PM16223: HANG AND NULLPOINTEREXCEPTION ON EXECUTING PORTLET ACTIONS

6.1.0.5 Download: WebSphere Portal and WCM V6.1.0 fix pack 5
6.1.0.6 Download: WebSphere Portal and WCM V6.1.0 fix pack 6

APAR status

• Closed as program error.

Error description

• Under some rare conditions a hang can be seen when Portlet
  actions are executed. These conditions are that the same action
  ID is submitted more than one time by a Portlet and garbage
  collection occours between the 2 actions.
  
  In the logs a NullPointerException can be seen:
  
  [6/9/10 9:13:07:007 CDT] 0000006a Servlet       E
  com.ibm.wps.engine.Servlet handleException EJPEJ0070E: An
  exception occurred while processing the request.
  java.lang.NullPointerException
  at com.ibm.wps.engine.phases.WPActionPhase$LockManager$Data
  .access$200(WPActionPhase.java:3462)
  at com.ibm.wps.engine.phases.WPActionPhase$LockManager
  .lockReleased(WPActionPhase.java:3453)
  at com.ibm.wps.engine.phases.WPActionPhase
  .processPortlets(WPActionPhase.java:2762)
  at com.ibm.wps.engine.phases.WPActionPhase
  .execute(WPActionPhase.java:668)
  at com.ibm.wps.state.phases.AbstractActionPhase
  .next(AbstractActionPhase.java:130)
  at com.ibm.wps.engine.Servlet.callPortal(Servlet.java:855)
  at com.ibm.wps.engine.Servlet.doGet(Servlet.java:617)
  at com.ibm.wps.engine.Servlet.doPost(Servlet.java:888)
  ...
  
  Some minutes later the Application Server will report hang
  threads:
  
  [6/9/10 9:22:41:135 CDT] 00000025 ThreadMonitor W   WSVR0605W:
  Thread "WebContainer : 31" (00000088) has been active for
  602945
  milliseconds and may be hung.  There is/are 1 thread(s) in
  total
  in the server that may be hung.
  at sun.misc.Unsafe.park(Native Method)
  at java.util.concurrent.locks.LockSupport.park(LockSupport
  .java:158)
  at java.util.concurrent.locks.AbstractQueuedSynchronizer
  .parkAndCheckInterrupt(AbstractQueuedSynchronizer.java:747)
  at java.util.concurrent.locks.AbstractQueuedSynchronizer
  .acquireQueued(AbstractQueuedSynchronizer.java:778)
  at java.util.concurrent.locks.AbstractQueuedSynchronizer
  .acquire(AbstractQueuedSynchronizer.java:1114)
  at java.util.concurrent.locks.ReentrantLock$FairSync
  .lock(ReentrantLock.java:201)
  at java.util.concurrent.locks.ReentrantLock.lock(ReentrantLock
  .java:262)
  at com.ibm.wps.engine.phases.WPActionPhase
  .processPortlets(WPActionPhase.java:2463)
  at com.ibm.wps.engine.phases.WPActionPhase
  .execute(WPActionPhase.java:668)
  at com.ibm.wps.state.phases.AbstractActionPhase
  .next(AbstractActionPhase.java:130)
  at com.ibm.wps.engine.Servlet.callPortal(Servlet.java:855)
  at com.ibm.wps.engine.Servlet.doGet(Servlet.java:617)
  at com.ibm.wps.engine.Servlet.doPost(Servlet.java:888)
  ...
  

Local fix

Problem summary

• Under some rare conditions a hang can be seen when Portlet
  actions are executed. These conditions are that the same action
  ID is submitted more than one time by a Portlet and garbage
  collection occurs between the 2 actions. In the logs a
  NullPointerException can be seen:
  
        [6/9/10 9:13:07:007 CDT] 0000006a Servlet       E
  com.ibm.wps.engine.Servlet handleException EJPEJ0070E: An
  exception occurred while processing the request.
  java.lang.NullPointerException at
  com.ibm.wps.engine.phases.WPActionPhase$LockManager$Data.access$
  200(WPActionPhase.java:3462)
        at
  com.ibm.wps.engine.phases.WPActionPhase$LockManager.lockReleased
  (WPActionPhase.java:3453)
        at
  com.ibm.wps.engine.phases.WPActionPhase.processPortlets(WPAction
  Phase.java:2762)
        at
  com.ibm.wps.engine.phases.WPActionPhase.execute(WPActionPhase.ja
  va:668)
        at
  com.ibm.wps.state.phases.AbstractActionPhase.next(AbstractAction
  Phase.java:130)
        at com.ibm.wps.engine.Servlet.callPortal(Servlet.java:855)
        at com.ibm.wps.engine.Servlet.doGet(Servlet.java:617)
        at com.ibm.wps.engine.Servlet.doPost(Servlet.java:888) ...
  S
  
     Some minutes later the Application Server will report hang
  threads:
  
        [6/9/10 9:22:41:135 CDT] 00000025 ThreadMonitor W
  WSVR0605W: Thread "WebContainer : 31" (00000088) has been active
  for 602945 milliseconds and may be hung.  There is/are 1
  thread(s) in total in the server that may be hung.
        at sun.misc.Unsafe.park(Native Method) at
  java.util.concurrent.locks.LockSupport.park(LockSupport.java:158
  )
        at
  java.util.concurrent.locks.AbstractQueuedSynchronizer.parkAndChe
  ckInterrupt(AbstractQueuedSynchronizer.java:747)
        at
  java.util.concurrent.locks.AbstractQueuedSynchronizer.acquireQue
  ued(AbstractQueuedSynchronizer.java:778)
        at
  java.util.concurrent.locks.AbstractQueuedSynchronizer.acquire(Ab
  stractQueuedSynchronizer.java:1114)
        at
  java.util.concurrent.locks.ReentrantLock$FairSync.lock(Reentrant
  Lock.java:201)
        at
  java.util.concurrent.locks.ReentrantLock.lock(ReentrantLock.java
  :262)
        at
  com.ibm.wps.engine.phases.WPActionPhase.processPortlets(WPAction
  Phase.java:2463)
        at
  com.ibm.wps.engine.phases.WPActionPhase.execute(WPActionPhase.ja
  va:668)
        at
  com.ibm.wps.state.phases.AbstractActionPhase.next(AbstractAction
  Phase.java:130)
        at com.ibm.wps.engine.Servlet.callPortal(Servlet.java:855)
        at com.ibm.wps.engine.Servlet.doGet(Servlet.java:617)
        at com.ibm.wps.engine.Servlet.doPost(Servlet.java:888) ...
  

Problem conclusion

• Fixed action ID synchronization.
  
  Failing Module(s):
     Engine: State handling
  
  Affected Users:
     All Users
  
  Version Information:
     Portal Version(s): 6.1.0.3
      Pre-Requisite(s): ---
       Co-Requisite(s): ---
  
  Manuel Steps:
     None
  
  Platform Specific:
     This fix applies to all platforms.
  
  
  PM16223 is part of Cumulative Fix 06 for Portal 6.1.0.3/6.1.5.0.
  
  The fix is available from Fix Central:
  
  http://www.ibm.com/eserver/support/fixes/fixcentral/swgquickorde
  r?apar=PM14900&productid=WebSphere%20Portal&brandid=5
  
  You may need to type or paste the complete address into your Web
  browser.
  

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

  WEBSPHERE PORTA

• Fixed component ID

  5724E7600

Applicable component levels

• R61C PSY

     UP