hosts fail cephadm check ceph分布式存告警

admin

health: HEALTH_WARN
            1 hosts fail cephadm check
; |; k' H5 S! |7 VCEPHADM_HOST_CHECK_FAILED: 1 hosts fail cephadm check
& m8 Q0 R( t4 ?: W3 ]  T* d# ceph health detail
. e7 {) |$ B: L3 z, G# q7 o8 I' Y/ tHEALTH_WARN 1 hosts fail cephadm check
[WRN] CEPHADM_HOST_CHECK_FAILED: 1 hosts fail cephadm check
6 R- ]/ Y3 w+ L9 e2 Y; D7 ^: T    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

To add the cephadm SSH key to the host:
7 u. ^! h* `0 k0 E4 y+ y2 h% {> ceph cephadm get-pub-key > ~/ceph.pub
4 o: C/ f* {6 F5 P+ U* o+ ]> ssh-copy-id -f -i ~/ceph.pub root@gm268-1
! w- K8 |' F6 e2 x3 M; e
# Z. I, a) K) [* mTo check that the host is reachable:
> ceph cephadm get-ssh-config > ssh_config
1 `5 p3 ^) ]4 W6 |3 }* l* l7 w" R> ceph config-key get mgr/cephadm/ssh_identity_key > ~/cephadm_private_key
- H5 f( Y& Z$ s/ @> chmod 0600 ~/cephadm_private_key
1 \! ]3 `1 j+ V" V# f1 d- b% Q> ssh -F ssh_config -i ~/cephadm_private_key root@gm268-1


# _' L% v. e. M% [' g( e
$ E" G. A" }; ~, }; v

admin

[root@gm268-5 ~]# ceph cephadm check-host gm268-1
podman|docker (/usr/bin/podman) is present
* O& S2 p2 h- L% R& L" _6 `5 Q3 dsystemctl is present
lvcreate is present
Unit chronyd.service is enabled and running
6 J- {+ }$ W' R: m7 QHostname "gm268-1" matches what is expected.
& a: {" h, J& [  B6 b0 m( F) @2 rHost looks OK
gm268-1 (None) ok
) n# s7 O6 K% X' `[root@gm268-5 ~]# ceph cephadm check-host gm268-2
podman|docker (/usr/bin/podman) is present
3 Y, Q0 }8 l$ U5 p7 |; W* Esystemctl is present
  f$ M& C+ ]: z, ^) g% ^4 X! M; ]lvcreate is present
  M) H! x) _7 D" x/ ~" Y) G- o" H9 oUnit chronyd.service is enabled and running
. ?. e. F: f+ `9 D& LHostname "gm268-2" matches what is expected.
5 j" I9 F3 q  Z8 h0 kHost looks OK
gm268-2 (None) ok
) ^1 o' l. Z  X+ n$ @[root@gm268-5 ~]# ceph cephadm check-host gm268-3
& V" E+ p0 G$ w$ K3 ?" Rpodman|docker (/usr/bin/podman) is present
systemctl is present
  f7 T' N0 T9 @: Q# G. J- x. J, N! ^lvcreate is present
" Y- B& l1 n- IUnit chronyd.service is enabled and running
Hostname "gm268-3" matches what is expected.
Host looks OK
! d, r% B; j5 M  W8 [: y& hgm268-3 (None) ok

admin

[root@gm268-1 lib64]# ceph -s
  cluster:
: f' Y# ~& {! d$ O    id:     ce68aab8-8f46-11ed-88c0-ac1f6b3a30b9
4 }& L1 Y1 P2 F- \1 Z    health: HEALTH_OK
0 b, A! K/ l& M5 R# W8 n3 ^: k
  services:
- e3 k" l( h& [& P5 l7 n* h    mon: 3 daemons, quorum gm268-1,gm268-2,gm268-3 (age 11m)
6 _8 {  U$ l! g    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
    osd: 41 osds: 41 up (since 10m), 41 in (since 2w); 1008 remapped pgs
9 k  p3 h  b9 K, p5 E9 l* C# H
  u+ O, N/ @, ]! X1 T$ ^$ b  data:
3 q+ A1 b& }& _/ o4 _) h    pools:   3 pools, 9217 pgs
( r- Q+ `& X5 u: U3 v# I    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
             52   active+remapped+backfilling

" \. f' h& E8 Y  io:
2 Q7 }6 \& X% z  x5 i. q; n    recovery: 735 MiB/s, 14 keys/s, 484 objects/s

[root@gm268-1 lib64]# ceph cephadm check-host gm268-1
podman|docker (/usr/bin/podman) is present
% @. N9 L1 r5 H( [$ h, _7 psystemctl is present
lvcreate is present
Unit chronyd.service is enabled and running
Hostname "gm268-1" matches what is expected.
  @1 A: C) w! B# r* S1 WHost looks OK
gm268-1 (None) ok