[LinuxPPS] offset 0 etc

Udo van den Heuvel udovdh at xs4all.nl
Sun Mar 18 11:09:50 CET 2012


On 2012-03-18 10:16, Dave Hart wrote:
> On Sun, Mar 18, 2012 at 07:59, Udo van den Heuvel <udovdh at xs4all.nl> wrote:
>> /usr/sbin/ntpd -n -u ntp:ntp -g -x
>>
>> all machines are on 24/24 for 99.9something % of the year.
>> So these do not have much impact.
> 
> -x
> Normally, the time is slewed if the offset is less than the step
(...)
> for other options. Note: The kernel time discipline is disabled with
> this option.

Very nice one. Thanks!

This very box now shows offset etc in ntptime output:

# ntptime
ntp_gettime() returns code 0 (OK)
  time d3103135.6463bec8  Sun, Mar 18 2012 11:02:29.392, (.392147286),
  maximum error 3760 us, estimated error 0 us, TAI offset 0
ntp_adjtime() returns code 0 (OK)
  modes 0x0 (),
  offset -20.680 us, frequency -25.230 ppm, interval 1 s,
  maximum error 3760 us, estimated error 0 us,
  status 0x2007 (PLL,PPSFREQ,PPSTIME,NANO),
  time constant 4, precision 0.001 us, tolerance 500 ppm,

The other box shows:

# ntptime
ntp_gettime() returns code 0 (OK)
  time d31031af.743126c0  Sun, Mar 18 2012 11:04:31.453, (.453875203),
  maximum error 13267 us, estimated error 283 us, TAI offset 0
ntp_adjtime() returns code 0 (OK)
  modes 0x0 (),
  offset 3729.313 us, frequency -57.395 ppm, interval 1 s,
  maximum error 13267 us, estimated error 283 us,
  status 0x2001 (PLL,NANO),
  time constant 4, precision 0.001 us, tolerance 500 ppm,

Why the PPSFREQ/PPSTIME omission?




More information about the discussions mailing list