[TriLUG] ntp questions

Joseph Mack mack.joseph at epa.gov
Wed Oct 13 09:48:52 EDT 2004


Ryan Leathers wrote:

> How much drift is "too much"?

2 secs. ntpd will die and you'll have to ntpdate and restart ntpd. 
The only time ntpd dies for me is on startup when the hardware clock
is too far offset. 
 
> How can I tell if the system clock is being correctly governed by ntp?

if ntp is running. look with ps or ntpq

make sure you update the hardware clock as part of shutdown.

> rootdelay=17.917, rootdispersion=23.312, peer=34696, refid=time-a.nist.gov,
> reftime=c517ab68.525afc36  Wed, Oct 13 2004  9:26:00.321, poll=10,
> clock=c517ac63.fc442048  Wed, Oct 13 2004  9:30:11.985, state=4,
> offset=-2.989, frequency=7.621, jitter=4.703, stability=0.010
> 
> I am curious about where the values reftime, clock, and offset come from.

offset is the difference between your local clock and the higher stratum clock
that ntp has decided is the best clock (the one with the '*' beside it in ntpq).

> I can't find a man entry for ntpq :(

ntp is poorly documented. It used to be worse.

Joe

-- 
Joseph Mack PhD, High Performance Computing & Scientific Visualization
LMIT, Supporting the EPA Research Triangle Park, NC 919-541-0007
Federal Contact - John B. Smith 919-541-1087 - smith.johnb at epa.gov



More information about the TriLUG mailing list