[SATLUG] ntp problems

Travis H. travis at nexus.subspacefield.org
Fri Nov 3 16:55:07 CST 2006


I've got every system running ntp, slaving off one master timehost,
that synchronizes with the outside world (this ensures that all
of them are synchronized with each other, because consistency
is more important to me than accuracy, because the former allows
me to correlate log file entries, but the latter just makes sure
I'm not late to work).

But on one client, my workstation, it first synchronizes with the
timehost, then it starts running off its local clock, which is
set up as a stratum 12 server, whereas the timehost is stratum 3.
The only reason that's there is if the time server is down, it
will run off the internal timer until it's back up.  So it
ends up with an offset from the server of about -29687 ms,
after just two and a half hours.

Anyone know how to troubleshoot this or have any guesses?
Maybe I just have a flaky RTC?
-- 
"Cryptography is nothing more than a mathematical framework for
discussing various paranoid delusions." -- Don Alvarez
<URL:http://www.subspacefield.org/~travis/> -><-


More information about the SATLUG mailing list