PM84130: MQ CICS BRIDGE BOUNCING AFTER MQGET FOR BRIDGE QUEUE RETURNS MQRC=2063

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• The error occurs if the MQ CICS bridge queue is protected by
  WMQ AMS (Advanced Message Security) and CKBR is triggered
  from the bridge queue.
  If privacy (encryption) is requested for the MQ CICS bridge
  queue the user ID which runs the CICS Bridge Monitor task
  (usually transaction CKBR) needs to be added to the WMQ AMS
  policy  "list of recipients" so it is capable of reading
  (browsing) messages.
  If this has not been configured this way MQGET against
  the MQ CICS bridge queue will fail with MQRC=2063 (error
  code introduced by WMQ AMS).
  WMQ AMS normally would move the message under access
  to the SYSTEM.
  PROTECTION.ERROR.QUEUE but CICS "wins" performing
  a backout of the message- i.e. putting it "back" to the
  MQ CICS bridge queue.
  This results in an endlessly bouncing process of
  - (re)starting the CICS bridge monitor transaction (CKBR)
  - attempting to READ (browse) messages  by CKBR, failing
    with MQRC=2063
  - backout of the message
  
  These are the system log messages of the situation :
  
  +DFHMQ0702 CICS11 CKBR 00094 CICS-MQ bridge monitor
             initialization complete.
  +DFHMQ0703 CICS11 CKBR 00094 WaitInterval=20,000,
             Auth=IDENTIFY Q=TEST.FILEA.CICS.BRIDGE.
  +DFHMQ0783 CICS11 CKBR 00094 Msg=BOTH, PassTktA=CICS11.
  +DFHMQ0792 CICS11 CKBR 00094 RouteMEM=N
  DRQZS0216E (AMREIN) Data unprotection failed processing
   object "pkcs7 enveloped data message", return code 12,
   reason 03353033. Additional information:
   function gsk_read_enveloped_data_content failed with
   x3353033, 00000008.
  +DFHMQ0710 CICS11 CKBR 00094 MQCC=2 MQRC=2063 returned
             for GET.
  DRQZS0209I (AMREIN) Message for TEST.FILEA.CICS.BRIDGE
   sent to error queue SYSTEM.PROTECTION.ERROR.QUEUE,
   reason 2063.
  
  However, the message is *no*t moved to the ERROR queue as
  stated by msgDRQZ0209I but what "really" happens is shown
  by the CICS message log (MSGUSR)
  
  DFHMQ0710 E 08/20/2012 14:01:34 CICS11 CKBR 00094 MQCC=2
   MQRC=2063 returned for GET
  DFHDU0203I 08/20/2012 14:01:34 CICS11 A transaction dump
   was taken for dumpcode: AMQI, Dumpid: 3/0003.
  DFHAC2236 08/20/2012 14:01:34 CICS11 Transaction CKBR
   abend AMQI in program DFHMQBR0 term ????. Updates to local
   recoverable  resources will be backed out.
  
  So, CICS preforms a backout of the message, which means it is
  put back to the bridge queue.
  And if TRIGGER TRIGTYPE(FIRST) is set for the bridge queue
  it will result in an immediate restart of the monitor
  transction(CKBR) with the same processing taking place
  over and over again.
  
  
  The MQ CICS bridge (CKBR code) has to changed to handle
  WMQ AMS error situations correctly.
  

Local fix

• *
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All.                                         *
  ****************************************************************
  * PROBLEM DESCRIPTION: CICS-WMQ Bridge can loop reading a      *
  *                      message from the bridge queue.          *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  With the introduction of WebSphere MQ AMS (Advanced Message
  Security) it is possible for the CICS-WMQ Bridge code to get
  into a loop reading a message from the bridge queue. This loop
  can occur when the WebSphere MQ AMS policy has not been setup to
  handle the userid which the CICS-WMQ bridge transaction runs
  under. WebSphere MQ AMS will pass an MQRC of 2063 back to the
  CICS-WMQ Bridge code, but the bridge code does not handle this
  specific MQRC and leaves the message on the bridge queue. If the
  bridge queue is defined with a TRIGGER TRIGTYPE(FIRST) then this
  sequence is repeated.
  
  Keywords: msgDRQZS0216E DRQZ0209I msgDFHMQ0710 MQ0710
  

Problem conclusion

• DFHMQBR0 has been changed to specifically handle an MQRC of
  2063.
  
  The CICS Transaction Server for z/OS V5 R1 CICS Messages and
  Codes Vol. 1 (GC34-2861-00) has had the new message added on
  page 1071 following message DFHMQ0725:
  
  DFHMQ0726 I date time applid tranid trannum
              CICS-WMQ Bridge message incompatible with WebSphere
              MQ AMS configuration - Bridge terminated.
  Explanation: WebSphere MQ Advanced Message Security is being
  used to protect WebSphere MQ queues, and when a CICS
  application attempted to access the queue, message DRQZS0209I
  reason 2063 was issued from Websphere MQ AMS.
  
  System action: CICS-WMQ bridge is terminated.
  
  User Response: Investigate WebSphere MQ AMS documentation for
  the cause of the 2063 reason code.
  
  Module: DFHMQBR0
  
  XMEOUT Parameters: date,time, applid, tranid, trannum
  
  Destination: Console and Transient Data Queue
  CMQM and Terminal End User.
  

Temporary fix

• FIX AVAILABLE BY PTF ONLY
  

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

  PM71673

• APAR is sysrouted TO one or more of the following:

  UK96157 UK96158

Modules/Macros

• DFHMQP0@ DFHMQP1@ DFHMQP3@ DFHMQR0@
  

Fix information

• Fixed component name

  CICS TS Z/OS V5

• Fixed component ID

  5655Y0400

Applicable component levels

• R800 PSY UK96157

     UP13/08/02 P F308

• R803 PSY UK96158

     UP13/08/02 P F308

Fix is available

• Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.