Re: [v6ops] draft-ietf-v6ops-ula-usage-recommendations - work or abandon?

[Owen DeLong <[e-mail]>]

> On Nov 4, 2015, at 4:52 PM, Gert Doering <[e-mail]> wrote:
> 
> Hi,
> 
> On Wed, Nov 04, 2015 at 04:37:44PM -0300, Owen DeLong wrote:
>>>> I actually agree with what David says below. However, I???d like to make sure that EVERYWHERE we document any form of IPv6 NAT, we also document that it is considered harmful.
>>> 
>>> "That some people in IETF consider it harmful"
>>> 
>>> Please do not speak for me.
>> 
>> Are you really claiming you can make a case that deployment of NAT is 
>> not harmful? I???m very skeptical that such a case can be made.
> 
> Widen your mind :-) - and indeed, I am making this case.  Multiple
> different scenarios where NAT is making life much easier (or enabling
> a solution at all that might not be otherwise possible).

In IPv6? Such as?

Even in IPv4, each and every one of those cases is actually harmful. It may also come with some  benefits which some people may believe outweigh the harm, but denying that it is also harmful is misguided at best and deceitful at worst.

> And none of this is the trivial case of conflicting RFC1918 IP address 
> usage in enterprise VPN scenarios.

Which doesn’t apply to IPv6. Note that my specific statement was about IPv6 NAT.

> Or the case of destination NAT used for load-balancing...  (still NAT,
> even if not N:1 source NAPT).

Unnecessary in IPv6 and not entirely harmless.

Owen