Please provide complete information as applicable to your setup.
• Hardware Platform (GPU) • DeepStream Version 6.0.1 • TensorRT Version 8.0.1.6 • NVIDIA GPU Driver Version 510.108.03 • Issue Type (bugs) • How to reproduce the issue ?
We run deepstream-app and disconnect the network of camera every 20 seconds, found it causes the host mem increase obviously. we use the yolov3 sample, source is configerd like
type=4
rtsp-reconnect-attempts=-1
rtsp-reconnect-interval-sec=10
We used docker for our testing without any modify with deepstream-app.
This porblem is occured in production environment,it is hard to update the SDK there.
We just reproduce the same issue with deepstream 6.3 in docker, getting a similar result.
there is a mistake with my discribe. we do not recover the net of camera immediately after the disconnect. Instead, we will do it just after the deepstream log shows “Resetting source 0” message.
please refer to this topic. to narrow down this issue, can you use this method to get a memory leak log? and can you use this method to get a valgrind memory leak analysis log? Thanks!
case1:
==2767== by 0x1200ED: reset_source_pipeline (deepstream_source_bin.c:1554)
==2767== by 0x11BCDA: watch_source_status (deepstream_source_bin.c:688)
case2:
==2767== by 0x663D6B7: gst_event_new_flush_stop (in /usr/lib/x86_64-linux-gnu/libgstreamer-1.0.so.0.1405.0)
==2767== by 0x11FF54: reset_source_pipeline (deepstream_source_bin.c:1540)
from the valReport.log, there is memory leak in gst_nvevent_new_stream_reset. it is not opensource. we will fix it in the latter versions. Thank you!
There is no update from you for a period, assuming this is not an issue any more. Hence we are closing this topic. If need further support, please open a new one. Thanks.
what is your company name?