Running Remote Clients

Assume that the computer shown in figure 8 is identified on the network by the hostname sun. Assume also that there is a network connection to another UNIX computer, with X available, with hostname hp.

Given the situation shown in figure 8, with the mouse pointer inside the xterm window, type the following commandsto the prompt:

 
		xhost + hp

		rlogin hp

		xterm -geometry +600+400 -display sun:0.0&

		logout

Figure 9 shows a transcript of the session taking place in the upper left xterm window on sun.

 
Figure 9: Transcript of xterm session

The screen shown in figure 8 should then change to resemble that shown in figure 10. Now there are two xterm client windows displayed on the same server but the one in the lower right window is running on the remote computer, hp, and provides terminal access to hp, NOT sun.

 
Figure 10: Running an xterm client over the network

Now for an explanation of the commands used in figure 9. Before a client running on a remote computer can connect to the server it must have the server's permission to do so. Within the server there is a remote access control list. The client program xhost is used to manipulate remote client access to a server. The command xhost + hp run on sun, connected to the server running on sun and appended hp to the server's remote access control list. The command rlogin hp provided a terminal session on hp, in order to invoke the command xterm -geometry +600+400 -display sun:0.0 & on hp. This command invokes an xterm client, with the window's upper left corner at position (600,400) relative to the screen's coordinate system, and displayed on screen 0 of server 0 on remote computer sun. If the geometry option were omitted, this second xterm window would be drawn directly on top of the first, at the default position (0,0) on the screen, rendering the first one invisible. The & backgrounds the whole command, allowing it to continue running after the user has logged out of hp.