Moin,
I tried to follow the thread and implement a solution that might
converge the points voiced:

- Emphasize RFC9715 following Paul Vixie's point
- Clarify that DNS-over-UDP refers to DNS-over-UDP transport as per
  RFC1035
- Given that RFC9715 effectively mitigates PMTUD dependence, made it a
  <bcp14>MAY</bcp14>, also clarifying that this (in addition) further
  _explicitly_ ensures not relying on PMTUD. I would argue that the
  <bcp14>MAY</bcp14> here is sufficient (despite the prior SHOULD NOT
  vs. MUST NOT discussion), as the whole discussion is effectively
  already covered by RFC9715, i.e., setting the option is, effectively,
  just making something explicit.

Arguably, though, it might even be sensible to just remove the PMTUD
part, with the reference to RFC9715 put more prominently, no?

Anyway; Pushed to git. Please take a look.

With best regards,
Tobias

On Thu, 2025-08-14 at 12:49 +0200, Joe Abley wrote:
> Hi Gorry,
> 
> On 14 Aug 2025, at 11:31, Gorry Fairhurst <[email protected]>
> wrote:
> 
> > I think we need to carefully explain what we mean as "UDP".
> > 
> 
> > [...]
> 
> 
> I think these are all helpful clarifications, thank you.
> 
> I think when we say "UDP transport" in the context of DNS, the
> meaning is pretty clear (for example, we don't mean DoQ), but it
> certainly doesn't hurt to be more precise, especially when we are
> touching on concepts that are ordinarily discussed more generally,
> and not in reference to a specific higher-layer protocol.
> 
> 
> Joe
> 
> _______________________________________________
> DNSOP mailing list -- [email protected]
> To unsubscribe send an email to [email protected]

-- 
Dr.-Ing. Tobias Fiebig
T +31 616 80 98 99
M [email protected]
Pronouns: he/him/his

_______________________________________________
DNSOP mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to