Hello,
I am attempting to run Isaac ROS VSLAM on a Jetson Orin NX 16GB board using Isaac ROS version 3.1. However, when trying to run the package, I received numerous errors about a missing GPU. After much detective work, I have narrowed down the problem to this: the container does not have the proper permissions to access the GPU. I am sure of this, because the nvidia-smi command results in “Unable to determine the device handle for GPU0002:00:00.0: Unknown Error”, while sudo nvidia-smi results in the expected summary:
±--------------------------------------------------------------------------------------+
| NVIDIA-SMI 540.3.0 Driver Version: N/A CUDA Version: 12.2 |
|-----------------------------------------±---------------------±---------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 Orin (nvgpu) N/A | N/A N/A | N/A |
| N/A N/A N/A N/A / N/A | Not Supported | N/A N/A |
| | | N/A |
±----------------------------------------±---------------------±---------------------+
±--------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| No running processes found |
±--------------------------------------------------------------------------------------+
My running hypothesis is that the Isaac ROS Dockerfile.user script is overriding permissions or providing incorrect permissions to the container. I noticed that the run_dev.sh script already makes the container privileged, so I am not sure what to try, as I don’t see any glaring errors in the Dockerfile.user script.
Please let me know if you have experienced this issue and if so, how you fixed it! Thank you in advance!