[ClusterLabs] corosync/pacemaker resources start after reboot - incorrect node ID calculated

[Ken Gaillot]

On Mon, 2021-09-27 at 23:39 +0000, Neitzert, Greg A wrote:

<snip>

> I am wondering if we should be adding something like this:
>  
>  
> nodelist {
>   node {
>     ring0_addr: m660b-qproc4-HA
>     nodeid: 1
>   }
>   node {
>     ring0_addr: m660b-qproc3-HA
>     nodeid: 2
>   }
> }
>  
> Where the hostnames above map to the 169.x.x.x addresses for each
> node of the cluster. 
> I think that will ensure a. the node ID is a stable value (always 1
> or 2 – not calculated by corosync) but also maps our ring addresses
> to the 169 addresses as well?

Yes, exactly
 
> Finally, am I correct that the hostnames listed in the nodelist above
> should be set in the /etc/hosts file to point to the 169 addresses
> for each host, NOT a hostname that resolves to 127.0.0.1?

Yes -- more specifically, an address on whichever interface the
corosync traffic should go over (cluster token passing and control
messages). Often this is a dedicated network so cluster traffic can't
be crowded out by a networking spike (which might otherwise lead to
fencing).
-- 
Ken Gaillot <kgaillot at redhat.com>