Re: [Ntp] comments on draft-mlichvar-ntp-ntpv5-03 / Message Format

[Miroslav Lichvar <[e-mail]>]

On Wed, Nov 24, 2021 at 10:38:10PM -0600, Dan Drown wrote:
> This email will be on the "4. Message Format" section.
> 
> Right now the flags/era/timescale fields come before root dispersion and
> root distance. This puts them at a different offset in the packet compared
> to their location in NTPv4. Was there a reason not to keep these at the same
> offset? It's not an implementation problem, just a cosmetic difference.

The reason was to not break the ordering of the fields by their
length. As the format of root delay/dispersion is different, I thought
moving them might have a small advantage that a combined NTPv4+NTPv5
implementation would be less likely to incorrectly interpret their
values as there would have to be two separate paths. But I wouldn't mind
the order to be changed if there is a good reason.

> For timescale offset, there is an overlap with the unknown value
> (0x8000/-1.0 seconds) during a negative leap second smear. Is this worth
> mentioning in the rfc?

Hm, good point. I think it depends on when the leap smear starts.
There is no +1.0 value. If we define 0x8000 in time16 to mean unknown,
at least it would be symmetric. +1.0/-1.0 would be rounded to
0x7fff/0x8001.

> Transmit Timestamp currently has: "In responses it is the time when a
> response to the client was transmitted. The specific response depends on the
> selected mode (basic or interleaved). The timestamp corresponds to the
> beginning of the transmission"
> 
> I'm suggesting this rewording: "In responses it is the beginning of the
> transmission of a response to the client. Which response it refers to
> depends on the selected mode (basic or interleaved). See Measurement Modes
> section below for detail."

Sounds good to me.

Thanks,

-- 
Miroslav Lichvar