PM80840: Rational Developer for System z (RDz) - Lock not released on host after Client ended from task manager

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• Rational Developer for System z - RDz - v8.0.3.
  Client used Windows taskmanager to end the RDz client because
  a Edit/save operation of a pds-member did not finish.
  
  The lock on the host for this member was not released.
  After a new connect of the client the lock had to be released
  by terminating the rDz threadpool that held the lock.
  
  This problem can also occur if a user keeps the RDz client
  logged onto the host overnight, and during the night the
  workstation is restarted because of automatic installations.
  The next day the user may still have a lock on the user's
  Source PDS-Library.
  ***************************************************************
  Local fix:
  To release the lock the applicable RDz threadpool has to be
  terminated.
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED: 01.All RDz users                             *
  *                 02.All RDz users editing z/OS UNIX files     *
  *                 03.All RDz users                             *
  ****************************************************************
  * PROBLEM DESCRIPTION: 01.data set lock not released when      *
  *                         client ends                          *
  *                      02.backup file causes problem when      *
  *                         parent folder is read-only           *
  *                      03.Multiple RDz RSE servers get         *
  *                         *timeout* status, which is never     *
  *                         reset.                               *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  01.data set lock not released when client ends due to NPE in
     finish processing java.lang.NullPointerException at org.
     eclipse.rse.internal.dstore.universal.miners.command.
     CommandMinerThread.sendExit(CommandMinerThread.java:1066)
     followed later by
     java.lang.NullPointerException at org.eclipse.rse.internal.
     dstore.universal.miners.command.CommandMinerThread.
     cleanupThread(CommandMinerThread.java:1114)
  02.creation of a backup file causes a problem when editing a
     file in a read-only folder (not allowed to create a new file)
  03.OpenRSE thread and MVSminer thread are in contention in
     writing traces onto standard.out.
  

Problem conclusion

• 01.NPE is caused by _stdOutputHandler being null. The code has
     changed to accept a null value.
  02.v8033 PTF UK81498 altered the way backups were created. The
     problem that was introduced is now resolved.
     problem bypass: add the following to rsed.envvars to disable
     the creation of the backup file
     _RSE_JAVAOPTS="$_RSE_JAVAOPTS -Dbackupfiles=false"
  03.Now, OpenRSE is changed so that it should not write any data
     onto standard.out
  

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

• FEKFCLIE FEKFDST0 FEKFDST1 FEKFDST2
  

Fix information

• Fixed component name

  RD/Z HOST

• Fixed component ID

  5724T0723

Applicable component levels

• R801 PSY UK96308

     UP13/08/05 I 1000

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.