Re: IPv6 only host NAT64 requirements?

> On 21 Nov 2017, at 12:01 pm, Brian E Carpenter <brian.e.carpenter@gmail.com> wrote:
> 
> On 21/11/2017 12:53, Ole Troan wrote:
>> Brian,
>> 
>>>>> One thing that we demonstrated very convincingly last week
>>>>> is that there are enough legacy devices and applications in
>>>>> existing user hosts that the co-existence plan is still needed.
>>>>> (And, BTW, we demonstated it for user hosts belonging to
>>>>> relatively sophisticated users.)
>>>> 
>>>> Huh?
>>>> Data please.
>>> 
>>> Well, I assume that Jen et al will produce a summary report, but
>>> the number of tickets at https://tickets.meeting.ietf.org/wiki/nat64
>>> tells me that ietf-nat64 did not provide a smooth co-existence
>>> experience for everybody. Imagine that each of those tickets equates
>>> to one support call per week, multiplied by the number of hotels
>>> in the world.
>> 
>> The purpose of the experiment was to get some real data.
>> Before jumping to conclusions let's get the report. No-one thought this would work without a single flaw.
>> And this experiment is not about providing IPv6 only service to hotels, it is about providing IPv6 only + NAT64 to the IETF network.
>> 
>> 32 tickets. Where people were encouraged to report anything. You cannot just assume those are real issues without looking at them.
>> E.g. you can discount 1151, 1175, 1185, 1160, 1174, 1172, 1159 straight away.
>> Some tickets may contain multiple issues and so on.
>> 
>> Looks like we're left with a handful of applications and some OS issues. We're not talking rocket science to get those fixed.
> 
> Sure, and we should try. But we can't fix legacy hosts in the hands of the general public.

But we can send feed back to the vendors of the products they are using that the product they are shipping does not work in this environment for this reason.  More and more of their clients will be behind some sort of IPv6 as a service and there products should work in that environment.  Eventually the fixed will get deployed.

I also think it is a myth that we tell ourselves that hosts won’t get updated.  They get updated all the time provided the vendor of the product puts out a fix.  Its getting the vendor to publish the fix that is the hard part.

Mark

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

-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742              INTERNET: marka@isc.org