[ntp:questions] offset slowly increases while ntpd is running?

matthew.garman at gmail.com matthew.garman at gmail.com
Wed Feb 7 15:21:59 UTC 2007

On Wed, Feb 07, 2007 at 09:35:16AM -0500, Richard B. gilbert wrote:
> You didn't mention the O/S you are running!

D'oh, that might be important, eh?  :)  Linux, specifically CentOS
4.4.  Kernel 2.6.9-42.ELsmp (i.e. Red Hat Enterprise Linux's patched
version of 2.6.9).

> Some flavors of Linux behave badly during periods of high disk
> usage.  It appears that interrupts are either masked or disabled
> for a period long enough to lose two or more clock ticks.  If you
> are running Linux and the clock is slow with respect to the server
> this may be your problem.

Interesting.  The problematic server shoudln't have any higher disk
load than our others, but I'll definately do some more research on

> It is also possible that server six has a really bad clock!  NTP
> cannot correct a clock with a frequency error greater then 500
> PPM.  If, without running ntpd, this machine gains or loses more
> than 43 seconds per day, it will never synch!  If the machine
> really keeps time this badly without running ntpd, ask Dell to fix
> it under warranty!

I'd like to try this over the weekend; this is what my gut feeling
told me could be wrong.

Thank you!

More information about the questions mailing list