PM80577: WHEN RUNNING A PRE-V710 VERSION OF CSQBRRSI WITH V710 VERSION SCSQAUTH, AN MQRC 2002 CAN OCCUR EVEN AFTER MQDISC.

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• The customer is running a batch job was running fine until
  they upgraded their early code and SCSQAUTH to V710. The batch
  job in question is issuing MQCONN, followed by some MQPUTs /
  PUT1 and then a MQDISC. It then attempts another MQCONN but
  unexpectedly fails with a 2002 MQRC_ALREADY_CONNECTED.
  .
  The change team looked at the code and can see the cause of
  the problem. The problem occurs when an application linked
  with a pre-V710 version of CSQBRRSI is run with a V710 version
  of the SCSQAUTH dataset (so that the application is using the
  V710 version of CSQBCON, CSQBDSC etc). The offset of the
  fRRSDisc flag in the BLOA has changed between V701 and V710,
  hence the MQRC 2002 may occur even though an MQDISC has been
  issued.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of WebSphere MQ for z/OS Version 7 *
  *                 Release 1 Modification 0.                    *
  ****************************************************************
  * PROBLEM DESCRIPTION: An application linked using CSQBRRSI or *
  *                      CSQBRSTB at a release prior to 710      *
  *                      receives MQRC_ALREADY_CONNECTED         *
  *                      (MQRC2002) when using MQCONN to         *
  *                      reconnect to the queue manager, even    *
  *                      if MQDISC was previously issued.        *
  *                      If the application was linked at        *
  *                      701 with the fix for PM70746 applied,   *
  *                      and MQCONN is called to reconnect to    *
  *                      the queue manager, it receives          *
  *                      MQRC_ENVIRONMENT_ERROR (MQRC2012).      *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  When an application linked with an RRS stub calls MQDISC to
  disconnect from the queue manager, a flag is set in the bloa to
  indicate this.
  An optimisation in the stubs uses this flag to check if the
  application has disconnected without the overhead of loading
  CSQBCON.
  Between 701 and 710 the offset of this flag within the BLOA was
  changed, causing this optimisation to incorrectly flag that
  the application is still connected when the application was
  compiled with 701 stub code, and connects to the queue manager
  using 710 adapter code.
  
  Additional Keywords: 2002 2012
  

Problem conclusion

• The adapter code is changed to set the flags required for the
  optimisation at the correct offsets for both the 701 and 710 RRS
  stubs.
  100Y
  CSQBCON
  CSQBDSC
  CSQBRRSI
  CSQBRSTB
  CSQBSRV
  CSQBTEXT
  CSQMEOTC
  CSQXACON
  CSQXDSC
  IMQS23DR
  

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:

  UK92602

Modules/Macros

•    CSQBCON  CSQBDSC  CSQBRRSI CSQBRSTB CSQBSRV
  CSQBTEXT CSQMEOTC CSQXACON CSQXDSC  IMQS23DR
  

Fix information

• Fixed component name

  WMQ Z/OS V7

• Fixed component ID

  5655R3600

Applicable component levels

• R100 PSY UK92602

     UP13/04/04 P F304

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.