[ntp:questions] Why is NTPD adjusting the system clock when in LOCAL(0) mode?

unruh unruh at wormhole.physics.ubc.ca
Thu Jun 16 20:44:07 UTC 2011


On 2011-06-15, Charles Elliott <elliott.ch at verizon.net> wrote:
> Hello:
>
>  
>
>                 I wrote a clock simulator program that, among other things,
> outputs the current time adjustment every second.  The only active lines in
> NTPD's conf file are:
>
>  
>
> # your local system clock, could be used as a backup
>
> # (this is only useful if you need to distribute time no matter how good or
> bad it is)
>
> server 127.127.1.0 iburst minpoll 4 maxpoll 4

Really  really really terrible idea. Get rid of it.
And if you are going to use ntp, you really need some time source it can
use to set your clock. It is pretty useless without that.

>
> # but it should operate at a high stratum level to let the clients know and
> force them to
>
> # use any other timesource they may have.
>
> fudge 127.127.1.0 stratum 12
>
>  
>
> Yet NTPD is constantly changing the curTimeAdjustment.  Here is the evidence
> from my program:
>
> CTA: 156003
>
> CTA: 156004
>
> CTA: 156003

no idea what this means.

>
> ..
>
>  
>
> I know that NTPD is doing this because when I stop it, my simulator outputs
>
> CTA: 156001
>
> CTA: 156001
>
> CTA: 156001
>
> ..
>
> until I restart NTPD, at which time it goes back to speeding up the system
> time.
>
>  
>
> On what basis could NTPD possibly adjust the system time when it is using
> the local clock as its only time source?  What standard is NTPD comparing
> the time to?
>
>  
>
> I am running Windows 7 64-bit and the version of NTPD I am using is ntpd
> 4.2.4p7 at copenhagen-o May 22 11:25:36 (UTC+02:00) 2009  (3).
>
>  
>
> Thanks in advance for your help.
>
>  
>
>  
>
>  
>
> Charles Elliott




More information about the questions mailing list