[ntp:questions] Keeping NTP Honest

Unruh unruh-spam at physics.ubc.ca
Sun Jul 12 20:21:39 UTC 2009


Evandro Menezes <evandro at mailinator.com> writes:

>On Jul 10, 2:40=A0pm, Unruh <unruh-s... at physics.ubc.ca> wrote:
>>
>> And you have at least a 1/5 chance that IT is the bad server. What do
>> you do then?

>Well, bar a false ticker, when it's ignored, it would keep NTP in
>shorter poll periods.  IOW, "honest".

>I wonder though if the right thing would be to configure 6 servers
>with half of them limited to 64s polling...


The way ntp works, faster polling also means worse rate estimation and
more annoyance of the providers of the time. The current setup is done
that way to try to minimize the rate error, so if your sconnection to
ntp goes down, your system can freewheel with the greatest accuracy.
decreasing the poll interval does not keep it more honest. It just
decreases the rate accuracy.
(Now, if like chrony, ntp operated as something other than a Markovian
system, with zero memory, it could get good rate estimates and good
offset with more frequent polling, but that was not the design decision
make many many years ago.




More information about the questions mailing list