PK10725: NODEAGENT MONITORING POLICY NOT ALWAYS HONORED. | |||||||||||||||||||||||||||||||||||||||||||
![]() |
|||||||||||||||||||||||||||||||||||||||||||
![]() APAR status Closed as program error. Error description When serverStartupSyncEnabled=true is set in a node, and the nodeagent monitoring policy is set to automatically start servers it governs, systems under heavy stress may see the servers which should be started automatically fail to do so. Heavy load may cause the synchronization process to take longer than the code waits. After the synchronization failure, the process designated to start the servers quits, without attempting any of the monitored servers.Local fix Problem summary **************************************************************** * USERS AFFECTED: All users of WebSphere Application Server * * V5.0 for z/OS * **************************************************************** * PROBLEM DESCRIPTION: During high system stress conditions, * * a Node Agent restart does not start * * Application Servers even though the * * monitoring policy is set to start. * **************************************************************** * RECOMMENDATION: * **************************************************************** A Node Agent has serverStartupSyncEnabled=true and the monitoring policy is set to automatically start the servers it governs. Under heavy stress automatic server start fails. In addition, the following exception error will appear in the Node Agent system log - FunctionName: com.ibm.ws.management.sync.NodeSync SourceId: com.ibm.ws.management.sync.NodeSync Category: ERROR ExtendedMessage: ADMS0012E: Unable to create administrative client connection: com.ibm.websphere.management. exception.AdminException: Admin client connection to deployment manager is unavailable at com.ibm.ws.management.sync.NodeSync.getAdminClient (NodeSync.java:461) at com.ibm.ws.management.sync.NodeSync.getCellRepositoryEpoch (NodeSync.java:339) at com.ibm.ws.management.sync.NodeSyncTask.doSync (NodeSyncTask.java:231) at com.ibm.ws.management.sync.NodeSyncTask.run (NodeSyncTask.java:149) at java.lang.Thread.run(Thread.java:513) ---- Begin backtrace for nested exception ...websphere.management.exception.ConnectorNotAvailableException at com.ibm.ws.management.RoutingTable.getParent (RoutingTable.java:160) at com.ibm.ws.management.sync.NodeSync.getAdminClient (NodeSync.java:452) at com.ibm.ws.management.sync.NodeSync.getCellRepositoryEpoch (NodeSync.java:339) at com.ibm.ws.management.sync.NodeSyncTask.doSync (NodeSyncTask.java:231) at com.ibm.ws.management.sync.NodeSyncTask.run (NodeSyncTask.java:149) at java.lang.Thread.run(Thread.java:513)Problem conclusion This problem is the result of a timing issue where the first node sync occurs before the Deployment Manager is discovered by the Node Agent. APAR PQ99401 eliminated previous occurrences of this problem by adding up to a 30 second delay for the Deployment Manager discovery to occur before attempting the first node sync. This APAR changes this delay to 2 minutes and also starts the servers even if the first node sync fails. APAR PK10725 is associated with SERVICE LEVEL W502034 of WebSphere Application Server V5.0 for z/OS.Temporary fix Comments
APAR is sysrouted FROM one or more of the following: APAR is sysrouted TO one or more of the following: PK10792 PK10793 Modules/Macros
Publications Referenced
|
Document Information |
Current web document: swg1PK10725.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 #: PK10725
IBM Group: Software Group
Modified date: Oct 4, 2005
(C) Copyright IBM Corporation 2000, 2009. All Rights Reserved.