Now a simple Failover Cluster got implemented.
It’s far from feature complete, also it’s possible to do this better!
Especially if you may have Acccess to Fibre Channel based SAN, you can improve Performance and dont have to stick to DRBD.
Also if you have Access to more Servers it may be a real Option to move the Database to other Servers, and only provide the Webservice from the Cluster and the Database Service from a second Cluster.
It’s always a question about the Equipment you have or got provided by the supplier.
if i would find economical reasons, i would try to implement the Application on a Solaris Cluster Platform with Zones.