Hi all,
we have a project which will have several wildfly cluster installations. Also in the same virtual lan (a development vlan) there are other wildfly clusters from totally unrelated projects.
Currently we are constantly logging exceptions in our cluster like this:
2015-01-14 15:00:26,005 ERROR [org.hornetq.core.server] (default I/O-5) HQ224018: Failed to create session: HornetQSecurityException[errorType=SECURITY_EXCEPTION message=HQ119031: Unable to validate user: HORNETQ.CLUSTER.ADMIN.USER]
at org.hornetq.core.security.impl.SecurityStoreImpl.authenticate(SecurityStoreImpl.java:145)
and like this:
2015-01-14 14:07:44,966 WARN [org.jgroups.protocols.TP$ProtocolAdapter] (INT-1,shared=udp) JGRP000031: master:server-1/ejb: dropping unicast message to wrong destination 96734980-0d24-bf33-4633-9eaf071dec99
I have closed all other wildfly servers from our projects and we keep receiving those errors. I used tcpdump and I noticed that the messages seem to originate from another wildfly server in the same virtual lan, from an unrelated project.
We have already tried setting a unique name in <broadcoast-groups>, <discovery-groups> and <cluster-connections> and also setting <cluster-user> and <cluster-password> in both places for hornetq-server.
Finally we tried changing multicast addresses and ports (increased everything by 1) in appropriate <socket-binding-group> (full-ha-sockets).
Still the error and warning kept showing. One thing to notice is that the other project's cluster is sending these messasges in server-1's 8180 port (offset 100).
What do I need to change in domain.xml to isolate my cluster?
