[ntp:questions] different Ntp servers...

Hal Murray hal-usenet at ip-64-139-1-69.sjc.megapath.net
Sun May 31 02:55:51 UTC 2009

In article <4a1c62c6$0$510$5a6aecb4 at news.aaisp.net.uk>,
 David Woolley <david at ex.djwhome.demon.co.uk.invalid> writes:
>Hal Murray wrote:
>> What do they suggest that you do with 2 servers when they
>> give back different answers?
>> Maybe you could take the average?  :)
>ntpd already does that, as long as they don't disagree too much.  Whilst 
>the stratum, and error statistics, are based on just one server, the 
>time is a mix of the time from all the truechimers.

Sorry.  I should have added a smiley or such.

I was thinking of the case where you had 2 servers and
one of them was off by many seconds (or many days).  One
was clearly broken, but how do you tell which one?

Taking the average will guarantee that you get a bad time.

It will only be half as bad as the bad server but that's still
horrible on the scale of what NTP normally gets.

These are my opinions, not necessarily my employer's.  I hate spam.

More information about the questions mailing list