On Thursday 15 September 2011 09:20 AM, David M. Lloyd wrote:
Client JNDI
-----------
Client JNDI is just a simple in-memory JNDI implementation. We will not
automatically bind anything to it at all, except in two possible cases:
1. A simple configuration file which describes what to bind, or 2. A
configuration instructing a binding list to be fetched from somewhere.
I haven't found clear details around this in various JIRAs on what this
actually means to an end user. In previous version(s) of JBoss AS and
other application servers that are available, a remote client (for
example a standalone Java app) works as follows for remote EJB invocations:
1) The (remote) client setup adds some app server specific jars to the
runtime classpath of the client
2) The code uses a app server specific or EJB3.1 spec standardised JNDI
names of EJBs to do a lookup using javax.naming.Context
3) Invokes on the returned bean view proxy.
In these above 3 steps, there is no other configuration or setup
required by the remote client to invoke on the bean using a JNDI name.
So with the Client JNDI that we are proposing in AS7, does this mean,
that in AS7 we are _mandating_ an additional step/instruction to
configure/setup the JNDI on the client side so that the invocation via a
jndi name works? Or is it going to be transparent to the users without
any explicit configuration instruction?
P.S: I am not talking about Java EE "application client" which is much
more than just remote invocation. Setting up a "application client"
always required additional configurations in all application servers
that I know of.
-Jaikiran