Project

General

Profile

Actions

Bug #45167

closed

cephadm: mons are not properly deployed

Added by Sebastian Wagner about 4 years ago. Updated over 3 years ago.

Status:
Can't reproduce
Priority:
Low
Category:
cephadm (binary)
Target version:
-
% Done:

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

cephadm run --name mon.hostname --fsid myfsid
INFO:cephadm:Using recent ceph image ceph/ceph:v15
"debug "2020-04-21T17:13:54.335+0000 7fed2debf6c0  0 set uid:gid to 167:167 (ceph:ceph)
"debug "2020-04-21T17:13:54.335+0000 7fed2debf6c0  0 ceph version 15.2.1 (9fd2f65f91d9246fae2c841a6222d34d121680ee) octopus (stable), process ceph-mon, pid 1
"debug "2020-04-21T17:13:54.335+0000 7fed2debf6c0  0 pidfile_write: ignore empty --pid-file
"debug "2020-04-21T17:13:54.343+0000 7fed2debf6c0  0 load: jerasure load: lrc load: isa
"debug "2020-04-21T17:13:54.347+0000 7fed2debf6c0 -1 Invalid argument: /var/lib/ceph/mon/hostname/store.db: does not exist (create_if_missing is false)

"debug "2020-04-21T17:13:54.347+0000 7fed2debf6c0 -1 error opening mon data directory at '/var/lib/ceph/mon/hostname': (22) Invalid argument

We need to be able to recover from this error.

Afterwards:

ceph orch apply mon hostname`, the `/var/lib/ceph/<fsid>` gets removed. When I re-add it, the directory comes back, but empty, and ceph orch ls shows this:

mon                1/2  0s ago     4m   storage-14b-0,storage-14b-1  mix                       mix

Related issues 3 (0 open3 closed)

Related to Orchestrator - Bug #45235: cephadm: mons are not properly undeployedCan't reproduce

Actions
Related to Orchestrator - Bug #45279: cephadm bootstrap: monmaptool --create: error writing to '/tmp/monmap': (21) Is a directoryCan't reproduce

Actions
Copied from Orchestrator - Documentation #45165: cephadm troubleshooting: recover from broken daemonsCan't reproduce

Actions
Actions

Also available in: Atom PDF