[Paraview] setup pvserver with remote rendering, was Segfault with OpenGL2

[Chuck Atkins]

What do the EGL settings in your CMakeCache look like?  You can find them
from the build directory:

$ grep '^EGL[^-]*=' CMakeCache.txt
EGL_INCLUDE_DIR:PATH=/usr/include
EGL_LIBRARY:FILEPATH=/usr/lib64/libEGL.so
EGL_gldispatch_LIBRARY:FILEPATH=/usr/lib64/libGLdispatch.so
EGL_opengl_LIBRARY:FILEPATH=/usr/lib64/libOpenGL.so

Your output will likely use different paths as this is from a Fedora
installation.

- Chuck

On Tue, Apr 12, 2016 at 9:39 AM, Harald Klimach <harald at klimachs.de> wrote:

> Hi,
>
> > Or I also noticed that you're running a very recent version of the
> NVidia driver.  I believe this actually has EGL support, which will allow
> you to build ParaView configured to use the GPU for off screen rendering
> with no X server necessary at all.  To do this, configure ParaView with the
> following CMake options:
> >       • -DVTK_RENDERING_BACKEND=OpenGL2
> >       • -DVTK_USE_OFFSCREEN_EGL=ON
> > You would then run pvserver on the remote machine with no need for an X
> server and connect to it with the GUI client in the same way.
>
> I gave this a try.
>
> But it results in an error box with:
>
> OpenGL drivers on the server side don't support
> required OpenGL features for basic rendering.
> Remote rendering will be disabled.
>
> OpenGL Vendor: NVIDIA Corporation
> OpenGL Version: 4.5.0 NVIDIA 361.28
> OpenGL Renderer: Tesla M2075/PCIe/SSE2
>
> What am I doing wrong?
> How can I enable the remote rendering. Are there other compilation flags I
> need to take care of?
>
> Also, this only works if I set this up manually (run pvserver on the
> remote box) and then create a ssh-tunnel
> to connect to it.
> If I use a command in the pvserver configuration of the client gui, to
> create the ssh tunnel and start the
> pvserver on the remote box, I get:
>
> OpenGL Vendor: Information Unavailable
> OpenGL Version: Information Unavailable
> OpenGL Renderer: Information Unavailable
>
>
> I guess, I am doing something wrong here, too:
>
> I have: Server Type = Client/Server
> Host=localhost
> Port=11772
> Command:  ssh -L 11772:remote-box:11772 user at remote-box pvserver
> --server-port=11772
>
>
>
> And if I do this command manually, and afterwards try to connect to the
> server, I get::
>
> ERROR: In
> /home/user/abs/paraview/src/ParaView-v5.0.1-source/VTK/Rendering/OpenGL2/vtkEGLRenderWindow.cxx,
> line 315
> vtkEGLRenderWindow (0x1660940): Only EGL 1.4 and greater allows OpenGL as
> client API. See eglBindAPI for more information.
>
> ERROR: In
> /home/user/abs/paraview/src/ParaView-v5.0.1-source/VTK/Rendering/OpenGL2/vtkOpenGLRenderWindow.cxx,
> line 575
> vtkEGLRenderWindow (0x1660940): GLEW could not be initialized.
>
> [shu:09048] *** Process received signal ***
> [shu:09048] Signal: Segmentation fault (11)
> [shu:09048] Signal code: Address not mapped (1)
> [shu:09048] Failing at address: (nil)
> [shu:09048] [ 0] /usr/lib/libc.so.6(+0x33330)[0x7f0c77394330]
> [shu:09048] *** End of error message ***
>
> Maybe this is due to a different environment in the tunneled login over
> the interactive login?
> Some settings that are available in the interactive shell, but not in the
> tunneled command:
> SSH_TTY
> ETS_TOOLKIT
>
> If I set DISPLAY=:0.0 in tunneling command, I get the same error as in the
> manual setup.
> Maybe, my compilation with EGL did not succeed?
>
> I used these flags:
>    -DVTK_USE_X:BOOL=OFF \
>    -DVTK_RENDERING_BACKEND:STRING=OpenGL2 \
>    -DVTK_USE_OFFSCREEN_EGL=ON \
>
> Thanks for your kind support!
> Harald
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20160412/58c971a5/attachment.html>