Re: [chrony-users] Newbie requesting help

[Miroslav Lichvar <mlichvar@xxxxxxxxxx>]

On Mon, Mar 21, 2016 at 02:16:10PM -0700, Stuart Maclean wrote:
> I was then under the impression that I would supply these two 'votes'
> for the correct time to ntpd/chrony??  But when I looked into how gpsd
> feeds its time parameters (ie a PPS event and a GPRMC string?) over SHM
> to ntpd, I see that not one but TWO clock sources are used.  Am I
> supposed to split my data into these two sources?  Given that the
> shmTime  struct has a 'clockTimestamp' AND a 'receiveTimestamp', do I
> need FOUR and not TWO timestamps to complete the shm implementation?

No, I think you just need two timestamps and one SHM refclock. gpsd
provides two SHM refclocks, one for NMEA-based timing data and the
other based on the PPS signal, but only the later is needed for
synchronization with chronyd or ntpd (assuming PPS is working) as it's
already combined with the NMEA data. SHM can't be used for PPS alone.

Alternatively, you could feed chronyd with the PPS signal as a SOCK
refclock in the PPS mode. In this case you would have two refclocks, a
SHM or SOCK refclock sending the second counter (similarly to the
first gpsd SHM refclock) and a SOCK refclock for PPS. The combining of
the two would now happen in chronyd instead of your program, the first
refclock would have the "noselect" option and the second would have
the "lock" option.

-- 
Miroslav Lichvar

-- 
To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "unsubscribe" in the subject.
For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "help" in the subject.
Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.