[ntp:questions] debugging cronic offset on Windows 2003 client
David J Taylor
david-taylor at blueyonder.co.uk.invalid
Wed Feb 1 13:51:50 UTC 2012
> Just deployed the 4.2.4p8 NTP client to 50 Windows 2003 SP2 servers, and
> all seem to be running fine...except...
> One or two of the clients develop an offset of about 16 seconds.
> Restarting the service puts them back in sync with the NTP server, but
> in a few minutes they're back to the old offset.
> What option can I add to the conf file to specify a log file? I don't
> seem to be able to modify the startup string in the service's
> properties. I tried adding "logfile <filename>" to the conf file, which
> didn't seem to work.
> How can I raise the debug level?
As a first step, I would check that the W32time service is not running,
and that there's nothing on those servers which might try and restart it.
I have little experience with Domains, but are those servers performing a
different role to the others? The NTP service normally write its entries
to the Windows Event Log, so you could look there for entries from "NTP".
I've written a little about basic fault-finding here:
Someone will doubtless comment that the current version is 4.2.6, but
let's try and get what you have working.
More information about the questions