Bug #43932: bin/cephadm: All daemons should call port_in_use - Orchestrator - Ceph

Actions

Copy link

Bug #43932

closed

bin/cephadm: All daemons should call port_in_use

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

Status:

Resolved

Priority:

Normal

Assignee:

Category:

cephadm (binary)

Target version:

% Done:

Source:

Tags:

Backport:

Regression:

Severity:

3 - minor

Reviewed:

Affected Versions:

ceph-qa-suite:

Pull request ID:

Crash signature (v1):

Crash signature (v2):

Description

Jan 26 18:38:04 monitor2 systemd[1]: ceph-6a888c04-4041-11ea-96dc-525400e6f794@rgw.realm1.default.aylrtn.service: Service RestartSec=10s expired, scheduling restart.
Jan 26 18:38:04 monitor2 systemd[1]: Stopped Ceph daemon for 6a888c04-4041-11ea-96dc-525400e6f794.
Jan 26 18:38:04 monitor2 systemd[1]: Starting Ceph daemon for 6a888c04-4041-11ea-96dc-525400e6f794...
Jan 26 18:38:04 monitor2 podman[22283]: Error: no container with name or ID ceph-6a888c04-4041-11ea-96dc-525400e6f794-rgw.realm1.default.aylrtn found: no such container
Jan 26 18:38:04 monitor2 systemd[1]: Started Ceph daemon for 6a888c04-4041-11ea-96dc-525400e6f794.
Jan 26 18:38:04 monitor2 systemd[1]: Started libpod-conmon-f076e67e3fa79d4bb597e4ed296d3963bf069eb3a72b0ee346ec104028a2db20.scope.
Jan 26 18:38:05 monitor2 systemd[1]: Started libcontainer container f076e67e3fa79d4bb597e4ed296d3963bf069eb3a72b0ee346ec104028a2db20.
Jan 26 18:38:05 monitor2 bash[22294]: debug 2020-01-26T17:38:05.461+0000 7f192a83fa80  0 framework: beast
Jan 26 18:38:05 monitor2 bash[22294]: debug 2020-01-26T17:38:05.461+0000 7f192a83fa80  0 framework conf key: port, val: 7480
Jan 26 18:38:05 monitor2 bash[22294]: debug 2020-01-26T17:38:05.473+0000 7f192a83fa80  0 deferred set uid:gid to 167:167 (ceph:ceph)
Jan 26 18:38:05 monitor2 bash[22294]: debug 2020-01-26T17:38:05.473+0000 7f192a83fa80  0 ceph version 15.0.0-9543-g1c7fc80ba1 (1c7fc80ba17319e7d50724ac7b32d47bdba4204a) octopus (dev), process radosgw, pid 1
Jan 26 18:38:05 monitor2 bash[22294]: debug 2020-01-26T17:38:05.809+0000 7f192a83fa80  0 starting handler: beast
Jan 26 18:38:05 monitor2 bash[22294]: debug 2020-01-26T17:38:05.813+0000 7f192a83fa80 -1 failed to bind address 0.0.0.0:7480: Address already in use
Jan 26 18:38:05 monitor2 bash[22294]: debug 2020-01-26T17:38:05.813+0000 7f192a83fa80 -1 ERROR: failed initializing frontend
Jan 26 18:38:05 monitor2 systemd[1]: ceph-6a888c04-4041-11ea-96dc-525400e6f794@rgw.realm1.default.aylrtn.service: Main process exited, code=exited, status=98/n/a
Jan 26 18:38:06 monitor2 systemd[1]: ceph-6a888c04-4041-11ea-96dc-525400e6f794@rgw.realm1.default.aylrtn.service: Unit entered failed state.
Jan 26 18:38:06 monitor2 systemd[1]: ceph-6a888c04-4041-11ea-96dc-525400e6f794@rgw.realm1.default.aylrtn.service: Failed with result 'exit-code'.
Jan 26 18:38:16 monitor2 systemd[1]: ceph-6a888c04-4041-11ea-96dc-525400e6f794@rgw.realm1.default.aylrtn.service: Service RestartSec=10s expired, scheduling restart.
Jan 26 18:38:16 monitor2 systemd[1]: Stopped Ceph daemon for 6a888c04-4041-11ea-96dc-525400e6f794.
Jan 26 18:38:16 monitor2 systemd[1]: ceph-6a888c04-4041-11ea-96dc-525400e6f794@rgw.realm1.default.aylrtn.service: Start request repeated too quickly.
Jan 26 18:38:16 monitor2 systemd[1]: Failed to start Ceph daemon for 6a888c04-4041-11ea-96dc-525400e6f794.
Jan 26 18:38:16 monitor2 systemd[1]: ceph-6a888c04-4041-11ea-96dc-525400e6f794@rgw.realm1.default.aylrtn.service: Unit entered failed state.
Jan 26 18:38:16 monitor2 systemd[1]: ceph-6a888c04-4041-11ea-96dc-525400e6f794@rgw.realm1.default.aylrtn.service: Failed with result 'exit-code'.

As bin/cephadm could successfully deploy the daemon, users are not getting any helpful error message, except for manually looking into the journald log files.

If we call port_in_use() before deploying the daemons, users will get a useful error message.