Error while Running Python Sample App (deepstream-test3)

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

• Hardware Platform (Jetson / GPU)
We are using AWS instance(g4dn.4xlarge) with T4 GPU with RDP setup( so that display export is not and issue)
• DeepStream Version
DeepStream 5.1
• TensorRT Version
TensorRT 7.2.3.4
• NVIDIA GPU Driver Version (valid for GPU only)
Nvidia Driver 460.32
• Issue Type( questions, new requirements, bugs)
Unable to Run Python Sample Application
• 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)
We were trying to run ‘deepstream-test3’ app from ‘deepstream_python_apps’ repo(from this link: GitHub - NVIDIA-AI-IOT/deepstream_python_apps: DeepStream SDK Python bindings and sample applications).
For this we were using a RTSP stream and this was the command:
“python3 deepstream_test_3.py <RTSP_URL>”
On running this, the program takes in around 5-6 frames and then stops, giving this error:

“gstname= video/x-raw
features= <Gst.CapsFeatures object at 0x7fdffbee84c8 (GstCapsFeatures at 0x7fde6c0f1c20)>
Frame Number= 0 Number of Objects= 0 Vehicle_count= 0 Person_count= 0
Frame Number= 1 Number of Objects= 0 Vehicle_count= 0 Person_count= 0
cuGraphicsGLRegisterBuffer failed with error(304) gst_eglglessink_cuda_init texture = 1
Frame Number= 2 Number of Objects= 0 Vehicle_count= 0 Person_count= 0
Frame Number= 3 Number of Objects= 0 Vehicle_count= 0 Person_count= 0
Frame Number= 4 Number of Objects= 0 Vehicle_count= 0 Person_count= 0
Frame Number= 5 Number of Objects= 0 Vehicle_count= 0 Person_count= 0
0:00:09.186704632 3054 0x2bc7ad0 WARN nvinfer gstnvinfer.cpp:1984:gst_nvinfer_output_loop: error: Internal data stream error.
0:00:09.186720578 3054 0x2bc7ad0 WARN nvinfer gstnvinfer.cpp:1984:gst_nvinfer_output_loop: error: streaming stopped, reason not-negotiated (-4)
Error: gst-stream-error-quark: Internal data stream error. (1): gstnvinfer.cpp(1984): gst_nvinfer_output_loop (): /GstPipeline:pipeline0/GstNvInfer:primary-inference:
streaming stopped, reason not-negotiated (-4)
Exiting app”

3

This log shows the display have some problems.

Have you installed openGL driver correctly? OpenGL | NVIDIA Developer

Can you refre to Deepstream cannot be displayed on the screen - Intelligent Video Analytics / DeepStream SDK - NVIDIA Developer Forums to check whether it can help you?

5

Hi,

I have checked the above links and it seems that the VM has proper display drivers.
Also, Just to make sure of this, I installed DeepStream on a local machine (i5 8th Gen, 16GB RAM, GTX1050 Ti 4GB, and a monitor) and got the same error.

I also tried this complete thing with DeepStream Triton docker and the same error occurs there as well.

6

Hi,

I have run the DeepStream python sample after exporting “GST_DEBUG=4”.

The sample run was DeepStream-test1 and this was the command: GST_DEBUG=4 python3 deepstream_test_1.py …/…/…/…/samples/streams/sample_720p.h264

I am attaching the complete output of the above command in gst_output_with_debug_messages.txt (85.3 KB).

Please help with the next steps.

7

Hello, were you able to debug this? I am facing the same issue and haven’t find any solutions yet

9

Seems like display error,
Please “rtsp-out” python example and let me know the result.

10

@shantanu.bakare In your log line 321:
ERROR nveglglessink ext/eglgles/gsteglglessink.c:2784:gst_eglglessink_configure_caps: Cuda Init failed

Your setting is not correct.