Re: [chrony-dev] handling of unstable sources

[Miroslav Lichvar <mlichvar@xxxxxxxxxx>]

On Wed, Jul 14, 2010 at 02:08:43PM +0200, Frank Muzzulini wrote:
> I also noticed that even directly after an update the server reports
> a root dispersion of ~0.5ms. I don't know however whether this is a
> general feature of ntpd or it is a configured value trying to
> reflect the potential absolute error of the received signal (DCF77
> in that case, which unlike GPS has no automatic correction for
> signal run times -- but than 0.5ms would be too less for any place
> more than 150km from the transmitter).

Looking at the sources, it seems that ntpd adds to the dispersion
a weighted average of the eight dispersions stored in the clock
filter, which include the 15ppm error. I think this will make most of
the root dispersion.

With a PPS refclock and poll 3 the peer dispersion reported by ntpd I
see here is 116 us.

$ ntpq -c "rv 31743"
associd=31743 status=973a conf, reach, sel_pps.peer, 3 events, sys_peer,
srcadr=PPS(0), srcport=123, dstadr=127.0.0.1, dstport=123, leap=00,
stratum=0, precision=-20, rootdelay=0.000, rootdisp=0.000, refid=PPS,
reftime=cfe83bea.0000000c  Wed, Jul 14 2010 15:36:42.000,
rec=cfe83bea.f405c917  Wed, Jul 14 2010 15:36:42.953, reach=377,
unreach=0, hmode=3, pmode=4, hpoll=3, ppoll=3, headway=0, flash=00 ok,
keyid=0, offset=0.000, delay=0.000, dispersion=0.116, jitter=0.001,
filtdelay=     0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00,
filtoffset=    0.00    0.00    0.00    0.00    0.00    0.00    0.00    0.00,
filtdisp=      0.00    0.12    0.24    0.36    0.48    0.60    0.72    0.84


-- 
Miroslav Lichvar

---
To unsubscribe email chrony-dev-request@xxxxxxxxxxxxxxxxxxxx with "unsubscribe" in the subject.
For help email chrony-dev-request@xxxxxxxxxxxxxxxxxxxx with "help" in the subject.
Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.