IBM Support

PK75950: INTERMITTENTLY A SERVANT WILL STOP SERVICING REQUESTS ON A RECYCLE WITH CONTROL_REGION_DREG_ON_NO_SRS=1 SET.

A fix is available

Subscribe

You can track all active APARs for this component.

 

APAR status

  • Closed as program error.

Error description

  • Intermittently a servant may become unresponsive to requests
    after a recycle (due to abend or cancel) when the server is
    configured with control_region_dreg_on_no_srs=1.  This is a
    timing problem, and will no always occur.  Servants experiencing
    this problem will fail to receive a j2ee.state.running
    notification and will not put out a CWWIM6002I message.  The
    following exception may also be seen:
    .
    error propagating notifs to servant <servant ID>;
    com.ibm.websphere.management.exception.ConnectorException:
     Local exception
      at
    com.ibm.ws390.management.connector.corba.CorbaConnectorClient.in
    voke(CorbaConnectorClient.java:859)
      at
    com.ibm.ws390.management.InterProcessAdminService.invoke(InterPr
    ocessAdminService.java:352)
      at
    com.ibm.ws390.management.proxy.ServerMBeanProxy.sendRunningNotif
    ication(ServerMBeanProxy.java:778)
      at
    com.ibm.ws390.management.proxy.ServerMBeanProxy.synchronizeThrea
    dMonitor(ServerMBeanProxy.java:675)
      at
    com.ibm.ws390.management.proxy.ServerMBeanProxy.handleInternalNo
    tification(ServerMBeanProxy.java:649)
      at
      at
    com.ibm.ws390.management.ProxyMBeanSupport.handleNotification(Pr
    oxyMBeanSupport.java:333)
      at
    com.ibm.ws.management.ControlAdminServiceImpl.startupServantJVM(
    ControlAdminServiceImpl.java:1384)
      at
    com.ibm.ws.management._ControlAdminServiceImpl_Tie.startupServan
    tJVM(_ControlAdminServiceImpl_Tie.java:184)
      at
    com.ibm.ws.management._ControlAdminServiceImpl_Tie._invoke(_Cont
    rolAdminServiceImpl_Tie.java:83)
    ...
    Caused by: org.omg.CORBA.INTERNAL:   vmcid: 0xc9c21000  minor
    code: 99  completed: No
      at java.lang.Class.newInstanceImpl(Native Method)
      at java.lang.Class.newInstance(Class.java:1300)
      at
    com.ibm.rmi.util.Utility.readSystemException(Utility.java:2013)
    ...
    This APAR will also resolve a NullPointerException directly
    associated with this problem:
    .
    java.lang.NullPointerException
     at com.ibm.ws390.orb.SRAggregator.hasNext(Unknown Source)
     at
    com.ibm.ws390.management.connector.corba.CorbaConnectorClient.re
    connect(CorbaConnectorClient.java:280)
    ...
    

Local fix

  • To ensure you avoid the issue entirely, remove the
    control_region_dreg_on_no_srs environment variable, or code it
    to 0 (which is the default).  Since this is a timing problem,
    the issue may not be seen on all installations, however.  If you
    experience the servant hang, a recycle of the server will be
    required.
    

Problem summary

  • ****************************************************************
    * USERS AFFECTED: All users of WebSphere Application Server    *
    *                 V7.0 for z/OS                                *
    ****************************************************************
    * PROBLEM DESCRIPTION: A servant may become unresponsive to    *
    *                      requests after a recycle when the       *
    *                      server is configured with               *
    *                      control_region_dreg_on_no_srs=1.        *
    ****************************************************************
    * RECOMMENDATION:                                              *
    ****************************************************************
    The function related to the control_region_dreg_on_no_srs
    setting causes the shutdown of listeners and the rejecting of
    requests by the Server until the minimum number of "Ready"
    servants has been restarted.  A timing window exists that may
    cause an internal request to fail during initialization. This
    request failure during initialization may lead to the servants
    failure to process new requests.
    
    The NullPointerException in SRAggregator was caused by a
    missing check in the hasNext method.
    

Problem conclusion

  • Code has been added to synchronize the servant startup code
    with the restarting of the listeners and allowing new requests
    to be processed.
    
    SRAggregator code has been updated to eliminate the
    NullPointerException in the hasNext method.
    
    APAR PK75950 is currently targeted for inclusion in Service
    Level (Fix Pack) 7.0.0.1 of WebSphere Application Server V7.0
    for z/OS.
    
    Please refer to URL:
    //www.ibm.com/support/docview.wss?rs=404&uid=swg27006970
    for Fix Pack availability.
    

Temporary fix

Comments

APAR Information

  • APAR number

    PK75950

  • Reported component name

    WEBSPHERE FOR Z

  • Reported component ID

    5655I3500

  • Reported release

    700

  • Status

    CLOSED PER

  • PE

    NoPE

  • HIPER

    NoHIPER

  • Special Attention

    NoSpecatt

  • Submitted date

    2008-11-18

  • Closed date

    2008-11-19

  • Last modified date

    2009-01-02

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

    PK69971

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

Modules/Macros

  • BBGUBINF BBOUBINF
    

Fix information

  • Fixed component name

    WEBSPHERE FOR Z

  • Fixed component ID

    5655I3500

Applicable component levels

  • R700 PSY UK41764

       UP08/12/05 P F812

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.

[{"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SS7K4U","label":"WebSphere Application Server for z\/OS"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"7.0","Edition":"","Line of Business":{"code":"LOB45","label":"Automation"}}]

Document Information

Modified date:
10 February 2022