Re: [Ntp] Smearing

[Tony Finch <[e-mail]>]

Hal Murray <[e-mail]> wrote:
>
> I think the protocol should use TAI and make available the leap info.
>
> That means an NTP client will need to know if it should smear.
>
> Plan A is to put that info in a flag in /etc/smear or such so that other
> programs can also use it, perhaps to unsmear or perhaps just to print a
> warning message when log files do/don't smear.
>
> Plan B would be to put that info into the kernel.  There are spare slots in
> struct timex and struct ntptimeval.  Then a flag in ntp.conf could tell ntpd
> to smear and it could pass that to the kernel.

Both plans don't answer the question of how the client gets the correct
configuration. Smearing is (typically) a feature of a data centre
environment and each client implicitly gets a time that matches its
neighbours by using the local NTP servers.

When there's a mixture of NTP4 and NTP5 clients, how can you be sure that
they get the same time? I think an NTP5 server should have some way to
say, "if you were talking NTP4 to me I would give you a smeared time", so
that an NTP5 client can match its neighbours.

> --------
>
> The next leap info is big enough that I don't want it in the basic
> packet format.

It doesn't need to be more than 4 bytes.

According to TF.460 (the definition of UTC) leap seconds can only happen
at the end of a month, so the leap indicator only needs to count months.

Leap seconds will no longer work when the earth is turning slowly enough
that we need one at the end of every month; we expect that to happen
before the year 4200, at which point DTAI will be about 4 hours.

https://www.ucolick.org/~sla/leapsecs/dutc.html

If you allow 15 bits for the count of months since 1972, that lasts until
after the year 4700. 14 bits for DTAI is over 4.5 hours; it might be
wise to add a sign bit in case things get really weird.

That leaves 2 bits spare which is just enough for the leap direction.
(Though it would be handy to have another bit to indicate smearing.)

Tony.
-- 
f.anthony.n.finch  <[e-mail]>  https://dotat.at/
protect and enlarge the conditions of liberty and social justice