[ntp:questions] Wrong time after changing hardware

Per Hedeland per at hedeland.org
Wed Jul 25 21:43:15 UTC 2007

In article <ywn96448mjyj.fsf at ntp1.isc.org> Harlan Stenn
<stenn at ntp.isc.org> writes:
>>>> In article <f87nle$eik$1 at hedeland.org>, per at hedeland.org (Per
>Hedeland) writes:
>H> I *think* that independent of this defalt value or one provided by -f, if
>H> the file does not exist ntpd will not create it.
>Per> Hm, maybe someone that actually *knows* could comment - I'm not able to
>Per> check it out myself ATM. ...
>I'm probably not going to dig and give a definitive answer because Dave is
>in the middle of changing how the driftfile is written and the dust hasn't

Well, I downloaded the 4.2.4p3 tarball, and the only ocurrences of the
string "ntp.drift" in source files are in comments in
ntpd/ntp_loopfilter.c. So, I think we can establish that my claim that

>>>[...] the ntpd "manual page" at
>>>http://www.eecis.udel.edu/~mills/ntp/html/ntpd.html wrongly claims that
>>>there is a default of /etc/ntp.drift, both in the description of the -f
>>>option and in the "Files" section.

is quite correct - i.e. there is no default of any kind, and ntpd won't
look for /etc/ntp.drift unless you tell it to. Now maybe someone can fix
the documentation?

>The current mechanism is something like:
>- if the previous drift compensation value differs from the current
>  value by more than clock_phi, do not write.
>- if we are in state 4 and the current clock_stability is greater than
>  a slowly decreasing wander threshold, reset the threshold and write
>  the driftfile.

This has of course nothing at all to do with whether there is a default
location for it - and changing that aspect would seem to be a pretty bad
idea at this point, or at least quite unmotivated. Especially since Dave
just asserted, as quoted in the post where I pointed out the doc error,
that "It has been the case for the life of the distribution that, if you
don't specify a frequency file, it will never be written."

--Per Hedeland
per at hedeland.org

More information about the questions mailing list