Hi,
Lucian Bargaoanu wrote:
>
> I've read about the bug in JBoss (reported as item #424287) some time ago
> and I thought that it might be the cause.
To me, it looks like that.
> But I was troubled by the fact that it didn't work even when I've set
> java.rmi.server.codebase to the path
I'm lost. Why do you need java.rmi.server.codebase?
> of the jar with the bean on the first JBoss. A file:// path which is in the
> local network so it is reachable from the second JBoss.
ok, but file:// path doesn't include the host/machine name. Even if all
filesystems are mounted together, the local and network paths are
different.
Don't I understand something.
>
> If you're not bored yet, the bigger picture.
> A client makes a request for a bean to a JBoss instance ( through a custom
> JNDI ). This one sends a mobile agent to
> a few JBoss fellows to ask them about this name. The agent returns with a
> list of servers wich have the name and
> server load information. The first JBoss chooses a server based on the load
> information and does a lookup to
> return the reference to the client ( through the custom JNDI ). I cannot
> have one JBoss instance knowing about
> all the home interfaces of all its JBoss fellows.
Aga. I got the idea. We were doing similar tasks. But maybe it is the
problem
in your "custom" jndi provider?
regards,
--
__________________________________________________
Alexander Kogan PTC www.ptc.com
[EMAIL PROTECTED] 140 Kendrick St. Needham MA 02494
_______________________________________________
JBoss-user mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/jboss-user