If I set up two WAS servers, says server1 and server2 and created a cluster to include them as cluster members for failover in that if any one server is down, the client request will be redirected to another server. Suppose the configuration has no problem, how can I detect the request is really redirected to another cluster member if I intentionally stop the WAS in one cluster member?
Another question is that the cluster setting is made in a WAS Administration window as stated in the on-line document at IBM website, does it mean that the configuration information is stored in one of the cluster members? If so, is it ture that the whole cluster will fail to server the clients' request if this cluster member is down? May be I was wrong with the concept of cluster provided with IBM WAS, the cluster information is stored in all cluster members, is it correct?
When you stop one server the web plugin redirects the incoming requests to the others nodes. If you work with standalone applications you will have to search again the homes in the JNDI tree.
About the configuration files you are wrong. The configuration files are in the manager, and then it distributes this files to the nodes. But only the files that belongs to the server are distributed. If you have two nodes: server1 and server2, inside the config folder of the server1 you will have the following structure with config files: config/cells/MYCELLNAME/nodes/server1/*
I’ve looked at a lot of different solutions, and in my humble opinion Aspose is the way to go. Here’s the link: http://aspose.com