IT44104: IBM MQ 9.3 queue manager might crash when using QALIAS of cluster queue as STREAMQ

APAR status

• Closed as program error.

Error description

• When an application puts messages into a local queue which has
  the STREAMQ set to a queue alias with cluster queue as target,
  the amqrrmfa process enters in a loop and crashes.
  
  Example configuration:
  QM1:
  DEFINE QLOCAL('CLUSTERQ') CLUSTER('CLT1')
  
  QM2:
  DEFINE QALIAS('CLUSTERQ') TARGET('CLUSTERQ') TARGTYPE(QUEUE)
  DEFINE QLOCAL('Q1')  STREAMQ('CLUSTERQ')
  
  When the issue occurs, the following FDCs are continuously
  generated:
  AMQ85977.0.FDC  2023/06/16 06:25:00 amqzxma0 QMgrHealth
  ZX005022 zxcProcessChildren      lrcW_S_FAST_PATH_APP_DEAD
  AMQ86195.0.FDC  2023/06/16 06:25:00 amqzlaa0 SharedAgent
  XC307030 xlsRequestMutex         xecL_W_SEM_OWNER_DIED
  AMQ85977.0.FDC  2023/06/16 06:25:00 amqzxma0 QMgrHealth
  ZX005025 zxcProcessChildren      zrcX_PROCESS_MISSING
  AMQ85977.0.FDC  2023/06/16 06:25:00 amqzxma0 QMgrHealth
  ZX028090 zxcHealthThread         STOP
  
  MQ trace for the affected process likely to show the following:
  
  06:27:22.974328   779451.1       CONN:000002 ------------{
  kpiGeneric
  06:27:22.974338   779451.1       CONN:000002 -------------{
  kpiStreamQUpdateChangeCount
  06:27:22.974341   779451.1       CONN:000002 --------------{
  kqiStreamQTableLock
  06:27:23.190210   779451.1       CONN:000002
  --------------------------------------------------{
  aouLockSynChain
  06:27:23.190211   779451.1       CONN:000002 !!! Invalid nesting
  count 51 !!! xlsRequestMutex
  06:27:23.190212   779451.1       CONN:000002      MtxName:
  hmtxSynChain_1_0 Id: 2066
  06:27:23.190214   779451.1       CONN:000002      MtxName:
  hmtxSynChain_1_0 Id: 2066
  06:27:23.190215   779451.1       CONN:000002 !!! Invalid nesting
  count 51 !!! xlsRequestMutex rc=OK FunctionTime=4
  06:27:23.190216   779451.1       CONN:000002
  --------------------------------------------------}
  aouLockSynChain rc=OK FunctionTime=6
  06:27:23.190255   779451.1       CONN:000002
  --------------------------------------------------{
  aouUnlockSynChain
  06:27:23.190256   779451.1       CONN:000002 !!! Invalid nesting
  count 51 !!! xlsReleaseMutex
  06:27:23.190257   779451.1       CONN:000002      MtxName:
  hmtxSynChain_1_3 Id: 2069
  06:27:23.190259   779451.1       CONN:000002 !!! Invalid nesting
  count 51 !!! xlsReleaseMutex rc=OK FunctionTime=3
  06:27:23.190260   779451.1       CONN:000002
  --------------------------------------------------}
  aouUnlockSynChain rc=OK FunctionTime=5
  06:27:23.527093   779451.1       CONN:000002 !!! Invalid nesting
  count 101 !!! xlsRequestMutex rc=OK FunctionTime=4
  06:27:23.527094   779451.1       CONN:000002 !!! Invalid nesting
  count 101 !!! xlsReleaseMutex
  06:27:23.527095   779451.1       CONN:000002      MtxName:
  qcBlock_2_8_8_9984 Id: 3059
  06:27:23.527096   779451.1       CONN:000002 !!! Invalid nesting
  count 101 !!! xlsReleaseMutex rc=OK FunctionTime=2
  06:27:23.527098   779451.1       CONN:000002 !!! Invalid nesting
  count 101 !!! kqiStreamQUpdateChangeCount
  

Local fix

• If above symptoms match, it is recommended to remove the STREAMQ
  attribute from the local queue.
  

Problem summary

• ****************************************************************
  USERS AFFECTED:
  All Users who try to define STREAMQ attribute set to a queue
  alias where the target queue is a cluster queue.
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  When we use an alias streaming queue name then MQ needs to
  additionally update an internal counter for the alias target
  queue. The update of the count eris not able to cope properly
  with an alias queue having to target itself which leads amqrrmfa
  process to enter a loop and crashes.
  

Problem conclusion

• The code has been modified to update the queue's internal
  counter if its streaming queue is equivalent to the updated
  streaming queue.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v9.3 LTS   9.3.0.15
  v9.x CD    9.3.4
  
  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

  MQ APPLIANCE M2

• Fixed component ID

  5900ALJ00

Applicable component levels