JR35144: MASTER SEQUENCE WHEN A SCHEDULED RUN IS STARTED AND THE PREVIOUS RUN IS STILL BUSY.

Download Fix Pack 2 for InfoSphere Information Server Version 8.1

APAR status

• Closed as program error.

Error description

• There is a problem in the Master sequence when a scheduled run
  is started and the previous run is still busy. In the old
  version it used to say that you can not start a new run as the
  master sequence is still running. Now it actually starts the
  run, skips all the jobs in the sequence that have already
  finished and then gives a error saying that you can not run the
  job again. The exact error is  Sequence job is being restarted
  after failure  which is not true as it never aborted. I just
  want to know if this is how it is suppose to work now. The other
  thing is we lose the runtime information of the instance of the
  sequence that is still busy running when a new schedule tries to
  start the job.  I liaised with Arron Harden on this issue and he
  was able to re produce.  He suggested a call be logged for
  engineering for a patch request.
  

Local fix

• The command used to start a scheduled job does not check whether
  the job is in a runnable state before attempting to start it.
  The behaviour of a normal job is different form that for a
  restartable job sequence. A normal job will fail to start if it
  is already running and log a message indicating that the job
  needs to be reset but the restartable job sequence will appear
  to star but will not run normally.
  
  We should fix this by changing the command used to start a
  scheduled job to use the same sequence of Job Control API calls
  as are used within Job sequnces. This will ensure that the
  schedule operation will fail if the job is not in a runnable
  state and a log message recorded indicating that the scheduler
  is at fault.
  Ewan Paton
  

Problem summary

• ****************************************************************
  USERS AFFECTED:
  DataStage developers using Scheduler
  ****************************************************************
  PROBLEM DESCRIPTION:
  The DSR.SCHED command used to start scheduled jobs did not check
  the state of the scheduled job before attempting to start it.
  This led to restartable job sequnces failing and confusing error
  message being logged for other job types.
  ****************************************************************
  RECOMMENDATION:
  Any users experiencing the problem described should install the
  patch
  ****************************************************************
  

Problem conclusion

• Fixed DSR.SCHED to check job status before attempting to start
  and log more appropriate message if the job is not in a runnable
  state.
  

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:

Modules/Macros

• SERVER
  

Fix information

• Fixed component name

  WIS DATASTAGE

• Fixed component ID

  5724Q36DS

Applicable component levels

• R810 PSN

     UP