[Ntp] Re: Paul Wouters' No Objection on draft-ietf-ntp-interleaved-modes-07: (with COMMENT)

[Miroslav Lichvar <[e-mail]>]

On Tue, Aug 20, 2024 at 11:38:01AM -0700, Paul Wouters via Datatracker wrote:
>        The server MAY separate the timestamps by IP addresses, but it
>         SHOULD NOT separate them by port numbers to support clients that
>         change their port between requests, as recommended in RFC 9109
>         [RFC9109].
> 
> Would this not interact badly if I fire up a cluster of machines behind
> a NAT router and they more or less do NTP requests at the same time. For
> example after a power failure ?

If the server separates timestamps by IP address and there are
multiple clients on one IP address using similar polling intervals,
yes, the interleaved mode will not work. That's expected. The protocol
will fall back to the basic mode. The same thing can happen when the
server has too many clients. There is no guarantee the client will
receive an interleaved response, ever.

>         If the conditions are not met (i.e. the request is not detected
>         to conform to the interleaved mode), the server MUST NOT respond
>         in the interleaved mode. The server MAY always respond in the
>         basic mode. In any case, the server SHOULD save the new receive
>         and transmit timestamps.
> 
> This reads like the server MAY also not respond, which is probably not
> what you meant?

Yes, the server is not required to respond. There can be rate limiting
or other reasons for not responding.

-- 
Miroslav Lichvar