[ClusterLabs] How to reload pacemaker_remote service?

[Gilbert, Mike]

Ken,

Thank you very much for the response. That method does allow us to reload the configs, but it's a bit heavy-handed for our use case. All we're wanting to do is rotate the log files. Is there any other mechanism we could use to achieve that goal?

Thanks!
Mike

On 6/9/20, 7:10 AM, "Users on behalf of Ken Gaillot" <users-bounces at clusterlabs.org on behalf of kgaillot at redhat.com> wrote:

    Currently it's not possible. However you should be able to put the
    cluster into maintenance mode, restart pacemaker_remote, then take the
    cluster out of maintenance mode.

    Test it to be sure. I believe the connection resource might be marked
    as failed, but the cluster should be able to reconnect and reprobe the
    state of all resources without fencing.

    On Tue, 2020-06-09 at 03:28 +0000, Gilbert, Mike wrote:
    > Hello all,
    >  
    > We are running Pacemaker 1.1.21-4 and are trying to figure out how we
    > can do the equivalent of “systemctl reload pacemaker_remote”. Does
    > anyone know what signal needs to get sent to the pacemaker_remoted
    > service to reload its config? Sending a SIGHUP appears to kill the
    > process.
    >  
    > Thanks for any help!
    > Mike
    -- 
    Ken Gaillot <kgaillot at redhat.com>

    _______________________________________________
    Manage your subscription:
    https://lists.clusterlabs.org/mailman/listinfo/users

    ClusterLabs home: https://www.clusterlabs.org/