[97131] in North American Network Operators' Group
Re: IPv6 Advertisements
daemon@ATHENA.MIT.EDU (Iljitsch van Beijnum)
Sat Jun 2 12:22:58 2007
In-Reply-To: <20070602152526.GB15392@kallisti.us>
Cc: NANOG list <nanog@merit.edu>
From: Iljitsch van Beijnum <iljitsch@muada.com>
Date: Sat, 2 Jun 2007 18:22:01 +0200
To: Ross Vandegrift <ross@kallisti.us>
Errors-To: owner-nanog@merit.edu
On 2-jun-2007, at 17:25, Ross Vandegrift wrote:
> Am I under some misconceptions about IPv6 routing policies here?
There are no IPv6 routing policies. Everyone gets to decide which
prefixes to accept and which to reject on their own. However, unlike
IPv4, there are currently pretty much only two flavors: /32 and /48.
So if you have a /32 and start announcing a bunch of /36s, or you
have a /48 and start announcing /52s, it's likely that at least some
people out there aren't going to accept those. (Nor /48s in blocks
that are carved up as /32.)
So I expect people who are in your position to start requesting
blocks larger than /32 or /48 in order to be able to deaggregate, or
even request multiple independent PI blocks. It will be interesting
to see what this means for the number of PI requests and speed at
which the global IPv6 routing table grows.
It would be nice if rather than fight about how difficult it should
be to occupy a slot in the routing table, with both "too difficult"
and "too easy" having painful consequences, we could work something
out using regional address blocks or something so it's not necessary
for a router on one side of the globe to have all the more specifics
that are only relevant on the opposite side of the globe. Obviously
we don't want metro addressing with mandatory interconnection all
over the place, but common sense suggests that there is some middle
ground where it's possible to have address space that's at least
portable within a certain region, but we get to prune the routing
tables elsewhere.
