[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Diagnosed]: Timeservice not running on Ubuntu 16.04

From: Gary E. Miller
Subject: Re: [Diagnosed]: Timeservice not running on Ubuntu 16.04
Date: Fri, 5 Jun 2020 15:28:26 -0700

Yo Dan!

On Fri, 5 Jun 2020 18:05:13 -0400
Dan Williams <> wrote:

> It required quite a few fixes, primarily verifying that all
> received sentences were reporting good fixes.

In gpsd land we call this: having a valid 3D fix.

>      - GPSD will not *Tell* you that these fixes are bad,

gpsd is not a client.  That is why the procedure says to look with
cgps to see that you have a valid fix.  All the client tell you if you
have a valid fix.

> NOR that
> it's why PPS won't output.  not even in debug level 6

I showed you in your debug logs where gpsd shows that.

Simple rule: No fix, no valid PVT.

That is why the procedures tell you to verify you had a fix before
proceeding to PPS.

gpsd tries hard to ensure that garbage in does not yield garbage out.

> - Lesson 3: GPSD can report a PPS signal to "gpsmon" without being in
> root.

Or rather, gpsmon can not open /dev/pps0 without being root.  But
that is distro/permission dependent.

> - Lesson 4: gpsd will fail back to ioctl mechanism even if it's run
> as root and has access to KPPS.  Even if your KPPS setup is faulty.

By design, as documented.

> - Lesson 5: The PPS errors "missing last_fixtime" do not prevent a
> pps fix. (visible at debug level 5 and above)

Correct, but the pps fix is not valid, so not used.  gpsd only uses
valid data.

> The gps receiver and computer I'm trying to set up are mounted on a
> vehicle, and I was debugging while parked.

Parked inside with bad skyview?  You logs only showed 3 sats in sight,
not enough for a 3D fix.

> Unfortunately for me, the GPS receiver (and IMU) only output valid
> sentences while it's calibrated, which effectively means: "while it's
> driving".

That would be a pretty bad bug in the oxts.

> For instance, our receiver can see 6 satellites, report 1m DOP,

No way you get 1m DOP with only 6 sats.  That would be another bug in
your oxts.

> and
> still not have a "valid" fix -- because it's waiting until the IMU &
> accelerometers are valid before reporting a valid fix.

That would be another bug in your receiver.

Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can't measure it, you can't improve it." - Lord Kelvin

Attachment: pgp23Cyi1qgz0.pgp
Description: OpenPGP digital signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]