Jetson Orin Nano nv_tegra_release differs from Jetpack Version

I am new to working with Linux, and this Jetson Orin Nano DTK so forgive me if these questions are dumb as I am trying to learn as I go here.

The end goal is to setup a OpenALPR Agent software on this so I can hook up an Axis LPR camera and read license plates. Unfortunately I am running into quite a few issues trying to get this squared away.

I have flashed Jetpack 5.1.2-b104 onto this which is running well however when I attempted to use the Jetson method on the link above, the package was incredibly outdated and the CPU arch was not correct therefore I opted for their agent:4.1.7-arm64 build which I am able to make some headway on however when I run the Docker container commands, I receive the following error after running the following command which is #4 on their Docker instructions.

sudo docker run -P -v openalpr-vol1-config:/etc/openalpr/ -v openalpr-vol1-images:/var/lib/openalpr/ -it openalpr/agent:4.1.7-arm64 alprlink-register

Registration successful.
ConfigWriter::writeConfig = /etc/openalpr
Data accepted for processingSend heartbeat succeeded
Restarting services...
unix:///var/run/supervisor.sock no such file
Error restarting openalpr-daemon

I then found this method on the Nvidia forums however when I run the following commands, I am told that it Could not determine Jetpack version. After looking at what is ran on the link itself it raised some concerns…

elif cat /etc/nv_tegra_release  | head -n 1 | grep "R35.*REVISION:\s*[3].*"; then
    # We are running JetPack 5.1.x
    OS_NAME="Nvidia Jetson JetPack 5.1"
    DEB_REPO_PATH="deb https://deb.openalpr.com/${REPO_PREFIX}jetson51${REPO_POSTFIX} jetson51 main"
    EXTRA_PACKAGES="libopencv"

Now when I run cat /etc/nv_tegra_release I get the following output.

R35 (release), REVISION: 4.1, GCID: 33958178, BOARD: t186ref, EABI: aarch64, DATE: Tue Aug 1 19:57:35 UTC 2023

Now if I run sudo apt-cache show nvidia-jetpack I get the following output.

Package: nvidia-jetpack
Version: 5.1.2-b104
Architecture: arm64

I suppose my next question is… is this correct and if so why does the nv_targe_release differ from the actual Jetpack version?

Hi,
5.1.2 is the overall version and r35.4.1 is the BSP version. Please check the release page:
JetPack SDK 5.1.2 | NVIDIA Developer

Jetpack contains the packages like Jetson Linux BSP, TensorRT, CUDA, …

Note sure if the docker can be successfully run on Jetson devices. This would need other users to check and share experience.

5.1.2 is the overall version and r35.4.1 is the BSP version. Please check the release page:

If I understand correctly, the 5.1.2 is the JetPack SDK version, and the r35.4.1 is the actual JetPack Linux version? If this is correct, is there a method to go to a previous Linux version without reflashing and if so, how would I go about doing that?

Hi,
Jetpack 5.1.2 matches r35.4.1 so your setup looks fine. The docker is a 3rdparty contribution and would need other users to share experience.

You may try our dockers as reference:
NVIDIA L4T Base | NVIDIA NGC

I apologize @DaneLLL but I’m slightly confused.

Are we concerned that Docker is not working appropriately on the Jetson, or is there something more? I ask because I have docker installed and the software has documentation for Docker specific installations on a Jetson. With that said, I even attempted to use their CURL methods, in which it stated the issue with the revision / release versioning on the nv_tegra_release.

Hi,
Do you know if the docker only can work with Jetpack 5.1.1(r35.3.1)? Can see the if condition in
https://deb.openalpr.com/install

  elif [ "$L4T_VERSION" = "35.3.1" ]; then

    OS_NAME="Nvidia Jetson JetPack 5.1"
    DEB_REPO_PATH="deb https://deb.openalpr.com/${REPO_PREFIX}jetson51${REPO_POSTFIX} jetson51 main"

If yes, please re-flash the system to Jetpack 5.1.1 through SDKManager. The system cannot be downgraded through commands.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.