Session information not retrieved when using session persistence with Multirow schema
 Technote (troubleshooting)
 
Problem(Abstract)
When session persistence is enabled with a Multirow schema, the session information is not retrieved when a failure occurs. In a clustered environment with database session persistence, using Multirow schema causes the session failover to not work properly. It appears that the data is gone, but the session ID carries over. When you deselect Multirow schema, the failover works properly. The Multirow schema option is used for storing larger session data.
 
Cause
This is the result of a code defect.
 
Resolving the problem
Initially it was determined that the session object should be serializable for the failover to work successfully. Later it was determined that even when the object is implemented serializable, the session failover works only when you use the getAttribute method, and not when you use the getAttributeName method. It is often desirable to use the getAttributeName method to get the name of the session keys. While Single row schema works fine, Multirow fails.

PQ85011 resolves this issue.

Note: If you use database persistence and set usingMultirow=true, then later switch to m2m, this causes a problem.

When using m2m, usingMultirow must always be set to false. PQ85011 addresses this problem.

 
Related information
Session Persistence
 
 
Cross Reference information
Segment Product Component Platform Version Edition
Application Servers Runtimes for Java Technology Java SDK
 
 


Document Information


Product categories: Software > Application Servers > Distributed Application & Web Servers > WebSphere Application Server > Sessions and Session Management
Operating system(s): Windows
Software version: 5.0.2.4
Software edition:
Reference #: 1165618
IBM Group: Software Group
Modified date: May 10, 2004