PM50560: ABEND04E RC00E50002 FROM DSNSRSUP AFTER AN ABEND04E RC00E20016 FROM DSNSVSTK

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• ABEND04E RC00E50002 from DSNSRSUP after an ABEND04E RC00E20016
  from DSNSVSTK.
  .
  If there are multiple SMC RECOVERY STACKs allocated, this
  problem is possible and results in a recovery failure.
  DB2 terminates with DSNV086E RC00E50727
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: All users of DB2.                            *
  ****************************************************************
  * PROBLEM DESCRIPTION: Following an ABEND04E RC00E20016 due    *
  *                      to a short on DBM1 storage situation,   *
  *                      a secondary ABEND04E RC00E50002         *
  *                      occurred during latch manager recovery  *
  *                      causing DB2 to terminate with           *
  *                      RC00E50727.                             *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  DB2 was running short on DBM1 virtual storage resulting in
  many storage abends (ie. ABEND04E RC00E20003 RC00E20016 etc).
  One thread had requested storage from a storage pool and
  DSNSVBK had invoked storage contraction in an attempt to
  free up unused storage.  While holding an LPVT latch,
  DSNSCON2 was attempting to release a PHB latch and received
  an ABEND04E RC00E20016 in DSNSVBK when resuming a latch
  waiter.  Due to the nested storage manager requests,
  abend recovery made two passes through recovery routine
  DSNSRSUP.  The LPVT latch was released on the first pass,
  but a latch tracking flag was not reset.  This caused the
  second pass to attempt to release the LPVT latch a second
  time which failed with ABEND04E RC00E50002 as the latch
  was not held.  This secondary abend during recovery caused
  DB2 to terminate with RC00E50727.
  

Problem conclusion

• DSNSRSUP has been changed to reset the latch held tracking
  flag when a latch is released in case a second or subsequent
  pass is made through the recovery routine.  This APAR does not
  address the virtual storage shortage issue.
  

Temporary fix

• *********
  * HIPER *
  *********
  

Comments

APAR Information

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

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

  UK73473 UK73474 UK73475

Modules/Macros

• DSNSRSUP
  

Fix information

• Fixed component name

  DB2 OS/390 & Z/

• Fixed component ID

  5740XYR00

Applicable component levels

• RA10 PSY UK73473

     UP11/11/18 P F111

• R810 PSY UK73474

     UP11/11/18 P F111

• R910 PSY UK73475

     UP11/11/18 P F111

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.