No EOS signal in the second video in runtime add or delete source

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

• Hardware Platform (Jetson / GPU) Tesla V100-PCIE
• DeepStream Version 6.1
• JetPack Version (valid for Jetson only) -
• TensorRT Version 8.4.1.5
• NVIDIA GPU Driver Version (valid for GPU only) 470.129.06
• Issue Type( questions, new requirements, bugs) No EOS signal in the second video in runtime add or delete source
• How to reproduce the issue? I have made the changes according to my requirement and attached the “deepstream_rt_src_add_del_exp.py”. To produce the issue run it in deepstream-6.1/sources/deepstream_python_apps/apps/runtime_source_add_delete
• Requirement details -

I have to process the videos one by one without loading the models again and again. Once a video gets a EOS the next video should start processing and it should continue like this.
For that I did some changes in the original deepstream_rt_src_add_del.py, as originally it was adding source after 10 sec.
After the processing of the first video when EOS message comes then my pipeline added the second source, but once the second finishes it does not give EOS and due to that the pipeline cannot add the new source,
deepstream_rt_src_add_del_exp.py (24.3 KB)

3

With your code, it will not do what you described.

Please add the correct logic with your code. You need to debug with your own code first.

4

According to the logic in the code, I am giving 5 as the Maximum number of sources, which should be added one by one until all 5 sources are added. I am not deleting the sources by stopping and releasing them, because that was giving some error.

5

But the problem is that the second stream is not giving the EOS message to add a new source.