Re: [chrony-dev] [GIT] chrony/chrony.git branch, master, updated. 1.26-31-g19b3c5b

[Miroslav Lichvar <mlichvar@xxxxxxxxxx>]

On Fri, Feb 03, 2012 at 08:59:09AM -0800, Bill Unruh wrote:
> On Fri, 3 Feb 2012, Miroslav Lichvar wrote:
> 
> >
> >Let me know if you think there are other useful values that should be
> >added. Also, I'm wondering if it would be a good idea to show "System
> >time" and "Frequency" as signed values instead of the slow/fast to
> >make parsing easier.
> 
> Perhaps. Also, an indication as to whether or not the system time offset is
> real (Ie the system has locked onto a source) or not to determine tht system
> time.  One can look for a * in the sources, but it might be easier to look at
> the tracking to tell if the time is reasonable.

I think that can already be determined from the Reference ID field. If
it's 0.0.0.0, chronyd is not locked. An interesting question is how to
determine if it's locked well, looking at the System time value alone
isn't very useful.

> >This is how tracking and sources reports now look like:
> >
> >System time     : 0.000000093 seconds slow of NTP time
> >#* SOC1                          0   4   377     7   -675ns[ -778ns] +/-  128ns
>                                                         ^^^^^
> The ns as part of the word makes it hard to parse. Since it is not reported in
> a fixed set of units, one must parse the units (ns, ms, us, s) to determine
> the offset automatically, and the lack of a separator makes it just a bit
> harder to do so.

Perhaps a chronyc option to enable a machine readable output where
floating point numbers instead of the units are printed would be
useful?

-- 
Miroslav Lichvar

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