Hi Clebert.
It would be nice to have documented requirements on firewall for fail-over configuration, in both community and EAP documentation. Or at least add note saying "Don't forget about your firewall settings." or something similar.

I met this issue on my Fedora 14, I wasn't able to configure fail-over as described in JBPAPP-5327, later I realized that examples multiple-failover and multiple-failover-failback are not working too. After turning firewall off, examples ended successfully.

Backup was not announced and later I received this error:
[java] HornetQServer_1 err:HornetQException[errorCode=3 message=Timed out waiting to receive initial broadcast from cluster]
[java] HornetQServer_1 err: at org.hornetq.core.client.impl.ServerLocatorImpl.createSessionFactory(ServerLocatorImpl.java:513)
[java] HornetQServer_1 err: at org.hornetq.core.client.impl.ServerLocatorImpl.connect(ServerLocatorImpl.java:447)
[java] HornetQServer_1 err: at org.hornetq.core.server.cluster.impl.ClusterManagerImpl$1.run(ClusterManagerImpl.java:846)
[java] HornetQServer_1 err: at org.hornetq.utils.OrderedExecutorFactory$OrderedExecutor$1.run(OrderedExecutorFactory.java:100)
[java] HornetQServer_1 err: at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
[java] HornetQServer_1 err: at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)