[ntp:questions] Re: How to set up a windows 2000 workstation as a NTP time server.
pmpradh at accusort.com
Fri Jan 23 20:20:06 UTC 2004
mayer at gis.net (Danny Mayer) wrote in message news:<3a2a0492.0401220851.4572da32 at posting.google.com>...
> "David J Taylor" <david-taylor at blueyonder.co.uk.not-this-bit> wrote in message news:<leNPb.6521$FH7.74825905 at news-text.cableinet.net>...
> > > Now, Why it is not correcting the time just by running ntpd?
> > > How frequently it (ntpd) synchronize the time with the time
> > > server specified in ntp.conf?
> > > I know there must be a way to change those parameters.. but
> > > I couldn't to find out. If it needs more parameters in
> > > ntp.conf to achive the automatic time synchronization ..
> > > please guide me through also the frequency of synchronization.
> > You need to start by ensuring that you box running ntpd against the
> > external sources is synchronising correctly. Did you stop the Win32Time
> > service? If not, then you have two services trying to sync your PC, and
> > they will conflict with each other.
> > The best check is with ntpq. Leave your system running for a couple of
> > hours, and then try a command like: ntpq -p <your server name>
> > C:\NTP>ntpq -p myserver
> > remote refid st t when poll reach delay offset
> > jitter
> > ==========================================================================
> > ====
> > *ntp0.isp.com .1PPS. 1 u 549 1024 377 27.861 0.693
> > 0.618
> > +ntp1.isp.com .1PPS. 1 u 548 1024 377 26.779 1.295
> > 1.089
> > The key things here are:
> > "*" against one of the remote servers means that is the chosen sync source
> > poll - 1024 - means that the remote is being queried once every 1024
> > seconds, the normal for a stable system
> > reach - 377 - means 100% connectivity with the remote source (it's an
> > octal value)
> > Your value of poll (256) suggests that ntpd hasn't yet reached stability.
> > Your reach values (340 200) means there has been no contact with the
> > server for the last few poll intervals.
> > Your values of delay (290 310) suggest the servers are two satellite hops
> > away - could you not find nearer ones?
> > This is just a starting point. I am not an expert on this, but I am more
> > Windows biased than many of the group members. I am surprised no-one else
> > has jumped in, quite honestly.
> > Cheers,
> > David
> You're doing quite well all by yourself. :-) Why interfere?
> If there is no contact with the server over the last few poll intervals,
> I'd check your network connectivity. Slow networks and broken networks
> can cause this.
Thank you for the reply both David and Danny.
On both of my machine W32Time has been disabled and there
aren't any time sync services installed other than ntpd.
Here is a dumb question.
Is it necessary that the primary windows 2000 workstation,
which is connected to internet and get synchronized with
time.nist.gov, has to be connected all the time?
I know/guess that the answer will be yes but I would like
to connect to time.nist.gov once day and get sync'ed.
After that I would like this workstation server as the
time source for rest of my windows 2000 workstations.
Is this possible?
Today I connect my primary workstation to internet and
it get synchronized with the time.nist.gov.
I disconnected this (primary)computer from internet. From my
other windows 2000 workstation I try to connect to primary win2k.
To synchronize the time (using ntpd).
As Danny mentioned, there could be network issues related
which I haven't ruled out yet.
I get stratum of 16 .. which means server unreacheable.
ntpq displays as
remote refid st t when poll reach delay offset jitter
ASI011177 .STEP. 16 u 78 1024 20 0.837 189474. 0.001
ind assID status conf reach auth condition last_event cnt
1 34036 b014 yes yes none reject reachable 1
I am definitely sure there is something wrong... $^#$#$*@
When I ping ASI011177 it comes back within less than 10ms.
Now what I am missing.. please guide me through.
Thank you in advance.
More information about the questions