> But if all the node executes at the same time trying to > select and delete .
The Quartz job runs, sure, but the JdbcLockingStrategy code prevents more than one node from performing cleanup at a time. If another node runs and can't acquire the lock, it immediately exists. > The chances of locks are more right? This database-driven locking code was designed exactly to prevent increased potential for deadlocks when multiple nodes run cleanup. There's a lot of discussion of this both in the issue I cited and the list archives. M -- You are currently subscribed to [email protected] as: [email protected] To unsubscribe, change settings or access archives, see http://www.ja-sig.org/wiki/display/JSG/cas-user
