PK63842: RECORD NOT FOUND ON EXEC CICS READ DATASET IS ISSUED ON A VSAM.

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• On the read update into request, the into area plus its length
  includes the ridfld area.
  The initial read update fails with an exclusive control conflict
  but this is not a problem, and CICS retries when the CI is
  freed.
  The read now works, but note at this point the read into has now
  overwritten the key. CICS then attempts to get the record lock
  (NQ) but this fails with a busy condition - another task must
  have the record lock and so this task must wait. Before the
  wait, we issue an endreq to free up the CI. Once we ontain the
  endreq, we must then do the read update all over again. But of
  course at this point the key has been overlaid and so we now
  get the record not found response.
  

Local fix

• If the ridfld area is not within the INTO area then all would
  be fine. It may be that the INTO area length is too great
  and so it appears that the ridfld is within the intended INTO
  area.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All.                                         *
  ****************************************************************
  * PROBLEM DESCRIPTION: EXEC CICS READ UPDATE INTO request      *
  *                      returns NOTFND response when record     *
  *                      lock initially held by another          *
  *                      request.                                *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  An EXEC CICS READ UPDATE INTO request against a KSDS or PATH
  over an ESDS has the same area assigned to both the INTO and the
  RIDFLD options. The initial READ UPDATE INTO fails as another
  task has the record lock. However the initial attempt has
  successfully read the file and updated the INTO area,
  overwriting the intended RIDFLD. Once the record lock has been
  obtained by this request, the update is attempted again. However
  when the RIDFLD is selected it now reflects the record data from
  the initial read, leading to a NOTFND response.
  

Problem conclusion

• DFHFCVS has been updated to ensure that a copy of the original
  RIDFLD (VSWAXKEY) is used for the retry request.
  

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:

  UK36426 UK36427 PK66828

Modules/Macros

•    DFHFCVS  DFHFCVS1
  

Fix information

• Fixed component name

  CICSTS V3 Z/OS

• Fixed component ID

  5655M1500

Applicable component levels

• R400 PSY UK36426

     UP08/05/15 P F805

• R500 PSY UK36427

     UP08/05/15 P F805

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.