Re: [gpsd-dev] 'internal error - too many satellites'

[Fred Wright]

On Tue, 29 Aug 2017, Zachary Bourdeau wrote:

> Is there reason why GPSD cannot accept more than 10 satellites at a time
> through the GPGSV sentence? It looks like that is whats happening. I am
> not sure if this is a bug or by design. I can turn off the GPGSV
> sentence if that would help, but i'm curious to see what feedback you
> folks have on this issue before I do. ?
>
> (92) 
> $GPGSV,3,1,11,10,76.6,281.3,30.9,15,20.4,46.4,30.9,18,66.8,63.3,24.4,21,43.6,122.3,34.3*70\x0d\x0a
> (89) 
> $GPGSV,3,2,11,24,17.6,87.2,31.8,27,50.6,255.9,30.4,32,30.9,189.8,26.3,08,33.8,302.3,*5B\x0d\x0a
> gpsmon:ERROR: internal error - too many satellites [72]!
> (52) $GPGSV,3,3,11,14,13.4,192.7,,16,,,,20,1.4,74.5,*71\x0d\x0a

There's certainly no 10-satellite limit.  I believe the "72" in the error
message is the real limit.  Also, I just ran gpsmon against a multi-system
GNSS receiver here with up to 20 satellites in view, 12 GPS (GPGSV) and 8
GLONASS (GLGSV), without any issues beyond the 12-satellite limit of the
view pane (which simply truncates the list).

First of all, are you using the latest source code?  There's no point in
trying to chase a bug that was already fixed.

Secondly, can you capture some data with gpscat that reproduces the
problem?

The "\x0d\x0a" in the above is fishy, since gpsmon doesn't normally report
the CR-LF that way.  Is the data being run through something that's
"decoding" the CR-LF into printable characters?  Even if it is, the
internal error isn't appropriate, but it might prevent correct results.

Fred Wright