Running Deepstream on Redhat Linux via Docker

Please provide complete information as applicable to your setup.

**• dGPU: NVIDIA A40 **
• DeepStream Version 7.1
**• NVIDIA GPU Driver Version: 565.57.01 CUDA Version: 12.7 **
**• OS info: NAME=“Red Hat Enterprise Linux”, VERSION=“8.10 (Ootpa)” **

Hello, I have developed a custom solution in deepstream 7.1. So far I have done the entire development through docker running using WSL2 on windows 11. Now I am trying to run that solution in my redhat OS environment that does have docker installed.

Just to test, I am using the default pre-built deepstream-app with my config
I get the app run successful message, but for some reason it doesn’t show any FPS logs or output when processing.

I have attached my logs and relevant files:
logs.txt (2.5 KB)
DockerFile.txt (718 Bytes)

Please note, the warnings in the logs right now are normal and were also showing up when I was developing the application on WSL2.
I have also made sure to convert my custom model again to have the correct Cuda Compute Capability Version

You must follow DeepStream On WSL — DeepStream documentation to setup the WSL2 environment.

I can get the FPS print.

Maybe your GPU is too powerful to finish the pipeline after 5 seconds. The default FPS print interval is 5 seconds in deepstream-app. If the processing time is more than 1 second, you may try to set “perf-measurement-interval-sec=1” in the deepstream-app configuration DeepStream Reference Application - deepstream-app — DeepStream documentation. If the processing time is less than 1 second, you can’t get any FPS print with deepstream-app.

Hi Fiona, my original question was not about WSL, it was about Redhat linux and for me the problem was in fact configuration. After making further modifications and making sure the container had access to all the requirements, the application works just fine.
Thank you for your help :)

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.