Re: [v6ops] Routers are hosts too!

[t.petch <[e-mail]>]

----- Original Message -----
From: <[e-mail]>
To: <[e-mail]>
Cc: <[e-mail]>; <[e-mail]>; <[e-mail]>
Sent: Sunday, January 10, 2016 2:07 PM
> > > But then RFC1122 does not mention classless addressing or Variable
> > > Length Subnet Masks - as I recall, they did not exist at that
time.
> > > Rather, RFC1122 deals with 32 bit Host addresses and Classful
Networks,
> > > ie 8 or 16 or 24 bit addresses, so while you might have more than
one
> > > entry in the cache for a given destination, they will all be the
same
> > > length so the concept of longest-prefix (and of course, the term
prefix
> > > has no meaning at that date) match does not come into
consideration.
> >
> > DO you mean that the longest-prefix match operation was not there
since
> > the earliest days of Internet?
>
> Reading up on the history of the Internet is recommended. See for
> instance
>
> https://en.wikipedia.org/wiki/Classless_Inter-Domain_Routing

Well, may be:-)

When I got involved with TCP/IP in 1992, subnets were fixed length
across any one network and longest match does not figure in any
description of routing at that time (Perlman, Interconnections, 1992;
IBM, TCP/IP Tutorial, 1989).  Before then, subnets did not exist - they
first appeared in RFC in 1984 (RFC950, 1985) so routing was purely based
on the network address, Class A, B, C.

Even in 1992, commercial products may not have had subnet support so
when designing an IP network, I had to check vendor literature for
subnet support before using it in a design.

Variable length subnets meant that the routing protocols had to carry
the length of the mask with the advertisement and you see this in RIPv2
(RFC1388, 1993) and this, coupled with CIDR, introduced me to the idea
of longest match in the forwarding algorithm.  I know of that from
RFC1349 (1992) and it will have been around earlier, but not by much -
as I say, it would have been alien to most at that time, certainly in
the commercial world, as opposed to the more academic/research oriented
IETF.

Does the history matter?  Well yes, if you are citing documents such as
RFC written at that time, ones which have not been superseded or updated
in that regard, such as RFC1122.  Host versus Router is one such, but
another is gateway (the term router is relatively recent if you look at
the RFC series as a whole) while the definition of the Network layer
generates divergent opinions on some IETF list; and then, when did the
Internet start?  April 1995 is my date, but I recognise that that
depends on your definition of Internet.  Best practice is to put in, or
cite, a definition so others are clear what you (think:-) you are
talking about.

Tom Petch

> Steinar Haug, AS2116