Re: [chrony-users] System Time Offset in Logs?

[Bill Unruh <unruh@xxxxxxxxxxxxxx>]

man chronyc

It says that Systemtime gives the time that that the system is trying to slew
off. Ie, the system has a time and chrony has a time, and makes up the
difference by increasing or decreasing he rate of the clock until that
difference is removed. This systemtime is how much time still needs to be
slewed off. It will change rapidly. Ie, each time you run tricking it will
give a different System time. It could not log that-- what value would it log?


William G. Unruh __| Canadian Institute for|____ Tel: +1(604)822-3273
Physics&Astronomy _|___ Advanced Research _|____ Fax: +1(604)822-5324
UBC, Vancouver,BC _|_ Program in Cosmology |____ unruh@xxxxxxxxxxxxxx
Canada V6T 1Z1 ____|____ and Gravity ______|_ www.theory.physics.ubc.ca/

On Tue, 24 Sep 2019, Zack Shivers wrote:

> > Measurements contains the difference between the system time and the time
> > obtained from the ntp packet, statistics contains the estimated offset as
> > determined by the least squares fit on the last N measurements. That is maybe
> > what you want.
>
> Hmm, when I look at measurements, it does not match.
>
> For the sake of testing, I manually offset the system time on the server:
> timedatectl set-time <date-5min-in-past>
>
> I quickly check 'chronyc tracking' on the client:
> Reference ID    : C0A80002 (burrata-com-1)
> Stratum         : 11
> Ref time (UTC)  : Tue Sep 24 22:09:08 2019
> System time     : 326.629364014 seconds fast of NTP time
> Last offset     : +0.000008316 seconds
> RMS offset      : 1.175688386 seconds
> Frequency       : 30.445 ppm fast
> Residual freq   : +8.221 ppm
> Skew            : 237.633 ppm
> Root delay      : 0.000233696 seconds
> Root dispersion : 0.000234681 seconds
> Update interval : 0.1 seconds
> Leap status     : Normal
>
> Now looking at the measurements log on the client, the number in the
> 'Offset' column of the measurements log briefly spikes to -3.27E+02
> for 10 samples, then drops to roughly -2E-6. Meanwhile, even minutes
> later, 'chronyc tracking' shows a 'System time` value in the hundreds
> of seconds.
>
> Why do these values not match all the time? If measurements logs the
> difference between system time and the NTP server system time, why do
> I not see large and slowly decreasing numbers in the measurements log?
>
>
> On Tue, Sep 24, 2019 at 2:21 PM Bill Unruh <unruh@xxxxxxxxxxxxxx> wrote:
> > On Tue, 24 Sep 2019, Zack Shivers wrote:
> >
> > > It appears that none of the logs contain the system time offset. Is there a way to enable logging this?
> > > The field I'm interested in is the one given by 'chronyc tracking' as 'System time'.
> >
> > Which logs are you writing out?
> >
> > Measurements contains the difference between the system time and the time
> > obtained from the ntp packet, statistics contains the estimated offset as
> > determined by the least squares fit on the last N measurements. That is maybe
> > what you want.
> > >
> >
> > --
> > To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
> > with "unsubscribe" in the subject.
> > For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
> > with "help" in the subject.
> > Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.
>
> --
> To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
> with "unsubscribe" in the subject.
> For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx
> with "help" in the subject.
> Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.

--
To unsubscribe email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "unsubscribe" in the subject.
For help email chrony-users-request@xxxxxxxxxxxxxxxxxxxx 
with "help" in the subject.
Trouble?  Email listmaster@xxxxxxxxxxxxxxxxxxxx.