Re: COEX GPS (model UBX-M8030 IC) on Raspberry Pi 4 Model B 2018 connection issues

[Gary E. Miller]

Yo Kerry!

On Tue, 20 Jul 2021 16:20:00 +0800
Kerry Hoath <kerry@gotss.net> wrote:

> Garry does not like systemd and is of the opinion that it makes gpsd 
> difficult to manage.

Fact, not opinion.  Just look at the list archives.  systemd(umb)
comes up every week.  Nothing esle comes close as a defect attractor.

> For this reason it is best to take gpsd out of the control of systemd 
> and have scripts that start and stop it yourself.

That is my opinion.  systemd(umber) maybe could work, but the people that
push it o nthe (l)users have never bothered.

> The reason that you should upgrade to the latest gpsd is that there
> are many fixes and improvements for newer protocols and gps units in
> the latest version

Yes, because I hate getting bug reports for bugs fixed 5 years ago.

> and distributions often bundle old buggy code.

OnlLy some distros bundle old code.  I'll not get into the argument
of what is best to bundle as there are good reasons on both sides,
depending on your needs.

> The newest version should talk to more GPS units and is considered to
> be the best to run in production.

Not knowing your "production", that would be a choice the user has to
make.

If someone had spent millions of dollars testing an old gpsd version for
space flight, then I would suggest they leave well alone.

> All the names Garry gives to systemD like SystemDreamon are just
> because he has had bad experiences with it and dislikes it so much it
> amuses him to come up with a different insult for it which I agree
> could confuse a new user not used to this.

"amuse" is not the correct word, closer to "derision".  I kept having
to escalate my derision before people got the idea I really was very
serious on what a pox systemd(umpkopf) is.

Have you ever heard a Gentoo or Devuan user complain about their init
systems?  I really doubt is.  Just on this list we get the weekly
ssytemd(ross) whinging.


> I have not tried GPSD on the raspberry Pi myself but plenty have and
> I believe the docs on the GPSD website and the latest code are highly 
> recommended.

Yes, except that doc is for serial (GPIO) conenction, and ths OP is
using a USB connection.  That has been confusing the OP since the
start.

> Basically it boils down to installing a recent GPSD, (perhaps
> compiling it yourself from source)

Only as a start.  I don't think the OP has found a gpsd bug.  So an
update is just to get on the same page.

Systemd(racula) is at least a small part of his problem, but I suspect
that having the flight controller also talking to the GPS is causing his
biggest issue.

gpsdebuginfo output would be a good start in diagnosis.  Then an up to
date ubxtool can answer many questions.

> figuring out which serial device on the Pi has your GPS on it, and 
> pointing GPSD at that device to see if it can communicate with a 
> hardware device on that port.

Which gpsdebuginfo will tell us.

> Hope this helps a little.

Feedback is always good.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        gem@rellim.com  Tel:+1 541 382 8588

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

pgpTfSEIXhL4u.pgp
Description: OpenPGP digital signature