PH41615: INTELLIGENT MANAGEMENT WEBSERVER PLUGIN IS SOMETIMES UNABLE TO ROUTE ONE HTTP SESSION REQUESTS TO THE SAME MEMBER SERVER.

APAR status

• Closed as program error.

Error description

• The Intelligent management webserver plugin is sometimes
  unable
  to route one HTTP session requests to the same member server
  after the collective member server is restarted. One session
  requests are routed to two member servers randomly.
  

Local fix

• Try to restart the member server(s) again.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:  All users of IBM WebSphere Application      *
  *                  Server Liberty- Virtual Enterprise/IM       *
  *                  Component                                   *
  ****************************************************************
  * PROBLEM DESCRIPTION: Intelligent Management enabled plugin   *
  *                      intermittently fails to route to        *
  *                      WebSphere Liberty collective member     *
  *                      server after a server restart           *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  Intelligent Management enabled plugin intermittently fails to
  route to WebSphere Liberty collective member server after a
  server restart. Symptoms include lack of traffic being routed to
  the affected member server after all the application endpoints
  have been published.  The root cause is a failure to publish a
  server's HTTPSession "cloneID" prior to publishing endpoints.
  
  When this issue occurs, the collective controller logs will
  always contain this message:
  
  RoutingInfoMa 3   Failed getting cloneID, set to default.
  Exception = java.util.NoSuchElementException: Node
  /sys.was.collectives/local/hosts/<users-
  host>/userdirs/opt/IBM/Liberty/wlp/usr/servers/<users-
  server>/sys.mbeans/WebSphere:name=com.ibm.ws.jmx.mbeans.sessionM
  anagerMBean/attributes/CloneID does not exist
  

Problem conclusion

• Collective member servers must publish their  HTTPSession
  "cloneId" to the collective controller before publishing
  application endpoint routing information.  The Intelligent
  Management runtime collective member code has been corrected to
  enforce the ordering of publishing
  information to the collective controller, including cloneID's
  and application routing endpoint information.
  
  The fix for this APAR is currently targeted for inclusion in fix
  pack 22.0.0.2.  Please refer to the Recommended Updates page for
  delivery information:
  http://www.ibm.com/support/docview.wss?rs=180&uid=swg27004980
  

Temporary fix

• Restarted the affected server(s) so that no "CloneID does not
  exist" message appears in the collective controller logs, and
  that routing resumes correctly.
  

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

  LIBERTY PROFILE

• Fixed component ID

  5724J0814

Applicable component levels