0 Replies Latest reply on Sep 5, 2007 3:46 PM by tnine

    Clustered MDB and Fault Tolerance

    tnine

      Hello all,
      I'm trying to configure my MDB's to run in an HA environment. I have half of the MDB working. The mdb performs the following tasks

      1. Read from HA JBoss MQ Queue
      2. Process message
      3. Push the message onto a different queue.

      I'm using 2 identical nodes (vmware instances, prodjboss1, and prodjboss2), and I have edited the file 'deploy/jms/hajndi-jms-ds.xml' to have the following

      prodjboss1

      java.naming.provider.url=${jboss.bind.address:localhost}:1100,prodjboss2:1100
      


      prodjboss2
      java.naming.provider.url=${jboss.bind.address:localhost}:1100,prodjboss1:1100
      


      Both my MDB's read from the correct master node. However, I'm trying to connect to a different output queue via my MDB, and I'm not getting the expected behavior. I'm using the following JNDI properties on both nodes when attempting to connect to the output queue

      java.naming.factory.initial=org.jnp.interfaces.NamingContextFactory
      java.naming.factory.url.pkgs=org.jboss.naming:org.jnp.interfaces
      java.naming.provider.url=jnp://prodjboss1:1100,prodjboss1:1100
      



      The lookup always works correctly on the node that's the master node, but fails on all slave nodes. Am I missing another property that I need to add in order to get automatic master discovery working and get all slaves connecting to the master MQ node to push data onto the queue?

      Thanks,
      Todd