I have a very basic 2-node cluster set up for a client, which allows
services to function even if a server fails or is downed for maintenance.
The "main" server is an older Compaq ML 330 P-III, which handles the
customer's needs very well. The "secondary" server is simply a Compaq EVO
desktop (remember this is for temporary\emergency use only). Between them
is a shared SCSI disk array containing their network drives, GW system,
etc, which is the cluster resource.

The challenge arises when the power returns after a power failure, and the
APC software powers on the servers. Because the EVO is a P-4, it reboots
sooner, and therefore becomes the master node of the cluster.

The short question is: Is there a way to "permanently" designate the ML330
as master node if it is available? In other words, services may fail over
to the EVO if the ML330 fails or goes offline, but I'd like them to return
to the ML330 when it comes back.