PK59523: DFHAP0001 AN ABEND (CODE 0C4/AKEX) HAS OCCURRED AT OFFSET X'FFFFFFFF' IN MODULE DFHFCN. 08/01/18 PTF PECHANGE

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• This APAR has been created do to a bug with PK58301 (CICS TS 2.3
  ) and PK58235 (CICS TS 3.1 and 3.2). The APAR code in DFHFCNO
  switches to FO before calling IGWARLS. It does not adjust R13
  around the IGWARLS call however, so this still addresses the CSA
  When IGWARLS is called, it will STM 14,12 12 bytes into R13
  CSAOSRSA - (as per standard MVS save area convention). Now, this
  is fine, if nothing else runs on QR before we return. However,
  on busier systems, QR is up for grabs and will dispatch other
  work. If this needs to invoke OS services, they will themselves
  STM into the same area. The result is a
  lost save area environment for the FCN task, with unpredictable
  results.
  .
  ADDITIONAL SYMPTOMS:
  DFHME0116 (Module:DFHMEME) CICS symptom string for message
            DFHFC0001 is PIDS/5655M1500 LVLS/640 MS/DFHFC0001
            RIDS/DFHFCFR PTFS/UK30187 AB/UAKEA ADRS/00001FEE
  .
  DFHSR0601 SYSNAME Program interrupt occurred with system task
            TCP in control
  .
  DFHME0116 SYSNAME (Module:DFHMEME) CICS symptom string for
            message DFHSR0601 is PIDS/5655M1500 LVLS/640 MS/DFHSR0
  .
  DFHSR0601 Program interrupt occurred with system task TCP
            in control.
  TCP's stack shows we were in DFHZOPN, but the program check
  occurred in VTAM module ISTRACEP. Problem was with the savearea
  we passed to VTAM out of our CSA. Some of the registers pointed
  into file control modules - DFHFCN IAMARLS
  .
  ADDITIONAL KEYWORDS:
  DFHTRP DFHTCP DFHSI1548 S0C1 0C1 DFHSR0601 TCP
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All                                          *
  ****************************************************************
  * PROBLEM DESCRIPTION: Various errors such as an abend S0C4 at *
  *                      offset x'FFFFFFFF' into DFHFCN.         *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  DFHFCN switches to the FO TCB before invoking IGWARLS, and
  returns to the QR TCB afterwards. It should reset R13 around the
  call, so that it is addressing a bespoke register save area, and
  not CSAOSRSA as provided for operating system calls to save
  their callers registers into when running under the QR TCB. If
  other work runs under QR while the IGWARLS call is made, and has
  need to use services that will save the registers into CSAOSRSA
  too, then DFHFCN's environment will be lost and unpredictable
  results will occur when DFHFCN resumes control. These can
  include S0C1s, S0C4s, etc.
  KEYWORDS: abends0c4 0c4 abends0c1 oc1 fcn vsam
  

Problem conclusion

• UK32591  UK32584
  DFHFCN has been changed to switch R13 to address its own
  save area around the call to IGWARLS.
  

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:

  PK59468 UK33588 UK33589

Modules/Macros

•    DFHFCN   DFHFCNO
  

Fix information

• Fixed component name

  CICSTS V3 Z/OS

• Fixed component ID

  5655M1500

Applicable component levels

• R400 PSY UK33588

     UP08/02/14 P F802

• R500 PSY UK33589

     UP08/02/14 P F802

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.