The RSA Via Lifecycle and Governance WildFly Clustering Setup document gives some examples of log messages to look for when the server is started to validate that connectivity was established.
This article gives some commands to help identify if connectivity between the hosts is functional by using some simple commands.
This is the log snippet from the Setup document:
2015-07-09 07:14:21,719 INFO [org.hornetq.core.server] (Thread-1 (HornetQ-server-HornetQServerImpl::serverUUID=ece149af-25f9-11e5-ab43-1d33c64eb3be-1144939731))
HQ221027: Bridge ClusterConnectionBridge@4b496d61 [name=sf.my-cluster.081e723a-25fb-11e5-b5e4-bb0b73dd86b3, queue=QueueImpl[name=sf.my-cluster.081e723a-25fb-11e5-b5e4-
bb0b73dd86b3, postOffice=PostOfficeImpl[server=HornetQServerImpl::serverUUID=ece149af-25f9-11e5-ab43-
1d33c64eb3be]]@41ac21e4 targetConnector=ServerLocatorImpl (identity=(Clusterconnection-
bridge::ClusterConnectionBridge@4b496d61 [name=sf.my-cluster.081e723a-25fb-11e5-b5e4-bb0b73dd86b3, queue=QueueImpl[name=sf.my-cluster.081e723a-25fb-11e5-
b5e4-bb0b73dd86b3, postOffice=PostOfficeImpl[server=HornetQServerImpl::serverUUID=ece149af-25f9-11e5-ab43-
1d33c64eb3be]]@41ac21e4 targetConnector=ServerLocatorImpl[initialConnectors=[TransportConfiguration(name=http-connector, factory=org-hornetqcore-
remoting-impl-netty-NettyConnectorFactory) ?port=8080&http-upgradeendpoint=http-acceptor&host=10-101-250-27&http-upgrade-enabled=true],
discoveryGroupConfiguration=null]]::ClusterConnectionImpl@1691175884[nodeUUID=ece149af-25f9-11e5-ab43-1d33c64eb3be, connector=TransportConfiguration(name=httpconnector,
factory=org-hornetq-core-remoting-impl-netty-NettyConnectorFactory)?port=8080&host=10-101-250-11&http-upgrade-endpoint=http-acceptor&http-upgradeenabled=
true, address=jms, server=HornetQServerImpl::serverUUID=ece149af-25f9-11e5-ab43-1d33c64eb3be])) [initialConnectors=[TransportConfiguration(name=http-connector,
factory=org-hornetq-core-remoting-impl-netty-NettyConnectorFactory) ?port=8080&httpupgrade-endpoint=http-acceptor&host=10-101-250-27&http-upgrade-enabled=true],
discoveryGroupConfiguration=null]] is connected
On both hosts, stop the application.
On the Host Controller open two windows / SSH sessions:
One running the receiver application:
java -cp /home/oracle/wildfly/modules/system/layers/base/org/jgroups/main/jgroups-*.jar org.jgroups.tests.McastReceiverTest -mcast_addr 231.7.7.7 -port 9876
One running a network trace watching for incoming traffic (replace eth0 with the interface that is in use in your server):
tcpdump -i eth0 udp port 9876
On the Domain Controller open two windows / SSH sessions:
One running the sender application:
java -cp /home/oracle/wildfly/modules/system/layers/base/org/jgroups/main/jgroups-*.jar org.jgroups.tests.McastSenderTest -mcast_addr 231.7.7.7 -port 9876
One running a network trace watching for outgoing traffic (replace eth0 with the interface that is in use in your server):
tcpdump -i eth0 udp port 9876
In the Domain controller sender application it will have a > prompt that you can type text and hit enter to send.
In a working environment you will see network traffic outbound, and see inbound traffic on the Host controller.
You will also see the text you typed show up in the receiver window.
In a non-working scenario you will see only the DC tcpdump window having traffic, and the receiver will not have any updates to its tcpdump or receiver window.
Here is an screenshot of what the test looks like when there is connectivity:
Image descriptionThis is what the results look like when there is no connectivity:
Image descriptionNote that the top two SSH windows do not show the Test message that was typed and the tcpdump window does not show any traffic from the test.
One issue that blocks the Multi-cast traffic is igmp snooping being enabled in the environment's network switch.