CEMT command examples

The following is a series of CEMT command examples showing how you can use CEMT to resolve unit of work failures in preparation for batch processing.

  1. A CEMT INQUIRE UOWDSNFAIL command for a specified data set shows in the following display that there are two failed units of work:
     CEMT INQUIRE UOWDSNFAIL DSN('RLS.ACCOUNTS.ESDS.DBASE1')
     STATUS:  RESULTS
      Dsn(RLS.ACCOUNTS.ESDS.DBASE1                    ) Dat Del
         Uow(AA6DB080C40CEE01)                         Rls
      Dsn(RLS.ACCOUNTS.ESDS.DBASE1                    ) Dat Ind
         Uow(AA6DB08AC66B4000)                         Rls

    The display shows a REASON code of DELEXITERROR (Del) for one unit of work, and INDEXRECFULL (Ind) for the other.

  2. A CEMT SET DSNAME(…) RETRY command, after fixing the delete exit error (caused because an XFCLDEL exit program was not enabled) invokes a retry of both units of work with the following result:
     SET DSNAME('RLS.ACCOUNTS.ESDS.DBASE1') RETRY
     STATUS:  RESULTS - OVERTYPE TO MODIFY
      Dsn(RLS.ACCOUNTS.ESDS.DBASE1                    ) Vsa       NORMAL
         Fil(0001) Val Bas Rec Sta         Ava     Ret
  3. Another CEMT INQ UOWDSNFAIL command shows that one of the two units of work failed again. It is the unit of work with the INDEXRECFULL error, as follows:
      INQUIRE UOWDSNFAIL
      STATUS:  RESULTS
       Dsn(RLS.ACCOUNTS.ESDS.DBASE1                    ) Dat Ind
          Uow(AA6DB08AC66B4000) Dat Ind                 Rls

    This type of error can occur when adding a value to a non-unique alternate index while trying to back out a change to a record; after the original update was made, other units of work have added values to the alternate index, filling all the available space. Such an error should be extremely rare. The solution is to define a larger alternate index record size for the data set.

  4. However, the data set with the remaining failed unit of work is needed urgently for a batch run, so you may decide not to complete the unit of work now, and to resolve the problem later. In the meantime, to allow the batch job to start, you issue SET DSN('RLS.ACCOUNTS.ESDS.DBASE1') RESETLOCKS to release the retained locks associated with the unit of work.
  5. To check that there are no retained locks remaining, you can issue another INQUIRE DSNAME command. The expanded display shows:
      Dsname(RLS.ACCOUNTS.ESDS.DBASE1)
      Accessmethod(Vsam)
      Action(              )
      Filecount(0001)
      Validity(Valid)
      Object(Base)
      Recovstatus(Recoverable)
      Backuptype(Static)
      Frlog()
      Availability( Available )
      Lostlocks()
      Retlocks(Noretained)
      Quiescestate()
      Uowaction(              )
      Basedsname(RLS.ACCOUNTS.ESDS.DBASE1)
      Fwdrecovlsn(RLS.ACCOUNTS.FWDRECOV.LOG)
  6. Diagnostic messages have been written to the CSFL transient data queue about the locks that were reset.
    DFHFC3003 01/04/95 12:08:03 CICSHDA1 Record not backed out because locks for 
               a backout-failed data set have been reset.
               Diagnostic information follows in message DFHFC3010. The record was
               updated by unit of work X'AA6DB08AC66B4000' for file ACCNT1  , base
               data set RLS.ACCOUNTS.ESDS.DBASE1
     
    DFHFC3010 01/04/95 12:08:03 CICSHDA1 Diagnostic information for unit of work
               X'AA6DB08AC66B4000' and file ACCNT1  .
               Update was a write-add made by transaction WKLY at terminal T583 
               under task number 00027.  Key length 4, data length 7, base ESDS 
               RBA X'00000DDF', record key X'00000DDF'


dfht2c00220.html | Timestamp icon Last updated: Monday, 5 June 2017