I have a deep learning model on working with camera stream and I run this model on jetson tx2. While I run my project, if I can’t get stream from camera, Jetson tx2 system get crash. Normally, if I can not get camera stream, gstreamer should to wait until it comes. But while gstreamer wait for camera stream, jetson tx2 system get crash in 5-7 minutes. My gstreamer version is 1.0 , Jatpack version is 4.4.1
when i run this command > head -1 /etc/nv_tegra_release
R32 (Release)
revision: 4.3
GCID: 21589087
Board: t186ref
EABI: aarch64
STEPS:
I run this: gst-launch-1.0 v412src device=/dev/video8 io-mode=mmap ! videorate ! video/x-raw, framerate=25/1 ! videoscale ! video/x-raw, width=720, height=576 ! nvvidconv ! omxh246enc bitrate=5000000 ! queue ! decodebin ! videoconvert ! ximagesink sync=false
Then when the image is cut for any reason (like unplugging) and the image comes back (like plugging it in), the image starts to come off intermittently. When reset, the image is taken in its normal flow again, but is there a way to solve this situation without resetting? How do I solve this problem?