Hi, I'm running a tomcat cluster with in-memory replication, two nodes in fact, with around 15 webapps all of which are being replicated, and I'm slightly surprised by the behaviour of the system when I start both nodes simultaneously up from cold.
Basically as each node starts up it deploys the webapps sat in its webapps directory and apparently tries to find another node to pull the current sessions from, but since both nodes are starting at the same time there are no nodes to pull from and so both nodes appear to sit around waiting for some timeout level of time (I assume for the stateTransferTimeout, which I've left as default 60 seconds). The net result of this is that the system can take 15 minutes to start up from cold....this is very very painful! Obviously I don't want to reduce the stateTransferTimeout because I want a starting node to be able to sync sessions if a single node is re-started. Is there some secret configuration way to do something more sensible, or am I forced to only start one node at a time (which will be slightly tricky to achieve in truth). Mark ________________________________________________________________________ This email has been scanned for all known viruses by the MessageLabs SkyScan service. --------------------------------------------------------------------- To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
