Re: Routing IPv4 with OSPFv3

Naiming Shen <naiming@REDBACK.COM> Mon, 19 August 2002 17:45 UTC

Received: from cherry.ease.lsoft.com (cherry.ease.lsoft.com [209.119.0.109]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA10756 for <ospf-archive@LISTS.IETF.ORG>; Mon, 19 Aug 2002 13:45:41 -0400 (EDT)
Received: from walnut (209.119.0.61) by cherry.ease.lsoft.com (LSMTP for Digital Unix v1.1b) with SMTP id <3.006DFA28@cherry.ease.lsoft.com>; Mon, 19 Aug 2002 13:47:02 -0400
Received: from DISCUSS.MICROSOFT.COM by DISCUSS.MICROSOFT.COM (LISTSERV-TCP/IP release 1.8e) with spool id 127535 for OSPF@DISCUSS.MICROSOFT.COM; Mon, 19 Aug 2002 13:46:58 -0400
Received: from 155.53.12.9 by WALNUT.EASE.LSOFT.COM (SMTPL release 1.0f) with TCP; Mon, 19 Aug 2002 13:46:57 -0400
Received: from redback.com (yoo-hoo.redback.com [155.53.12.43]) by prattle.redback.com (Postfix) with ESMTP id 0A9A839B5A6; Mon, 19 Aug 2002 10:47:00 -0700 (PDT)
Message-ID: <20020819174700.0A9A839B5A6@prattle.redback.com>
Date: Mon, 19 Aug 2002 10:46:59 -0700
Reply-To: Mailing List <OSPF@DISCUSS.MICROSOFT.COM>
Sender: Mailing List <OSPF@DISCUSS.MICROSOFT.COM>
From: Naiming Shen <naiming@REDBACK.COM>
Subject: Re: Routing IPv4 with OSPFv3
To: OSPF@DISCUSS.MICROSOFT.COM
In-Reply-To: Mail from Vincent Jardin <jardin@6WIND.COM> dated Mon, 19 Aug 2002 11:45:39 +0200 <3D60BE42.2B403ECD@6wind.com>
Precedence: list

 ] Arthur Dimitrelis a écrit :
 ] >
 ] > Greetings,
 ] >
 ] > I'm interested in people's opinions and experiences in using OSPFv3 (aka
 ] > OSPF for IPv6) for routing IPv4.
 ] >
 ] > My understanding from reading the OSPFv3 specification (RFC2740) is that
 ] > OSPFv3 is designed with a degree of protocol independence. By this I
 ] > mean that you could, in principle, use OSPFv3 to route any protocol
 ] > family you chose, just so long as you were able to map addressing
 ] > information to the topology (and of course your OSPF code knew how to
 ] > set up the correct forwarding state for your given protocol family). The
 ] > topology-to-address mapping for IPv6 is done using OSPFv3's
 ] > Intra-area-prefix-LSAs, but the spec makes no mention of how you might
 ] > map IPv4 addressing information to a network topology.
 ] >
 ] > So, my questions are:
 ] > - Regarding OSPFv3 - is the bulk of it protocol agnostic, or is it just
 ] > my imagination? Was it the intention of the protocol authors to create a
 ] > protocol that could easily route multiple protocol families, or just
 ] > IPv6?
 ] > - Is there any interest out there in using OSPFv3 to route IPv4?
 ]
 ] According to me, if there is a single routing protocol for both IPv4 and
 ] IPv6, the routers that support only an IPvX could be detected. Some
 ] tunnels could be used to route through them. The endpoints of these
 ] tunnels could be the routers that support both IPv4 and IPv6.

You probably can detect the both ends of the link support IPv4 and
IPv6, but thats usually not enough. You need to know if this is a
IPv6 only link or not. It's ok to tunnel IPv6 to bridge islands, but
it's not ok to also forward IPv4 traffic through the tunnel(assume
the tunnel is for IPv6).

 ]
 ] My understanding of the section 3.8 of the Integrated IS-IS RFC-1195 is
 ] that IS-IS supports this kind of feature that is described by the
 ] ISO/10589.

I don't think current IS-IS spec/implementations support automatic
encapsulation stuff. A "dual" router in ISIS must set the
"protocols supported" field to be identical on every link of the
router. In other words you can not have a router with some links
to be v4 only, some to be v6 only and others are dual.

One proposal to solve this problem is to use "multi topology"
of IS-IS, see draft:

http://www.ietf.org/internet-drafts/draft-ietf-isis-wg-multi-topology-04.txt

thanks.

 ]
 ] However, who needs this repairing function ?
 ]
 ] Regards,
 ]   Vincent

- Naiming