[Hipsec] Rendezvous Server Concerns

pekka.nikander@nomadiclab.com (Pekka Nikander) Thu, 03 February 2005 14:50 UTC

From: pekka.nikander@nomadiclab.com
Date: Thu, 03 Feb 2005 14:50:00 +0000
Subject: [Hipsec] Rendezvous Server Concerns
In-Reply-To: <42027267.5090503@mitre.org>
References: <42027267.5090503@mitre.org>
Message-ID: <32ff0289f98cec5dfdbfd350e021043e@nomadiclab.com>
X-Date: Thu Feb 3 14:50:00 2005

> It seems that this approach will prevent a mobile HIP node from being
> reachable by *all* non-HIP nodes, whenever it moves and its current IP
> address is not the IP address stored in DNS. This seems like an
> unacceptable outcome.

Nothing prevents you from using some non-HIP mechanism to remain
reachable by non-HIP nodes.  For example, if you can arrange one
of your rendezvous servers (or some other node) to forward packets
to you and packets sent by you, i.e. what is sometimes called a
HIP proxy, you can certainly make your A / AAAA record to point
to that node.  But that is beyond the scope of core HIP, and
beyond the scope of the current WG charter.

Two questions:

- Have you read what the architecture draft says about rendezvous?
   Does that make sense to you?

   http://www.ietf.org/internet-drafts/draft-ietf-hip-arch-02.txt

- Are you familiar with more advanced rendezvous concepts, like
   Hi3 or rendezvous-based NAT traversal?

   http://www.watersprings.org/pub/id/draft-nikander-hiprg-hi3-00.txt

Note that the advanced rendezvous issues are being discussed
at the *Research* Group side, as they go beyond the current WG
charter.

--Pekka Nikander