[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2522?page=c... ] prasenjit B commented on HHH-2522: ---------------------------------- Hi , I have seen your post on Cache Provider implementation for IBM WebSphere's dynamic cache I am new to hibernate and currently working on one engagement with WAS 6.1. The enterprise application is built on WAS 6.1 cluster (horizontal) with two nodes and the Hibernate 3.2.5 is used for the persistence .We have tried CMTTransactionFactory and webSphereExtendedJTATransation factory initially but the save and update operations are throwing some challenges as the Container managed transaction is not been synchronized with the persistence layer and we are getting an error "Could not register synchronization for container transaction" from Hibernate. However the following configuration is supported by the application environment. <session-factory> <property name="hibernate.session_factory_name"> HibernateSessionFactory </property> <property name="hibernate.transaction.factory_class"> org.hibernate.transaction.JTATransactionFactory </property> <property name="hibernate.transaction.flush_before_completion">true</property> <property name="org.hibernate.transaction.JTATransactionFactory"> org.hibernate.transaction.WebSphereTransactionManagerLookup </property> <property name="hibernate.transaction.manager_lookup_class"> org.hibernate.transaction.WebSphereTransactionManagerLookup </property> The configuration was OK in the stand-alone server in the development .But when we deploy the same in the cluster ,the application started showing the old values/transaction details , although the individual data elements are updated in the back-end Database without any problem. According to me ,we might be facing some caching related issues as we have not configured the second level cache for Hibernate. Your article has given me a pointer to resolve the issues.For this implementation I have changed the package from package org.hibernate.cache to com.ibm.hb.cache .would there be any problem? I would therefore be grateful if you let me know what change do I need to do for the HBM and the CFG files. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2522?page=c... ] prasenjit B commented on HHH-2522: ---------------------------------- Thanks Joris, At this point I have two Concerns 1.We can not use CMTTransactionFactory and webSphereExtendedJTATransation factory for the configuration.If we do that we get error with synchronization issues with Save or Update operations.We have used .However using org.hibernate.transaction.WebSphereTransactionManagerLookup and org.hibernate.transaction.JTATransactionFactory ,we could eleminate such issues.will there be any problem related to the performance if we don't use the former combination?If yes ,How do we eleminate that? 2.I would like to implement the 2nd level cache to check if such problems are eleminated.To start with ,I have included two classes of yours in my application.What chages do I need to do for the Hibernate config and the hbms to to include the 2nd level cache support using WebSphere's dynamic cache ? Regards -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira

[ http://opensource.atlassian.com/projects/hibernate/browse/HHH-2522?page=c... ] Mike Youngstrom commented on HHH-2522: -------------------------------------- I noticed you didn't implement the lock and unlock methods. Was this because Websphere's dynamic cache doesn't provide a locking facility? Or is a locking facility not needed with dynamic cache? I'm trying to determine if this cache implementation is going to run into problems with concurrent access. -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: http://opensource.atlassian.com/projects/hibernate/secure/Administrators.... - For more information on JIRA, see: http://www.atlassian.com/software/jira