JMS Client

[Maruthi Chokkanathan]

Hi All,
I have a doubt in JMS. In JMS document, it is specified that JMS(instead of RMI) can be useful in the case of server lies behind the firewall/NAT .
Is it correct?
If So,How it solves the problem. Because in the weblogic jms thin client documentation, it is specified that JMS uses internally RMI. In the case of JBOSS,JNDI is implemented using RMI. How JMS avoids the firewall/NAT issues?
After some R&D, I came to know that JMS is not the replacement for RMI.It can be considered as top layer over RMI. If so, how (can i exactly use) JMS solves this firewall/Nat issue..

Thanks in advance
Maruthi

[Roger Chung-Wee]

The underlying wire protocol is not specified by the JMS specification. Typically, a JMS client communicates with a JMS provider via an open TCP/IP socket or via HTTP. HTTP tunneling can be enabled to get through a firewall (as every message will be wrapped in HTTP).

[Maruthi Chokkanathan]

Hi Roger,
Thanks for the reply. You are right. But in JMS this should be achieved without the HTTP tunelling. Am I right?
Actual Concept in JMS is, there is no need to know the details of the destination (or source) . All we want to do is to send the details to the JMS Messaging system. It will take care the remaining things.Am I right?
Where they are handling this firewall,NAT issues?
Using HTTP Tunelling for JMS, I can use HTTPINVOKER Layers (which actually act as a wrapper for JNDI/RMI Look Up)know?

Thanks
Maruthi