users@glassfish.java.net

Clustering Concepts

From: <glassfish_at_javadesktop.org>
Date: Mon, 14 Jul 2008 11:41:53 PDT

Hello,

I am a newbie to GF and I am not very clear on certain clustering related concepts.

- I want to setup a cluster spanning server instances that are running on 2 machines. This is what I have done so far.

1. Installed GFv2UR2 on both these machines. Use the setup-cluster.xml script to start a domain (with clustering enabled) on each of these machines. The domain is named "domain1".

2. Started node-agents on each of these machines (example: NA-MAC1, NA-MAC2)

3. Start 2 instances per machine managed by those node agents.

- Now, this is where my understanding goes out of the window. How do I start a single cluster that spans all these instances (2instances x 2machines)?

- I want to deploy few EJBs on the cluster and I want to lookup these EJBs over JNDI from a stand alone Java client. Is there a way by which I could use the connection parameters of the "cluster" for the JNDI lookup and the EJB instances get created on one of the 4 instances?

Any help would be appreciated...

Thanks,

Shreyas
[Message sent by forum member 'sshinde2' (sshinde2)]

http://forums.java.net/jive/thread.jspa?messageID=286522