I am trying to use Pacemaker and create a multi-node cluster which is geographically separated nodes. I have the following questions..
. Is it possible without a shared data resources? Each of these nodes have a database server such as MySql running and in the event of failover, my plan is to replicate data to the node that takes over
How would the IP addressing work? Is it completely handled by the Pacemaker framework or anything else need to be done from the setup point of view. So far, I have only configured, two node clusters that are co-located, ie on the same subnet. I am not sure how the IP address allocation would work over the WAN.
Thanks
