Re: NTP with only GPS/PPS flags PPS as falseticker

[Gary E. Miller]

Yo Nick!

On Thu, 19 Mar 2020 00:56:42 +0000
"Nick Burkitt" <address@hidden> wrote:

> I'm using NTPsec 1.1.8+, but the pps driver doesn't seem to be 
> available:
> 2020-03-18T23:40:49 ntpd[1468]: CONFIG: Unknown driver name pps

Easy to check:
    # ntpd --list
    [...]
    This version was compiled with the following clock drivers:
           LOCAL  SPECTRACOM    TRUETIME     GENERIC     ARBITER
           MODEM        NMEA         PPS       HPGPS         SHM
         TRIMBLE      ONCORE         JJY       ZYFER        GPSD

But, I thought you were using gpsd, thus the NTPsec SHM driver is
what you want.

> I'm using my own kernel driver instead of the pps-gpio driver, but it 
> does the same thing, only with a hardware interrupt line instead of a 
> GPIO pin.

You probably do not want the NTPsec PPS driver, but the NTPSec NMEA
driver with PPS enabled.  But that does not support KPPS, which you just
said you are using.  Which is why you are using gpsd, so a wild goose
chase.

> But I don't think the problem lies with PPS.

"Thinking" (guessing) is not good.  Easy to know:

     # ppstest /dev/pps0

> It's the fact that the
> GPS clock is marked as a falseticker, nearly always. I've tweaked the
> time1 value to get the offset (as shown by ntpmon) to around 10-15
> ms, with a jitter of around 5 ms, but still ntpd doesn't like my GPS

I'll have to ask, yet again:

    have you tried to set minsane?
    whnat do you ntpd logs show?

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can't measure it, you can't improve it." - Lord Kelvin

pgpjBeJoN7Ram.pgp
Description: OpenPGP digital signature