[ntp:questions] Re: NTP "maximum error" on local network
Richard B. Gilbert
rgilbert88 at comcast.net
Thu May 18 17:55:16 UTC 2006
ajit.warrier at gmail.com wrote:
> ------------------ client 1 -----
> # ntpq -pcrv
> remote refid st t when poll reach delay offset
> jitter
> ==============================================================================
> *192.168.2.2 .NIST. 1 u 8 16 377 0.316 -0.004
> 0.049
> assID=0 status=0644 leap_none, sync_ntp, 4 events,
> event_peer/strat_chg,
> version="ntpd 4.2.0a at 1:4.2.0a+stable-8-r Fri Sep 9 16:44:48 UTC 2005
> (1)"?,
> processor="i686", system="Linux/2.6.14-ck1", leap=00, stratum=2,
> precision=-18, rootdelay=0.311, rootdispersion=23.428, peer=30804,
> refid=192.168.2.2,
> reftime=c8170ae1.7219220f Thu, May 18 2006 11:01:53.445, poll=4,
> clock=0xc8170b1d.39cd0bb6, state=4, offset=-0.008, frequency=212.134,
> noise=0.006, jitter=0.214, stability=0.176
>
> ----------- client 2 --------------
>
> # ntpq -pcrv
> remote refid st t when poll reach delay offset
> jitter
> ==============================================================================
> *192.168.2.2 .NIST. 1 u - 16 377 0.327 -0.010
> 0.103
> assID=0 status=06b4 leap_none, sync_ntp, 11 events,
> event_peer/strat_chg,
> version="ntpd 4.2.0a at 1:4.2.0a+stable-8-r Fri Sep 9 16:44:48 UTC 2005
> (1)"?,
> processor="i686", system="Linux/2.6.14-ck1", leap=00, stratum=2,
> precision=-18, rootdelay=0.327, rootdispersion=22.889, peer=7436,
> refid=192.168.2.2,
> reftime=c8170e61.c989df11 Thu, May 18 2006 11:16:49.787, poll=4,
> clock=0xc8170e75.fd7a13c2, state=4, offset=-0.010, frequency=183.551,
> noise=0.005, jitter=0.103, stability=0.165
>
> -------- server --------------
>
> # ntpq -pcrv
> remote refid st t when poll reach delay offset
> jitter
> ==============================================================================
> *LOCAL(1) .NIST. 0 l 18 64 377 0.000 0.000
> 0.004
> assID=0 status=0544 leap_none, sync_local_proto, 4 events,
> event_peer/strat_chg,
> version="ntpd 4.2.0a at 1.1190-r Fri Sep 23 21:02:05 EDT 2005 (1)"?,
> processor="i686", system="Linux/2.6.16.14", leap=00, stratum=1,
> precision=-18, rootdelay=0.000, rootdispersion=12.198, peer=11860,
> refid=NIST, reftime=c8170f79.d6b87379 Thu, May 18 2006 11:21:29.838,
> poll=10, clock=0xc8170f8b.90c7b02d, state=4, offset=0.000,
> frequency=0.000, noise=0.000, jitter=0.004, stability=0.000
>
> -------------------------------------------------------------
>
> I am trying to build an isolated system, disconnected from the rest of
> the internet, so I set the REFID to "NIST" and stratum to 0. If this is
> a stupid thing to do, please let me know :)
>
I wouldn't go so far as to call in "stupid" but it is not correct and it
is not necessary. (But don't ask what I would have said 25 or 30 years
ago!) If you don't specify the refid, I believe it defaults to ".LCL.",
".LOCL." or something similar. That simply says that the reference is
your local clock. .NIST. is not correct because it implies that the
time comes from the National Institute of Standards and Technology when,
in fact, it does not.
Even if your local clock was originally set to a time taken from NIST,
there is a 99.99999999% probability that the time has since deviated
from the correct value. It is not unusual for a local clock to drift as
much as four seconds per day. Some local clocks are much worse than that.
More information about the questions
mailing list