IT34591: Destination agent generates an ABEND file when processing a managed transfer that has timed out

APAR status

• Closed as program error.

Error description

• An IBM MQ Managed File Transfer agent is acting as the
  destination agent for a managed transfer that has been
  configured with a transfer recovery timeout. When processing the
  managed transfer, the agent generates an ABEND file containing
  the information shown below:
  
  ...
  Level: p910-006-200703
  ...
  Thread: 41 (ReceiverEventListenerTransferIdentifier)
  Class: com.ibm.wmqfte.thread.FTEThread
  ...
  Method: uncaughtException
  Probe: ABEND_001
  Cause: java.lang.NullPointerException
  java.lang.NullPointerException
    at
  com.ibm.wmqfte.transfer.eventlistener.ReceiverTransferEventListe
  ner$EventRunnable.run (ReceiverTransferEventListener.java:175)
    at java.lang.Thread.run(Thread.java:820)
    at com.ibm.wmqfte.thread.FTEThread.run(FTEThread.java:70)
  
  and then shuts itself down.
  

Local fix

Problem summary

• ****************************************************************
  USERS AFFECTED:
  This issue affects all users of IBM MQ Managed File Transfer who
  have agents that process managed transfers which have a transfer
  recovery timeout specified.
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  MQ 9.0.1 CD introduced some new functionality that allowed
  managed transfers that have been in recovery for longer than a
  specified period of time to be automatically cancelled. When
  this happens, the source agent for the managed transfer will
  send a message to the destination agent indicating that the
  managed transfer has been cancelled, and then mark the managed
  transfer as "Failed".
  
  Some more information about this functionality can be found in
  the Setting a timeout for recovery of stalled transfers topic in
  the MQ sections of IBM Knowledge Center. The URI for this topic
  in the MQ 9.2.x section of IBM Knowledge Center is
  https://www.ibm.com/support/knowledgecenter/en/SSFKSJ_9.2.0/com.
  ibm.mq.adm.doc/recovery_timeout.html
  
  
  Now, the following sequence of events led up to the issue
  reported in this APAR occurring:
  
  - A source agent received a managed transfer request, and the
  transfer recovery timeout functionality was enabled for that
  managed transfer.
  - As part of the processing of that managed transfer request,
  the source agent sent an initial transfer negotiation message to
  the destination agent, and waited for a response..
  - After waiting for the period of time specified by the agent
  property "maxTransferNegotiationTime", the source agent hadn't
  received the response from the destination agent and so put the
  managed transfer into recovery.
  - The source agent then continued to wait for the response from
  the destination agent.
  - After waiting for the period of time specified by the transfer
  recovery timeout, the source agent still hadn't received any
  response from the destination agent. As a result, it sent
  another message to the destination agent, indicating that the
  managed transfer had been cancelled, and marked the managed
  transfer as "Failed".
  
  - Some time later, the main CommandHandler thread within the
  destination agent picked up the initial transfer negotiation
  message from the source agent, registered the managed transfer
  with its state store and then started an internal thread to
  perform the negotiation.
  - Shortly afterwards, the destination agent's CommandHandler
  thread received the message indicating that the managed transfer
  had been cancelled and so removed details of it from its
  internal state store.
  - A few milliseconds later, the internal thread tried to get
  details about the negotiated parameters for the managed transfer
  from the state store, and use them during the negotiation
  process. However, the transfer was no longer present in the
  state store. As a result, the agent generated an ABEND
  containing the following information:
  
  Thread: 41 (ReceiverEventListenerTransferIdentifier)
  Class: com.ibm.wmqfte.thread.FTEThread
  ...
  Method: uncaughtException
  Probe: ABEND_001
  Cause: java.lang.NullPointerException
  java.lang.NullPointerException
    at
  com.ibm.wmqfte.transfer.eventlistener.ReceiverTransferEventListe
  ner$EventRunnable.run()
    at java.lang.Thread.run()
    at com.ibm.wmqfte.thread.FTEThread.run()
  
  and then shut itself down.
  

Problem conclusion

• To resolve this issue, IBM MQ Managed File Transfer has been
  updated so that:
  
  - If an internal thread within a destination agent is
  negotiating a managed transfer.
  - And the internal thread detects that the managed transfer is
  no longer in the agent's state store when it tries to access the
  negotiated parameters which should be used.
  
  then the internal thread will stop the negotiation process. This
  ensures that the managed transfer is stopped cleanly, and
  prevents the ABEND reported in this APAR from occurring.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v9.1 LTS   9.1.0.8
  v9.2 LTS   9.2.0.2
  v9.x CD    9.2.2
  
  The latest available MQ 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 MFT V9.1

• Fixed component ID

  5724H7272

Applicable component levels