PM65489: WEBSPHERE SCHEDULER POLL DAEMON ACTIVE IN ALL SERVANTS FOR A SERVER ON ZOS

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• Typically the WebSphere Scheduler service requires only one
  instance being active in a cluster of servers.  However,
  WebSphere for z/OS servers are normally configured with
  multiple servants and the scheduler poll daemon
  does not restrict itself to operating in only one
  Servant region.  This can lead to multiple servants attempting
  to run the same task.
  
  One of the possible symptoms can be the following warning:
  com.ibm.websphere.scheduler.TaskPending: SCHD0057W: The task
  information for task ID 1664701 for scheduler MyScheduler
  (MyScheduler) cannot be updated in the database because the
  task has already been run by another Scheduler, the task
  information was changed, or the task was cancelled.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED:  All users of IBM WebSphere Application      *
  *                  Server V6.1                                 *
  ****************************************************************
  * PROBLEM DESCRIPTION: All Servants attempts to fire the       *
  *                      same task within a z/OS server          *
  *                      resulting in SCHD0057W                  *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  The scheduler service will be started on all servers in the
  cluster. A lease is required for the scheduler service
  to start processing the tasks and for ensuring that only one
  scheduler instance will fire the task. Typically the
  server that comes up first will obtain the lease.
  
  WebSphere z/OS platform supports multiple servants for Workload
  Managament (WLM) and it was found that  all of these servants
  within a z/OS server were able to obtain the leases. This was
  possible because the scheduler uses the server name for the
  lease owner name and all of these servant shared the same name
  and hence they were all able to obtain the lease and also can
  fire the tasks.
  
  This can result in the following messages in a multi servant
  z/OS environment:
  
  com.ibm.websphere.scheduler.TaskPending: SCHD0057W: The task
  information for task ID 1664701 for scheduler MyScheduler
  (MyScheduler) cannot be updated in the database because the
  task has already been run by another Scheduler, the task
  information was changed, or the task was cancelled.
  
  Now with this APAR, the lease owner name has been made unique
  to uniquely identify each servant and thereby ensuring only one
  servant will obtain the lease at any point of time.
  

Problem conclusion

• The WebSphere scheduler component has been updated such that
  the scheduler poll daemon is active on only on servant in a
  z/OS server envrionment.
  
  APAR PM65489 is currently targeted for inclusion in WebSphere
  Application Server V6.1 Fix Pack 6.1.0.45.
  
  Please refer to URL:
  //www.ibm.com/support/docview.wss?rs=404&uid=swg27006970
  for Fix Pack availability.
  

Temporary fix

Comments

APAR Information

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

  PM58610

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

Fix information

• Fixed component name

  WEBSPHERE FOR Z

• Fixed component ID

  5655I3500

Applicable component levels

• R610 PSY UK81646

     UP12/09/23 P F209

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.