[ntp:questions] ntpd - gpsd communication

Steven Sommars stevesommarsntp at gmail.com
Mon Jun 8 22:06:55 UTC 2020


I monitor many NTP servers and have observed off-by-N second errors (N
often 1, 2, 3) for years.  Some of these errors are understood: fuzzing
errors in ntpd, late NMEA arrival times.  But many of these errors have
unknown causes.  These errors can recur on a given server with the same
offset.  On May 22 I noticed a system running ntpsec + GPSD (
khronos.mikieboy.net,) was in error by two seconds.  The same system was
also off by two seconds on  January 15.  I asked about this on the ntpsec
mailing list, which happens to include several GPSD folks.  Their answer
was that a GPSD error caused the problem, but was fixed in GPSD 2.0     No
bug description seems to exist, so I'm uncertain whether the errors I saw
were related.

The off by N-second errors are often seen soon after NTP server
initialization.

I lack detailed configuration information for many of the affected systems,
unfortunately.



On Mon, Jun 8, 2020 at 5:02 AM Miroslav Lichvar <mlichvar at redhat.com> wrote:

> On 2020-06-08, ah at avtodoria.ru <ah at avtodoria.ru> wrote:
> > понедельник, 8 июня 2020 г., 10:52:36 UTC+3 пользователь Miroslav
> > Lichvar написал:
> >> If you need something to report a large offset to ntpd via SHM, you
> >> could try this program for testing leap seconds:
> >>
> >> https://github.com/mlichvar/leapshm
>
> > Thank you ver much. Please can you explain the params program uses.
> > Second can be path to unix socket or something else starting not with
> > "/" to use SHM. What about third param ?
>
> It's the number of seconds before the hardcoded leap second time (1 Jul
> 2015 00:00:00 UTC) where the time fed by the program should start. E.g.
> if it was 600, it would start 10 minutes before the midnight of the leap
> second. You would probably want to start with a larger number (if you
> don't want to test a leap second) and also change the LEAP_TIME constant
> to a current date.
>
> --
> Miroslav Lichvar
>
> _______________________________________________
> questions mailing list
> questions at lists.ntp.org
> http://lists.ntp.org/listinfo/questions
>


More information about the questions mailing list