[ntp:questions] poll interval - Clarification! - RFC compliance question

Steve Kostecke kostecke at ntp.org
Mon Jun 16 20:06:25 UTC 2008


On 2008-06-16, Anton Persson A <anton.a.persson at ericsson.com> wrote:

> The question I had was, in other words, should I expect the poll
> interval reported by ntpq to increase when using ntpd 4.2.4?

Yes.

> And, if it is to be expected, how fast should it increase?

Please see http://doc.ntp.org/4.2.4/ntpd.html#poll for the discussion of
Poll Interval Control pertaining to NTP 4.2.4

You may also wish to review the Poll Rate Control section of
http://www.cis.udel.edu/~mills/ntp/html/rate.html (please note that all
documentation found in this tree applies to the current NTP development
snapshot).

You will also find the documentation for your release of NTP in the
./html directory of the appropriate source tarball.

> (This is in the case when ntpd is in synchronization with a server and
> the communication breaks down, the server is "down".)

When time sources become unreachable ntpd will continue to discipline
the clock using the currently calculated frequency correction value.
If that ntpd is configured with Orphan Mode, or with the Undisciplined
Local Clock, it will be able to continue serving time to others.

-- 
Steve Kostecke <kostecke at ntp.org>
NTP Public Services Project - http://support.ntp.org/




More information about the questions mailing list