[GRASS-user] disappearing d.mon

[Markus Neteler]

Hi,

On Wed, May 12, 2021 at 4:26 PM Benjamin Ducke <benducke at fastmail.fm> wrote:
>
> On 12/05/2021 13:30, Dave Roberts wrote:
> > Colleagues,
> >
> >      Finally, after probably a year I now have no errors about
> > incompatible versions of wxwidgets, wxpython, and GRASS.  However, when
> > I execute
> >
> > d.mon wx0
> >
> > it pops up the box, but which then spontaneously disappears after about
> > 10-15 seconds.  It leaves behind the MONITORS/wx0 file.  Honestly I'm
> > pretty much at my wits end with wxwidgets.
> >
> > I run manjaro(arch) linux with the the custom build by Sylvain Poulain at
> >
> > https://github.com/giscan/AUR-grass
> >
> > uname -a
> > Linux sonnyboy 5.4.116-1-MANJARO #1 SMP PREEMPT Sun May 2 11:10:35 UTC
> > 2021 x86_64 GNU/Linux
> >
> > Thanks in advance for any help, Dave
>
> Same here:
> - Arch Linux
> - grass 7.8.5
> - GTK 3.24.29
> - wxGTK2/3 3.0.5
> - system Python 3.9.4
>
> After a 'd.mon start=wxN', the monitor comes
> up, I can use 'd.*' commands to draw into it.
> I can resize and move the window, and I can
> interact with the UI elements in the icon and
> status bars. However: When I move the mouse
> cursor inside the monitor's map canvas, d.mon
> crashes and closes its window (after a delay
> of a few seconds).
>
> I can then use 'd.mon stop=wxN' and start
> the monitor again, with the same behaviour
> as above.

I was able to reproduce this problem in a fresh Manjaro installation
in Virtualbox.

I have created a new bug report:
https://github.com/OSGeo/grass/issues/1576

Now someone with gdb and child process debugging skills is needed to
get a proper backtrace.

best,
Markus