Reviewer: Wes Hardaker Review result: Not Ready I reviewed this document as part of the Security Directorate's ongoing effort to review all IETF documents being processed by the IESG. These comments were written primarily for the benefit of the Security Area Directors. Document authors, document editors, and WG chairs should treat these comments just like any other IETF Last Call comments. Document: draft-ietf-dconn-domainconnect-async Reviewer: Wes Hardaker Review Date: 2026-10-29 Summary: Overall, a very strong -00 draft! Just some nits Major Concerns: None Minor Concerns: Additional nits, comments and suggestions below Nits and comments: Section 3: The terminology would be helpful to be indented that is being specifically defined in a bulleted list. Specifically, client ID, response type, and OAuth. Is the DC force term used anywhere? Section 4: The list of steps is almost humorous to read because the left-hand side before the colon almost identically mirrors the right-hand side. I wonder if both are really needed. For section 6.2 in the last paragraph, it should be stated that the DNS provider must give that secret over a properly protected channel. Even better if it’s out of band. Section 6.3 works used to have some challenging user interface implementation issues In section 6.3 underneath table 2, there is discussion about being successful and some notions of what to do upon errors. For example, does the paragraphs later starting with the word similar also apply to error handling in this paragraph or different? Why is the response code in table 7 use stars sometimes and x’s in the other times? In section 6.6, beginning with the sentence that says “the DNS provider must”, it talks about parameters being accepted in a query string or the body of this POST. I wonder, though, if you really mean “and” in particular, because the example uses properties only in the URL, but it looks like other properties are expected in the POST body. And if this is the case, then what happens when there’s a conflict if they are both to be used? In the end of section 6.6, it shows a structure of how errors are supposed to be returned, including various records. This JSON structure is not specified anywhere that I saw, and either a reference to where it is specified should be included, or it should be defined more formally as fields and values that must be present and returned in a dictionary. Minor comment: when HTTP codes are used like 501 and other ones, it would be helpful if ( you could include the code name as well (maybe in ()s) For the security considerations section 8.1, in the last paragraph, why is it “should therefore maintain” instead of “must therefore maintain”? This may be a valid choice, I’m asking generically. One other security consideration to consider inserting would be what happens when there are multiple asynchronous requests in flight and how to deal with race conditions and conflicts because of it, especially if the attacker controls one path, although that situation should be incredibly rare For the security considerations section, I would probably state that the considerations of other documents should be included too, including the synchronous version as well as OAuth security considerations sections