[omniORB] potential bug: two ORBs with the same endPoint

[Igor Lautar]

Hi,

Server:
 - using omniINSPOA, endpoint set to giop:tcp:localhost:<port>
 - activate object with such POA and <objectID>

Client:
 - endpoint set to giop:tcp:localhost:<port>
 1) get object reference using corbaloc:
     - corbaloc::localhost:<port>/<objectID>

throws CORBA::OBJECT_NOT_EXIST

 2) get object reference using IOR (IOR:xyz...)

connects(!), but first remote method call throws CORBA::OBJECT_NOT_EXISTS

note that ORB_init returned ok, even though endPoint <port> was already used 
by first (server) ORB.

I would expect ORB_init to fail, if port is already binded. Is there a special 
resaon why this is not done, or is this a bug?

Regards,
Igor

PS. If I change endPoint port in client, everythink works as it should