ConMan CMD_IDENTIFY_REQUEST Error

Question

I am running a 2 instance WLS5.1 SP10 cluster on one Solaris 7 box.
          After firing up the 2nd instance, I get the following error after about
          1 minute:
          Wed Nov 21 11:29:08 EST 2001:<I> <ConMan> CMD_IDENTIFY_REQUEST for this
          JVM has the wrong destination address
          -3666656829734979952S10.4.109.94:[7001,7001,7002,7002,7001,-1]
          Wed Nov 21 11:29:08 EST 2001:<I> <RJVM> Signaling peer
          -3666656829734979952S10.4.109.94:[7001,7001,7002,7002,7001,-1] gone:
          weblogic.rmi.ConnectException: PeerGone
          - with nested exception:
          [weblogic.rjvm.PeerGoneException: Peer requested connection shutdown]
          Wed Nov 21 11:29:08 EST 2001:<E> <ConMan> Attempt to sendMsg using a
          closed connection
          Wed Nov 21 11:29:08 EST 2001:<E> <RJVM> Exception on send :
          weblogic.rmi.ConnectException: Attempt to sendMsg using a closed
          connection
          Wed Nov 21 11:29:08 EST 2001:<I> <Cluster> Adding server
          -3666656829734979952S10.4.109.94:[7001,7001,7002,7002,7001,-1] to
          cluster view
          The same sort of error occurs when running a client against the app
          server. I've seen similar posts about this error, and they indicate the
          problem can occur when starting up an instance before it shuts down. In
          my case, I know the old instance has been shutdown before starting a new
          one.
          The error always occurs on instance 1. Regardless of whether or not I
          start 1 or 2 first, the error always shows up on instance 1.
          If only one instance is running, my test client works fine.
          Can somebody please help me? This is pretty frustrating. Thanks.
          Greg
          

Answer 1

This should not make a difference. Setting a weblogic specific property on
          the command line will take precedence over any other property within the
          properties files. So you will cascade through precedence as follows:
          command line
          per server properties file
          per cluster properties file
          global weblogic properties
          It could be that the command line property was not being set properly, it
          couldn't find anything in the properties for a bind address property, and on
          a multi-homed machine simply grabbed the first available IP.
          Patrick
          "Greg Crider" <[email protected]> wrote in message
          news:[email protected]...
          > I figured it out. In my startup script, I specified the value of
          > weblogic.system.bindAddr, but it didn't put it in the instance
          > weblogic.properties file. Even though it appeared to be set (after
          > examining weblogic.log), I added it to the weblogic.properties file as
          > well. Once I did this, it worked fine.
          >
          > Why did this make a difference?
          >
          > >
          > > I am running a 2 instance WLS5.1 SP10 cluster on one Solaris 7 box.
          > > After firing up the 2nd instance, I get the following error after about
          > > 1 minute:
          > >
          > > Wed Nov 21 11:29:08 EST 2001:<I> <ConMan> CMD_IDENTIFY_REQUEST for this
          > > JVM has the wrong destination address
          > > -3666656829734979952S10.4.109.94:[7001,7001,7002,7002,7001,-1]
          > > Wed Nov 21 11:29:08 EST 2001:<I> <RJVM> Signaling peer
          > > -3666656829734979952S10.4.109.94:[7001,7001,7002,7002,7001,-1] gone:
          > > weblogic.rmi.ConnectException: PeerGone
          > > - with nested exception:
          > > [weblogic.rjvm.PeerGoneException: Peer requested connection shutdown]
          > > Wed Nov 21 11:29:08 EST 2001:<E> <ConMan> Attempt to sendMsg using a
          > > closed connection
          > > Wed Nov 21 11:29:08 EST 2001:<E> <RJVM> Exception on send :
          > > weblogic.rmi.ConnectException: Attempt to sendMsg using a closed
          > > connection
          > > Wed Nov 21 11:29:08 EST 2001:<I> <Cluster> Adding server
          > > -3666656829734979952S10.4.109.94:[7001,7001,7002,7002,7001,-1] to
          > > cluster view
          > >
          > > The same sort of error occurs when running a client against the app
          > > server. I've seen similar posts about this error, and they indicate the
          > > problem can occur when starting up an instance before it shuts down. In
          > > my case, I know the old instance has been shutdown before starting a new
          > > one.
          > >
          > > The error always occurs on instance 1. Regardless of whether or not I
          > > start 1 or 2 first, the error always shows up on instance 1.
          > >
          > > If only one instance is running, my test client works fine.
          > >
          > > Can somebody please help me? This is pretty frustrating. Thanks.
          > >
          > > --
          > > Greg
          >
          > --
          > Greg
          >
          > GREGORY K. CRIDER, Emerging Digital Concepts
          > Systems Integration/Enterprise Solutions/Web & Telephony Integration
          > (e-mail) [email protected]
          > (web) http://www.EmergingDigital.com
          > (phone) 703-335-0974
          > (cell) 703-851-5073
          > (fax) 703-365-0223