Hi,
• Hardware Platform (Jetson / GPU): laptop RTX 3070ti
• DeepStream Version: 7.0
• TensorRT Version: 8.6.1.6
• NVIDIA GPU Driver Version (valid for GPU only): 535.183.01
• Issue Type( questions, new requirements, bugs): nvtracker makes my program hangs at uninitialization.
I tried to simplify at maximum the reproduction of the error that happens in my program, but that is reproducible with gst-launch:
xhost +
docker container run \
-it \
--rm \
--name dstest \
--runtime nvidia \
-v /tmp/.X11-unix:/tmp/.X11-unix \
-e DISPLAY=:1 \
nvcr.io/nvidia/deepstream:7.0-gc-triton-devel \
gst-launch-1.0 \
videotestsrc ! \
nvvideoconvert ! \
'video/x-raw(memory:NVMM),format=RGBA' ! \
m.sink_0 nvstreammux name=m attach-sys-ts=TRUE sync-inputs=FALSE width=1920 height=1080 batched-push-timeout=40000 enable-padding=1 live-source=1 batch-size=1 ! \
nvtracker ll-config-file=/opt/nvidia/deepstream/deepstream/samples/configs/deepstream-app/config_tracker_NvDCF_perf.yml ll-lib-file=/opt/nvidia/deepstream/deepstream/lib/libnvds_nvmultiobjecttracker.so ! \
nvvideoconvert ! \
autovideosink
When closing the window or pressing Ctrl+C, the program hangs with
[NvMultiObjectTracker] De-initialized
With 7.1 docker image, gst-launch shuts down correctly, but I need to use 7.0 for now.
In my program in C++ (with other elements like rtspsrc or v4l2src, nvinfer, nvdsosd, …), the same happens after I quit the main loop, the hanging happens on
gst_element_set_state (pipeline, GST_STATE_NULL);
The problem doesn’t occur if nvtracker is not present.
I tried to send EOS to some elements, but no luck. Is there a workaround to shutdown nvtracker without hanging my program with DS7.0?
Thanks