Re: ntpd not keeping time

[debian@clouded.org]

This might be useful to someone out there..


Run as root on the problem system:

# ntptime
ntp_gettime() returns code 5 (ERROR)
  time c1e3d62c.cf888000  Thu, Jan 30 2003 11:49:48.810, (.810677),
  maximum error 16384000 us, estimated error 16 us
ntp_adjtime() returns code 5 (ERROR)
  modes 0x0 (),
  offset 0.000 us, frequency 0.000 ppm, interval 4 s,
  maximum error 16384000 us, estimated error 16 us,
  status 0x41 (PLL,UNSYNC),
  time constant 0, precision 1.000 us, tolerance 512 ppm,
  pps frequency 0.000 ppm, stability 512.000 ppm, jitter 200.000 us,
  intervals 0, jitter exceeded 0, stability exceeded 0, errors 0.



Run as root on working systems:

# ntptime
ntp_gettime() returns code 0 (OK)
  time c1e3d490.690d1000  Thu, Jan 30 2003 11:42:56.410, (.410356),
  maximum error 460134 us, estimated error 741 us
ntp_adjtime() returns code 0 (OK)
  modes 0x0 (),
  offset -26.000 us, frequency 94.460 ppm, interval 4 s,
  maximum error 460134 us, estimated error 741 us,
  status 0x1 (PLL),
  time constant 6, precision 1.000 us, tolerance 512 ppm,
  pps frequency 0.000 ppm, stability 512.000 ppm, jitter 200.000 us,
  intervals 0, jitter exceeded 0, stability exceeded 0, errors 0.


# ntptime
ntp_gettime() returns code 0 (OK)
  time c1e3d49d.dc3b6000  Thu, Jan 30 2003 11:43:09.860, (.860281),
  maximum error 356570 us, estimated error 34979 us
ntp_adjtime() returns code 0 (OK)
  modes 0x0 (),
  offset 7319.000 us, frequency -69.290 ppm, interval 4 s,
  maximum error 356570 us, estimated error 34979 us,
  status 0x1 (PLL),
  time constant 6, precision 1.000 us, tolerance 512 ppm,
  pps frequency 0.000 ppm, stability 512.000 ppm, jitter 200.000 us,
  intervals 0, jitter exceeded 0, stability exceeded 0, errors 0.



The builds on all systems are nearly identical - configs, packages, etc.
So now I'm suspecting some sort of incompatibility between the cheap
motherboard I chose and either libc, the kernel, or ntpd.

The board, btw, is a Gigabyte GA-7VKML.

Or maybe I'm just going wild with speculation again. Help please? :)



On Wed, 29 Jan 2003, nate wrote:

> > Hello all. I've got this weird problem with ntpd. Up until now I've never
> > had ntpd fail me. But on this one box I've got the clock speeding up by 30
> > minutes each day. I've got ntpd installed and configured properly with
> > working ntp servers. In fact, ntpdate sets the clock properly, but ntpd is
> > failing to keep the clock in sync with the ntp servers.
>
>
> try 'ntptrace <host>'
>
> sample:
> defiant:/home/aphro# ntptrace redhat.aphroland.org
> redhat.aphroland.org: stratum 2, offset 0.001617, synch distance 0.07616
> time.nist.gov: stratum 1, offset 0.001895, synch distance 0.00000, refid 'ACTS'
>
> nate
>
> (redhat.aphroland.org is not accessable from the internet so don't bother
> trying to query it :) )
>
>
>
>