PM44669: DFHSM0102 STORAGE VIOLATION CODE X'0D11' DFHWUIPG

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• Attempting to install a CPSM Resource Defintion Group
  into a collection of CICS regions. The resource being installed
  was a library definition which was 'invalid'.  The failure
  then produced a storage violation in CICS.
  .
  The trace entry for the error is:
  ---------------------------------------------------------------
  SM 0D11 SMMF  *EXC* - Storage_check_failed_on_freemain_request -
  FUNCTION(FREEMAIN)
  ...
  3-0000  C3F0F0F2 F3F1F3F9 C6C5C5C4 C2C1C3D2
  C6C5C5C4 C2C1C3D2 00000000 00070003
  *C0023139FEEDBACKFEEDBACK........*
  0020  00000039 00000002 C2C9D5E2 E3C5D9D9
  00000000 00000000 00000000 D9C5E2C7
  *........BINSTERR............RESG*
  ....
  4-0000  00000000 00000000 C2C9D5E2 E3C5D9D9
  *........BINSTERR                *
  ---------------------------------------------------------------
  .
  The overlaid storage starts has more entries than would fit in
  the storage that was GETMAINed for them.
  A couple of the entries look like this:
  ---------------------------------------------------------------
  398411B0 C2C9D5E2 E3C5D9D9 C2C9D5E2 E3C5D9D9 |BINSTERRBINSTERR|
  398411C0 C3C9C1F1 C1C1F0F2 00000000 00000000 |CIA1AA02........|
  398411D0 C3E2E3C7 D5C5D9C3 01000000 09000000 |CSTGNERC........|
  398411E0.:3984127F.--All bytes contain X'00'
  39841280 00000000 C2C9D5E2 E3C5D9D9 C2C9D5E2 |....BINSTERRBINS|
  39841290 E3C5D9D9 C3C9C1F1 C1C1F0F3 00000000 |TERRCIA1AA03....|
  398412A0 00000000 C3E2E3C7 D5C5D9C3 01000000 |....CSTGNERC....|
  ---------------------------------------------------------------
  That is:
  x'30' bytes of data (d' 48') for the first bit of data
  x'A4' bytes of data (d'160') for the bytes of x'00'
  = x'D4' bytes         (d'212') per entry
  With x'39' "BINSTERR" entries that equals x'2F34' bytes of data.
  Backing up in the trace, data was GETMAINed for x'1A80' bytes.
  So, the number of entries far exceeds the capacity to hold them.
  .
  Mixed with the CICS traces were x'39' (decimal 57) BINSTERR
  entries appear to come from.  Here are a selection of trace
  entries starting at the "Restful" process that has the problem
  ---------------------------------------------------------------
  WUEYU ENTRY Restful CPSM Request RESGROUP,CICSGRP1,CICSGRP1,...
  WUEYU DATA  Input parms RESGROUP,CICSGRP1,CICSGRP1,NO,INSTALL
  WUEYU DATA  entryPerformAction RESGROUP,CICSGRP1,CICSGRP1,...
  ERM   ENTRY ASSEMBLER-APPLICATION-CALL-TO-TRUE(EYU9XLAP)
  ERM   EXIT  ASSEMBLER-APPLICATION-CALL-TO-TRUE(EYU9XLAP)
  ...
  AP A33F WUEYU DATA
  performActionFailed
  RESOURCE(RESGROUP)
  CONTEXT(CICSGRP1)
  SCOPE(CICSGRP1)
  CSD_RESOURCE(NO)
  ACTION_NAME(INSTALL     )
  API_FUNCTION(PERFORM SET     )
  API_RESPONSE1(1038)
  API_RESPONSE2(1361)
  AP A348 WUEYU DATA
  entryObtainFeedback
  AP A363 WUEYU DATA
  entryGrowErrorInfoStorage
  AP A34A WUEYU DATA  before
  before feedback GETMAIN
  ...
  The next thing in the trace is the GETAIN for that storage:
  ---------------------------------------------------------------
  EIP   ENTRY GETMAIN
  TASK-23139 KE_NUM-01B2 TCB-QR   /007C1E88 RET-B99A85CE
  ---------------------------------------------------------------
  This GETMAIN was done by module DFHWUIPG 660 +x'1BECE'
  (that is at BASE level).
  .
  The CPSM entry seems to acknowledge this information in
  Data Fields #3 and #4 in the following trace entry:
  ---------------------------------------------------------
  23139 QR    AP A34B WUEYU DATA
  after feedback GETMAIN
  ...
  2-0000  39841008
  3-0000  00001A80
  ---------------------------------------------------------
  .
  There are 57 iterations
  that include this set of trace entries:
  ---------------------------------------------------------
  AP A361 WUEYU DATA -
  obtainFeedbackPostFetch2
  AP A357 WUEYU DATA -
  entryProcessBinsterr
  ... (same info)
  AP A358 WUEYU DATA -
