IT29370: AMQP channel ends unexpectedly due to NullPointerException

APAR status

• Closed as program error.

Error description

• The AMQP channel ends with an error in the log:
  
  AMQCO2001E: An error (probe: XR014005) occurred and
  a Failure Data Capture (FDC) file has been written to file
  '/var/mqm/errors/AMQP0.FDC'.
  AMQCO1006I: Channel 'SYSTEM.DEF.AMQP' has stopped
  
  and an FDC generated with details:
  
   Probe Id          :- XR014005
  
  java.lang.NullPointerException
   at
  com.ibm.mq.MQXRService.MQConnection.jmqiGet(MQConnection.java:10
  57)
   at
  com.ibm.mq.MQXRService.WorkerContext.browseMessageFromTransmitQu
  eue(WorkerContext.java:563)
   at
  com.ibm.mq.MQXRService.OutboundInFlightMessage.getMqTransmitQueu
  eMessage(OutboundInFlightMessage.java:180)
   at
  com.ibm.mq.MQXRService.AMQPSubscriptionWrapper.captureBufferedMe
  ssages(AMQPSubscriptionWrapper.java:327)
   at
  com.ibm.mq.MQXRService.AMQPServerSessionV10.sendMessageBacklog(A
  MQPServerSessionV10.java:2024)
   at
  com.ibm.mq.MQXRService.MQTTServerSession.prepareAsynchronousWork
  (MQTTServerSession.java:912)
   at
  com.ibm.mq.MQXRService.AMQPServerWireContext.prepareAsynchronous
  Work(AMQPServerWireContext.java:198)
   at
  com.ibm.mq.communications.NonBlockingWorker.synchronousWorkCompl
  eted(NonBlockingWorker.java:615)
   at
  com.ibm.mq.communications.NonBlockingWorker.run(NonBlockingWorke
  r.java:422)
   at java.lang.Thread.run(Thread.java:809)
  

Local fix

Problem summary

• ****************************************************************
  USERS AFFECTED:
  This issue affects users of the IBM MQ AMQP service, where the
  service experiences very high JVM heap utilisation while
  messages are being sent to subscribing clients.
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  When a heavily-loaded v8.0 or v9.0 AMQP service approaches the
  limit of available heap memory, it can preemptively close some
  client connections and associated internal objects in order to
  free up space. This avoids it from hitting a serious runtime
  error (such as a java.lang.OutOfMemoryError) that could result
  in the AMQP service ending unexpectedly.
  
  There was a small timing window which meant that if the internal
  objects were closed at the same time as the AMQP service was
  getting a message from the transmit queue to send to an AMQP
  client, then a method was called on a null object. This resulted
  in a NullPointerException to be generated, which caused the AMQP
  service to stop unexpectedly.
  

Problem conclusion

• Additional checks have been added to the IBM MQ AMQP service to
  ensure that internal calls are not made on objects that could be
  null at any point. These checks prevent NullPointerExceptions
  from occurring if client connections and internal objects are
  being closed to free up memory at the same time as messages for
  those clients are being retrieved from the queue manager.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v8.0       8.0.0.13
  v9.0 LTS   9.0.0.8
  
  
  The latest available maintenance can be obtained from
  'WebSphere MQ Recommended Fixes'
  http://www-1.ibm.com/support/docview.wss?rs=171&uid=swg27006037
  
  If the maintenance level is not yet available information on
  its planned availability can be found in 'WebSphere MQ
  Planned Maintenance Release Dates'
  http://www-1.ibm.com/support/docview.wss?rs=171&uid=swg27006309
  ---------------------------------------------------------------
  

Temporary fix

Comments

APAR Information

• 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

  IBM MQ BASE MP

• Fixed component ID

  5724H7251

Applicable component levels