I am currently working on deepstream_imagedata_multistream.py to RTSP stream on it. When I start the python code with RTSP memory usage starts increasing slowly after 30 minutes the output gets stuck but memory is still increasing.
After that, I have tested with the video files the same issue again in that also. I am working on jetson nano with deepstream 5.1 and jetpack version 4.5
Can you tell us the steps of reproducing your test?
For this, I have cloned the Github repo from GitHub - NVIDIA-AI-IOT/deepstream_python_apps: A project demonstrating use of Python for DeepStream sample apps given as a part of SDK (that are currently in C,C++)..
After I have run all the steps as per repo documentation. then I have run command
python3 deepstream_imagedata-multistream.py rtsp://127.0.0.1/video1 frames
After this step, the script will show Output on display. Then I started the system monitor and Check memory occupied by python. In the system monitor, python memory was increasing slowly. And It is not decreasing after that. That’s why after some time the output gets stuck.
were you able to generate the issue on your end ?
Yes. Seems I can reproduce the memory leak with the steps you provided. We will check this issue.
Is there any update on the memory leak issue?
The memory leak is fixed with DS 6.0 GA.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.