Re: [gpsd-dev] Need p/d help with gpsd and or gps HW - no response after gps polling

[Steve Lang]

We found out what was causing the problem. It appears to be user
error. User was executing:

gpsmon /dev/gpsd0 while gpsd was running.

ls -al /dev/gpsd0
/dev/gpsd0 -> ttyAMA0

I informed user that 2 programs can not access the same serial port at
the same time.

Once user stopped doing this the problem stopped,

Regards,

--Steve

On Wed, Jul 13, 2016 at 10:30 PM, Steve Lang <address@hidden> wrote:
> On Wed, Jul 13, 2016 at 1:40 PM, Gary E. Miller <address@hidden> wrote:
>> Yo Steve!
>>
>> On Tue, 12 Jul 2016 22:47:19 -0500
>> Steve Lang <address@hidden> wrote:
>>
>>> HW Config: Pi 3, Adafruit GPS hat, and LNA
>>> SW Config: latest Jessi / clockmaker install script abd SW updates to
>>> support installing the following scons options:
>>
>> OK.  What kernel version?  What did you do to get systemd out of the way?
>
> Kernel version is 4.4.13-v7+
>
> I'll take a look at systemd. The service scripts were created by
> clockmaker and appear to be in-order but will take a look around at
> the common problems related to this in the forums.
>
>> I suspect that is your problem.  Systemd has been sabotaging many people.
>
> Great, I'll look into this.
>
>>
>> I have the same config, except I use Gentoo stable and kernel
>> 4.4.14-v7+.  No problems for me.  You can see how it works well here:
>>
>>     https://pi3.rellim.com/week/
>
> I have a working one also. I built 2 Pi3 GPS's. The difference between
> the 2 are physical location and the gps hat used. Same exact software
> setup though. The working hat is the uptronics hat where the poorly
> working one is the adafruit.
>
>>> scons timeservice=yes nmea0183=yes fixed_port_speed=9600
>>> fixed_stop_bits=1 xgps=yes leapfetch=yes qt=yes gpsdclients=yes
>>
>> The Adafruit is a mediatek chip, so you should also enable the
>> "mtk3301" support.
>>
>
> Fantastic, I'll add this into the build options. Thank you for
> pointing this out.
>
>>> Symptoms: gpsd appears to hang if gpsd does not receive data
>>
>> That is the design of gpsd.  Nothing in means nothing out.
>>
>> B why would you ever not have data in?
>>
>>> When the problem happens, ntp moves off the gps shm source to an
>>> external provider and the following is logged:
>>
>> That all looks good.
>>
>>> The last 4 lines are repeated for the remainder of the log file.
>>
>> Yeah, you still have PPS, but no serial data.
>
> I suspect that this was the case but was not sure.
>
>>
>>> ----------Cut------------
>>> gpsd:DATA: /dev/gpsd0 returned zero bytes
>>> gpsd:DATA: /dev/gpsd0 will be repolled in 0.010000 seconds
>>> ---------Cut-------------
>>
>> Yup, no serial data.  No data in, thus no data out.  As designed.
>>
>>> Due to lack of logging in this path I cant tell where we ended up and
>>> if an attempt was planned to resume IO to this GPS.
>>
>> Nothing to log.  No data in, thus no data out.
>>
>>> I ran gpsmon when this condition is happening.which seems to return
>>> valid data ..
>>
>> Forget gpsmon, it will just confuse the issue.  Use cgps.  If gpsd is
>> disconnected then gpsmon may directly connect to the GPS and mislead you
>> into thinking gpsd is fine.
>
> Thank you again for your time and incite into how gpsd works. So far
> the problem has not re-appeared but now I have the tools to move
> forward and possibility get this hat into a working state or at least
> have an idea of where the problem is located.
>
> Your help is very much appreciated over here.
>
> Regards,
>
> --Steve
>
>>
>> RGDS
>> GARY
>> ---------------------------------------------------------------------------
>> Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
>>         address@hidden  Tel:+1 541 382 8588