[ntp:questions] Query about NTP accuracy

Terje Mathisen "terje.mathisen at tmsw.no" at ntp.org
Sun May 24 13:14:13 UTC 2009


Jan Ceuleers wrote:
> Terje Mathisen wrote:
>> This is why you want ntpd to do the initial step, after first doing 
>> the full vote collection from all configured servers, not ntpdate 
>> which picks the first server to respond!
> 
> If that is so, then the ntpdate manpage is wrong:

ntpdate is deprecated, so it doesn't really matter.
> 
> ntpdate sets the local date and time by polling the Network Time Proto‐
> col (NTP) server(s) given as the server arguments to determine the cor‐
> rect time. It must be run as root on the local host. A number  of  sam‐
> ples  are  obtained  from each of the servers specified and a subset of
> the NTP clock filter and selection algorithms are applied to select the
> best  of  these.
> 
> (http://www.eecis.udel.edu/~ntp/ntp_spool/html/ntpdate.html)
> 
> In other words, before ntpd was capable of being a good ntpdate 
> replacement, best practice was to use a script to extract server lines 
> from ntpd.conf and pass all of those on the ntpdate command line.

This only happened late in ntpdate's life, originally it did behave the 
way I described. However, it still wouldn't matter because that 
particular DB server was hardcoded to query just one (and in this case 
broken) server. :-(

Terje

-- 
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"




More information about the questions mailing list