PQ78222: STOP COMMAND ISSUED AGAINST WEBSPHERE SERVER MAY NOT STOP SERVERUSER MAY HAVE TO ISSUE CANCEL COMMAND

 A fix is available

Obtain the fix for this APAR



APAR status
Closed as program error.

Error description
Customer issues the MVS 'STOP' command against a WebSphere
server, to 'stop' the server after it has completed it's current
tasks.  The STOP command will not stop the WebSphere Servers,
if there are certain java threads active in the WebSphere Appli-
cation server.  The theads that are not being 'stopped' are 'non
WebSphere managed threads, and have tracebacks similar to the
following:
.
ThreadUtils_BlockingSection
sysTimeout
JVM_Timeout
Java_java_net_PlainSocketImpl_socketAccept
java/net/PlainSocketImpl.socketAccept(Ljava/net/SocketImpl;)V
java/net/PlainSocketImpl.accept(Ljava/net/SocketImpl;)V
java/net/ServerSocket.implAccept(Ljava/net/Socket;)V
com/ibm/disthub/impl/net/tcp/IMBTcpServerSocket.accept()
       Lcom/ibm/disthub/impl/net/IMBSocket;
com/ibm/disthub/impl/server/Acceptor.threadMain()V
mmipSelectInvokeJavaMethod
      (com/ibm/disthub/impl/client/DebugThread.run)
INVOKDMY
EXECJAVA  (Xリ.5￱?ハ)
mmipExecuteJava  (com/ibm/disthub/impl/client/DebugThread.run)
xeRunDynamicMethod
.
another Thread Stack that we have seen is:
pthread_cond_timedwait
condTimedWait
sysMonitorWait(owner = 0)
lkMonitorWait
JVM_MonitorWait
java/lang/Object.wait(J)V
com/ibm/disthub/impl/server/CSConnectionReaper.threadMain()V
mmipSelectInvokeJavaMethod
    (com/ibm/disthub/impl/client/DebugThread.run)
INVOKDMY
EXECJAVA  (Xリ.5ラP)
mmipExecuteJava  (com/ibm/disthub/impl/client/DebugThread.run)
xeRunDynamicMethod
threadRT0
xmExecuteThread
.
Local fix
Currently the local fix is to issue the MVS 'CANCEL' command
against the WebSphere APplication server.
Problem summary
****************************************************************
* USERS AFFECTED: All users of WebSphere Application Server    *
*                 V5.0 for z/OS                                *
****************************************************************
* PROBLEM DESCRIPTION: When stop-server is issued, controller  *
*                      shuts down after significant delay      *
*                      (> 100 minutes) but servant regions     *
*                      shut down immediately.                  *
****************************************************************
* RECOMMENDATION:                                              *
****************************************************************
With memory-to-memory replication (DRS) enabled, the controller
does not shutdown in an acceptable amount of time and
com/ibm/disthub* threads are the only remaining java threads
that are active. Problem is that during the shutdown of the
DRS component, DRS does not issue a stop to the broker that
it is using for replication. Since those remaining threads are
not marked as Daemon threads, the Controller does not shutdown
until some specified timeout has taken place.

Please note that in the pre-Cumulative Fix W502000 version of
WebSphere Application Server V5.0 for z/OS, the broker was
initialized in the servant regions, therefore at those levels of
code the servant regions will exhibit similar behavior
- com/ibm/disthub* threads remaining in the address space
preventing the servant from shutting down. As of Cumulative Fix
W502000, the broker is only initialized in the controller and
therefore the current behavior will be seen.
Problem conclusion
The shutdown of session-to-session replication will also
shutdown the broker that it is utilizing, as a result, the
broker will terminate properly and will not prevent the
Controller region from stopping.

Please note:

The following two messages may appear in the Control Region logs
when Memory-to-Memory replication is enabled and the server is
stopped:

"com.ibm.disthub.impl.util.SocketThreadPoolException:
       doWrite() did not die"

Uncaught exception exc=java.lang.InterruptedException
 : JVMLK007: operation interrupted
   at java.lang.Object.wait(Native Method)
   at java.lang.Object.wait(Object.java(Compiled Code))
   at com.ibm.disthub.impl.util.SafeQueue.dequeue
   (SafeQueue.java:113)
   at com.ibm.hplwps.reference.WASLogRecorder.run
   (WASLogRecorder.java:570)
   at java.lang.Thread.run(Thread.java:513)

These messages however can be ignored as they are informational.
These messages will be surpressed in APAR 
PQ85199.

APAR PQ78222 is associated with SERVICE LEVEL W502003 of
WebSphere Application Server V5.0 for z/OS.
Temporary fix Comments
APAR information
APAR number PQ78222
Reported component name WEBSPHERE FOR Z
Reported component ID 5655I3500
Reported release 500
Status CLOSED PER
PE NoPE
HIPER NoHIPER
Special Attention NoSpecatt
Submitted date 2003-09-08
Closed date 2004-02-26
Last modified date 2004-04-03

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:

Modules/Macros
BBOUBINF          

Publications Referenced

Fix information
Fixed component name WEBSPHERE FOR Z
Fixed component ID 5655I3500

Applicable component levels
R500 PSY UQ85594    UP04/03/02 P F403

  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.


Document Information


Current web document: swg1PQ78222.html
Product categories: Software > Application Servers > Distributed Application & Web Servers > WebSphere Application Server for z/OS
Operating system(s):
Software version: 500
Software edition:
Reference #: PQ78222
IBM Group: Software Group
Modified date: Apr 3, 2004