On 6/13/22 07:39, farhad kh wrote:
i upgraded my cluster to 17.2 and locked process upgrade
i have error
[root@ceph2-node-01 ~]# ceph -s
cluster:
id: 151b48f2-fa98-11eb-b7c4-000c29fa2c84
health: HEALTH_WARN
Reduced data availability: 32 pgs inactive
Degraded data redundancy: 32 pgs undersized
services:
mon: 3 daemons, quorum ceph2-node-03,ceph2-node-02,ceph2-node-01 (age
4h)
mgr: ceph2-node-02.mjagnd(active, since 11h), standbys:
ceph2-node-01.hgrjgo
osd: 12 osds: 12 up (since 43m), 12 in (since 21h)
data:
pools: 1 pools, 32 pgs
objects: 0 objects, 0 B
usage: 434 MiB used, 180 GiB / 180 GiB avail
pgs: 100.000% pgs not active
32 undersized+peered
^^ Nowadays the mgr is a critical component. Especially in cephadm
deployments. Probably your cluster is fine, but the manager is not. At
least, that is my experience when 100% pgs not active is reported.
What does a "ceph versions" give? So we can chech wat daemons have and
what daemons have not been upgraded (yet).
The cluster is not progressing with the update as it is in WARN state.
And possibly because the manager is not working correctly.
Disk One has not yet been upgraded to a new version, and the upgrade
process has stopped altogether
How can I solve this problem? What is the cause?
You might be able to stop the active manager and let the standby take
over and see if that improves things.
Gr. Stefan
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
