On Tue, Apr 01, 2025 at 08:43:05AM +0200, Tobias Fiebig wrote:
> I know I have a thing for picking stuff like this... but...
> 
> What does the list think about taking a shot at some revisions to
> RFC1930/BCP6? I would argue that this is somewhat in-scope for GROW.
> 
> Specifically, I'd argue that it might make sense to revisit the strict
> multi-homed requirement; As of now, iirc, the only RIR requiring that
> is RIPE.
> 
> It also strictly conflicts with common use-cases of ASNs:
> 
> - IIRC, roughly 1/3rd of ASes in the GRT is single homed

>From the Route collector's (RIPE RIS, RouteViews, etc) perspective,
observing a single upstream doesn't mean that the ASN is single-homed.
BGP has the curious property that as distance increases, visibility of
path diversity decreases (due to best path selection).

I don't think we can operate under the assumption that 1/3rd of the ASes
in the global Internet routing system are single-homed.

> - Various anycast use-cases are incompatible with that requirement

Can you elaborate a bit more beyond stating a simple assertion?

> - Technically, IXes are not really compatible with RFC1930 atm either

Can you elaborate a bit more beyond stating a simple assertion?

> Furthermore, we might remove the language on IDRP having been deployed
> by now...
> 
> So... general thoughts on this?

I'm all for updating things that need updating, so don't take the above
as objections.

Kind regards,

Job

_______________________________________________
GROW mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to