Please provide complete information as applicable to your setup.
**• Hardware Platform (Jetson)
**• DeepStream Version 7.0
I’m experiencing delays and encountering the error message “gstnvtracker: Unable to acquire a user meta buffer” when running deepstream-app 7.0 in RTSP stream mode, even at 10 fps, with potential FPS being constrained. Could anyone recommend adjustments to the user-meta-pool-size parameter or other optimizations to mitigate this issue and enhance performance? Any advice would be greatly appreciated. Thank you.
I have look through the guide but what is causing the 5-7 second delay observed when streaming RTSP output through a sink in our DeepStream application, which sends analytics (such as face detection, gender, and age estimation) to MQTT?
Additionally, on our DeepStream 7 machine, a warning ‘**PERF: 9.99 (9.99) gstnvtracker: Unable to acquire a user meta buffer. Try increasing user-meta-pool-size’ appears in the terminal when RTSP Streaming (Option 4) output sink-2 is enabled and the application is running.
Could you attach the whole pipeline you are using? This 5-7 second delay may not be caused by the nvtracker. You can refer to the Enable Latency measurement for deepstream sample apps to get the latency of the plugins.
We’re not changing the default deepstream-app setup. We’re just adding our own models (both primary and secondary) and tweaking the analytics_done_buf_prob setting to extract frame metadata
################################################################################
# Copyright (c) 2020, NVIDIA CORPORATION. All rights reserved.
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
################################################################################
[application]
enable-perf-measurement=1
perf-measurement-interval-sec=1
[tiled-display]
enable=1
rows=1
columns=1
width=1280
height=720
gpu-id=0
[source0]
enable=1
#Type - 1=CameraV4L2 2=URI 3=MultiURI
type=1
num-sources=1
#uri=file:///home/glueck/Videos/test.mp4
#uri=file://../../streams/rajeshStevenRosini.mp4
#uri=rtsp://root:Glueck321@192.168.1.18/axis-media/media.amp?streamprofile=H264
uri=rtsp://root:Glueck321@10.0.1.30/axis-media/media.amp?videocodec=h264&camera=1
#drop-frame-interval=2
gpu-id=0
camera-width=1280
camera-height=720
#camera-fps-n=10
#camera-fps-d=1
camera-v4l2-dev-node=0
[streammux]
gpu-id=0
batch-size=1
batched-push-timeout=40000
## Set muxer output width and height
width=1920
height=1080
[sink0]
enable=0
#Type - 1=FakeSink 2=EglSink 3=File
type=2
sync=0
source-id=0
gpu-id=0
[osd]
enable=0
gpu-id=0
border-width=3
text-size=15
text-color=1;1;1;1;
text-bg-color=0.3;0.3;0.3;1
font=Arial
[primary-gie]
enable=1
gpu-id=0
# Modify as necessary
#model-engine-file=../../models/Primary_FaceDetector/resnet18_detector.etlt_b1_gpu0_fp16.engine
batch-size=1
bbox-bg-color0=0;1;0;0.3
#Required by the app for OSD, not a plugin property
#bbox-border-color0=1;0;0;1
#bbox-border-color1=0;1;1;1
#bbox-border-color2=0;0;1;1
#bbox-border-color3=0;1;0;1
gie-unique-id=1
config-file=config_infer_primary_facedetectir.txt
[sink1]
enable=0
type=3
#1=mp4 2=mkv
container=1
#1=h264 2=h265 3=mpeg4
codec=1
#encoder type 0=Hardware 1=Software
enc-type=0
sync=0
bitrate=2000000
#H264 Profile - 0=Baseline 2=Main 4=High
#H265 Profile - 0=Main 1=Main10
profile=0
output-file=out.mp4
source-id=0
##RTSP Output
[sink2]
enable=1
#Type - 1=FakeSink 2=EglSink 3=File 4=RTSPStreaming 5=Overlay
type=4
#1=h264 2=h265
codec=1
