephadm check ceph分布式存告警

admin

health: HEALTH_WARN
            1 hosts fail cephadm check
4 c' {% a8 |# u8 h' B+ P- e1 FCEPHADM_HOST_CHECK_FAILED: 1 hosts fail cephadm check
& b! G5 l) z: O. B7 Y7 e( ]/ u# ceph health detail
6 J( E% d8 F) U! K1 [% dHEALTH_WARN 1 hosts fail cephadm check
[WRN] CEPHADM_HOST_CHECK_FAILED: 1 hosts fail cephadm check
    host gm268-1 failed check: Failed to connect to gm268-1 (gm268-1).
Please make sure that the host is reachable and accepts connections using the cephadm SSH key

6 g, i/ }% E! B) dTo add the cephadm SSH key to the host:
5 M6 {/ |. m: J> ceph cephadm get-pub-key > ~/ceph.pub
+ {7 M  w' D2 `9 [3 h> ssh-copy-id -f -i ~/ceph.pub root@gm268-1
! u/ h- a4 D# Q/ U5 r) B# a
To check that the host is reachable:
  q$ H, s- S; x; a3 A  W$ A> ceph cephadm get-ssh-config > ssh_config
> ceph config-key get mgr/cephadm/ssh_identity_key > ~/cephadm_private_key
2 E9 p& o+ i, c2 ]> chmod 0600 ~/cephadm_private_key
, [' L! J+ G4 y; s> ssh -F ssh_config -i ~/cephadm_private_key root@gm268-1
6 m% f# S0 ]6 i" O7 `9 j' C

/ q% U4 ^/ U: n8 B% j3 g7 Q

admin

[root@gm268-5 ~]# ceph cephadm check-host gm268-1
podman|docker (/usr/bin/podman) is present
systemctl is present
, z' T4 e- X6 zlvcreate is present
Unit chronyd.service is enabled and running
0 J. a2 O/ s+ v4 `- G( l) y: Q$ ]8 bHostname "gm268-1" matches what is expected.
% a0 ~3 A* [5 E9 [; iHost looks OK
1 G$ n* X% K# G/ K9 _3 B* ggm268-1 (None) ok
/ [. t3 N* p/ d3 r7 S& ?[root@gm268-5 ~]# ceph cephadm check-host gm268-2
podman|docker (/usr/bin/podman) is present
systemctl is present
lvcreate is present
Unit chronyd.service is enabled and running
2 g  ^' k/ u$ {7 oHostname "gm268-2" matches what is expected.
Host looks OK
gm268-2 (None) ok
[root@gm268-5 ~]# ceph cephadm check-host gm268-3
podman|docker (/usr/bin/podman) is present
8 V( x5 H7 W( Wsystemctl is present
% F* E8 V& w" @0 ~lvcreate is present
# r5 G8 r" U" h# k# b  yUnit chronyd.service is enabled and running
5 K. V7 B/ F) ?. q( L3 K6 B/ BHostname "gm268-3" matches what is expected.
Host looks OK
- w, X. A! d1 k! e( agm268-3 (None) ok

admin

[root@gm268-1 lib64]# ceph -s
  o8 `7 p  D8 o' s, e0 h  cluster:
4 R# d" f9 u/ P% S5 W, D    id:     ce68aab8-8f46-11ed-88c0-ac1f6b3a30b9
5 C& U+ e, Q6 ]+ Z    health: HEALTH_OK
9 q/ G, a( e7 p! @8 T7 h# `% D
  services:
" f5 d- G5 m/ n7 ?! w0 n) {, i    mon: 3 daemons, quorum gm268-1,gm268-2,gm268-3 (age 11m)
    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
* O0 f7 I# G1 e7 I5 o& ]6 x- I    osd: 41 osds: 41 up (since 10m), 41 in (since 2w); 1008 remapped pgs

  data:
5 C/ G' @. X" c' S; r  a' W    pools:   3 pools, 9217 pgs
9 R& P5 o. W% K, \  P, r9 ^- j    objects: 10.14M objects, 14 TiB
    usage:   28 TiB used, 569 TiB / 597 TiB avail
    pgs:     1026683/20286478 objects misplaced (5.061%)
             8207 active+clean
             958  active+remapped+backfill_wait
3 H# r8 O! h9 g3 m             52   active+remapped+backfilling

  io:
5 k0 `$ g. V6 L0 F. H6 A% {) P    recovery: 735 MiB/s, 14 keys/s, 484 objects/s

[root@gm268-1 lib64]# ceph cephadm check-host gm268-1
8 g: e, c* U7 c# d# `9 mpodman|docker (/usr/bin/podman) is present
6 n0 r: N" o# Esystemctl is present
/ i, F5 ~0 K- Vlvcreate is present
Unit chronyd.service is enabled and running
/ p  \6 O. i4 y3 s% p# [: @Hostname "gm268-1" matches what is expected.
Host looks OK
gm268-1 (None) ok
$ b1 {1 Q4 M7 s/ h( L! w& T. t  G