[LinuxPPS] time_pps_kcbind return EOPNOTSUPP

[Alexander Gordeev]

В Mon, 31 May 2010 16:43:36 +0200
Udo van den Heuvel <udovdh at xs4all.nl> пишет:

> On 2010-05-30 21:15, William S. Brasher wrote:
> > On Sun, 30 May 2010, Udo van den Heuvel wrote:
> >> On 2010-05-30 10:53, Udo van den Heuvel wrote:
> >>> So what is going on?
> >>
> >> See below:
> [zippp]
> > 
> > The current ntp has pps in the nmea driver (127.127.20.x).   You do not 
> > need to include the pps driver (127.127.22.x) in ntp.conf, it seems.  Try
> > removing the pps driver from ntp.conf and restarting ntpd...
> 
> Did so and a few other tweaks:
> 
> [root at epia wide-dhcpv6-20080615]# ntpq -np -crv
>      remote           refid      st t when poll reach   delay   offset
> jitter
> ==============================================================================
> o127.127.20.0    .GPS.            0 l   14   16  377    0.000    0.126
>  0.012
> +194.109.22.18   193.67.79.202    2 u   10   64  377   23.231   -2.436
> 25.103
> +194.109.20.18   193.67.79.202    2 u    2   64  377   19.254   -4.335
> 20.220
> +193.67.79.202   .PPS.            1 u   30   64  377   23.229   -1.731
> 23.264
> *193.79.237.14   .PPS.            1 u   19   64  377   23.251   -2.232
> 19.623
> +83.163.219.98   .DCFa.           1 u   12   64  377   43.214  -10.462
> 25.438
> +134.221.205.12  .PPS.            1 u   25   64  377   23.215   -1.802
> 23.196
> associd=0 status=04fd leap_none, sync_uhf_radio, 15 events, kern,
> version="ntpd 4.2.6p1 at 1.2158-o Sat May 29 06:32:36 UTC 2010 (1)",
> processor="i686", system="Linux/2.6.34-ts13", leap=00, stratum=1,
> precision=-20, rootdelay=0.000, rootdisp=0.581, refid=GPS,
> reftime=cfae4718.69e81cc5  Mon, May 31 2010 16:32:56.413,
> clock=cfae4726.8176760f  Mon, May 31 2010 16:33:10.505, peer=61365, tc=4,
> mintc=5, offset=0.126, frequency=-5.725, sys_jitter=0.012,
> clk_jitter=0.000, clk_wander=0.036
> [root at epia wide-dhcpv6-20080615]# ntptime
> ntp_gettime() returns code 5 (ERROR)
>   time cfae4732.904230c8  Mon, May 31 2010 16:33:22.563, (.563510396),
>   maximum error 5367 us, estimated error 0 us
> ntp_adjtime() returns code 5 (ERROR)
>   modes 0x0 (),
>   offset 105.533 us, frequency -5.695 ppm, interval 4 s,
>   maximum error 5367 us, estimated error 0 us,
>   status 0x2007 (PLL,PPSFREQ,PPSTIME,NANO),
>   time constant 4, precision 0.001 us, tolerance 500 ppm,
>   pps frequency 0.000 ppm, stability 0.000 ppm, jitter 0.000 us,
>   intervals 0, jitter exceeded 0, stability exceeded 0, errors 0.
> [root at epia wide-dhcpv6-20080615]# ntpq -ccv
> associd=0 status=0000 , no events, clk_unspec,
> device="NMEA GPS Clock",
> timecode="$GPRMC,143429,A,5150.2309,N,00457.3699,E,000.0,098.5,310510,000.7,W*62",
> poll=4850, noreply=0, badformat=0, baddata=0, fudgetime1=0.000,
> stratum=0, refid=GPS, flags=5
> 
> 
> So it looks better (time-wise) but ntptime still shows `error`.
> The error has to go away. Error is not there as long as ntpd is
> starting, then status is OK. When local PPS is entering the game, the
> status becomes error.
> 
> So what is the next step?
> 
> I am also in #ntp on freenode.

I see at least one problem: missing PPSSIGNAL flag in the output. This
means that kernel consumer doesn't receive PPS timestamps for at least
two minutes. It probably received them in the at first because PPSTIME
and PPSFREQ are set (ntpd usually does this).

-- 
  Alexander
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 490 bytes
Desc: not available
Url : http://ml.enneenne.com/pipermail/linuxpps/attachments/20100531/d82b92c6/attachment.pgp