Mr. Doering, I could not agree with you more!
This sort of operational nonsense will limit the wider acceptance of IPv6! I
am responsible research and for the documentation and implementation of IPv6
for a Fortune 200 company. We have locations worldwide. The allocation of
unstable end network addresses complicates the deployment and support of IPv6.
Essentially, this means we would need to ask ISPs for a stable IPv6 block and
frequently we run into problems with the ISP not even understanding what we
need even though their network backbone and endpoints are running IPv6! It
took us a long time to be able to obtain a /48 that actually works at our main
datacenter where we already have commercial fiber. Finally, due to our major
contract with the provider we were able to escalate to support engineering that
understood the requirements and got it working. Having ISPs randomly change
the /64 that is allocated to end user networks creates confusion and
operational problems for exactly the people who least understand what is going
on and why there are connectivity problems. End users have become used to
having a stable internal IPv4 address space for decades now we want them to
switch to an unstable IPv6 internal network address space? I don't believe
that ULA multi-homing is a solution either.
-----Original Message-----
From: Gert Doering <[email protected]>
Sent: Thursday, October 24, 2019 11:50 PM
To: Fernando Gont <[email protected]>
Cc: Michael Sturtz <[email protected]>; [email protected]
Subject: Re: ipv6-ops Digest, Vol 159, Issue 1
Hi,
On Thu, Oct 24, 2019 at 09:02:44AM -0300, Fernando Gont wrote:
> As noted in the draft, the renumbered home network is one of many
> possible scenarios where the renumbering event occurs. While we can
> certainly recommend stable prefixes, I do think that the network
> should be robust in the presence of such events.
Right. This is missing in the whole "only bad ISPs will ever give their
customers prefixes that are not stable" discussing - customers *change* ISPs,
and this should be as painless as it is in the IPv4+NAT world.
Thus, anything relying or implicitly assuming "IPv6 addresses are stable"
(in an unmanaged SoHo network) must be very much discouraged.
Maybe even dual-/48 multihoming can be made to work one day (and no, it is not
even working well in theory today, but even less so in practice with the CPE
implementations you can buy today).
Gert Doering
-- Operator
--
have you enabled IPv6 on something today...?
SpaceNet AG Vorstand: Sebastian v. Bomhard, Michael Emmer
Joseph-Dollinger-Bogen 14 Aufsichtsratsvors.: A. Grundner-Culemann
D-80807 Muenchen HRB: 136055 (AG Muenchen)
Tel: +49 (0)89/32356-444 USt-IdNr.: DE813185279
