Re: [gpsd-dev] ntpshmmon

[Nuno Gonçalves]

Hi,

address@hidden:~$ sudo ntpshmmon -c 0.1
ntpshmmon version 1
#      Name   Seen@                Clock                Real
    L Prec
sample NTP0 1445944686.190394092 1445944686.190303261 1445944686.075000047 0  -1
sample NTP0 1445944686.240551357 1445944686.190303261 1445944686.075000047 0  -1
sample NTP0 1445944686.290560999 1445944686.190303261 1445944686.075000047 0  -1
sample NTP0 1445944686.340675889 1445944686.190303261 1445944686.075000047 0  -1
sample NTP0 1445944686.390733864 1445944686.190303261 1445944686.075000047 0  -1
sample NTP0 1445944686.440748547 1445944686.190303261 1445944686.075000047 0  -1
sample NTP0 1445944686.490911978 1445944686.190303261 1445944686.075000047 0  -1
sample NTP0 1445944686.540972995 1445944686.190303261 1445944686.075000047 0  -1
sample NTP0 1445944687.190219188 1445944687.190141606 1445944687.075000047 0  -1
sample NTP0 1445944687.240370286 1445944687.190141606 1445944687.075000047 0  -1
sample NTP0 1445944687.290436260 1445944687.190141606 1445944687.075000047 0  -1
sample NTP0 1445944687.340443027 1445944687.190141606 1445944687.075000047 0  -1


I still don't understand.

Does this example work as supposed?

It looks to me "1445944686.075000047" is a single signal, and we are
reading duplicates of it.

I'm using -c 0.1 just to display this behaviour, but the problem
started to me without the -c option, and it was printing the same
signal twice, 0,5s appart.

Thanks,
Nuno

On Tue, Oct 27, 2015 at 3:08 AM, Gary E. Miller <address@hidden> wrote:
> Yo Nuno!
>
> On Tue, 27 Oct 2015 00:01:55 +0000
> Nuno Gonçalves <address@hidden> wrote:
>
>> Without using any option ntpshmmon was producing sometimes 2 results
>> per second for the same signal, and other times just 1 as supposed.
>
> Well, supposed for some cases and not for some others.
>
>> The code also have a comment saying "* we're ignoring duplicates via
>> timestamp,".
>
> Yup.
>
>> But the code uses the shm_stat.tvc timestamp for this, which is the
>> collection time.
>
> Yup.  We only collect once per pulse, so any of the time stamps ensures
> uniqueness.
>
>> So this is why I receive duplicates.
>
> Well, that would be a bug.
>
>> I would be glad to fix it if I know what is the intended behaviour.
>
> You understand the intended behavior now?
>
> RGDS
> GARY
> ---------------------------------------------------------------------------
> Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
>         address@hidden  Tel:+1 541 382 8588