jetpack 6.2.1
In the file /etc/cdi/nvidia.yaml, the devices generated by nvidia-ctk cdi list do not include nvidia.com/gpu=tegra.
As a result, when a Pod is created in the Kubernetes cluster, it fails with:
failed to create task for container: failed to create shim task: OCI runtime create failed: could not apply required modification to OCI specification: error modifying OCI spec: failed to inject CDI devices: unresolvable CDI devices nvidia.com/gpu=tegra: unknown
*** Please note that this reply is generated by LLM automatically ***
Based on the provided context, it appears that the issue you are experiencing is related to the Jetson Orin Nano device and its compatibility with Kubernetes. The error message you provided suggests that there is an issue with the device tree generated by the do command, which is causing the Pod creation to fail in the Kubernetes cluster.
After reviewing the context, I did not find any specific information that directly addresses this issue. However, I can provide some general insights and potential troubleshooting steps.
Device Tree Generation: The do command is likely generating the device tree for the Jetson Orin Nano device. The device tree is a data structure that describes the hardware components of a system. If the device tree is not generated correctly, it can cause issues with the system’s functionality.
Kubernetes Compatibility: Kubernetes is a container orchestration system that relies on the underlying hardware and software infrastructure to function correctly. If the device tree is not compatible with Kubernetes, it can cause issues with Pod creation and deployment.
JetPack 4.6.1: JetPack is a software development kit (SDK) provided by NVIDIA for their Jetson modules. JetPack 4.6.1 is a specific version of the SDK that may have its own set of compatibility issues with Kubernetes.
To troubleshoot this issue, you may want to try the following steps:
Verify Device Tree Generation: Check the device tree generated by the do command to ensure that it is correct and complete. You can use tools like dtc (Device Tree Compiler) to verify the device tree.
Check Kubernetes Version: Ensure that you are running a compatible version of Kubernetes with JetPack 4.6.1. You can check the Kubernetes version by running kubectl version.
Consult NVIDIA Documentation: NVIDIA provides documentation and support for their Jetson modules, including JetPack. You can consult the NVIDIA documentation to see if there are any known issues or compatibility problems with Kubernetes.
Seek Community Support: You can also seek support from the Kubernetes community or NVIDIA forums to see if others have experienced similar issues.
In summary, while I did not find a direct solution to your issue, I hope these troubleshooting steps and insights help you to identify and resolve the problem.
*** Please give the thumbs up if you get answers you like. Or provide feedback to help us improve the answer. ***
Hi,
By default Kubernetes is not supported on default Jetpack release. This would need community support. Would see if other users can share experience.