Re: Traceback from gpscsv

[Gary E. Miller]

Yo Hans!

On Wed, 14 Jun 2023 23:43:32 +0200
Hans Mayer <gpsd@ma.yer.at> wrote:

> # python
> Python 3.9.2 (default, Feb 28 2021, 17:03:44)
> [GCC 10.2.1 20210110] on linux

Python 3.9 is starting to near EOL on many distros...

> this server has no gpsd running

Not really a supported mode for gpscsv.

> in the past gpscsv was running successfully but now it stops with a 
> backtrace

Can you provide the FULL backtrace?

> mayer# gpscsv -n 1 -c SAT --header 0 pi4.home.yer.at:gpsd:/dev/serial0

So you are connecting toa gpssd on a remote host.

> But other programs like xgps or cgps are working fine.

Clearly a python bug, and xgps/cgps are not python.  I should be able
to fix that. gpscsv assumes your eceiver is sendint the time, and it is
not.  But when you reset your receiver, it no longer knows the time.

> I thought maybe some updates did break this program. Therefore i
> started from scratch downloading gpsd, compiling it and installing
> it. But this didn't help. (but I am on the latest version now)

gpscsv is relatively new, so a bug like that is not unexpected.

> Randomly I fixed this issue.

More like avoided the issue, the fix requires a Python patch.

> The fix was I send a "RESET" to the ZED-F9P as described in 
> https://gpsd.io/ubxtool-examples.html

Which puts the recevier back to defaults.

> In the meantime I can also reproduce the issue. After sending a 
> "WARMBOOT" to the u-blox the problem exists.  After waiting this 
> WARMBOOT has finished gpscsv generated this nasty traceback.

When the receiver restarts, it does not know the time.  Thus the missing
time in the data[] dictionary.

> So what does WARMBOOT do ?

That requires a long and detailed answer.  I suggest you read the
complete description in the u-blox doc:

    'ZED-F9P_InterfaceDescription_(UBX-18010854).pdf'

    Section 3.10.19.1 Reset Receiver / Clear Backup Structures

> As far as I understand "RESET" will set
> all values to default values from layer 7 to lower layers.

Nope.  Read the doc.

> But a
> WARMBOOT after a RESET should change anything in my opinion. Or does
> the RESET only write from default layer 7 to RAM but not to BBR and
> flash ?

Nope.  Read the doc.

> Then the question is: what changed ?

Between what and what?

> The interfacedescription PDF in 3.10.19.1 describes hot start, warm 
> start and cold start and different reset types. How does this reflect
> to RESET, COLDBOOT and WARMBOOT in
> https://gpsd.io/ubxtool-examples.html  ?

One and the same.

You really do not want to reset/coldboot/warmboot your receiver just
before using gpscsv.  You just told the receiver to delete all the
data that you want gpsdcsv to collect.

And totally unrelated to your bug in the gpscsv Python code.

I just pushed a fix to git head for the missing time in the sky report.

Please give it a try.

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

pgpL5ZujktFCw.pgp
Description: OpenPGP digital signature