[PVE-User] How to restart ceph-mon?

[Alwin Antreich]

On Fri, Feb 21, 2020 at 03:29:08PM +0100, Marco Gaiarin wrote:
> Mandi! Alwin Antreich
>   In chel di` si favelave...
> 
> > Yes, that looks strange. But as said before, it is deprecated to use
> > IDs. Best destroy and re-create the MON one-by-one. The default command
> > will create them with the hostname as ID. Then this phenomenon should
> > disappear as well.
> 
> Done, via web interface, with a little glitch.
> 
> I've stopped and dropped the monitor, but these don't stop (and drop)
> the manager, and so creating a new mon va webinterface lead to:
> 
>  Created symlink /etc/systemd/system/ceph-mon.target.wants/ceph-mon at hulk.service -> /lib/systemd/system/ceph-mon at .service.
>  INFO:ceph-create-keys:ceph-mon is not in quorum: u'synchronizing'
>  INFO:ceph-create-keys:ceph-mon is not in quorum: u'synchronizing'
>  INFO:ceph-create-keys:ceph-mon is not in quorum: u'probing'
>  INFO:ceph-create-keys:ceph-mon is not in quorum: u'electing'
>  INFO:ceph-create-keys:ceph-mon is not in quorum: u'electing'
>  INFO:ceph-create-keys:ceph-mon is not in quorum: u'electing'
>  INFO:ceph-create-keys:ceph-mon is not in quorum: u'electing'
>  INFO:ceph-create-keys:ceph-mon is not in quorum: u'electing'
>  INFO:ceph-create-keys:Key exists already: /etc/ceph/ceph.client.admin.keyring
>  INFO:ceph-create-keys:Key exists already: /var/lib/ceph/bootstrap-osd/ceph.keyring
>  INFO:ceph-create-keys:Key exists already: /var/lib/ceph/bootstrap-rgw/ceph.keyring
>  INFO:ceph-create-keys:Key exists already: /var/lib/ceph/bootstrap-mds/ceph.keyring
>  INFO:ceph-create-keys:Talking to monitor...
>  TASK ERROR: ceph manager directory '/var/lib/ceph/mgr/ceph-hulk' already exists
> 
> probably because the task try also to fire up a mgr, that was just
> created.
> 
> 
> Anyway, nothing changed. On a rebooted node:
> 
> 	root at capitanmarvel:~# ps aux | grep ceph[-]mon
> 	ceph        2725  0.5  0.2 522224 98428 ?        Ssl  feb18  21:14 /usr/bin/ceph-mon -i capitanmarvel --pid-file /var/run/ceph/mon.capitanmarvel.pid -c /etc/ceph/ceph.conf --cluster ceph --setuser ceph --setgroup ceph
> 
> on a node when i do a 'systemctl restart ceph-mgr@<ID>.service':
> 
> 	root at hulk:~# ps aux | grep ceph[-]mon
> 	ceph     4166380  0.8  0.1 466648 55676 ?        Ssl  15:19   0:03 /usr/bin/ceph-mon -f --cluster ceph --id hulk --setuser ceph --setgroup ceph
I don't see this in the systemd unit files for Ceph. Also my test
systems do not have the pid file either. Maybe this is something from
an previous upgrade?

  systemctl cat ceph-mon@<id>.service
You can check with the above command how each Ceph service or target
should be started.

--
Cheers,
Alwin