[ntp:questions] Re: ntp + ppskit + oncore M12+ - I'm getting closer! :)

Ryan googlegroups at draziwDOTnet.net
Thu Aug 28 04:35:07 UTC 2003


Ulrich,

It's been a while since I've done any tweaking/changes - so I don't even
remember exactly how it was.  I'll paste what things look like now.  If
there is anything you think I should try - I'm game. :)

Thanks,
Ryan (googlegroups-AT-draziw-dot-net on/off as spam hits it)


ntptime
ntp_gettime() returns code 0 (OK)
  time c2f80050.27d199ac  Wed, Aug 27 2003 21:15:12.155, (.155542898),
  maximum error 4137 us, estimated error 1 us, TAI offset -1073743032
ntp_adjtime() returns code 0 (OK)
  modes 0x0 (),
  offset 146.539 us, frequency 8.489 ppm, interval 4 s,
  maximum error 4137 us, estimated error 1 us,
  status 0x2001 (PLL,NANO),
  time constant 4, precision 0.945 us, tolerance 496 ppm,
  pps frequency 8.266 ppm, stability 0.000 ppm, jitter 0.000 us,
  intervals 0, jitter exceeded 0, stability exceeded 0, errors 0.

ntpq -p
     remote           refid      st t when poll reach   delay   offset
jitter
==============================================================================
 LOCAL(0)        LOCAL(0)        10 l   60   64  377    0.000    0.000
0.001
*GPS_ONCORE(0)   .GPS.            0 l    5   16  377    0.000    0.146
0.002

snip of ntp.conf:
server  127.127.1.0     # local clock
fudge   127.127.1.0 stratum 10

#pps /dev/oncore.pps.0 assert hardpps    # PPS device
restrict 127.127.30.0 mask 255.255.255.0
server 127.127.30.0 prefer
fudge 127.127.30.0 stratum 0
#ATOM PPS
#server 127.127.22.0
#fudge 127.127.22.0 flag3 1 refid GPS
-=-
snip of /var/log/messages:
Aug 26 23:15:47 r2 ntpd[1050]: ntpd 4.1.80-rc1-r Mon Aug 11 11:16:11 PDT
2003 (1)
Aug 26 23:15:47 r2 ntpd[1050]: precision = 1.000 usec
Aug 26 23:15:47 r2 ntpd[1050]: no IPv6 interfaces found
Aug 26 23:15:47 r2 ntpd[1050]: kernel time sync status 2040
Aug 26 23:15:47 r2 kernel: adjtimex: ntpd could be using obsolete
ADJ_TICKADJ
Aug 26 23:15:48 r2 ntpd[1050]: frequency initialized 8.266 PPM from
/etc/ntp/drift
Aug 26 23:18:35 r2 ntpd[1050]: time reset -0.318632 s
Aug 26 23:18:35 r2 ntpd[1050]: kernel time sync disabled 2041
Aug 26 23:18:35 r2 kernel: adjtime: ntpd used obsolete
ADJ_OFFSET_SINGLESHOT




On 27 Aug 2003 13:32:28 +0200, Ulrich Windl
<Ulrich.Windl at RZ.Uni-Regensburg.DE> wrote:

>Still wondering: In your previous article the kernel flags did not
>show that the kernel PLL is using the PPS info. I still don't know how
>the Oncore driver handles PPS, but I might guess that your ntpd runs
>without the additional blessings from PPS.
>
>Regards,
>Ulrich
>
>Ryan <googlegroups at draziwDOTnet.net> writes:
>
>> Many more hours of running - looks like my offset runs .020 to .035 most
>> of the time.  Also looks like I'm off by ~4ms (?) vs some fairly solid
>> S1 servers.  Any tips?  (Improving accuracy, and why I'm off so much
>> from good servers? - and anything else to optimize without installing
>> *bsd. ;)  )
>> 
>> Thanks,
>> Ryan
>> 
>> ntpq -p
>>      remote           refid      st t when poll reach   delay   offset
>> jitter
>> ==============================================================================
>>  LOCAL(0)        LOCAL(0)        10 l   37   64  377    0.000    0.000
>> 0.015
>> *GPS_ONCORE(0)   .GPS.            0 l   15   16  377    0.000   -0.048
>> 0.015
>> -bb1-l0.snfc21.p tick.usno.navy.  2 u   20   64  377   35.293    4.196
>> 0.526
>> +time.nist.gov   .ACTS.           1 u   58   64  377   56.302    4.434
>> 0.173
>> +timekeeper.isi. .GPS.            1 u   59   64  377   26.242    4.438
>> 0.433
>> -tick.ucla.edu   .PSC.            1 u    5   64  367   28.422    4.985
>> 0.312
>> 
>> .

.



More information about the questions mailing list