[ntp:questions] Which number do you believe?

David L. Mills mills at udel.edu
Fri Jan 18 17:00:08 UTC 2008


Erik,

The jitter reported on the associations billboard applies only to that 
association; each one could be different. The jitter reported on the 
systems billboard applies to the system clock, which combines the 
association jitter in a weighted average along with the selection 
jitter. See the specification.

Dave

Erik Soosalu wrote:
> So in looking at the results of different commands for monitoring my NTP server, I'm trying to figure out what's "right"
> 
> In a script, I run ntpq -p, ntptime, ntpdc -c kern and ntpq -c rv one right after another so I can get the results from the system as close as possible to each other.
> 
> An example of what's bugging is 
> 
> from ntpq -p
>      remote           refid      st t when poll reach   delay   offset  jitter
> ==============================================================================
> *GPS_NMEA(1)     .GPS.            0 l    3   16  377    0.000    0.000   0.031
> 
> and from ntpq -c rv
> 
> status=2444 leap_none, sync_uhf_clock/PPS, 4 events, event_peer/strat_chg,
> version="ntpd 4.2.0 at 1.1161-r Sun Feb 26 05:12:14 UTC 2006 (1)",
> processor="i386", system="FreeBSD/5.5-RELEASE", leap=00, stratum=1,
> precision=-15, rootdelay=0.000, rootdispersion=0.703, peer=44873,
> refid=GPS, reftime=cb39fdba.34aabb6a  Thu, Jan 17 2008 11:15:54.205,
> poll=4, clock=cb39fdc7.270256be  Thu, Jan 17 2008 11:16:07.152, state=4,
> offset=0.000, frequency=-2.122, jitter=0.002, stability=0.001
> 
> Why are the jitter numbers different?
> 
> Ultimately what I'm looking to do is catch a couple of stats so that I can do some light monitoring of the box and check to see if there has been any failures along the way.
> 
> Thanks
> Erik
> 
> 
> _________________________________________________________________




More information about the questions mailing list