Yo Hal! On Fri, 18 Jan 2019 16:58:31 -0800 Hal Murray via devel <[email protected]> wrote:
> Gary said:
> > There is no TLS session between NTP client and NTP server. UDP
> > only. And yet the NTP server must provide a new cookie every client
> > request. Since there is no TLS session, the master key used can not
> > be from the TLS session.
>
> The NTP server extracts S2C and C2S and AEAD number by using the
> master key to decrypt them from the cookie. (That's the whole point
> of the cookie.)
Yes! And the point is?
> So it can use them to make new cookies.
No, you use the master key to make new cookies.
> > Here is another way. If each connection used a different master
> > key, then the NTPD server would need to store state for each client
> > to know what master key to use. Instead the NTPD server just
> > generates a new master key every day or so.
>
> I think you are confusing things with that use of "master key".
Yup, that is what I have been saying!
> Each TLS session sets up new working keys.
Yes, just for THAT session. And only for NTS-KE connections. No
NTPD client to NTPD server TLS session ever exists.
> I'm not familiar with the details.
I am.
> More info in Richard's recent msg.
Which I disagreed with.
> Each end has whatever
> it takes to use RFC 5705 to make more keys without exchanging any
> packets.
Yes, each end has, as the Proposed RFC calls it: the current master key.
This master key is used to make new C2S and S2C as required, on demand.
RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
[email protected] Tel:+1 541 382 8588
Veritas liberabit vos. -- Quid est veritas?
"If you can’t measure it, you can’t improve it." - Lord Kelvin
pgpWczdWJTgCv.pgp
Description: OpenPGP digital signature
_______________________________________________ devel mailing list [email protected] http://lists.ntpsec.org/mailman/listinfo/devel
