Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!ucsd!ucbvax!bloom-beacon!LARRY.MCRCIM.MCGILL.EDU!mouse
From: mouse@LARRY.MCRCIM.MCGILL.EDU
Newsgroups: comp.windows.x
Subject: Re:  The client of a client?
Message-ID: <9008100739.AA28303@Larry.McRCIM.McGill.EDU>
Date: 10 Aug 90 07:39:14 GMT
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Lines: 34

> I am logged on to a machine A (a Sun running X windows).  I need to
> connect to a machine B from A.  On B I have a process running and
> this process will display images on the server A.

> My problem is that I am unable to log into B directly from A.
> However, I can log into B indirectly through another machine C (that
> is, from A log on to C, and from C log on to B).  But this causes
> problems since machine B now cannot connect directly to A for
> displaying the images.

> My question: Is there anyway of informing machine B that A can be
> reached through C?  Machine C happens to be a VAX in my case - would
> this make a difference?

If all the machines in question speak TCP/IP, it's perfectly feasible
to have a small dummy program on C which accepts the connection from B
and in turn connects to A, then just passes bytes back and forth.
There is a program called xscope[%] available which was designed to
debug protocol problems, but can be turned to this use, or someone used
to writing network applications can toss a specialized program together
in short order.

If you have to deal with differing or other protocols, the problem may
be harder, though a similar approach should be workable.

[%] There is another program called xscope available that does
    something else unrelated - something oscilliscope-like, I think.  I
    don't know where either sort of xscope can be obtained, except
    presumably for the usual archive sites.

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu