IT27120: MFT Agent generates an FFDC from MonitorWork indicating that a NullPointerException occurred.

APAR status

• Closed as program error.

Error description

• An IBM MQ Managed File Transfer agent has been configured with a
  number of resource monitors. Each monitor polls a different
  directory looking for files, and submits managed transfer
  requests to the agent to move the files that they find.
  
  Intermittently, the agent generates an FFDC containing the
  following information:
  
  ****************************************************************
  ********
    Class:      com.ibm.wmqfte.monitor.management.MonitorWork
    Method:     generateFFDC
    Probe:      FFDC_005
    Cause:      java.lang.NullPointerException
    java.lang.NullPointerException
         at
  com.ibm.wmqfte.monitor.persist.impl.MonitorPersistStoreImpl.getM
  onitorMatcherData(MonitorPersistStoreImpl.java:116)
         at
  com.ibm.wmqfte.monitor.management.MonitorWork.execute(MonitorWor
  k.java:340)
         at
  com.ibm.wmqfte.monitor.management.MonitorImpl.run(MonitorImpl.ja
  va:147)
         at
  com.ibm.wmqfte.monitor.management.MonitorTimerTask.run(MonitorTi
  merTask.java:78)
         at java.util.TimerThread.mainLoop(Timer.java:566)
         at java.util.TimerThread.run(Timer.java:516)
  
  ****************************************************************
  ********
  
  After the FFDC had been generated, the monitors stop submitting
  managed transfer requests to the agent.
  

Local fix

• The agent needs to be stopped and restarted once the issue has
  occurred,
  

Problem summary

• ****************************************************************
  USERS AFFECTED:
  This issue affects users of IBM MQ Managed File Transfer, who
  have agents configured to use resource monitors.
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  The configuration for a resource monitor, along with its history
  (the list of items it has seen before), are stored in messages
  on the SYSTEM.FTE.EVENT.<agent name> queue hosted on the agent
  queue manager. Whenever a monitor performs a poll, it connects
  to the queue manager and gets the message containing its history
  from this queue. During the poll, the monitor updates the
  history contained within the message. At the end of the poll,
  the monitor puts the message back onto the queue and disconnects
  from the agent queue manager.
  
  The design of the agent means that all resource monitors will
  use a singleton class when accessing the messages on the
  SYSTEM.FTE.EVENT.<agent name> queue. The methods on the class
  are synchronized, which means that only one monitor can access
  the SYSTEM.FTE.EVENT.<agent name> queue at a time.
  
  Now, if an agent lost connectivity to its agent queue manager,
  it would start "recovery processing". This involved putting all
  active managed transfers into recovery and stopping all resource
  monitors.
  
  As part of this processing, the agent would remove some internal
  state related to the connection to the agent queue manager
  within the singleton class used by resource monitors. If the
  state was cleared just before a resource monitor started a poll,
  then when the monitor tried to use it to get messages from the
  SYSTEM.FTE.EVENT.<agent name> queue, a NullPointerException
  would occur. The agent would catch the NullPointerException, and
  generate an FFDC containing the following information:
  
  ****************************************************************
  ********
    Class:      com.ibm.wmqfte.monitor.management.MonitorWork
    Method:     generateFFDC
    Probe:      FFDC_005
    Cause:      java.lang.NullPointerException
    java.lang.NullPointerException
         at
  com.ibm.wmqfte.monitor.persist.impl.MonitorPersistStoreImpl.getM
  onitorMatcherData(MonitorPersistStoreImpl.java:116)
         at
  com.ibm.wmqfte.monitor.management.MonitorWork.execute(MonitorWor
  k.java:340)
         at
  com.ibm.wmqfte.monitor.management.MonitorImpl.run(MonitorImpl.ja
  va:147)
         at
  com.ibm.wmqfte.monitor.management.MonitorTimerTask.run(MonitorTi
  merTask.java:78)
         at java.util.TimerThread.mainLoop(Timer.java:566)
         at java.util.TimerThread.run(Timer.java:516)
  
  ****************************************************************
  ********
  
  The NullPointerException would cause the monitor to stop. If the
  agent subsequently reconnected to the agent queue manager, then
  the monitor would not be restarted and would no longer submit
  any managed transfer requests to the agent.
  

Problem conclusion

• To resolve this issue, logic has been added to the singleton
  class used by resource monitors to ensure that the internal
  state is valid, and that the singleton class is still connected
  to the agent queue manager, before trying to use it. If the
  internal state is null, indicating that the singleton class is
  no longer connected to the agent queue manager, then an internal
  exception will be thrown. This causes the agent to go into
  "recovery processing", if it is not already doing so.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v9.0 LTS   9.0.0.6
  v9.1 CD    9.1.2
  v9.1 LTS   9.1.0.2
  
  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 M/P

• Fixed component ID

  5724H7261

Applicable component levels