IT17773: Queue manager becomes unresponsive eventually reporting log full errors: AMQ6709 and AMQ7463

IBM MQ for HP NonStop Server V8.1 fix pack 8.1.0.1

APAR status

• Closed as program error.

Error description

• Queue manager hangs during MQPUT processing on behalf of an
  application putting to a remote clustered queue.
  
  Eventually the checkpointer component also hangs, and so the
  whole queue manager becomes unresponsive and the following
  errors are generated:
  
    AMQ6709: The log for the queue manager is full
    AMQ7463: The log for queue manager EMPMIB1 is full
  
  These errors are seen soonest when the queue manager is doing
  larger quantities of persistent messaging, which consumes log
  space.
  
  FDCs also appear:
  
    KN749015   kqiDeregisterCLUSSDR
    XC307100   xlsRequestMutex
  
  If at this point the user ends the queue manager, the following
  FDCs also appear:
  
    PS058054   psiStopAllTasks
    PS058099   psiStopAllTasks
    XC307030   xlsRequestMutex
    AT030050   atmLockDataMutex
    AT036010   atmRecoverTransaction
    XC308010   xlsReleaseMutex
  

Local fix

Problem summary

• ****************************************************************
  USERS AFFECTED:
  Users of MQ where administrative changes are being made to
  remove cluster channels or switch which channel is in use for a
  cluster transmission queue, while the queue manager is trying to
  move messages over those channels.
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  There existed a small timing window within the queue manager
  MQPUT operation, in which it would notice a change to the
  channel definition, and retry some of the steps.  In doing this
  the code tried to obtain a lock already held by the thread, and
  this caused a permanent hang on that thread.
  
  Before retrying, the MQ code should have released the lock.
  
  The effect was wider than the original MQPUT, because the lock
  was then never released, and eventually the checkpointer
  component of the queue manager was not able to continue with its
  work.  This led to the queue manager's log becoming full.
  

Problem conclusion

• The MQ product code has been corrected so it will release the
  lock that ought to have been released.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v7.5       7.5.0.8
  v8.0       8.0.0.7
  v9.0 CD    9.0.2
  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

  WEBSPHERE MQ 7.

• Fixed component ID

  5724H7240

Applicable component levels