Hi Rick,
Rick Moen writes:
> Quoting Olaf Meeuwissen (paddy-hack@???):
>
>> I think it's fair to point out that systemd-timesyncd only promises
>> Simple NTP (SNTP). How good a job it does of that is another matter
>> but at least it explains some of the "quirks" you mention below.
>
> Put that way, fair point. _But_, the larger context is that a
> systematically better job of time sync, using better code and also (in
> the case of ntimed-client), _smaller_ code, can be performed instead,
> gaining all the advantages of a real NTP client.
I had no issues with the larger context so refrained from commenting on
that.
> Or, to put it a different way, with several excellent genuine NTP
> clients to choose among, I deny the existence of a compelling use-case
> for any SNTP client, let alone one that's more than a little slipshod.
> (systemd-timed's security history, which I didn't get into, is less than
> reassuring.)
Fully agree. When I looked into timing issues on a Debian server I
still maintain, I noticed that systemd-timesyncd only promised SNTP and
when I read up on that wondered why anyone in their sane mind would want
to use that instead of NTP.
> The SNTP protocol is what Windows users settle for, because that's what
> they're offered by defaul, and mostly they don't know that they're
> missing out. On Linux, we don't need to settle.
We definitely don't and the large collection of alternatives, in terms
of distributions as well as programs, attests to that :-)
Hope this helps,
--
Olaf Meeuwissen, LPIC-2 FSF Associate Member since 2004-01-27
GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13 F43E B8A4 A88A F84A 2DD9
Support Free Software https://my.fsf.org/donate
Join the Free Software Foundation https://my.fsf.org/join