Re: [IPv6] I-D Action: draft-ietf-6man-eh-limits-02.txt

> -----Original Message-----
> From: ipv6 <ipv6-bounces@ietf.org> On Behalf Of Bob Hinden
> Sent: Thursday, March 02, 2023 11:07 AM
> To: Ole Troan <otroan=40employees.org@dmarc.ietf.org>
> Cc: Bob Hinden <bob.hinden@gmail.com>; Tom Herbert <tom=40herbertland.com@dmarc.ietf.org>; IPv6 List <ipv6@ietf.org>
> Subject: Re: [IPv6] I-D Action: draft-ietf-6man-eh-limits-02.txt
> 
> Ole,
> 
> > On Mar 2, 2023, at 7:58 AM, Ole Troan <otroan=40employees.org@dmarc.ietf.org> wrote:
> >
> > Tom,
> >
> >> It's a classic "chicken and egg problem". Extension headers would on
> >> be deployable if there are real use cases, but we can't develop real
> >> use cases unless they are deployable.
> >
> > There might be an aspect of that, but I don’t think that’s the whole truth.
> > It is quite telling that there isn’t a single universally “interesting” extension header.
> 
> I continue to think that RFC9268 "IPv6 Minimum Path MTU Hop-by-Hop Option” is both interesting.

Yes, and also the augmented version that appears in 'draft-templin-intarea-parcels'.

>
> Bob
> 
> 
> > That’s applicable to intermediate nodes. You could almost generalise it to any node.
> >
> >>> As an implementor I am quite unhappy when headers aren’t at fixed and predictable offsets in the packet.
> >>
> >> As an implementer I am quite unhappy that some vendors and operators
> >> have taken arbitrary liberty to block protocols thereby ossifying the
> >> Internet and otherwise make protocol conformance optional for the
> >> purposes of their financial gain. The effect is that we need to
> >> reverse engineer these implementations and policies to provide
> >> workarounds to continue to give end users the best service and the
> >> benefits of an evolving Internet.
> >>
> >> In any case, the problems and perceptions about extension headers have
> >> been well documented. Now the question is: what is IETF going to do
> >> about it? This draft and hbh-processing are attempting to address at
> >> least part of the problem.
> >
> > It’s not like there is a raft of use cases / solutions waiting for extension headers to magically start working.
> > I’ve come to the conclusion that it’s just the wrong layer. 3.5 isn’t a layer. ;-)
> > Limited domains, sure, there’s some special cases there. Although they could also be resolved by TLVs in tunnel headers.
> >
> > But I’m not going to stop anyone who thinks spending effort on EHs is useful.
> >
> > Best regards,
> > Ole
> > --------------------------------------------------------------------
> > IETF IPv6 working group mailing list
> > ipv6@ietf.org
> > Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> > --------------------------------------------------------------------