IT15913: Queue manager becomes unresponsive error log reports AMQ7463 queue manager log is full

APAR status

• Closed as program error.

Error description

• The queue manager is unresponsive to runmqsc commands and is
  unable to process messages.
  
  An FDC is generated with the following information:
  
  Probe Id          :- HL010004
  Component         :- hlgReserveLogSpace
  Program Name      :- amqzlaa0
  Major Errorcode   :- hrcE_LOG_FULL
  Probe Description :- AMQ6709: The log for the queue manager is
  full.
  .
  MQM Function Stack
  zlaMainThread
  zlaProcessMessage
  zlaProcessMQIRequest
  zlaMQPUT
  zsqMQPUT
  kpiMQPUT
  kqiPutIt
  kqiPutMsgSegments
  apiPutMessage
  aqmPutMessage
  aqhPutMessage
  aqqWriteMsg
  aqqWriteMsgData
  aqlReservePutSpace
  almReserveSpace
  hlgReserveLogSpace
  xcsFFST
  
  
  The Queue manager error log reports:
  
  AMQ7463: The log for queue manager 'QueueManagerName' is full.
  
  After the queue manager is forcibly shut down, the queue
  manager starts but takes an unexpectedly long time to replay
  the queue manager transaction log records.
  
  No explanation for this slowdown is observed in the MQ error
  logs or FDC files prior to the AMQ6709 message being issued.
  

Local fix

Problem summary

• ****************************************************************
  USERS AFFECTED:
  Users of IBM MQ version 8 or 9
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  The amqzmuc0 MQ process performs checkpoint processing activity
  to ensure efficient space utiliztation of the log. If this
  thread hangs, the transaction log fills. The time it takes to
  fill depends on the log size and transaction workload, but
  without the checkpoint process, it will eventually fill and the
  queue manager will stop.
  
  If this happens an FDC is produced with the following
  characteristics:
  Major Errorcode   :- hrcE_LOG_FULL
  Probe Description :- AMQ6709: The log for the queue manager is
  full.
  
  MQ is designed to monitor the health of the checkpoint process,
  however a coding error meant that this monitoring was not
  performing correctly, and as such no warning was issued to the
  user that the checkpoint process had terminated prior to the log
  becoming full.
  

Problem conclusion

• The MQ queue manager code has been updated to correctly monitor
  the health of the checkpoint process. If no activity from the
  checkpoint process is observed for a period of 5 minutes, then
  an FDC will be cut with the following expected characteristics:
  
  	Probe Id          :- AD163010
  	Major Errorcode   :- MQRC_UNEXPECTED_ERROR
  	Probe Description :- AMQ6125: An internal WebSphere MQ
  error has occurred.
  	Comment1          :- Unexpected delay in checkpoint thread
  
  The error log will also include an entry:
  	AMQ6125: An internal WebSphere MQ error has occurred.
  
  If this FDC is seen from a running queue manager, the queue
  manager should be quiesced and restarted in order to restart the
  checkpoint process.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v8.0       8.0.0.6
  v9.0 CD    9.0.1
  v9.0 LTS   9.0.0.1
  
  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

  WMQ BASE MULTIP

• Fixed component ID

  5724H7251

Applicable component levels

• R800 PSY

     UP