1 Reply Latest reply on Jan 4, 2006 3:48 AM by belaban

    local_addr is null

    okros

      Hi,

      I have a JBoss 4.0.2 cluster consisting of two servers.

      In some time after deployment an ear file I got a warning messages:

      2005-10-22 23:00:17,219 WARN [org.jgroups.protocols.FD - FD.up] I was suspected, but will not remove myself from membership (waiting for EXIT message)
      2005-10-22 23:00:19,730 WARN [org.jgroups.protocols.FD - FD.up] I was suspected, but will not remove myself from membership (waiting for EXIT message)
      2005-10-22 23:00:20,226 WARN [org.jgroups.protocols.pbcast.CoordGmsImpl - CoordGmsImpl.handleSuspect] I am the coord and I'm being am suspected -- will probably leave shortly
      2005-10-22 23:00:20,228 INFO [org.jboss.ha.framework.interfaces.HAPartition.lifecycle.KaparPartition-2 - HAPartitionImpl.suspect] Suspected member: kapalk1:7800 (additional data: 20 bytes)
      2005-10-22 23:00:20,285 WARN [org.jgroups.protocols.pbcast.GMS - GMS.installView] checkSelfInclusion() failed, kapalk1:7800 (additional data: 20 bytes) is not a member of view [kapalk2:7800 (additional data: 20 bytes)|2] [kapalk2:7800 (additional data: 20 bytes)]; discarding view
      2005-10-22 23:00:20,286 WARN [org.jgroups.protocols.pbcast.GMS - GMS.installView] I (kapalk1:7800 (additional data: 20 bytes)) am being shunned, will leave and rejoin group (prev_members are [kapalk1:7800 (additional data: 20 bytes) kapalk2:7800 (additional data: 20 bytes) ])
      2005-10-22 23:00:22,376 WARN [org.jgroups.blocks.ConnectionTable - ConnectionTable$Connection.sendLocalAddress] local_addr is null


      The last warning message about the local_addr is null is repeating periodically fooding log files.

      It is our first cluster project and I have no idea what these warning messages mean.
      Can anybody help me?

      Applied database is Oracle 9.2. Operation system on JBoss computers is RedHat 9.

      Thanks,
      Peter

        • 1. Re: local_addr is null
          belaban

          You're using a TCP-based configuration. You could simply replace FD with FD_SOCK in cluster-service.xml, plus in all other clusters you're running