Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
Nvidia Jetson ORIN AGX 32GB industrial module provided by Auvidea
• DeepStream Version
Deepstream SDK 6.1
• JetPack Version (valid for Jetson only)
• TensorRT Version
• Issue Type( questions, new requirements, bugs)
I built a custom Deepstream docker image for my application mainly inspired by This and saved it in our private container registry. I pulled the docker image and ran it on 2 different Jetsons( model specified above) using docker compose with the following docker-compose.yml configuration:
version: '3' services: jetson-deepstream:latest command: "/deepstream-custom-app -c config_deepstream_app.txt -t" container_name: deepstream build: context: . dockerfile: docker/deepstream/Dockerfile.deepstream volumes: - deepstream_configuration_textfile.txt:/config_deepstream_app.txt restart: on-failure ports: - "2000:2000" depends_on: - rabbitmq #defined in actual yml file runtime: nvidia deploy: resources: reservations: devices: - driver: nvidia count: 1 capabilities: [gpu] networks: jetson_docker:
The 2 devices were flashed with the same way and contain the same components installed on the host system however on one of them the docker container exits with error 255: “gst_element_factory_make: no such element factory “nvstreammux”!”. There are other gst & nv plugins missing inside the container even though it was pulled from the docker container registry, and the same image worked as expected on the similar device.
The only way we could see the same behavior on the correct environment is when we ommitted --runtime=nvidia from the docker run command, and then we obtain the same error on both environments.
Forcing the value of docker default-runtime to “nvidia” does not solve the issue.
Looking for ideas on how to tackle this or potential issues.
NOTE: the Nvidia jetsons are operating in a remote environment and internet access is very limited.
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)