[ntp:questions] Second attempt at GPS-18 based NTP server
David J Taylor
david-taylor at blueyonder.co.uk.invalid
Fri Feb 10 16:25:36 UTC 2012
> Hello Again,
> Okay, everyone has said I definitely need to use a custom kernel, and I
> have now done that, and ntpdc -c kern now reports the following:
> pll offset: -1.091e-06 s
> pll frequency: -3.196 ppm
> maximum error: 0.019912 s
> estimated error: 1e-06 s
> status: 2107 pll ppsfreq ppstime ppssignal nano
> pll time constant: 7
> precision: 1e-09 s
> frequency tolerance: 496 ppm
> pps frequency: -3.196 ppm
> pps stability: 0.022 ppm
> pps jitter: 1.261e-06 s
> calibration interval: 128 s
> calibration cycles: 797
> jitter exceeded: 522
> stability exceeded: 0
> calibration errors: 5
> Which looks fairly reasonable, as the status line is talking about
> ppstime and ppssignal. Not sure about 5 calibration errors...
> But then I look at ntpq -p and it says this:
> remote refid st t when poll reach delay offset
> +server1.terrybu 126.96.36.199 2 u 70 128 377 6.356 -0.122
> -188.8.131.52 184.108.40.206 2 u 70 128 377 5.441 -0.582
> +gamma.h3x.no 220.127.116.11 3 u 1 128 377 6.066 -0.228
> *GPS_NMEA(0) .PPS. 0 l 2 64 377 0.000 -0.001
> I'm fairly sure that there should be a lower case "o" to the left of
> GPS_NMEA, and there is not.
> Am I right to be worried?
> Thanks again!
Paul, I had the same problem about the tally code. NTP 4.2.4 (as supplied
with FreeBSD 8) produced a "*" when PPS /was/ in operation. NTP 4.2.6p5
produced a "o", as does the current development version - ntpd 4.2.7p255.
All these as viewed from a remote PC running Windows NTP 4.2.7p241.
Your 2107 status looks good to me. I also see calibration errors - about
25 in ten hours since a cold start.
I hope that helps.
More information about the questions