  exitProcessBinsterr
  ... (same info)
  AP A360 WUEYU DATA -
  obtainFeedbackPreFetch2
  ---------------------------------------------------------
  .
  Then exit the Restful process that caused the overlay:
  ---------------------------------------------------------
  23139 QR    AP A377 WUEYU EXIT
  Restful CPSM Request
  ---------------------------------------------------------
  .
  The task takes several more 'restful' steps
  to discard the request and begins freeing storage.
  That is when the overlay is detected:
  ---------------------------------------------------------
  EIP   ENTRY FREEMAIN
  TASK-23139 KE_NUM-01B2 TCB-QR   /007C1E88 RET-B99BA1A2
  .
  SMMF  *EXC* - Storage_check_failed_on_freemain_request -
  (see the detail at the beginning of this update)
  ---------------------------------------------------------
  The FREEMAIN was done in DFHWUIPG +x'2DAA2' at BASE.
  .
  A second dump may be generated after the task ends
  as the XM process tries to free the storage again.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of CICSPlex SM V4R1M0              *
  ****************************************************************
  * PROBLEM DESCRIPTION:    Through a CMCI client like CICS      *
  *                      Explorer, you attempt to install a      *
  *                      RESGROUP containing resources which are *
  *                      not valid at the CICS release of one or *
  *                      more target MASes.  The WUI server to   *
  *                      which the CMCI client connects encoun-  *
  *                      ters a storage violation during a CICS  *
  *                      freemain in program DFHWUIPG.           *
  ****************************************************************
  * RECOMMENDATION: After applying the PTF that resolves this    *
  *                 APAR, all MASes and Single Server (SMSS)     *
  *                 regions which are configured as CMCI         *
  *                 servers must be restarted.  Note that the    *
  *                 restarts do not need to occur at the same    *
  *                 time.                                        *
  ****************************************************************
     When processing error feedback information after a failed
  action, the CMCI server module DFHWUEYU (WUEYU - the CPCI to
  CPSM API Interface) gets a block of CICS storage sufficient to
  contain the primary FEEDBACK instance and up to 31 areas
  describing the error encountered processing a single resource.
  If more than 31 related error resources are returned, a block
  sufficient to contain the actual number of error resources is
  acquired, the contents of the original block are copied into the
  new block, and the old block is freed.
     A logic error in DFHWUEYU causes the amount of additional
  storage needed to be calculated based on zero error resources.
  If the actual number is greater than 31, resources beyond the
  31st overwrite storage following the block.  This may result in
  abend ASRA (S0C4) if the block lies at the end of a page of
  storage and the following page is invalid, or in a storage vio-
  lation during FREEMAIN because the trailing check zone is over-
  written.
  

Problem conclusion

•    Module DFHWUEYU was modified to use the actual number of
  error resources returned by a failed action request to cal-
  culate the amount of storage required to contain the error
  data.
  

Temporary fix

• FIX AVAILABLE BY PTF ONLY
  

Comments

APAR Information

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

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

  UK71551

Modules/Macros

•    DESWUEYU DFHWUEYU
  

Fix information

• Fixed component name

  CICS TS Z/OS V4

• Fixed component ID

  5655S9700

Applicable component levels

• R60M PSY UK71551

     UP11/09/04 P F109

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.