Re: [gpsd-dev] Snap 3.15 shipped as promised; things to do next

[Eric S. Raymond]

Brian T (gmail) <address@hidden>:
> 
> 
> On 6/3/2015 9:23 AM, Eric S. Raymond wrote:
> >Brian T., would you be able to work with me on developing a script to
> >reproduce this problem?  If we runit at a high enough debug level I
> >think something will shake loose.
> 
> Here is a trace with gpsd v3.14, without the patch, debug level 5, gpsd 
> watching ttyS4 and ttyUSB1, then I cut the power to the device providing 
> ttyUSB1, and this is what I get in the syslog
> 
> Jun  5 17:53:19 gpsd[3382]: gpsd:CLIENT: => client(0): 
> $GPRMC,225320.000,A,3838.7079,N,08957.7429,W,0.1,0.0,050615,0.0,W*68\x0d\x0a
> Jun  5 17:53:19 gpsd[3382]: gpsd:PROG: Changed mask: 
> {ONLINE|TIME|LATLON|SPEED|TRACK|MODE|PACKET|CLEAR|PPSTIME} with reliable 
> cycle detection
> Jun  5 17:53:19 gpsd[3382]: gpsd:IO: <= GPS: 
> $GPGGA,225320.000,3838.7079,N,08957.7429,W,1,04,4.8,193.18,M,-33.6,M,,*59^M
> Jun  5 17:53:19 gpsd[3382]: gpsd:PROG: GPGGA sentence timestamped 225320.00.
> Jun  5 17:53:19 gpsd[3382]: gpsd:PROG: GPGGA ends a reporting cycle.
> Jun  5 17:53:19 gpsd[3382]: gpsd:CLIENT: => client(0): 
> $GPGGA,225320.000,3838.7079,N,08957.7429,W,1,04,4.8,193.18,M,-33.6,M,,*59\x0d\x0a
> Jun  5 17:53:19 gpsd[3382]: gpsd:PROG: Changed mask: 
> {ONLINE|TIME|LATLON|ALTITUDE|STATUS|MODE|PACKET|REPORT|PPSTIME} with reliable 
> cycle detection
> Jun  5 17:53:19 gpsd[3382]: gpsd:PROG: time to report a fix
> Jun  5 17:53:19 gpsd[3382]: gpsd:CLIENT: => client(0): 
> {"class":"TPV","device":"/dev/ttyS4","mode":3,"time":"2015-06-05T22:53:20.000Z","ept":0.005,"lat":38.645131667,"lon":-89.962381667,"alt":193.180,"epv":112.700,"track":0.0000,"speed":0.051}\x0d\x0a
> Jun  5 17:53:19 gpsd[3382]: gpsd:INFO: closing GPS=/dev/ttyUSB1 (10)
> Jun  5 17:53:19 gpsd[3382]: gpsd:PROG: no /etc/gpsd/device-hook present, 
> skipped running DEACTIVATE hook
> Jun  5 17:53:19 gpsd[3382]: gpsd:INFO: reconnection attempt on device 1
> Jun  5 17:53:19 gpsd[3382]: gpsd:PROG: no /etc/gpsd/device-hook present, 
> skipped running ACTIVATE hook
> Jun  5 17:53:19 gpsd[3382]: gpsd:INFO: opening read-only GPS data source type 
> 0 and at '/dev/ttyUSB1'
> Jun  5 17:53:19 gpsd[3382]: gpsd:ERROR: device open of /dev/ttyUSB1 failed: 
> No such file or directory - retrying read-only
> Jun  5 17:53:19 gpsd[3382]: gpsd:ERROR: read-only device open of /dev/ttyUSB1 
> failed: No such file or directory
> Jun  5 17:53:19 gpsd[3382]: gpsd:ERROR: /dev/ttyUSB1: device activation 
> failed.
> 
> I know from previous debugging, that at this point gpsd_await_data() on line 
> 2180 in gpsd.c returns AWAIT_FAILED, and gpsd exits with   
> "exit(EXIT_FAILURE)"
> 
> Not sure if you need more, but I can provide just about anything you request.
> 
> Thanks!
> 
> -Brian T.

I have lost most of the context on this bug, for which I apologize. I've
been very distracted by my work as the tech lead for NTPsec for most of
the last year.

Is your problem still present in the tip version of GPSD?

I expect to ship 3.16 before the end of the week in order to get some
urgent bug fixes fielded. After that, I'll try to budget time to
chase this down with you if it's still a problem.
-- 
                <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>