Re: IPv6 only host NAT64 requirements?

[Alexandre Petrescu <[e-mail]>]

Le 22/11/2017 à 01:01, Brian E Carpenter a écrit :
> in line...
> 
> On 21/11/2017 22:40, Ola Thoresen wrote:
>> On 20. nov. 2017 20:37, Brian E Carpenter wrote:
>>
>>> On 21/11/2017 02:36, Ole Troan wrote:
>>> ...>> [Med] These are generic statements, Ole. We are talking about the IETF case.
>>>>> * The IETF has no control on the hosts that connect to the IETF network,
>>>>> * IETF attendees who are using corporate devices, have no control on these hosts
>>>>>
>>>>> So, how forcing devices to use "IPv6+nat64" will help here?
>>>> Eat own dogfood. Many IETF people are developers or work for companies having applications not working.
>>>> As I said there were a minimum of applications that didn't work. Corporate VPNs largely did. Jen has the final numbers.
>>> However, as long as even one application, such as one VPN, or one
>>> literal IPv4 address, fails, that represents millions of failure cases
>>> if we consider the whole world (e.g. imagine every hotel network in
>>> the world running IPv6+NAT64 only). That simply isn't viable. Dual
>>> stack in every hotel room in the world is viable, from the hotel guests'
>>> point of view. Operators might not like it, but users wouldn't care.
>>
>> In hotel rooms and other "public" or "guest" networks, there are so many
>> things that fail already, due to NAT, misconfigured firewalls,
>> unmaintained blocklists, SSL-proxies and whatnot, so you can hardly
>> expect any services other than basic web-surfing without https to work.
>> Not that this is an ideal situation today, but i do not believe that "if
>> even one VPN or one liter IPv4 address fails" should be a showstopper
>> for introducing this.
> 
> Hotels won't make that choice, but the providers of hotel networks will,
> entirely based on their perception of the number of help desk calls
> they will have to handle for any given change. Since they haven't even
> made the move from IPv4 to dual stack, I think we'll wait a long time
> before they attempt any form of IPv6-only. Sad but true.
> 
>> Possibly, or even probably, introducing IPv6 (-only and NAT64) will even
>> make the situation better for a lot of people, as you get rid of all the
>> horrible NAT solutions for services that are already dual stacked.
> 
> But those horrible NAT solutions are up and running today.

It will be noticed that on NAT64 ESSID the experience is worse than on 
IPv4, at least for VPN sessions.

Alex

  For this
> class of operator, any change is bad news, unfortunately. What is their
> incentive to make such a change?
> 
>      Brian
> 
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> [e-mail]
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------
>