[1800] in linux-net channel archive
Re: Need help to connect to ISP!
daemon@ATHENA.MIT.EDU (Ed Carp @ TSSUN5)
Fri Feb 2 15:59:54 1996
Date: Fri, 2 Feb 1996 11:10:39 -0600
From: ecarp@tssun5.dsccc.com (Ed Carp @ TSSUN5)
To: ecarp@netcom.com, tfries@umr.edu
Cc: abel@netvision.net.il, hjstein@bogart.nnt.com, linux-net@vger.rutgers.edu
> From erc@dal1820.computek.net Fri Feb 2 11:06 CST 1996
> Subject: Re: Need help to connect to ISP!
> To: ecarp@netcom.com
> Date: Fri, 2 Feb 1996 10:47:45 -0600 (CST)
> Cc: abel@netvision.net.il, hjstein@bogart.nnt.com, linux-net@vger.rutgers.edu
> From: tfries@umr.edu (Todd Fries)
> Mime-Version: 1.0
> Content-Transfer-Encoding: 7bit
>
> > > But what do you do about the fact that the IP address is dynamic? If
> >
> > If the IP address is dynamic, you're pretty much sunk. But that wasn't
> > my point - my point was that you only need one static IP, rather than three.
>
> Huh? I'm successfully masquerading anyone who plugs into my network over
> a dynamic ip address. Please explain.
I think we are talking apples and oranges here. If their IP address is dynamic,
IN THE ABSENSE OF ANYONE RUNNING SOCKS OR DOING IP MASQUERADING, they are sunk.
> All they know is their dos telnet program telnets to the unix machine on
> campus and anywhere else, and the win95 machine is able to use netscape,
> MS browser, telnet, ftp, etc... just fine, no configuration involved other
> than making my machine (internal network ip) the gateway.
>
> Would someone please explain to me what the benifit of socks is over
> masquerading?
SOCKS runs on the earlier kernels, and you don't have to patch and rebuild your
kernel. With a production box, this is important, especially in a heterogeneous
environment, where you might not be able to do IP masquerade with a non-linux
kernel.
