When to follow RFC and when to innovate. [was about auto route-adds]

David (david@killerlabs.com)
Tue, 2 Feb 1999 14:54:48 -0800 (PST)

On Tue, 2 Feb 1999, Fred Reimer wrote:
> I have not been following what this is all about yet, but the short answer
> is no. If it comes down to deciding how things should be implemented we
> should NOT just follow the practices of the router vendors, even if they all
> do the same thing. We should follow the RFCs that describe how IP
> addressing and routing protocols are supposed to work. If vendors implement
> an "option" such as zero-subnet-mask, then we can implement that "option"
> also, but it should not be part of the core implementation.

You must also consider that some of these RFCs mentioned are very antiquated
and don't suit today's world. CIDR was created because of such limitations.

Using a supernet in the antiquated 'Class C' domain is quite acceptable and
necessary. I would -very- much rather add one /17 route instead of 128 class
/24 routes.

Times change, requirements change, old standards need revised. Otherwise
things like routing tables could be one heck of a huge mess.

-d

-- 
  Look, Windows 98  Buy, lemmings, buy!  MCSE, Must Consult Someone Experienced
__ (c) 1998 David Ford.  Redistribution via the Microsoft Network is prohibited
\/  for linux-kernel: please read linux/Documentation/* before posting problems




-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/