Re: [EXTERNAL] Re: Limited Domains:

[Toerless Eckert <[e-mail]>]

On Sat, Apr 17, 2021 at 02:30:52AM +0200, Robert Raszuk wrote:
> Hi Brian,
> 
> 
> > On that argument the SPRING WG should never have been chartered and the
> > 6MAN WG should never have approved RFC8754. Also, we should never have
> > defined diffserv in 1998. And NAT, of course, would be excluded by
> > definition, and RFC1918 from 1996 would need to be obsoleted.
> 
> I beg to differ in the perspective on this.

I am sure Brian is dead serious undoing all the pragmatic workarounds
we did over the decades. NOT.

> I would rather see 6man understanding the reality, understanding that
> hardware engineering and innovation there have moved on and that protecting
> original father's of IPv6 fixed size header wall garden does not make sense
> any more.
> 
> Yet we still see orthodox protection over technically obsolete dogmas.

I think there is just a very justifyable desire not to break stuff
that works in a trillion dollar environment. But i also hope that most
folks by now will recognize that we have already been stretching the
applicability of IPv6 packet headers way beyond its original understood
goal of "IPv4 with longer addresses" and through the learning effort
burned many options (Flow ID) and created a lot of technology debt.

The desire by IPv6 evangelists to make IPv become the one thing that
fits all is well understandable, but when we did introduce an IPv6
version of SR, i think we recognized how much we are stretching what
 IPv6 with its basically 50 year old quite inflexible header and
address design can accommodate.

> As we see in parallel there is equally interesting discussion on use of the
> last 64 bits of IPv6 address happening. This is pretty interesting
> considering that even most conservative folks agree that IPv6 address
> should be assigned to the interfaces. So I define a logical interface which
> is an anchor to internal switching vector and different processing
> function.

All these ongoing proposals to get more out of the IPv6 header remind
me of an old colleague who was moving out of his home after i think
20 years telling me "i have exhausted all possible renovations in this
place. Time to move on". (Hi Steve ! ;-)

And see my other posts wrt. how to maintain all functionality and 
"backward compatibility" (the way i define it).

> Last RFC1918 just talks about address allocation - not sure how is this
> related to any of this discussion ... you mean that the notion that private
> addresses are not routable globally is the key point here ? I thought the
> concern is with router's processing capabilities not with where such
> processing is subject to happen.

scoped addresses are a key element to building limited domains. Repeating
myself, RFC1918 for example spurred a lot of easily built, mostly industrial
networking with IP, just using "generic" 10.0.x.y addresses, same for
every instance of a product. IPv6 has taken that away.  Just one example.

Cheers
    Toerless
> 
> Best,
> R.
> 
> PS. Btw FWIW my interest and former activity in 6man list on SR topics
> where not so much motivated in that I like SR. Much more I oppose hardcore
> resistance to protect something which technology wise is all pretty much
> obsoleted.
> 
> And I fully understand why this is going on like this - to make sure new
> features do not break existing IPv6 world ... it is just that protecting
> something which technically is already addressed and keeping innovation
> gated is IMO not the best strategy for networking.

> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> [e-mail]
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------


-- 
---
[e-mail]