[ntp:questions] Bug 2341 - ntpd fails to keep up with clock drift at poll>7
stenn at ntp.org
Mon Dec 2 20:05:54 UTC 2013
I think the problem here is that folks are setting a "high floor" for
minpoll and then noticing the value they picked is too high.
ntpd will, I believe, increase the poll interval when that seems like a
safe thing to do. Left to its own devices, ntpd will notice that a
larger-than-expected offset was applied and will try and decrease the
poll interval to make sure that doesn't happen. I'd be real curious to
learn what the error variables show in this case, as it's something we'd
want to properly track for the General Timestamp API project (please see
http://nwtime.org/projects/timestamp-api/ for more info), especially if
these timestamps were used for kernel timekeeping.
Harlan Stenn <stenn at ntp.org>
http://networktimefoundation.org - be a member!
More information about the questions