Re: gpsd switched to wrong driver SVN 6397

[Michael R. Davis]

GPSD Folks,
Well gpsd switched my BU-303 to the "Trimble TSIP" driver again for no reason and never recovered.  I guess my next step is to take that driver out of my compile.  I was running for five days straight without a hitch until this.  Why would gpsd even think about changing the driver on a GPS that has been running for five days?

Last Fix before driver switch:
TPV, Time: 1255970295.47, Lat: 38.951514039, Lon: -77.352128077, Speed: n/a, Heading: n/a

$ telnet localhost 2947
Trying 127.0.0.1...
Connected to localhost.localdomain (127.0.0.1).
Escape character is '^]'.
{"class":"VERSION","release":"2.40dev","rev":"$Id: gpsd_json.c 6377 2009-10-01 23:24:21Z esr $","proto_major":3,"proto_minor":1}
?DEVICES;
{"class":"DEVICES","devices":[{"class":"DEVICE","path":"/dev/ttyUSB0","activated":1256050073.45,"flags":1,"driver":"Trimble TSIP","native":1,"bps":4800,"parity":"N","stopbits":1,"cycle":1.00}]}

$ svn info
Path: .
URL: svn://svn.berlios.de/gpsd/trunk
Repository Root: svn://svn.berlios.de/gpsd
Repository UUID: 3da76a2a-08e2-0310-9c3e-bc9091dd3f7d
Revision: 6397
Node Kind: directory
Schedule: normal
Last Changed Author: esr
Last Changed Rev: 6397
Last Changed Date: 2009-10-11 20:58:43 -0400 (Sun, 11 Oct 2009)

--- On Tue, 10/6/09, Michael R. Davis <mrdvt92@...> wrote:

> From: Michael R. Davis <mrdvt92@...>
> Subject: Re: [Gpsd-dev] gpsd switched to wrong driver SVN 6377
> To: gpsd-dev@...
> Date: Tuesday, October 6, 2009, 1:42 PM
> > "I have never heard of / seen
> this before."
> > like to see some logs including packet
> > sniffer showing the failure
>
> I've had the BU-303 running ever since without a glitch.
>
>
>      
>

     
_______________________________________________
Gpsd-dev mailing list
Gpsd-dev@...
https://lists.berlios.de/mailman/listinfo/gpsd-dev

[Eric S. Raymond-2]

Michael R. Davis <mrdvt92@...>:
> Well gpsd switched my BU-303 to the "Trimble TSIP" driver again for
> no reason and never recovered.  I guess my next step is to take that
> driver out of my compile.  I was running for five days straight
> without a hitch until this.  Why would gpsd even think about
> changing the driver on a GPS that has been running for five days?

Well...gpsd must think it sees a TSIP packet in the data stream.

The way the dispatch logic works is per-packet.  The packet getter
is constantly trying to match one of the known packert types.  Whenever
it gets a match, it switches to the driver for that trye.

The only scenario I cann even imagine leading to this is that a burst
of line noises confused gpsd into thinking it saw a TSIP packet.  That's
not strange; what's strange is that it never switched *back* in response
to following SiRF packets.

Chris, is there something funky about our state machine that keeps
us from climbing out of that hole?
--
                <a href="http://www.catb.org/~esr/">Eric S. Raymond</a>
_______________________________________________
Gpsd-dev mailing list
Gpsd-dev@...
https://lists.berlios.de/mailman/listinfo/gpsd-dev

[Evrard Patrick]

 
Helo

I think that my email does not belong to your community


-----Original Message-----
From: gpsd-dev-bounces@...
[mailto:gpsd-dev-bounces@...] On Behalf Of Eric S. Raymond
Sent: Monday, October 26, 2009 1:11 PM
To: gpsd-dev@...
Subject: Re: [Gpsd-dev] gpsd switched to wrong driver SVN 6397

Michael R. Davis <mrdvt92@...>:
> Well gpsd switched my BU-303 to the "Trimble TSIP" driver again for
> no reason and never recovered.  I guess my next step is to take that
> driver out of my compile.  I was running for five days straight
> without a hitch until this.  Why would gpsd even think about
> changing the driver on a GPS that has been running for five days?

Well...gpsd must think it sees a TSIP packet in the data stream.

The way the dispatch logic works is per-packet.  The packet getter
is constantly trying to match one of the known packert types.  Whenever
it gets a match, it switches to the driver for that trye.

The only scenario I cann even imagine leading to this is that a burst
of line noises confused gpsd into thinking it saw a TSIP packet.  That's
not strange; what's strange is that it never switched *back* in response
to following SiRF packets.

Chris, is there something funky about our state machine that keeps
us from climbing out of that hole?
--
                <a href="http://www.catb.org/~esr/">Eric S. Raymond</a>
_______________________________________________
Gpsd-dev mailing list
Gpsd-dev@...
https://lists.berlios.de/mailman/listinfo/gpsd-dev
_______________________________________________
Gpsd-dev mailing list
Gpsd-dev@...
https://lists.berlios.de/mailman/listinfo/gpsd-dev