How to setup automatic failover for a cluster?
To enable automatic failover, two things must happen. Firstly, automatic failover is currently only available for the RoundRobin criterion. If the cluster is using any other criterion, then there is no failover support. Secondly, the client must install a BindInterceptor for the failover feature to be activated. This BindInterceptor is called com.inprise.vbroker.naming.NamingBindInterceptor and it takes the com.inprise.vbroker.orb.ORB as a parameter. This bind interceptor performs the necessary magic to failover to other cluster members when the current member fails.