Deployed an overcloud with CephClusterName: central, overriding the default "ceph". Deployment failed [1]. Looks like the podman exec command that is executed needs to be able to handle if the conf file isn't called /etc/ceph/ceph.conf.
[1]
2021-02-09 19:50:50.368991 | 2442019d-dca6-a223-294d-000000007e4b | FATAL | Enforce minimum Ceph cl
ients version to Mimic | undercloud -> 192.168.24.22 | error={"changed": true, "cmd": ["podman", "exec",
"e20408480409", "ceph", "osd", "set-require-min-compat-client", "mimic"], "delta": "0:00:00.472878", "e
nd": "2021-02-09 19:50:49.163912", "msg": "non-zero return code", "rc": 1, "start": "2021-02-09 19:50:48
.691034", "stderr": "Error initializing cluster client: ObjectNotFound('error calling conf_read_file',)\
nError: exec session exited with non-zero exit code 1: OCI runtime error", "stderr_lines": ["Error initi
alizing cluster client: ObjectNotFound('error calling conf_read_file',)", "Error: exec session exited wi
th non-zero exit code 1: OCI runtime error"], "stdout": "", "stdout_lines": []}
[WARNING]: ('undercloud -> 192.168.24.22',
'2442019d-dca6-a223-294d-000000007e4b') missing from stats
host: undercloud, task: None
I think train DCN (which deploys multiple clusters with diff names) is broken because this task was backported:
https:/