Re: [gpsd-users] gpsd 3.12 cgps 3.12 - time out with NL-402U

[Gary E. Miller]

Yo Eric!

On Thu, 12 Mar 2015 09:45:44 -0400
"Eric S. Raymond" <address@hidden> wrote:

> Philipp Seelig <address@hidden>:
> > > How long did you wait?  This is sort of expected behavior.  I have
> > > seen the same thing for a long time.  cgps will timeout until the
> > > GPS starts sending data, which in a cold start could be many
> > > minutes.
> > > 
> > 
> > That is exactly what I misunderstood. I thought cgps would _always_
> > keep running, as long as a gps device was plugged. Which is why I
> > guessed that something is wrong with gpsd.
> > 
> > Sorry for the inconvenience.
> > 
> > Is there a easy way to determine if a gps device is present (even if
> > this device doesn't send any valid data yet)?
> 
> Another question is whether this means cgps's timeout should be
> removed.

I too find the current bahavior annoying, especially on startup while
doing the long wait for autobaud.  But I think timeouts should be
reserved for loss of connection to the server.  Two minutes is about
right, like any other TCP timeout.

I've been wondering if gpsd should send a keepalive if no other data to send.
Maybe every 10 seconds?  Maybe the keepalive could also mention what
devices are in use by gpsd?

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97701
        address@hidden  Tel:+1(541)382-8588

pgptvlMBzbQFo.pgp
Description: OpenPGP digital signature