Failing to launch the rviz2 GUI with Autoware running on Jetson Xavier NX
Hello! I've been trying to run the Autoware.Auto 3D perception stack tutorial in a Jetson Xavier NX (Ubuntu 18.04) with the ADE environment with no success.
I have followed the installation steps from here.
As my system is running under the ARM64 architecture, I first sourced the .aderc-arm64
file (from within the AutowareAuto folder), and then ran ade start
as stated on this post. In this case it fails to open rviz2 and throws the following error message:
[ERROR] [rviz2]: Failed to create an OpenGL context. BadValue (integer parameter out of range for operation) [ERROR] [rviz2]: RenderingAPIException: Unable to create a suitable GLXContext in GLXContext::GLXContext at /tmp/binarydeb/ros-dashing-rviz-ogre-vendor-6.1.7/obj-aarch64-linux-gnu/ogre-master-ca665a6-prefix/src/ogre-master-ca665a6/RenderSystems/GLSupport/src/GLX/OgreGLXContext.cpp (line 60) [ERROR] [rviz2]: rviz::RenderSystem: error creating render window: RenderingAPIException: Unable to create a suitable GLXContext in GLXContext::GLXContext at /tmp/binarydeb/ros-dashing-rviz-ogre-vendor-6.1.7/obj-aarch64-linux-gnu/ogre-master-ca665a6-prefix/src/ogre-master-ca665a6/RenderSystems/GLSupport/src/GLX/OgreGLXContext.cpp (line 60) [ERROR] [rviz2]: Unable to create the rendering window after 100 tries terminate called after throwing an instance of 'std::runtime_error'
what(): Unable to create the rendering window after 100 tries Aborted (core dumped)
I've also tried making the changes suggested in the .aderc-arm64
file (mainly setting ADE_DISABLE_NVIDIA_DOCKER=false
or even removing it completely) and also what was suggested here. In this case, ade start
command throws an error.
I also tried to check the conditions described here to run basic GUI applications within the Docker container with an NVIDIA GPU:
- I ran the command
ls -la /dev | grep nvidia
to see how my GPU is listed in the system and I could not findnvidia0
as it is expected. The command returnedtty1
andttyTCU0
. I guess the problem lies around here. - I also checked if the
nvidia-container-toolkit
is installed in my system with the commandapt policy nvidia-container-toolkit
, and everything seems to be fine.
Any idea on how to solve the issue from point number 1?
I would appreciate the help as I've been stuck in this issue for several days now.
Can you paste here what you get when you run the following commands on the host?