We connect to the Finesse server using XMPP as a Server application to receive agent state and dialog notifications. We are using Smack Libs to manage connection retry attempts. If we are to handle failover scenarios wherein the primary finesse server fails, should our code be handling the connection to the secondary or the other server in the cluster? Or can we still try connecting to the Finesse primary server and it can automatically direct the connection to an active server.
Would you please advice on how this should be approached?