To help with the WG keeping abreast of issues resolved by the (BRSKI) design team, we will post a summary of issues worked on during the Tuesday meetings. To remind everyone is welcome to join the design team meetings which are most Tuesdays at 11am Eastern at whereby.com/sandelman. There are calendar invites in the archives. There will be design team meeting on Oct.31 or Nov. 7. We will resume on Nov.14, and I will resend an invite for that date.
Who: Toerless, Michael, Steffen, Thomas, Matthias, Marco Calipari. 1. https://github.com/anima-wg/anima-brski-prm/issues/79 discovery of registrar with BRSKI-PRM function set (one realization is that there will be Registrar/Registrar-Agent communication/configuration which likely will go beyond the standards we are defining. We have no experience here, and there will need to be experiments with running code before we can have standardization here. i.e. a Registrar and Registrar-Agent are part of a single product offering) 2. https://github.com/anima-wg/anima-brski-prm/issues/106 "registrar-agent signatures" check consistent use closed with text in section 6.3.6 3. https://github.com/anima-wg/anima-brski-prm/issues/116 Security Considerations - nonce explanation closed, everyone happy with text 4. https://github.com/anima-wg/anima-brski-prm/issues/117 Security Considerations - Misuse of Registrar-Agent Credentials nothing changed in the text, agreed that description was good enough. issue closed. 5. https://github.com/anima-wg/anima-brski-prm/issues/128 "domain/site" terminology definition and consistent usage We had a long discussion about whether or not the term "site" needs to be go into the document. Does it mean the same as "domain" or not? It has been in slides, and often was used to refer to the multiple "downstairs" sites. We also discussed that "site" is a place where multicast DNS works (which is kinda of standing the defintion on its head) We think we will remove all references to site, and try to emphasize the nomadic connectivity of the Registrar-Agent using some new term. 6. https://github.com/anima-wg/anima-brski-prm/issues/129 Need to add IANA registration for brski-pledge Probably this is redundant, and we need to move all the discovery IANA stuff to the (new) discovery document. If we keep the Agent discovery of Pledge in PRM, then we need to keep the IANA registry for the .well-known and mDNS in that document. Much conversation dealing with lack of feedback from mediaman about Media Type process. There is no dependancy from discovery document about media types, but it is a bit hidden /implicit. 7. https://github.com/anima-wg/anima-brski-prm/issues/130 product-serial-number for discovery Suggested text is now in -01 of discovery. https://www.ietf.org/archive/id/draft-ietf-anima-brski-prm-09.html#name-discovery-of-pledge-by-regi We believe that it is okay to do **discovery** by manufacturer-specific serial number without qualifying it with the manufacturer. While this results in false positives (every manufacturer has serial number 001 present), but in that unlikely result, it is okay, because the Registrar-Agent can filter through the results. More discussion about how much we can ask for, and how will this get integrated into supply processes. Definitely a topic for discussion at IETF118. -- Michael Richardson <[email protected]> . o O ( IPv6 IøT consulting ) Sandelman Software Works Inc, Ottawa and Worldwide
signature.asc
Description: PGP signature
_______________________________________________ Anima mailing list [email protected] https://www.ietf.org/mailman/listinfo/anima
