Re: IPv6 only host NAT64 requirements?

On 11/21/17, 4:59 AM, "ipv6 on behalf of Manfredi, Albert E"
<ipv6-bounces@ietf.org on behalf of albert.e.manfredi@boeing.com> wrote:

>-----Original Message-----
>From: Jen Linkova [mailto:furry13@gmail.com]
>
>> People from the real world is coming to IETF saying 'we are deploying
>> Ipv6-only hosts' [1] [2]
>
>References:
>[1[ 
>https://datatracker.ietf.org/meeting/100/materials/slides-100-v6ops-sessa-
>ipv6-only-deployment-at-cisco/
>[2] https://datatracker.ietf.org/meeting/99/materials/agenda-99-v6ops/
>
>Okay. So, people can follow whatever approach they prefer, but my first
>reaction was that switching over to IPv6 *by location* seems an odd way
>to go about it. If it were up to me, I'd say not to use that model. It's
>more a matter of end to end functions used, by hosts in a given subnet.
>If a given subnet can demonstrate that all of its hosts' needs can be
>met, end to end, with IPv6 only, then great. Shut off IPv4 in that one
>subnet. But "end to end" sessions are frequently not confined to one
>location, and it also seems unlikely that all hosts in all subnets, in a
>given location, have the same needs.

Not sure what your goal is here in complaining about someone else’s
deployment model.

A building on a campus probably is a single subnet. Not sure how you’d
find out what doesn’t work over IPv6 (plus NAT64) without trying it.

>
>The hotel model that Brian offered seems to reflect what should be the
>most generic reality. Diverse devices, diverse needs, location isn't the
>primary factor to be considered. I realize that there's some sort of
>management appeal, to make this a "by location" issue.

Hospitality networks have specific engineering requirements which may not
be generalizable. For one, they have different users and devices daily:
unlike an enterprise network, devices that tech support fixes don’t stay
fixed (new devices come in). A support call to an ISP subscriber might
cost the same or less than a hotel support call, but if the call is
repeated every time there’s a new user, the hotel has a very different
problem. Plus, of course, hotels are pretty much all Wi-Fi, which may not
be true of enterprise or campus or data center networks.

Lee

>
>Bert
>
>--------------------------------------------------------------------
>IETF IPv6 working group mailing list
>ipv6@ietf.org
>Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
>--------------------------------------------------------------------
>