Start a 4 nodes cluster deploy the attached clusterbench-ee10.ear application put the cluster under some load (1000 users - 1 request per user every 4 secs): this is key in reproducing the issue, with just a few http request you won't see anything Note: you might have to restart nodes in the cluster to observer the issue; hence, run the load for ~5min, restart your nodes, run load for other 5min and you should be finally able to observe the message flooding the logs... if you are familiar with JMETER you can use this test plan JMETER-1.jmx to generate the necessary load to trigger the log: just open it with the GUI and press the start green button; you can start a 4 nodes local cluster started like: ./server1/bin/standalone.sh --server-config=standalone-ha.xml -Djboss.default.multicast.address=230.0.0.190 -Djboss.node.name=wildfly1 -Djboss.socket.binding.port-offset=100 ./server2/bin/standalone.sh --server-config=standalone-ha.xml -Djboss.default.multicast.address=230.0.0.190 -Djboss.node.name=wildfly2 -Djboss.socket.binding.port-offset=200 ./server3/bin/standalone.sh --server-config=standalone-ha.xml -Djboss.default.multicast.address=230.0.0.190 -Djboss.node.name=wildfly3 -Djboss.socket.binding.port-offset=300 ./server4/bin/standalone.sh --server-config=standalone-ha.xml -Djboss.default.multicast.address=230.0.0.190 -Djboss.node.name=wildfly4 -Djboss.socket.binding.port-offset=400