Re: [gpsd-dev] ✘master/slave gpsd

[Gary E. Miller]

Yo Eric!

On Thu, 26 Apr 2018 17:46:49 -0400
"Eric S. Raymond" <address@hidden> wrote:

> > The man page implies this should work?  I looked at the data with
> > tcpdump.  The slave gpsd connects to the master gpsd.  The gpsd
> > sends the VERSION JSON, then nothing happens.  The gpsd slave is
> > asked to WATCH data by the slave cgps, the the slave gpsd never
> > sends a WATCH to the master gpsd.  
> 
> It used to work.  I think there's a test in the regression suite that
> exercises it.

The only test I see is test/daemon/tcp-torture.log, and it passes.

I mentions:

"Savannah tracker bug  #36409: "GPSD fails to start get GPS data from
tcp://location:port".

Which looks like my bug.  tcp-torture.log is indeed run by 'scons
check', but only as a normal log, not as a gpsd master/slave case.

This was bug 36409 in 2013.  It appears to have been closed as 'fixed'
but the original reporter never responded to the fix.  The test case
does not exercise the error case.  

The bug descibed using 'socat' to send data to slave gpsd, it did not test the
full master/slave gpsd protocol.

I reopened that bug.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can’t measure it, you can’t improve it." - Lord Kelvin

pgpc8X1tWAYF7.pgp
Description: OpenPGP digital signature