Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!ub!csn!boulder!recnews
From: satz@cisco.com (Greg Satz)
Newsgroups: comp.dcom.sys.cisco
Subject: Re: Router Discovery Protocol
Message-ID: <9106232155.AA13655@wolf.cisco.com>
Date: 23 Jun 91 20:57:55 GMT
Sender: news@colorado.edu
Organization: University of Colorado, Boulder
Lines: 15
In-Reply-To: Your message of "Sun, 23 Jun 91 23:03:22 +0100."
             <9106232103.AA14034@geveke.nl>
Return-Path: <cisco-request@spot.Colorado.EDU>

You mention that MTU discovery can take a short cut when used in
conjunction with cisco routers running IGRP as the information is present
in the closest router. This is true. Unfortunately the world is not made up
of cisco routers alone. There is no guarantee nor is it probable that the
complete path (or set of paths) will traverse a single vendor's router. The
IETF implemented a generic MTU discovery mechanism that will work
regardless of routing protocol and vendor. This is a requirement of the
IETF. The MTU discovery working group was sensitive to offered load of
their protocol and documented when it should be used to minimize wasted
bandwidth.

What you point out is an enhancement we could make as a value added feature
in a future release of our software.

Greg