[winswitch] Question about Xpra, OpenGL, and Xming-Mesa
michael.sgv at gmail.com
Fri Dec 1 17:32:30 GMT 2017
Thank you for the prompt response. I added the "+extension GLX" to the
command line, so now my command in the file looks like this:
xvfb = Xvfb +extension GLX +extension Composite -screen 0 5760x2560x24+32
-dpi #96 -nolisten tcp -noreset -auth $XAUTHORITY
Unfortunately, the application is still refusing to start and gives me the
same "no OpenGL found in context" error.
Yesterday I was digging around the Xpra site and I saw that there seems to
be an issue with nvidia graphics cards and OpenGL (
https://xpra.org/trac/wiki/Xdummy). The graphics card for my local windows
PC (which, in X terminology, I guess would be called the "server"?) is an
Nvidia GeForce 960M, and I saw on the Xpra site that VirtualGL is a
potential solution for this. I just spent the past 6 hours reading the
documentation for VirtualGL on their site (
https://cdn.rawgit.com/VirtualGL/virtualgl/2.5.2/doc/index.html), but the
documentation was incredibly confusing. For instance, I could not for the
life of my figure out this image:
Shouldn't the "application" be sitting in on the "client "side, since in X
terminology the client side is the remote server where I'm running the
application? And, following the X terminology, shouldn't I have to download
VirtualGL for the "server" side--in this case, my local Windows PC? There
was no "server" downloads on the OpenGL site for Windows PC's, only for
"client" software which linked to a software called "Exceed", of what *I
think *is a commercialized version of Xming? And in fact, on the latest
download files for VirtualGL, the client software for Windows was
completely gone--there was just a "VirtualGLUtils" package that seemed to
just provide benchmarking tools.
Terminology confusion aside, I tried installing VirtualGL for my remote
Ubuntu 16.04 linux box (the "client"?) just because I had no idea what to
do at this point, and then tried running the following (with Xming and X11
tunnelling enabled over Putty):
"vglrun java -jar MyApp.jar"
but then VirtualGL gave me the error:
"[VGL] ERROR: Could not open display 0."
So at this point, I'm, completely confused. Any help/clarification you
could provide would be greatly appreciated; perhaps I should also ask over
at the VirtualGL support thread and see if they say anything.
Thank you for your time,
On Thu, Nov 30, 2017 at 8:35 PM, Antoine Martin via shifter-users <
shifter-users at lists.devloop.org.uk> wrote:
> On 01/12/17 01:11, Michael Segev via shifter-users wrote:
> > Hello,
> > My situation is as follows:
> > I'm currently on a Windows 10 machine that has Xpra, Xming-Mesa (Xming
> > OpenGL support), and PuTTY installed. I am trying to run a GUI java app
> > that uses OpenGL on a remote Linux (Ubuntu 16.04) machine and have it
> > display back on my PC using Xpra.
> > I can successfully run the application when I just use Xming-mesa.
> > when I add Xpra into the equation, the GUI java app fails to start and
> > throws this error:
> > OpenGL: ~~ERROR~~ RuntimeException: No OpenGL context found in the
> > thread
> > But if i shut down the Xpra server and just connect using Xming-mesa, the
> > application starts fine; so I'm somehow inclined to believe that there is
> > some conflict here involving Xpra and OpenGL.
> > Any advice you could provide to fix this would be greatly appreciated.
> The easiest way is to add "+extension GLX" to your xvfb command line in
> This will be the default for Ubuntu 16.04 in the upcoming 2.2 release.
> This solution is not accelerated.
> For more options, including OpenGL acceleration see:
> > Thank you.
> > _______________________________________________
> > shifter-users mailing list
> > shifter-users at lists.devloop.org.uk
> > http://lists.devloop.org.uk/mailman/listinfo/shifter-users
> shifter-users mailing list
> shifter-users at lists.devloop.org.uk
More information about the shifter-users