Blacklisted plugins in DeepStream Docker Container

• Hardware Platform (Jetson / GPU): RTX 3060 Laptop
• DeepStream Version: 6.3-triton-multiarch
• TensorRT Version: 8.5.3.1-1+cuda11.8 (dpkg -l | grep tensorrt)
• NVIDIA GPU Driver Version (valid for GPU only): NVIDIA-SMI 535.103 | Driver Version: 537.13 | CUDA Version: 12.2

First of all I think its worth mentioning that I am running WSL 2 with Ubuntu 22.04 in Windows 11, with all the necessary (i guess) Nvidia dependencies to run the ngc containers. So I have pulled a fresh DeepStream Docker container image ( nvcr.io/nvidia/deepstream:6.3-triton-multiarch), ran the container using: “docker run --gpus all -it --rm --net=host --runtime=nvidia --privileged -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY -w /opt/nvidia/deepstream/deepstream-6.3 nvcr.io/nvidia/deepstream:6.3-triton-multiarch” and proceded to run the installation scripts “./user_additional_install.sh” and “./user_deepstream_python_apps_install.sh” inside /opt/nvidia/deepstream/deepstream-6.3. Then I went to test the deepstream_test_1.py with the command “python3 deepstream_test_1.py /opt/nvidia/deepstream/deepstream-6.3/samples/streams/sample_720p.h264” and got the following:

(gst-plugin-scanner:11318): GStreamer-WARNING **: 01:11:36.165: Failed to load plugin ‘/usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream/libnvdsgst_udp.so’: librivermax.so.0: cannot open shared object file: No such file or directory

(gst-plugin-scanner:11318): GStreamer-WARNING **: 01:11:36.171: Failed to load plugin ‘/usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream/libnvdsgst_preprocess.so’: /usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream/libnvdsgst_preprocess.so: undefined symbol: cuGraphicsEGLRegisterImage

(gst-plugin-scanner:11318): GStreamer-WARNING **: 01:11:36.335: Failed to load plugin ‘/usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream/libnvdsgst_infer.so’: /usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream/libnvdsgst_infer.so: undefined symbol: cuGraphicsEGLRegisterImage

(gst-plugin-scanner:11318): GStreamer-WARNING **: 01:11:36.348: Failed to load plugin ‘/usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream/libnvdsgst_deepstream_bins.so’: /usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream/libnvdsgst_infer.so: undefined symbol: cuGraphicsEGLRegisterImage
Creating Pipeline

Creating Source

Creating H264Parser

Creating Decoder

Unable to create pgie
Creating EGLSink

Playing file /opt/nvidia/deepstream/deepstream-6.3/samples/streams/sample_720p.h264
Traceback (most recent call last):
File “deepstream_test_1.py”, line 258, in
sys.exit(main(sys.argv))
File “deepstream_test_1.py”, line 201, in main
pgie.set_property(‘config-file-path’, “dstest1_pgie_config.txt”)
AttributeError: ‘NoneType’ object has no attribute ‘set_property’

Because the pgie could not be created I went to verify the plugin and discovered that it was blacklisted, along with a few others:

gst-inspect-1.0 nvinfer: No such element or plugin ‘nvinfer’

gst-inspect-1.0 -b:

Blacklisted files:
libnvdsgst_deepstream_bins.so
libnvdsgst_infer.so
libnvdsgst_preprocess.so
libnvdsgst_udp.so
libcustom2d_preprocess.so

Total count: 5 blacklisted files

I don’t understand why these essential DeepStream plugins are blacklisted in a fresh DeepStream container. How can I fix this? Without them I can’t create/run a proper DS Pipeline. Any help would be much appreciated. Thanks in advance!

I am afraid WSL can’t be supported currently.

Only supports docker for linux systems.

Ok, so I have repeated the same steps, but now in a Jetson Nano and with the same DeepStream Docker Container (nvcr.io/nvidia/deepstream:6.3-triton-multiarch). When trying to run “python3 deepstream_test_1.py /opt/nvidia/deepstream/deepstream-6.3/samples/streams/sample_720p.h264” I got a lot more elements being unable to be created in the pipeline due to the plugins being blacklisted, such as the streammux and the nvinfer. How can I fix this? Is there another container option? Ideally I wanted to achieve something as close as possible as the environment used in the “Building Real-Time Video AI Application” Nvidia course.

As this table, Jetson nano only support DS-6.0.1.

Due to hardware limition, some libraries on jetson are sharing between docker and host.

It means regardless of docker or host, nano can only support ds-6.0.1

1 Like

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