Hi all, There were some discussions at RIPE91 about supporting the following draft. There is now a WG call to adopt the draft, so responses on the GROW mailing list to get the draft adopted would be greatly appreciated: https://mailarchive.ietf.org/arch/browse/grow/?q=draft-romijn-grow-rpsl-regi... For those that don't know the draft, it can be found on the IETF datatacker: https://datatracker.ietf.org/doc/draft-romijn-grow-rpsl-registry-scoped-memb... This draft addresses the ambiguity of references between sets in RPSL, where currently one set includes another by its RPSL primary key. As the same primary key may exist in multiple IRR registries, these references are ambiguous, while the referred objects sometimes have very different contents. The draft adds a new attribute where references to other sets must be scoped to a specific IRR registry. Specific resolving and validation rules ensure that we keep backwards compatibility with older RPSL objects, authoritative servers, and resolvers. At the same time, it provides benefits even with partial deployment With kind regards, James Bensley (he/him) [CompanySignature] Inter..link GmbH | Boxhagener Straße 80, 10245 Berlin, Germany | Managing Directors: Marc Korthaus, Theo Voss | Commercial Register: Amtsgericht Charlottenburg, HRB 138876 | VAT ID: DE281288887 | Email: hello@inter.link<mailto:hello@inter.link> | Web: inter.link<https://inter.link>
Wasn't this ambiguity already solved by introduction of hierarchical AS-SET?
Hello, On 31 Mar 2026, at 12:06, Serhii via routing-wg <routing-wg@ripe.net> wrote
Wasn't this ambiguity already solved by introduction of hierarchical AS-SET?
That was also a good addition, but does not solve this problem. Keys are still not unique among all IRRs. Not all do (or can) validate against an aut-num's maintainer. Hierarchical AS-sets also do not allow graceful upgrades, i.e. you need everyone who uses your as-set to update the reference if you change from AS-EXAMPLE to AS64500:AS-EXAMPLE, and have two in the mean time. There is no enforcement of their consistency during this time. It might be useful for me and James to gather some specific numbers on all this, but that will take a bit of time. Sasha
participants (3)
-
James Bensley -
lists@at.encryp.ch -
Sasha Romijn