[IPv6]Re: Fwd: New Version Notification for draft-herbert-deprecate-eh-00.txt

[Brian E Carpenter <[e-mail]>]

On 29-Dec-25 22:33, Ole Trøan wrote:
> 
>>
>>> As I final thought, I would be reluctant to support this draft until the IETF has a Standards Track definition of a limited domain.
>>
>> It would be quite easy to edit section 1 of RFC 8799 into a formal definition, although it seems to me that "limited domain" is pretty much self-defining. There's also some good text in draft-wkumari-intarea-safe-limited-domains.
> 
> Let me suggest a requirement.
> When Internet traffic transits a limited domain, the domain MUST forward packets transparently, independent of the IP protocol type.

The important word in that sentence is "transits". I agree that if you're offering transit service, you should be completely transparent. I hope there's nothing in RFC 8799 that suggests otherwise.

But I will quote two sentences from the RFC:

"For example, the Internet is not transparent to IPv6 extension headers [RFC7872], and Path MTU Discovery has been unreliable for many years [RFC2923] [RFC4821]. IP fragmentation is also unreliable [FRAG-FRAGILE], and problems in TCP MSS negotiation have been reported [IPV6-USE-MINMTU]."

That's the problem, right there.
  
> Which is why the limited domain protocols like SRv6 uses encapsulation.
> If we disagree on that I am not sure if we’re the _Internet_ ETF anymore?

My concern is that by *not* recognizing that limited domains exist, we are making the transparency problem worse.

Regards
    Brian

> 
> Ole
>