Hello, I’m attempting to run my Yolov5 model using several RTSP streams. For a long, there are no problems, and real-time functionality is perfect. However, after a few hours, the cameras begin to freeze and go dark one by one, as seen in the picture below. The ones that become frozen occasionally might begin to function once more. When I reload the app, it resumes operating as intended. When I was keeping an eye on the system resources, I saw that there were no utilization peaks. Has anyone else encountered this problem before?
Can it be caused by a camera connection that drops off and deepstream fails to attempt a reconnection, or it might have something to do with the memory?
Thank You,
Chaki
NVIDIA RTX2070• Hardware Platform (Jetson / GPU)**
6.1• DeepStream Version**
8.2.5.1• TensorRT Version**