Re: [gpsd-users] syslog error message with new version - kernel PPS failed Connection timed out

[hans mayer]

Yo Gary,

>> Yo Garry,
> Just one 'r'.  Gary.

Big sorry, Gary.


> The ERROR is alarmist, I just changed git head to drop that to a WARN.
> It just means no time stamp was available, thus a timeout.  This is 
normal
> on some GPS if no sat lock.

That's exactly my problem. I don't get a time stamp
but I have a 3D fix with about 10 satellites.

> Well, none of this is a problem.  A problem would be if you do not
> get good time from /dev/pps1.

I don't get any time.


Kind regards
Hans

--



Am 18.08.2015 21:58, schrieb Gary E. Miller:
> Yo hans!
>
> On Tue, 18 Aug 2015 21:31:06 +0200
> hans mayer <address@hidden> wrote:
>
> > Yo Garry,
>
> Just one 'r'.  Gary.
>
> >   > Odd, the source for pps-ktimer says it is a client, not a source.
> >
> > It's a client as it has no own time source. It takes the
> > kernel time to generate the 1PPS information.
>
> Whcih makes it a server to my mind.  All in your point of view.
>
> >   > And how exactly did you try it?
> >
> > This is the way I try to use gpsd ( without success ) with
> > the PPS signal of gps-hardware with pps-gpio" kernal module.
>
> Different topic.  Already asked and answered in previsou emails.
>
> > For test reasons in combination with pps-ktimer I tried this:
> >
> > gpsd -n -F /var/run/gpsd.sock
> > -P /var/run/gpsd.pid /dev/ttyS2 /dev/pps1
>
> That should work, assuming the sysclock is pretty close to the
> /dev/ttyS2 clock.  Otherwise it should fail, so a really poor test.
>
> > And this is the syslog:
> >
> > Aug 18 21:14:37 gps0 kernel: [343435.561377] pps_ldisc: PPS line
> > discipline registered
> > Aug 18 21:14:37 gps0 kernel: [343435.572084] pps pps2: new PPS source
> > serial2
> > Aug 18 21:14:37 gps0 kernel: [343435.580213] pps pps2: source
> > "/dev/ttyS2" added
> > Aug 18 21:14:37 gps0 gpsd[29883]: gpsd:ERROR: KPPS:/dev/ttyS2 kernel
> > PPS failed Interrupted system call
> > Aug 18 21:14:37 gps0 gpsd[29883]: gpsd:ERROR: KPPS:/dev/pps1 kernel
> > PPS failed Interrupted system call
> > Aug 18 21:14:37 gps0 gpsd[29883]: gpsd:ERROR: KPPS:/dev/ttyS2 kernel
> > PPS failed Interrupted system call
> > Aug 18 21:14:37 gps0 gpsd[29883]: gpsd:ERROR: KPPS:/dev/pps1 kernel
> > PPS failed Interrupted system call
> > Aug 18 21:14:37 gps0 gpsd[29883]: gpsd:ERROR: KPPS:/dev/ttyS2 kernel
> > PPS failed Interrupted system call
> > Aug 18 21:14:40 gps0 gpsd[29883]: gpsd:ERROR: KPPS:/dev/ttyS2 kernel
> > PPS failed Connection timed out
> > Aug 18 21:14:43 gps0 gpsd[29883]: gpsd:ERROR: KPPS:/dev/ttyS2 kernel
> > PPS failed Connection timed out
> > Aug 18 21:14:46 gps0 gpsd[29883]: gpsd:ERROR: KPPS:/dev/ttyS2 kernel
> > PPS failed Connection timed out
>
> The /dev/ttyS2 fails as expected, as it should, because of the deficient
> RasPi tty's.   Already asked and answered in previous emails.
>
> The ERROR is alarmist, I just changed git head to drop that to a WARN.
> It just means no time stamp was available, thus a timeout.  This is normal
> on some GPS if no sat lock.
>
> The EINTR just means that no edge was detected, I just changed that to
> an INFO log.
>
> So really nothing to see here, all as it should be.
>
> > I tried to figure out if there is maybe an issue with
> > this "pps-gpio" module. But the errors are the same.
>
> Well, none of this is a problem.  A problem would be if you do not
> get good time from /dev/pps1.
>
> > Has this warning any relevance for general functionality
> > gpsd:WARN: KPPS:/dev/pps0 missing PPS_CAPTURECLEAR, pulse may be
> > offset
>
> Nope, that is a known bug in the pps-gpio driver.  I submitted a kernel
> page 6 months ago.  So, a different, unrelated issue.
>
> RGDS
> GARY
> ---------------------------------------------------------------------------
> Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
>         address@hidden  Tel:+1(541)382-8588