Last Call Review of draft-ietf-tsvwg-udp-options-39
review-ietf-tsvwg-udp-options-39-dnsdir-lc-blacka-2025-02-15-00

This specification is well-written, well-organized, and clear.  This
specification sits at a layer below the DNS and thus does not depend on DNS in
any way.  However, DNS is a significant *user* of UDP and may benefit from the
adoption of UDP transport options.  In fact, the draft mentions DNS in this
context:

Section 5, paragraph 2:
> Among the use cases where this approach could be of benefit are
> request-response protocols such as DNS over UDP [He24]

He24 is "draft-heard-dnsop-udp-opt-large-dns-responses", which describes using
the MRDS and FRAG UDP options to (optionally) return larger DNS responses over
UDP without incurring the same security and usability issues as standard IP
fragmentation.  I don't know if DNS implementations would take advantage of this
or not, but the idea is reasonable. This is perhaps the most obvious application
of UDP transport options to DNS, but may not be the only options that could be
used.

I can't think of anything that would prevent DNS implementations from using UDP
transport options if they were available.