PK83434: Updates to the WebSphere MQ messaging provider, and interoperation between WebSphere MQ and the default messaging

Fixes are available

APAR status

Closed as program error.

Error description

Fixes to the WebSphere MQ messaging provider, and
interoperation between the default messaging provider for
WebSphere Application Server and WebSphere MQ

Local fix

Problem summary

****************************************************************
* USERS AFFECTED:  Users of IBM WebSphere Application Server   *
*                  Version 7.0 in combination with WebSphere MQ*
****************************************************************
* PROBLEM DESCRIPTION: Problems using the WebSphere MQ         *
*                      messaging provider, and the             *
*                      WebSphere MQ link or                    *
*                      WebSphere MQ Server functionality       *
*                      of the default messaging provider       *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
The following issues are addressed in this APAR that might be
seen when using WebSphere MQ in combination with
WebSphere Application Server Version 7.0.

Problems with the WebSphere MQ messaging provider:
- Once a value has been specified in the "Queue manager or
  Queue sharing group name" property of a JMS queue for the
  WebSphere MQ messaging provider, the property cannot be
  set back to an empty value.
  If an empty value is entered and OK is pressed, then the
  original value is restored when you next view the queue.
- The default value for the "Subscription wildcard format"
  property of a connection factory for the WebSphere MQ
  messaging provider is incorrect. The default is "Character
  wildcards". The default should be "Topic level wildcards".
- Inconsistent transactional behavior is seen for
  a message-driven bean application on z/OS using
  bean-managed transactions (or setting the NotSupported
  transaction attribute on the onMessage method) consuming
  messages from a WebSphere MQ queue using an activation
  specification with a "Bindings" or "Bindings, then client"
  Transport.
  Messages passed to the MDB's onMessage might be rolled back
  to the WebSphere MQ queue even though the onMessage method
  completed successfully. Exceptions may also be seen in the
  joblogs of the application server.
- When WebSphere MQ trace is enabled in an application server,
  there is a small possibility of a deadlock between
  a thread containing the following stack:
  "com/ibm/msg/client/commonservices/trace/Trace.setOn"
  and other threads calling WebSphere MQ trace or FFDC methods.
- If a WebSphere MQ queue manager is stopped and restarted
  while messages are being processed by a message-driven
  bean application deployed against an activation specification
  on z/OS, consumption of messages might not automatically
  resume after the queue manager becomes available again.
- The "Append RFH version 2 headers to messages sent to this
  destination" check-box for a JMS Queue in the WebSphere MQ
  messaging provider is always checked, even if it has
  previously been unchecked and the configuration was saved.
  However, the runtime operation of the JMS Queue correctly
  reflects the value of the check-box when it was last saved.
- JMS Topic objects in the WebSphere MQ messaging provider
  incorrectly have an "As connection" option for the
  following properties:
  "Broker durable subscription queue"
  "Broker durable subscriber connection consumer queue"
  "Broker publication queue manager"
  These properties do not exist on either the connection
  factory or topic connection factory.
  When "As connection" is specified, the default
  value described in the WebSphere MQ information center
  is used at runtime.
- After enabling SSL security on a connection factory
  in the WebSphere MQ messaging provider on z/OS, a message
  listener port fails to start with the following exception
  java.io.NotSerializableException:
              com.ibm.websphere.ssl.protocol.SSLSocketFactory
- A deadlock might be observed for a message-driven bean
  listening to a WebSphere MQ queue or topic via an
  activation specification.
  The stack of one of the threads that is deadlocked contains
  the following method:
   com/ibm/ejs/jms/JMSDiagnosticModule.ffdcDumpDefaultDefault
  The deadlock is most likely to occur after pausing and
  resuming a message endpoint.
- The runtime component for the WebSphere MQ messaging
  provider is incorrectly started in node agents.
  This introduces an unnecessary overhead, and might result
  in FFDCs being logged.
- A message-driven bean application might stop consuming
  messages after a connection error to the queue manager occurs.
  An exception similar to the following is logged:
  J2CA0079E: Method cleanup has detected an internal illegal
   state and is throwing an IllegalStateException. The
   exception is: java.lang.IllegalStateException: cleanup:
   illegal state exception. State = STATE_INACTIVE MCW
  This is due to an attempt to invalidate a connection in the
  connection pool from a JMS exception listener thread.

Problems with the default messaging provider interoperating
with WebSphere MQ:
- An application consuming messages from a bus
  destination, that were received over a WebSphere MQ link
  from a queue manager, might stop receiving messages.
  FFDCs are logged containing exceptions with WebSphere MQ
  reason code 2298 (MQRC_FUNCTION_NOT_SUPPORTED) and
  2120 (MQRC_CONVERTED_MSG_TOO_BIG)
- When stopping a messaging engine to which applications
  connect and consume messages from a WebSphere MQ queue
  via a WebSphere MQ Server bus member destination,
  FFDCs might be logged containing RMQCursorException,
  SISessionUnavailableException, RMQResourceException and
  SIMPErrorException exceptions, from
  RMQCursor.cancel(), RMQCursor.next(), SibRaListener.stop(),
  RMQConsumerKey.close(), SibRaListener.close(),
  RMQAsynchThread.AsynchRunnable.runAsynchConsumer() or
  RMQAsynchThread.AsynchRunnable.run().

Problem conclusion

The fix for this APAR addresses the problems described
above in the WebSphere MQ messaging provider, and the default
messaging provider.

The fix for this APAR also updates the WebSphere MQ
Resource Adapter shipped with WebSphere Application Server to
k000-L090615.1

The APAR is targeted for delivery in WebSphere Application
Server fix pack 7.0.0.5.

Temporary fix

Comments

APAR Information

APAR number
PK83434
Reported component name
PLAT MSG COM
Reported component ID
620800101
Reported release
300
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt
Submitted date
2009-03-26
Closed date
2009-06-23
Last modified date
2009-06-23

APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:

Fix information

Fixed component name
PLAT MSG COM
Fixed component ID
620800101

Applicable component levels

R300 PSY
UP

[{"Business Unit":{"code":"BU053","label":"Cloud & Data Platform"},"Product":{"code":"SSEQTP","label":"WebSphere Application Server"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"7.0","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
24 October 2021

Tips

PK83434: Updates to the WebSphere MQ messaging provider, and interoperation between WebSphere MQ and the default messaging

Fixes are available

Subscribe

APAR status

Closed as program error.

Error description

Local fix

Problem summary

Problem conclusion

Temporary fix

Comments

APAR Information

APAR number

Reported component name

Reported component ID

Reported release

Status

PE

HIPER

Special Attention

Submitted date

Closed date

Last modified date

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:

Fix information

Fixed component name

Fixed component ID

Applicable component levels

R300 PSY

Document Information

Share your feedback

Need support?