A fail over problem in new cluster mode (keepOldFailoverModel:false) of JBoss Messaging

Solution Unverified - Updated -

Issue

There's a fail over problem:

  1. Kill (kill -9) a JBoss node (nodeA).
  2. Shut down other node (nodeB) cleanly in twice the time of NodeStateRefreshInterval.
  3. Failover to last node (nodeC) doesn't occur occasionally. In same time, polling job stays on nodeA. After that, nodeA cannot boot correctly.
    When the problem occurred, error messages logged repeatedly in nodeC:
2013-01-29 12:43:17,926 ERROR [org.jboss.messaging.core.impl.postoffice.MessagingPostOffice] (Thread-71) Error refreshing state of node: 3
java.lang.IllegalStateException: Failover node 2 for node Quarantined Node[1], failover[2], crashed[true] is not alive!
    at org.jboss.messaging.core.impl.postoffice.MessagingPostOffice.processClusterState(MessagingPostOffice.java:640)
    at org.jboss.messaging.core.impl.postoffice.MessagingPostOffice.access$3700(MessagingPostOffice.java:108)
    at org.jboss.messaging.core.impl.postoffice.MessagingPostOffice$StateMonitor.run(MessagingPostOffice.java:4271)

Environment

  • Red Hat JBoss Enterprise SOA Platform (SOA-P)
    • 5.3

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.

Current Customers and Partners

Log in for full access

Log In

New to Red Hat?

Learn more about Red Hat subscriptions

Using a Red Hat product through a public cloud?

How to access this content