I have already tried bumping up the timeouts to very high numbers and that doesnt help. There are no concurrent transactions that are accessing this same fqn. However, as described above, it is a nested (with a requires_new propagation) transaction.The external transaction is suspended when this happens in the internal transaction. This object/fqn is never accessed by the external transaction though. In fact the cache isnt even touched in the external transaction. There is never a problem when the same operation is done in a single transaction (non-nested) View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4250728#... Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&a...