[ntp:questions] enable calibrate?

unruh unruh at invalid.ca
Sun Nov 13 22:17:10 UTC 2011


On 2011-11-13, Richard B. Gilbert <rgilbert88 at comcast.net> wrote:
> On 11/13/2011 12:59 AM, Doug Calvert wrote:
>> Hello,
>>   I just ran across the "enable calibrate" feature in the documentation.
>> Unfortunately there is little information on how it works, how to use
>> it. What do I do after I enable calibrate?
>
> What problem are you trying to solve?
>
> "Enable calibrate" appears to be very new and sparsely documented.  This 
> is not the sort of software that most people want to have within a mile 
> of their systems!
>
> If you have have a copy, there's nothing to stop you from playing with 
> it.  I would discourage playing with it on a production system!
>
> If you run it and it blows up in your face you can't say you weren't warned!

Primarily he wants to know what it is and does. Hard to play and
experiment if you have no idea what it is supposed to do. 
And it is not very new (it is already in ntp 4.2.4 which is about 4
years old by now) but IS poorly documented. 

>From man ntp_misc
calibrate
    Enables the calibrate feature for reference clocks. The default for
    this flag is disable. 

Of course, what the calibrate feature of a reference clock is is also a
bit mysterious.

>From the source (ntp_refclock.c), the only purpose seems to be to run the sentence
if (peer != sys_peer)
     pp->fudgetime1 -= pp->offset * FUDGEFAC;
else
     pp->fudgetime1 -= pp->fudgetime1 * FUDGEFAC;

But of course what fudgetime1 is for, and why subtracting
FUDGEFAC*fudgetime1 from it is useful, I have no idea. (Not the most
well documented of source codes).




More information about the questions mailing list