PM87976: After a grid restart, occasionally one or more catalog servers are encountering OOM exception from DCS.

APAR status

• Closed as program error.

Error description

• After a grid restart, occasionally one or more catalog servers
  are encountering OOM exception from DCS.
  

Local fix

• N/A.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:  Users with topologies that generate a large *
  *                  amount of                                   *
  *                  placement work units within a window of one *
  *                  minute.                                     *
  ****************************************************************
  * PROBLEM DESCRIPTION: The catalog server might                *
  *                      run out of memory due to a high number  *
  *                      of canceled TimerTasks that have not    *
  *                      been purged from java.util.Timer.       *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  Tracking placement work units involves the use of TimerTask
  objects on the catalog server.  These tasks are frequently
  canceled.
  However, java.util.Timer does not purge the objects at the time
  of
  cancellation.  As a result, the build up
  of thousands of TimerTask references being held by the Timer
  might occur.
  The build up of these objects might cause an OutOfMemoryError on
  the catalog server.
  

Problem conclusion

• This fix purges the java.util.Timer on an interval.
  

Temporary fix

Comments

APAR Information

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

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

Fix information

• Fixed component name

  WS EXTREME SCAL

• Fixed component ID

  5724X6702

Applicable component levels

• R850 PSY

     UP

• R860 PSY

     UP