Hi , Thanks for your suggestion. As you said, I have set fetchpersistantstate to FALSE. Now the records are NOT missing in the database during portal startup failure. I want to know by setting fetchpersistentstate to FALSE, will this impact on cache clustering. Please validate my understanding 1. I want to know the difference between fetchpersistentstate and fetchinmemorystate. My understanding is below (i). if fetchinmemorystate is TRUE - It acquires the initial state from the existing members in the cluster during startup (ii). if fetchpersistent state is TRUE - Reads the data from database and loads the data into its cache (iii). if fetchinmemorystate is TRUE and if fetchpersistent state is TRUE a.it acquires the persistent state from the existing member in the cluster b.Deletes the record(s) from the DB. c.Persist the data fetched from the existing member in the cluster to DB d.Retrieves the data from DB and reloads the data into its cache Is my understanding on the above flow is correct? 2. Can we set both fetchpersistentstate and fetchinmemorystate to TRUE. If we can, please explain how it will work and the need(advantage) behind this setting. 3. What is the impact on cache clustering if fetchpersistentstate is set to FALSE? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4141578#... Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...