Skip to main content

Early Review of draft-ietf-dnsop-integration-00
review-ietf-dnsop-integration-00-secdir-early-farrell-2025-09-03-00

Request Review of draft-ietf-dnsop-integration
Requested revision No specific revision (document currently at 02)
Type Early Review
Team Security Area Directorate (secdir)
Deadline 2025-09-12
Requested 2025-07-17
Requested by Peter Thomassen
Authors Swapneel Sheth , Andrew Kaizer , Bryan Newbold , N. Johnson
I-D last updated 2026-04-07 (Latest revision 2026-04-07)
Completed reviews Secdir Early review of -00 by Stephen Farrell (diff)
Artart Early review of -00 by Barry Leiba (diff)
Dnsdir Early review of -01 by David C Lawrence (diff)
Comments
(Apologies for the earlier email, I just learned about this Datatracker capability.)

During this draft's Adoption Call, there was some concern about the breadth of review in case mostly DNS people would look at it. Russ Housley therefore had suggested [1] we request an Early Review -- here you are.

The review is not particularly urgent, so I put down a date in ~2 months. Thanks!

[1]: https://mailarchive.ietf.org/arch/msg/dnsop/0oKiCtGR3IpNIbPjanYl6AVvv1o/
Assignment Reviewer Stephen Farrell
State Completed
Request Early review on draft-ietf-dnsop-integration by Security Area Directorate Assigned
Posted at https://mailarchive.ietf.org/arch/msg/secdir/D2vT7XobOCQAx3_Nw9hF1BHUo1A
Reviewed revision 00 (document currently at 02)
Result Has issues
Completed 2025-09-03
review-ietf-dnsop-integration-00-secdir-early-farrell-2025-09-03-00
The document has issues but I'd guess those'll be fixed as things evolve. (This being
an early review.)

The main issue I see is scope - I'm not clear what the document's scope is or should be. 
The text is very general, but the examples discussed seem rather limited, e.g. with no
mention of the handling of DNS names in e.g. email, ssh, anti-spam blocklists or things
like infratructure tools like cPanel. Put another way, if the scope is intended to be
as broad as the text, then there're a lot more quirks to using DNS names that are in the
current text. But perhaps some much more limited scope might be easier to finish and 
still useful. I'd encourage the authors/WG to consider that. I don't myself have any
strong opinion, but I'd say that not fixing this issue might lead to much pain later;-)

As a near-nit at this stage, I'd also note that various UTAH WG documents discuss
handling names in TLS which is a subset of this I guess.