IT39632: Queue manager fails to start due to amqzmuc0 crash while generating FDC KN467078 in ReconcileDurableSubscribers

APAR status

• Closed as program error.

Error description

• Queue manager fails to start due to amqzmuc0 ending abruptly
  because of memory exception(SIGSEGV) when generating a failure
  data capture (FDC) record with probe Id KN467078 in component
  kqiReconcileDurableSubscribers. The following FDC records were
  generated during the failure:
  
  amqzmuc0  44592 23 KN467078 kqiReconcileDurableSubscribers
  krcE_UNEXPECTED_ERROR
  amqzxma0   44581  1 ZX005020 zxcProcessChildren
  zrcX_PROCESS_MISSING
  amqzxma0  44581  1 ZX000040 zxcExecutionController
  STOP_ALL
  
  Probe Id          :- KN467078
  Component         :- kqiReconcileDurableSubscribers
  Program Name      :- amqzmuc0
  Process           :- 44592
  Process(Thread)   :- 44666
  Thread            :- 23    DurSubsMgr
  Major Errorcode   :- krcE_UNEXPECTED_ERROR
  Probe Type        :- INCORROUT
  Probe Description :- AMQ6125E: An internal IBM MQ error has
  occurred
  FDCSequenceNumber :- 0
  
  MQM Function Stack
  zmuThreadMain
  zmuDurableSubscriptionManagerTask
  kpiDurableSubscriptionManagerTask
  kqiReconcileDurableSubscribers
  xcsFFST
  
  Probe Id          :- ZX005020
  Component         :- zxcProcessChildren
  Program Name      :- amqzxma0
  Process           :- 44581
  Process(Thread)   :- 44581
  Thread            :- 1    ECMain
  Probe Description :- AMQ5008S: An essential IBM MQ process 44592
  (zllCRIT)
     cannot be found and is assumed to be terminated.
  FDCSequenceNumber :- 0
  Arith1            :- 44592 (0xae30)
  Comment1    :- zllCRIT
  
  MQM Function Stack
  amqzxma0
  zxcExecutionController
  zxcStartupLogic
  zxcProcessChildren
  xcsFFST
  
  Probe Id          :- ZX000040
  Component         :- zxcExecutionController
  Program Name      :- amqzxma0
  Process           :- 44581
  Process(Thread)   :- 44581
  Thread            :- 1    ECMain
  Major Errorcode   :- STOP_ALL
  Minor Errorcode   :- OK
  Probe Type        :- HALT6110
  Probe Severity    :- 1
  Probe Description :- AMQ6110S: An internal IBM MQ error has
  occurred.
  FDCSequenceNumber :- 1
  
  MQM Function Stack
  amqzxma0
  zxcExecutionController
  xcsFFST
  

Local fix

• Workaround 1:
  
  Note: This procedure affects existing subscriptions, some
  subscriptions might be lost and
             some old subscriptions might be restored.
  
  * Backup the queue file 'SYSTEM!DURABLE!SUBSCRIBER!QUEUE'
  * Replace the queue file 'SYSTEM!DURABLE!SUBSCRIBER!QUEUE' from
  queue manager backup
  * Start the QM
  * Delete/Create/Modify the subscriptions as required
  
  Workaround 2:
  
  Note: This procedure deletes all existing subscriptions and the
  user needs to newly create the subscriptions
  
  * Replace the queue file 'SYSTEM!DURABLE!SUBSCRIBER!QUEUE' from
  a newly created temporary queue manager
  * Start the QM
  * Create subscriptions as required
  

Problem summary

• ****************************************************************
  USERS AFFECTED:
  Users using MQ publish/subscribe messaging
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  The system queue SYSTEM.DURABLE.SUBSCRIBER.QUEUE stores
  persistent copy of the durable subscriptions in the queue
  manager. When the QM starts, the QM reads and processes the
  messages and recreates the subscriptions.
  During this processing, the QM detected an invalid subscription
  type from the message read and the MQ generated a FDC. However
  when dumping the subscription details in the FDC the QM process
  amqzmuc0 encountered a memory exception due to a defect in code
  that dumps the subscription in question. This resulted in
  failure of critical MQ process amqmuc0 and QM startup.
  

Problem conclusion

• MQ code has been modified to prevent the memory exception.
  
  The fix only prevents the memory exception and allows the QM to
  continue the startup.
  The publish/subscribe feature might be disabled due to the bad
  subscription message that triggers the issue.
  The user needs to recreate the subscriptions as suggested in the
  workaround section to
  resolve the publish/subscribe issue.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v9.0 LTS   9.0.0.13
  v9.1 LTS   9.1.0.11
  v9.2 LTS   9.2.0.6
  v9.x CD    9.2.5
  
  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

  5724H7271

Applicable component levels