CARLA 0.9.6 with Apollo 5.0 Contatiner issue

[marcusvinicius178]

CARLA version: 0.9.6 Platform/OS: Ubuntu 20.04 Problem you have experienced: Hi I am following this tutorial to set up CARLA with Apollo 5.0: https://github.com/AuroAi/carla_apollo_bridge. In "USAGE" step, I have an error after run this script inside the carla-apollo container:

python examples/manual_control.py

The Error:

libGL error: No matching fbConfigs or visuals found libGL error: failed to load driver: swrast X Error of failed request: BadValue (integer parameter out of range for operation) Major opcode of failed request: 152 (GLX) Minor opcode of failed request: 3 (X_GLXCreateContext) Value in failed request: 0x0 Serial number of failed request: 101 Current serial number in output stream: 102

I have already tried these approaches https://github.com/AuroAi/carla_apollo_bridge/issues/30 but Any solution...How to fix this LibGl error?

Thanks in advance

-->

[ldh2020]

https://github.com/AuroAi/carla_apollo_bridge/issues/11 Here can solve your problem.

[marcusvinicius178]

Hi @idh2020 thanks very much for your attention. I have already checked this solution previously. However the Docker commands suggested by this guy, did not work on my system. He suggested here https://github.com/SoonminHwang/dockers/issues/1#issue-350292669, to run the carla-server container using:

docker run -it \ -v /tmp/.X11-unix:/tmp/.X11-unix:rw \ --privileged \ -e DISPLAY=$DISPLAY \ -v /usr/lib/nvidia-384:/usr/lib/nvidia-384 \ -v /usr/lib32/nvidia-384:/usr/lib32/nvidia-384 \ --device /dev/dri \ YOUR_DOCKER_IMAGE_NAME

And later inside the docker image:

export PATH="/usr/lib/nvidia-384/bin":${PATH} export LD_LIBRARY_PATH="/usr/lib/nvidia-384:/usr/lib32/nvidia-384":${LD_LIBRARY_PATH}

In my case I am using nvidia-470 and in the folder /usr/lib I have the folder "nvidia" not the folder "nvidia-470".... Then I have modified the command above to my case:

Command 1

docker run -it \ -v /tmp/.X11-unix:/tmp/.X11-unix:rw \ --privileged \ -e DISPLAY=$DISPLAY \ -v /usr/lib/nvidia:/usr/lib/nvidia \ -v /usr/lib32/nvidia:/usr/lib32/nvidia \ --device /dev/dri \ carla-server

Command 2

export PATH="/usr/lib/nvidia/bin":${PATH} export LD_LIBRARY_PATH="/usr/lib/nvidia/usr/lib32/nvidia":${LD_LIBRARY_PATH}

But after issue on terminal the Command 1 I got this error:

I am not an expert in Docker. Would you have some hint to fix this command and adjust it to my machine I think in this way I will be available to fix this issue.

[ldh2020]

Hi @IDH2020 thanks very much for your attention. I have already checked this solution previously. However the Docker commands suggested by this guy, did not work on my system. He suggested here SoonminHwang/dockers#1 (comment), to run the carla-server container using:

docker run -it -v /tmp/.X11-unix:/tmp/.X11-unix:rw \ --privileged -e DISPLAY=$DISPLAY -v /usr/lib/nvidia-384:/usr/lib/nvidia-384 -v /usr/lib32/nvidia-384:/usr/lib32/nvidia-384 --device /dev/dri YOUR_DOCKER_IMAGE_NAME

And later inside the docker image:

export PATH="/usr/lib/nvidia-384/bin":${PATH} export LD_LIBRARY_PATH="/usr/lib/nvidia-384:/usr/lib32/nvidia-384":${LD_LIBRARY_PATH}

In my case I am using nvidia-470 and in the folder /usr/lib I have the folder "nvidia" not the folder "nvidia-470".... Then I have modified the command above to my case:

Command 1

docker run -it -v /tmp/.X11-unix:/tmp/.X11-unix:rw \ --privileged -e DISPLAY=$DISPLAY -v /usr/lib/nvidia:/usr/lib/nvidia -v /usr/lib32/nvidia:/usr/lib32/nvidia --device /dev/dri carla-server

Command 2

export PATH="/usr/lib/nvidia/bin":${PATH} export LD_LIBRARY_PATH="/usr/lib/nvidia/usr/lib32/nvidia":${LD_LIBRARY_PATH}

But after issue on terminal the Command 1 I got this error:

I am not an expert in Docker. Would you have some hint to fix this command and adjust it to my machine I think in this way I will be available to fix this issue.

I tested the method with nvidia-410 was OK, but I didn't know 470 how to do. Can you launch manual_control.py in local instead of in the container? If you didn't launch manual_control.py in the container, there won't be this problem. But, as shown in the picture, you cannot launch the container. Is it?

[marcusvinicius178]

Yes, locally the script works, however, I cannot work with the simulator locally, because the CARLA 0.9.6 is pretty old to install locally (I tried but the installation is not possible, some old libraries and compilers required are not even more available for ubuntu 20.04...I have packages conflicts installation that crashes the O.S) Unfortunately, Apollo 5.0 stack has a bridge available just with CARLA 0.9.6 using Docker....so locally won't work.

And yes, you are right, these docker commands do not work...I would like to understand what is wrong with these commands and adjust them in order to fix this communication with the LibGL issue.

[ldh2020]

Maybe you can change the version of carla to 0.9.10. I test this is OK, and you have to modify the code of lidar in apollo-bridge because of having some update in lidar of 0.9.10. You can find what different of lidar data between 0.9.10 and 0.9.6 in the manual_control.py.

[marcusvinicius178]

Hi @ldh2020 would you know How to use Localization instead ground truth? My doubt is in the last comment of issue below. However any answer. I guess Auro.AI stopped working on This bridge.

https://github.com/AuroAi/carla_apollo_bridge/issues/19

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.