This is a test of the nested upcall feature in TAO.  The client will
start off by calling the server and providing a callback object and a
time-to-live number.  The server will callback the client (using the
callback object) and decrement the time-to-live number. The client
will turn around and call the server again.

The above process of calling each other will continue until the
time-to-live number reaches zero.

Both client and server can be run in multi-threaded mode.  A
multi-threaded client can be used to check the client side leader
follower model, while the multi-threaded server can be used to check
the integration of the leader follower model of the TP reactor and the
client side leader follower model in the ORB.

In addition, the server can be used with the thread-per-connection
model.