ceph reduced data availability

[cikoshel]

hi. need help. ceph cannot be use after all server shutdown.

root@host1-sa:~# ceph -v
ceph version 12.2.5 (dfcb7b53b2e4fcd2a5af0240d4975adc711ab96e) luminous (stable)
root@host1-sa:~#
root@host1-sa:~# ceph -s
2019-07-13 15:57:06.959518 7f919bfff700 0 -- 192.168.2.1:0/3506265993 >> 192.168.2.3:6804/2644 conn(0x7f918400b540 :-1 s=STATE_CONNECTING_WAIT_CONNECT_REPLY_AUTH pgs=0 cs=0 l=1).handle_connect_reply connect got BADAUTHORIZER
2019-07-13 15:57:06.959621 7f919bfff700 0 -- 192.168.2.1:0/3506265993 >> 192.168.2.3:6804/2644 conn(0x7f918400b540 :-1 s=STATE_CONNECTING_WAIT_CONNECT_REPLY_AUTH pgs=0 cs=0 l=1).handle_connect_reply connect got BADAUTHORIZER
cluster:
id: f5c2333f-ed34-4fe1-a02a-191e78e98b51
health: HEALTH_WARN
Reduced data availability: 45 pgs inactive, 23 pgs peering
clock skew detected on mon.host2-sa, mon.host3-sa

services:
mon: 3 daemons, quorum host1-sa,host2-sa,host3-sa
mgr: host3-sa(active), standbys: host1-sa, host2-sa
osd: 5 osds: 5 up, 5 in

data:
pools: 1 pools, 64 pgs
objects: 323k objects, 1290 GB
usage: 2849 GB used, 2739 GB / 5588 GB avail
pgs: 17.188% pgs unknown
53.125% pgs not active
23 peering
19 active+clean
11 activating
11 unknown

 

[Alwin]

Try to restart Ceph's MONs services one at a time and see if the state changes. Also make sure the time i synced and the same on all Ceph nodes.