Docker system setup not working

Hello.

I have Nvidia Jetson Nano L4T R32.7.6. During the bash install.sh command of the Jetson containers system setup I’ve encountered an error:

Could not find a version that satisfies the requirement black>=23.0.0 (from -r /home/nvidia/jetson-containers/requirements.txt (line 8)) (from versions: 18.3a0, 18.3a1, 18.3a2, 18.3a3, 18.3a4, 18.4a0, 18.4a1, 18.4a2, 18.4a3, 18.4a4, 18.5b0, 18.5b1, 18.6b0, 18.6b1, 18.6b2, 18.6b3, 18.6b4, 18.9b0, 19.3b0, 19.10b0, 20.8b0, 20.8b1, 21.4b0, 21.4b1, 21.4b2, 21.5b0, 21.5b1, 21.5b2, 21.6b0, 21.7b0, 21.8b0, 21.9b0, 21.10b0, 21.11b0, 21.11b1, 21.12b0, 22.1.0, 22.3.0, 22.6.0, 22.8.0)
No matching distribution found for black>=23.0.0 (from -r /home/nvidia/jetson-containers/requirements.txt (line 8))

If I change the requirements for Black to 22.8.0 I get a similar Pyflakes error.

Additionally Black requires Python version 3.10 or greater. Can I upgrade it on this rustic glass house of a system?

Sincerely
Emil

Hi,

Which container do you want to use?
You might find the prebuilt in the link below so you don’t need to build it on your own:

Thanks.

Below is the dockerfile, should explain it all. I understand that there’s ros humble desktop which contains demo nodes, but I’d have to write my own packages anyway. Would building them inside a container and then preserving its state be a better idea?

I also tried specifying the name of the ROS2 package instead of giving a github link, also no luck and errors about not finding the package.

FROM dustynv/ros:humble-ros-core-l4t-r32.7.1

ENV ROS_DISTRO=humble
ENV ROS_WORKSPACE=/workspace

COPY ros2_install.sh /ros2_install.sh

RUN chmod +x /ros2_install.sh && \
    curl -sSL https://raw.githubusercontent.com/ros/rosdistro/master/ros.key \
      -o /usr/share/keyrings/ros-archive-keyring.gpg && \
    apt update && \
    apt install -y \
        iputils-ping \
        vim \
    && rm -rf /var/lib/apt/lists/*

RUN /ros2_install.sh https://github.com/ros2/demos/tree/humble/demo_nodes_cpp

Hi,

The container for R32 is no longer updated.
So it’s recommended to try the prebuilt container and try to install the required package within the container.

Thanks.