Re: [Ntp] Quick review of WGLC for status change for draft-ietf-ntp-update-registries

[Danny Mayer <[e-mail]>]

On 8/16/22 3:03 AM, Miroslav Lichvar wrote:
> On Mon, Aug 15, 2022 at 04:27:47PM -0400, Danny Mayer wrote:
>> On 8/15/22 6:57 AM, Miroslav Lichvar wrote:
>>> The IANA registry includes the values from RFC 5906, which nobody is
>>> using. ntpd is using its own values. That's one of the issues that the
>>> draft in the subject is supposed to address.
>>>
>> Are you saying that the ones in the IANA registry are wrong or that RFC5906
>> is wrong? Or both?
> The IANA registry matches the values in RFC 5906, but they don't match
> the values used by ntpd. There doesn't seem to be an implementation
> using the IANA/RFC5906 values. The document discussed here is trying
> to update the IANA table to include both sets of values in order to
> prevent conflicts with newly specified extension fields, as already
> happened with NTS.
>
> My understanding is that RFC5906 was based on the ntpd implementation,
> similarly to RFC5905, so other implementations could interoperate with
> it. I think the discrepancy can be easily explained as an oversight.
> There are macros used in the code and it's not very obvious how the
> values are encoded in the type field.
>
I had a conversation with Harlan about this. It turns out that the 
Autokey version currently implemented in the ntpd reference 
implementation is using what is called a v1 version of the extension 
field while RFC5906 is documenting the v2 version of the extension 
field. I was not aware of this when I put it into the draft of RFC5906. 
Dave Mills never mentioned that to me. My apologies for that.

Danny