Log into a node running ceph-mon. Typically this will be ncn-s001/2/3.
Check the ceph health.
ceph health detail
Example Output:
HEALTH_WARN insufficient standby MDS daemons available
[WRN] MDS_INSUFFICIENT_STANDBY: insufficient standby MDS daemons available
have 0; want 1 more
This output explicitly states that you need at least 1 more to clear the alert.
Determine which MDS daemons are down.
ceph orch ps --daemon_type mds
Example Output:
NAME HOST STATUS REFRESHED AGE VERSION IMAGE NAME IMAGE ID CONTAINER ID
mds.cephfs.ncn-s001.lhoocr ncn-s001 stopped 4m ago 18h <unknown> registry.local/ceph/ceph:v15.2.8 <unknown> <unknown>
mds.cephfs.ncn-s002.nywheq ncn-s002 stopped 4m ago 18h <unknown> registry.local/ceph/ceph:v15.2.8 <unknown> <unknown>
mds.cephfs.ncn-s003.jdufcg ncn-s003 running (18h) 4m ago 18h 15.2.8 registry.local/ceph/ceph:v15.2.8 5553b0cb212c 4df61111d738
IMPORTANT: Depending on the configuration and the number of MDS daemons, the number of MDS daemons in a stopped or error state may vary.
Start the stopped MDS daemon(s).
ceph orch daemon start <MDS daemon name>
Repeat for each stopped MDS daemon.
Check the status of the cluster.
ceph health detail
Expected Output:
HEALTH_OK
IMPORTANT: If the daemon is not starting using the method above, please refer to Manage Ceph Services