users@genericjmsra.java.net

Generic RA not reconnecting to Websphere MQ 5.3

From: Janardhanan, Ajith \(AJANARDH\) <"Janardhanan,>
Date: Thu, 5 Apr 2007 14:16:41 -0400

I have an MDB that reads messages from a IBM Websphere MQ Queue 5.3 and
writes to the disk. This MDB uses generic RA to connect to MQ. The
appserver is Glassfish 9.1 (build b33e-beta). MQ and appserver are on
the same physical machine. XATransaction is enabled.

 The JMSAdmin set up is:
  def xaqcf(MYQCF) transport(BIND) qmanager(MYHOST)
  def q(MyQueue) queue(MQUEUE.QUEUE) qmanager(MYHOST)

The deployment descriptor for MDB has these lines:
            <activation-config-property>
 
<activation-config-property-name>ReconnectAttempts</activation-config-pr
operty-name>
 
<activation-config-property-value>18000</activation-config-property-valu
e>
          </activation-config-property>
          <activation-config-property>
 
<activation-config-property-name>ReconnectInterval</activation-config-pr
operty-name>
 
<activation-config-property-value>30</activation-config-property-value>
          </activation-config-property>

When the MQ Queue manager goes down and then come back online; the MDB
reconnects only rarely. Most of the time the reconnect attempt is
failed. The appserv logfile shows this:

[#|2007-04-05T17:48:03.543+0000|INFO|sun-appserver9.1|com.sun.genericjms
ra|_ThreadID=31;_ThreadName=MQQueueAgent (MyMDB);|Reconnect
attempt->62|#]

[#|2007-04-05T17:48:03.638+0000|INFO|sun-appserver9.1|com.sun.genericjms
ra|_ThreadID=31;_ThreadName=MQQueueAgent (MyMDB);|Reconnect attempt
failed.Now sleeping for30000|#]


Any input to get it to work is greately appreciated.

Thanks
Ajith