[ClusterLabs] Pacemaker remote node ofgline after reboot

[Ken Gaillot]

On 05/11/2017 03:45 PM, Ignazio Cassano wrote:
> Hello, I installed a pacemaker cluster with 3 nodes and 2 remote nodes
> (pacemaker remote). All nodes are centos 7.3. The remote nodes are
> online and pacemaker resources are running on them.  When I reboot e
> remote pacemaker node it does not return online and Its pacemaker
>  resources remain blocked. The only solution I found is removing the
> remote node (pcs resource delete node) and creating it again. In this
> case its resources become immediately available. 
> Please, any help?
> Ignazio

Did you do "systemctl enable pacemaker_remote"?

By reboot, do you mean gracefully or simulating a failure? A graceful
exit should cause all resources on the remote node to stop. If failure,
the cluster needs to be able to fence the remote node before it can
recover its resources.