|Another one I just found: invocation of finder methods now require that
|all cached instances are synched with the db. This is all fine in a
|non-clustered environment.. but what happens in a cluster!?!? Client
|calls container, the container must then tell all nodes to synch their
|cached instances to the db, and then the call can proceed. For each
|finder invocation you will hence get N-1 more calls where N is the
|number of nodes in the cluster. The only way to avoid this is to make
|sure that once a tx comes to a container it stays there. We were gonna
|do this anyway, but I know some other servers (Hi Cedric! :-) that allow
|the tx to move into other nodes. Which truly messes things up now.
you probably mean bean<->tx in one container. I.e. that for a given bean
and a given tx there is only one container.
but "all tx in one container" isn't realistic
marc
|
|/Rickard
|
|--
|Rickard �berg
|
|Email: [EMAIL PROTECTED]
|http://www.telkel.com
|http://www.jboss.org
|http://www.dreambean.com
|
|
|--
|--------------------------------------------------------------
|To subscribe: [EMAIL PROTECTED]
|To unsubscribe: [EMAIL PROTECTED]
|Problems?: [EMAIL PROTECTED]
|
|
--
--------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Problems?: [EMAIL PROTECTED]
