Deepstream-app, deepstream_test3, test5, with multi-RTSP streams

[FreedomLiX]

1. I can run deepstream-app -c deepstream_app_config.txt , with 8 rtsps , the issue is the OSD plugin on the srceen display, balck, green, or frozen. the FPS printf on the terminal : 0.00, 0.00, 24.30, 25.01, 0.00,0.00,0.00,0.00;

2. I use command line : gst-launch-1.0 uridecodebin uri=rtsp:admin：admin123@192.168.96.121/Streaming/Channels/101 !nveglglessink, test every rtsp, I can get the screen display normal.

3. with ./deepstream-test3, test5 with multi-rtsp streams, (the total rtsp num is 8), like uri=rtsp://admin:admin123@192.168.96.121:554/Streaming/Channels/101;uri=rtsp://admin:dixin_Admin123@192.168.96.122/Streaming/Channels/101;uri=rtsp://admin:dixin_Admin123@192.168.96.123/Streaming/Channels/101;uri=rtsp://admin:dixin_Admin123@192.168.96.124/Streaming/Channels/101, etc. I Can get the num_freams_in_batch in NvDsbachMeta , batch_meta->num_frams_in_batch, batch_meta ->max_frames_in_batch in appsink . when the pipline runing, the issue is , num_freams_in_batch always less than max_freams_in_batch, like 3, 4, 6, < 8; the OSD plugin, display on the srceen, balck, green, or frozen.

4. without OSD-dispaly sink, num_freams_in_batch still less than max_freams_in_batch.

Please provide complete information as applicable to your setup.
**• Hardware Platform GeForce RTX2080
**• DeepStream 6.1
**• TensorRT Version 8.4
**• NVIDIA GPU Driver Version 510
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
deepstream_app_config.txt (1.8 KB)

[yingliu]

Can you share the output of “nvidis-smi” when running 8 steams?

[FreedomLiX]

1. GPU 0 , which deepstream-app used.
2. GPU-Util 35%~50%.

[Fiona.Chen]

Seems there is heavy packet lost with your rtsp sources, please check your rtsp servers and network status( For 8 rtsp streams at the same time). And please make sure the “live-source=1” is set in your streammux settings in deepstream-app configuration file.

https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_ref_app_deepstream.html

[FreedomLiX]

I really appreciate your help.

[FreedomLiX]

@Fiona.Chen

rtsp servers and network status( For 8 rtsp streams at the same time

1. I check my rtsp servers and network status , my switch is 100 Gigabit switch. 8 rtsp streams at the same time.
2. still black, green, frozen on the display .
3. pipeline block ? or RTX2080Ti can’t hold multi-rtsp streams more than 3, or 4 ?

[FreedomLiX]

1. This .png file, my pipeline detail .
   rtsp==> urideccodebin =>
   rtsp==> urideccodebin =>
   rtsp==> urideccodebin => streammux => nvinfer =>tee => appsink
   -------------------------------------------------------.--------------- =>render
   rtsp==> urideccodebin =>
   rtsp==> urideccodebin =>

2. streammux, live_source = 1 ;

[Fiona.Chen]

You can monitor the CPU and GPU usage during your app running to find out the bottleneck.

@FreedomLiX

still black, green, frozen on the display .

This is caused by packet lost during network transferring.

[FreedomLiX]

appreciated for your suggestion,

[FreedomLiX]

It seems like, ! WOW, I GOT IT !, Just replace the Network Switch , 1000M!

[Fiona.Chen]

Glad to know the issue is resolved.