This may be stupid, but I can't seem to find any info that indicates what this particular library is used for. At first I thought I had to have this on my java swing client classpath in order for my app to load balance to 2 jboss nodes. However, we don't have it there and the load balancing seems to work. From the docs, I am guessing the info on the smart proxy stuff is working, meaning that when my client first connects to any clustered jboss node, that node sends back a list of all nodes as well as the load balance policy. Is this correct?
If so, then what is jbossha-client.jar for? Is it to build your own load balancer for jboss nodes? Can you use this library to gather node information, and thus provide your own load balancer front end? Please explain if someone knows... thanks.
It is for the load-balancing for fat java client, if you don't have it and it works, then maybe you should check if you are REALLY using clustering right now.
For example, create a dummy exception on the server (in one of your bean method) and look at the stack trace generated on the client: either post it here or look if there is some kind of HAInvoker in it.