RuntimeError: Frame didn't arrive within 5000 Inside Docker container with Nvidia Jetson ORIN

[kais-bedioui]

---

Required Info
Camera Model	{ D400 }
Firmware Version	v2.50.13
Operating System & Version	Ubuntu 20.04
Kernel Version (Linux Only)	5.10.104-tegra
Platform	NVIDIA Jetson& Docker
SDK Version	{ 2.50.13 }
Language	python }
Segment	{Other }

Issue Description

Hello,
I have a FRAMOS D455e stereo camera connected to Nvidia Jetson ORIN with Jetpack 5.1 via ethernet port (not PoE).
I successfully created a Docker image to run my Computer Vision application using the camera following the instructions unpacked from FRAMOS_D400e_Software_Package_Linux64_ARM_latest and extracted with the debian package under ~/librealsense2/scripts/Docker_FRAMOS/Dockerfile.
I can successfully run rs-enumerate-devices, rs-color and rs-depth inside the docker container.
The python example scripts I managed to run and get the expected output from:

python3 librealsense2/wrappers/python/examples/python-tutorial-1-depth.py # changed import pyrealsense2 as rs ---> import pyrealsense2.pyrealsense2 as rs

BUT when I apply the same technique to read color frames, as exhibited in the script `opencv_viewer_example.py’ I get this error:

RuntimeError: Frame didn't arrive within 5000

How can I proceed? Thank you

[JerryChang]

hello kais-bedioui,

it looks a timeout failure,
could you please refer to Applications Using V4L2 IOCTL Directly to use V4L2 IOCTL to verify basic camera functionality.

[kais-bedioui]

Hi @JerryChang ,
Thank you for your input. I didn’t get to try your suggestion but I have found the solution as mentioned in this github issue #13571
In short, one should run the docker container with Nvidia Runtime i.e.
docker run --runtime nvidia [more options] librealsense-docker [command]
I will close down this ticket.
Thanks