Why deepstream-paraller-infer program blocked

[lmw0320]

Docker images of deepstream-l4t: 6.1.1-triton, under jetson Xavier NX with corresponding version as shown below:

I have setup the environment and run the code successfully as per official command. But when I reduce the source from 4 to 1 in the source.csv file, and found the program block without any error report.

If I set the source.csv as below, it can be worked successfully:

Hope reply, thanks in advance!

[spolisetty]

Hi,

We are moving this post to the Jetson Xavier NX forum to get better help.

Thank you.

[lmw0320]

I just redirect this topic to Accelerated Computing / Intelligent Video Analytics / DeepStream SDK, thanks.

[Fiona.Chen]

Can you provide the complete configurations of your modification?

[lmw0320]

dec_parallel_infer.yml (6.5 KB)

Here is my main configure yaml

[Fiona.Chen]

Where are the other configuration files? If you changed the sources, you need to change the other configurations too.

[lmw0320]

I used two models with the following configuration files as below, and I just remove two source urls in the csc file without any change in other configuration files. I’m not clear if any setting need to be revised accordingly ?
config_infer_primary.txt (556 Bytes)
config_infer_primary(1).txt (556 Bytes)

[lmw0320]

I got the solution…I need to reset the src-ids of branch in each primary-pgie。 This should be following the setting in sources.csv …