[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: IPv6 multihoming

That would not provide multihoming as much as provide direct access to the whole internet in v6 format to anyone deciding to switch to v6 which would be a major incentive for a lot of end-users and enterprise clients instead of dealing with the few existing islands of v6 and running translation to reach the v4 internet. Also I doubt that its a valid statement to say that an organization receiving a /48 will still need to announce its smaller v4 blocks once it starts migrating the end customers to the /48 v6 range. I know several organization currently announcing a bunch of /24s and /22s in v4 which could easily be replaced by a single /48 in v6.


BR,
Vlad

On Mon, Jan 25, 2010 at 11:34 PM, Brian E Carpenter <brian.e.carpenter@gmail.com> wrote:
On 2010-01-26 04:54, Andy Davidson wrote:
> On 25/01/2010 08:55, Vlad Ion wrote:
>> I propose that the 6to4 ip conversion space from ipv4 addresses to
>> 2002::ipv6 space will be redefined as provider independent address
>> space. This way whoever wants to implement ipv6 with multi-homing can
>> simply redefine their existing IPv4 addresses in IPv6 6to4 format and
>> have multi-homing in ipv6.
>
> Sorry, this scares me.

Well yes. Let's try a thought experiment that doesn't break 6to4:

Proclaim that everyone who has a valid global IPv4 prefix owns
the PI IPv6 prefix 4444:V4ADDR::/48.

Then we have automatically imported all 318333 IPv4 prefixes
(according to routeviews) into IPv6.

I'm not quite sure how this would provide multihoming.

   Brian


> It is not difficult to get IPv6 PI from the RIRs I have experience with.
>
> For example, in Europe, obtaining a single /48 IPv6 PI is a quick
> process.  Obtaining it in this way means that unused v4 is not recycled
> as spoofable v6, and that organisations with tens of v4 unjoined
> prefixes need not announced tens of unjoined v6 prefixes when they migrate.
>
> I commend you for thinking about ways to encourage networks to adopt v6,
> but I think that education and advocacy is more future-proof than
> migrating the v4 swap to v6.
>
> Andy
>
>