PK15961: HIGH CPU UTILIZATION CAUSED BY THE DATA REPLICATION SERVICE

 Fixes are available

5.0.2.17: WebSphere Application Server 5.0.2 Cumulative Fix 17 for Solaris
5.0.2.17: WebSphere Application Server 5.0.2 Cumulative Fix 17 for Windows
5.1.1.10: WebSphere Application Server V5.1.1 Cumulative Fix 10 for Windows
5.1.1.10: WebSphere Application Server V5.1.1 Cumulative Fix 10 for AIX
5.0.2.17: WebSphere Application Server 5.0.2 Cumulative Fix 17 for HP-UX
5.0.2.17: WebSphere Application Server 5.0.2 Cumulative Fix 17 for AIX
5.0.2.17: WebSphere Application Server 5.0.2 Cumulative Fix 17 for Linux
5.1.1.10: WebSphere Application Server V5.1.1 Cumulative Fix 10 for HP-UX
5.1.1.10: WebSphere Application Server V5.1.1 Cumulative Fix 10 for Linux
5.0.2.16: WebSphere Application Server 5.0.2 Cumulative Fix 16 for HP-UX
5.0.2.16: WebSphere Application Server 5.0.2 Cumulative Fix 16 for Solaris
5.0.2.16: WebSphere Application Server 5.0.2 Cumulative Fix 16 for Windows
5.0.2.16: WebSphere Application Server 5.0.2 Cumulative Fix 16 for AIX
5.0.2.18: WebSphere Application Server 5.0.2 Cumulative Fix 18 for Solaris
5.0.2.18: WebSphere Application Server 5.0.2 Cumulative Fix 18 for Windows
5.0.2.18: WebSphere Application Server 5.0.2 Cumulative Fix 18 for HP-UX
5.0.2.18: WebSphere Application Server 5.0.2 Cumulative Fix 18 for AIX
5.0.2.18: WebSphere Application Server 5.0.2 Cumulative Fix 18 for Linux
5.0.2.16: WebSphere Application Server 5.0.2 Cumulative Fix 16 for Linux
5.1.1.10: WebSphere Application Server V5.1.1 Cumulative Fix 10 for Solaris



APAR status
Closed as program error.

Error description
If you configured DRS replication service in WebSphere V5.x
cluster environment, you may experience a high CPU utilization
for WebSphere AppServer JVM where you configured DRS service.
.
If you take several javacores from the Application Server JVM
which has a high CPU utilization, you may observe this Java
stack in them:
.
"Alarm : 2" (TID:40BC2E68, sys_thread_t:8E10CD8, state:R, native
ID:1B8010) prio=5
  at
java.util.LinkedList$ListItr.hasNext(LinkedList.java(Compiled
Code))
  at
com.ibm.ws.drs.DRSPeerCoordinator.pingPeers(DRSPeerCoordinator.j
ava(Compiled Code))
  at
com.ibm.ws.drs.DRSCacheApp.pingPeers(DRSCacheApp.java(Compiled
Code))
  at
com.ibm.ws.drs.DRSPeerCoordinator.alarm(DRSPeerCoordinator.java(
Compiled Code))
  at com.ibm.ejs.util.am._Alarm.run(_Alarm.java(Compiled Code))
  at
com.ibm.ws.util.ThreadPool$Worker.run(ThreadPool.java(Compiled
Code))
Local fix
N/A. The only way how to recover from a high CPU utilization
is to recycle the JVM/AppServer.
Problem summary
****************************************************************
* USERS AFFECTED: All users of WebSphere Application Server    *
*                 using HTTP Session memory-to-memory          *
*                 replication.                                 *
*                                                              *
****************************************************************
* PROBLEM DESCRIPTION: High CPU utilization may occur when     *
*                      using the data replication service.     *
****************************************************************
* RECOMMENDATION:                                              *
*                                                              *
****************************************************************
When the data replication service (DRS) determines that one of
its partner servers is no longer available a message is sent
to the remaining partners. High CPU utilization has been
observed when performing this procedure.
The problem was caused by using a list iterator to notify
partner servers.
Problem conclusion
The process of notifying partner servers has been changed to
avoid using a list iterator and use the linked list directly.
This will avoid the high CPU utilization in the list iterator.

This fix for this APAR is targeted for inclusion in fixpack
5.0.2.16 and 5.1.1.10. Please refer to the Recommended Updates
page for delivery information:

http://www.ibm.com/support/docview.wss?rs=180&uid=swg27004980
Temporary fix Comments
APAR information
APAR number PK15961
Reported component name WAS NETWRK DEPL
Reported component ID 5630A3601
Reported release 10I
Status CLOSED PER
PE NoPE
HIPER NoHIPER
Special Attention NoSpecatt
Submitted date 2005-11-29
Closed date 2006-02-06
Last modified date 2006-02-08

APAR is sysrouted FROM one or more of the following:

APAR is sysrouted TO one or more of the following:

Modules/Macros
DRS          

Publications Referenced

Fix information
Fixed component name WAS NETWRK DEPL
Fixed component ID 5630A3601

Applicable component levels
R00A PSY    UP
R00H PSY    UP
R00I PSY    UP
R00S PSY    UP
R00W PSY    UP
R10A PSY    UP
R10H PSY    UP
R10I PSY    UP
R10S PSY    UP
R10W PSY    UP


Document Information


Product categories: Software > Application Servers > Distributed Application & Web Servers > WebSphere Application Server > General
Operating system(s):
Software version: 10I
Software edition:
Reference #: PK15961
IBM Group: Software Group
Modified date: Feb 8, 2006