#encoder type 0=Hardware 1=Software
enc-type=0
sync=0
bitrate=4000000
#H264 Profile - 0=Baseline 2=Main 4=High
#H265 Profile - 0=Main 1=Main10
profile=0
# set below properties in case of RTSPStreaming
rtsp-port=8554
udp-port=5400
[tracker]
enable=1
# For NvDCF and DeepSORT tracker, tracker-width and tracker-height must be a multiple of 32, respectively
tracker-width=640
tracker-height=384
ll-lib-file=/opt/nvidia/deepstream/deepstream/lib/libnvds_nvmultiobjecttracker.so
# ll-config-file required to set different tracker types
# ll-config-file=/opt/nvidia/deepstream/deepstream/samples/configs/deepstream-app/config_tracker_IOU.yml
ll-config-file=/opt/nvidia/deepstream/deepstream/samples/configs/deepstream-app/config_tracker_NvDCF_perf.yml
# ll-config-file=/opt/nvidia/deepstream/deepstream/samples/configs/deepstream-app/config_tracker_NvDCF_accuracy.yml
# ll-config-file=/opt/nvidia/deepstream/deepstream/samples/configs/deepstream-app/config_tracker_DeepSORT.yml
gpu-id=0
enable-batch-process=1
enable-past-frame=1
display-tracking-id=1
[secondary-gie0]
enable=1
model-engine-file=/opt/nvidia/deepstream/deepstream-7.0/samples/models/genderv11/gender11.caffemodel_b16_gpu0_fp16.engine
gpu-id=0
batch-size=1
bbox-bg-color0=0;1;0;0.3
#Required by the app for OSD, not a plugin property
bbox-border-color0=1;0;0;1
bbox-border-color1=0;1;1;1
bbox-border-color2=0;0;1;1
bbox-border-color3=0;1;0;1
gie-unique-id=2
operate-on-gie-id=1
operate-on-class-ids=0;
config-file=config_infer_genderv11_classifier.txt
[secondary-gie1]
enable=1
model-engine-file=/opt/nvidia/deepstream/deepstream-7.0/samples/models/age_model/age.model_b16_gpu0_fp16.engine
gpu-id=0
batch-size=1
gie-unique-id=3
operate-on-gie-id=1
operate-on-class-ids=0;
config-file=config_infer_age_classifier.txt
[secondary-gie2]
enable=0
model-engine-file=../../models/Secondary_Emotion/emotion_2convs_iter_3000.caffemodel_b16_gpu0_fp16.engine
gpu-id=0
batch-size=16
gie-unique-id=4
operate-on-gie-id=1
operate-on-class-ids=0;
config-file=config_infer_emotion_classifier.txt
[tests]
file-loop=1
OK. Just from your config file, you need to set the source type= 4 when you use rtsp source and set live-source=1 for streammux. You can refer to our Guide to set some parameters related to rtsp source.
##RTSP Output
[sink2]
enable=1
#Type - 1=FakeSink 2=EglSink 3=File 4=RTSPStreaming 5=Overlay
type=4
#1=h264 2=h265
codec=1
#encoder type 0=Hardware 1=Software
enc-type=1
sync=0
bitrate=4000000
#H264 Profile - 0=Baseline 2=Main 4=High
#H265 Profile - 0=Main 1=Main10
profile=0
# set below properties in case of RTSPStreaming
rtsp-port=8554
udp-port=5400
latency=8000
can i set source as rtsp for webcam? And I’ve tried it but no visuals popping, after a bit of fine tuning, the delay is reduced around 5 to 7 seconds, but can it be reduced more?
OK. But what you attached earlier was using rtsp source.
You can compare the difference between direct display to the screen and RTSP Output first. If the delay is caused by the rtsp output, there is basically no room for reduction.
i tried on the rtsp on source, had a 10 seconds delay, so as you mentioned up there, there’s nothing we can do about it is it?, But as i said, im not using the rtsp source as my main camera, this time we are using a logitech webcam, with this i can’t reduce the delays?
This basically proves that the delay is caused by RTSPStreaming sink. Just from your config file, you can using software encoder enc-type=1. This may take a long time.