Date: Wed, 12 Dec 2001 23:53:40 -0500 From: The Anarcat <anarcat@anarcat.dyndns.org> To: Kevin Oberman <oberman@es.net> Cc: devin-freebsdquestions@rintrah.org, freebsd-questions@FreeBSD.ORG Subject: Re: ntpdate works, ntpd does not Message-ID: <3C183454.2000105@anarcat.dyndns.org> References: <200112121946.fBCJkRd11801@ptavv.es.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Kevin Oberman wrote:
>>Date: Wed, 12 Dec 2001 12:39:46 -0500
>>From: devin-freebsdquestions@rintrah.org
>>Sender: owner-freebsd-questions@FreeBSD.ORG
>>
>>I believe ntpd will poll the server over a period of some time to get a feel
>>for drift. I know when I set it up it took several hours before it would
>>adjust the clock. How long are you waiting before you expect ntpd to make adjustments?
>
> More specifically, it polls once every 64 seconds it has enough
> samples to be statistically significant. This usually takes 5
> samples. It then starts drifting the clock, if needed.
So that's what.. around 5'20"?
> To watch this happen, use the command "ntpq -p". When a server is
> tagged with a symbol, it's considered for use. The one marked '*' is
> the one being used and those marked '+' are also considered "good".
Ok. Let's take it from here:
[sorry, darn mozilla autofills everything...]
su-2.05# ntpq -p && date && ps jax -o start | grep ntpd\ -p
remote refid st t when poll reach delay offset
jitter
==============================================================================
time.nist.gov 0.0.0.0 16 u - 64 0 0.000 0.000
4000.00
Wed Dec 12 23:37:15 EST 2001
root 4593 1 4593 c06ef9c0 0 S<s ?? 0:00.74 ntpd -p
/var/run 11:27PM
root 4680 4465 4679 c076a2c0 2 R+ p0 0:00.05 grep ntpd
-p 11:37PM
so ntpd was started at 11:27PM, 10 minutes before the command was ran,
and the character before the hostname is " ". This means, from the man page:
space (reject) The peer is discarded as unreachable, synchro-
nized to this server (synch loop) or outrageous synchro-
nization distance.
So ntpd does reject the host. Maybe it's the synch loop thing I can't
figure out though..
One thing I read from the tcpdump output I found strange, though.. It
sends a packet containing a "stratum 0" instruction, somehow. Maybe it's
related?
The ntpdc "peers" command leads to a different result, surprisingly.
Indeed, it seems the character codes are different between ntpq and ntpdc:
ntpdc> peers
remote local st poll reach delay offset disp
=======================================================================
=time.nist.gov 5.0.0.0 16 64 0 0.00000 0.000000 0.00000
ntpdc(8):
...
a "=" means the remote server is being polled
in client mode,
...
And, BTW, the more I wait, the more the "unreach" counter from the
ntpdc(8) "peers" command increments.
> IF the source and the sync to the local system are stable, the time
> between polls will increase to 1024 seconds between polls.
>
> R. Kevin Oberman, Network Engineer
> Energy Sciences Network (ESnet)
> Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
> E-mail: oberman@es.net Phone: +1 510 486-8634
Thanks,
A.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3C183454.2000105>