Multiple streammux call back fails

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• 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)

Hi,

I am using deepstream 6.1.1 to create my custom pipeline, with reference to the deepstream-parallel-inference., and everything works fine.
Now when i am using the deepstream-6.3 version docker image.
The model loads but the second demux after the tee, does not gets called by the pad
-probe call back which was not a problem with the older version of deepstream i.e 6.1.1

why do you need the second demux? plesae refer to the pipeline in the sample deepstream_parallel_inference_app.

What is the issue in having the second demux? I need it for my custom use case.

My Point is that, with the above pipeline my code is working absolutely fine.
I need to know why something that is working fine, on 6.1.1 is not working on 6.3?? Can we just look into the actual issue?
@fanzh can i get a reply soon, I have been stuck with this for days!

  1. is the pad-probe call back sink probe or src probe? is the second demux element called “Streammux-Demuxer” ?
  2. can you simplify the code to narrow down this issue? for example, can it work after removing nvinfer and tracker? you and add more probe function to check which element did not receive data.

sink!

yes it is “streammux-Demuxer”

The second parallel streammux after the first demux does not get called in the 6.2/6.3 deepstream version.
Which works fine in the 6.1.1 version

@fanzh, waiting for your reply!

  1. do you have some new findings after debugging?

could you provide a simplified code to reproduce this issue?

All the models get loaded from all the parallel-streammux.

I removed the nvinfer, and tracker from one of the streammux, and still the same .

Right, the first parallel streammux works, but all the other streammux in paralllel does not get called.

did you try increasing nvstreammux’ buffer-pool-size? could you provide a simplified code to reproduce this issue?

@fanzh
Well, I am using “New Streammux”!

@fanzh
I can share my code Privately!
You can tell me how, you want the code to be shared with you.

you can use forum email. please click forum avatar-> email.

@fanzh, I have done can you please check?

OK, I will check.

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