AWS IoT MSGBroker Pipeline takes very long to Start

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) Jetson
• DeepStream Version 5.1
• JetPack Version (valid for Jetson only) 4.5
• TensorRT Version 7.1
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs) questions
• How to reproduce the issue ?

I’m currently working on a custom pipeline created with the python-bindings.

The general architecture of the pipeline goes like this:

uridecodebins (3) → streamMuxer → pgie → tracker → demuxer (3) ->nvvidconvert → nvdsosd → tee → transformer → glsink
→ msgconv → msgbroker

The MQTT configuration was successfully built following this tutorial How to integrate NVIDIA DeepStream on Jetson Modules with AWS IoT Core and AWS IoT Greengrass.

The app worked fine at first with the desired output, and the MQTT messages been published to the desired TOPIC.

After a few days and without any changes in the code or configuration, the code hangs for around 20 minutes in the “Using winsys x 11” log after the pipeline loop starts.

I will note that I changed of place (IP, Internet connection)

After 20 minutes the app runs okay.

Any idea how can I fix this?

• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

Seems network issue. It has nothing to do with deepstream.

Fiona, thanks for your reply. Is there any network requisites? I have 200Mbps in the place that takes 20 minutes to run… so what kind of network issue could we encounter.

The matter here is that if I want to deploy our application to every customer, those network requirements must be clear for us developers. Don’t you think so?

The app can resume, so the status is caused by the AWS server.