Re: [lisp] Distinguished Name comments

["Alberto Rodriguez-Natal (natal)" <[e-mail]>]

Hi Dino,

Changes look mostly good to me, thanks! Just one comment, how about this wording for the last paragraph of section 6?

> It is RECOMMENDED that each use-case register their Distinguish Names with a unique Instance-ID. For any use-cases which require different uses for Distinguish Names within an Instance-ID MUST define their own Instance-ID and structure syntax for the name registered to the Mapping System. See the encoding procedures in [I-D.ietf-lisp-vpn] for an example.

Also, please consider double checking that we are consistent with names (capitalizations, dashes, etc) through the document. I think the official spellings are “EID-Prefix” and “Distinguished Name”, it might be worth to scan the document and update where needed.

Thanks!
Alberto

From: Dino Farinacci <[e-mail]>
Date: Wednesday, December 13, 2023 at 11:22 PM
To: Alberto Rodriguez-Natal (natal) <[e-mail]>
Cc: [e-mail] <[e-mail]>
Subject: Re: Distinguished Name comments
Thanks for the comments Roberto. See comments inline.

> First my apologies, this is long overdue. I’m trying to catch up with my shepherd’s review of draft-ietf-lisp-name-encoding and I have some comments/suggestions on the current draft.

I have submitted draft-ietf-lisp-name-encoding-03 to reflect your comments with a diff file attached.

> >    • Sec 3: You describe that when a DN is used as an EID, an exact match is performed (which is correct). However, this is described in the format section of the document, shouldn’t this be discussed somewhere else (maybe on its own section)? I know this is a very short document, but having that behavior described in the format section seems odd to me. No strong opinion though.

I have added a new section.

>     • Sec 4: You say it in the title of the section already, but it might be interesting that on the body of the section you mention that the listed use-cases are examples and more importantly that we explicitly say that that other use-cases not listed are possible as well. Typo: s/ascii/ASCII

Made the change.

>     • Sec. 5: I think that this section should talk about unique Instance-IDs (IIDs), not VPNs, so that the text is more general while preserving the considerations about name collisions. We can point to the VPN draft to mention one example of how a particular use-case is registering DNs in unique IIDs, see also the next point on this.

Made the change.

>    • Sec. 8: If we talk about IIDs in Section 5, there is no need to keep the VPN draft as a Normative Reference and could be moved to Informative, easing the RFC process.

Fixed.

Thanks again,
Dino