[Idr] Re: Publication has been requested for draft-ietf-idr-deprecate-as-set-confed-set-16

[Robert Raszuk <[e-mail]>]

Hi Jeff,

So when you aggregate you recommend to lose originating ASNs ?

What if those aggregatable PI addresses have nothing in common with address
blocks of the upstream ISPs ?

This entire draft looks to me like recommendation to duck tape indicator
lights in cars only because today's cops have issues in understanding them
and some (even most) users don't know how to use them properly ?

Thx,
R.


On Mon, Jan 27, 2025 at 9:45 PM Jeffrey Haas <[e-mail]> wrote:

>
>
> > On Jan 27, 2025, at 2:52 PM, Robert Raszuk <[e-mail]> wrote:
> >
> > All,
> >
> > Imagine it is 2040 and magic happened and IPv6 really got popular ... So
> out of the blue there are millions of stub networks getting IPv6 PI blocks
> as well their own ASNs.
> >
> > Those end networks peer with more then one ISP and dynamically advertise
> their blocks.
> >
> > Tell me how upstream ISPs could aggregate those announcements ?
>
> They would aggregate them as they would today.  This draft doesn't get rid
> of aggregation, only impacts the generation of AS_SETS as part of the
> procedure.
>
> If you have some peculiar impression the draft gets rid of aggregation,
> feel free to flag the text that made you think that.
>
> -- Jeff
>
> P.S. I don't have to imagine it's 2040. I can imagine it's today and
> someone's SRv6 SID zoo is trying to escape into the global table. :-)
>
>
>