NR cannot instrument tracing with python agent on non-web application

[jonathan-fileread]

Description
NOTE: # ( Describe the problem you're encountering. )

Hello! We've setup a python agent in our kubernetes container that runs a worker which polls from a queue.

The dockerfile entrypoint looks something like this -

CMD ["python3", "/app/workerservice"]

We get logs but we do not receive distributed traces.

We have NEW_RELIC_DISTRIBUTED_TRACING_ENABLED set to true in the environment variables.

New relic is injected via our main.py, looking something like this.

import newrelic.agent

newrelic.agent.initialize()

@newrelic.agent.background_task()

async def main():

insert-some-components-here

if __name__ == '__main__':

asyncio.run(main())

Here's our permalink.
permalink: https://onenr.io/0EjOoDd18j6

python agent, newrelic==9.1.2 on python,

we ran the CLI diagnostic, but to no avail.

this is also a non-web-app worker we are looking to monitor - there are no frameworks used, just plain python.
[TIP]: # ( Do NOT share sensitive information, whether personal, proprietary, or otherwise! )

Expected Behavior
NOTE: # ( Tell us what you expected to happen. )

our goal is to instrument the agent to send distributed traces to the traces directory. we have a lot of dependencies to our worker services that we'd like to monitor.

Troubleshooting or NR Diag results
NOTE: # ( Provide any other relevant log data. )
[TIP]: # ( Scrub logs and diagnostic information for sensitive information )

Steps to Reproduce
NOTE: # ( Please be as specific as possible. )
[TIP]: # ( Link a sample application that demonstrates the issue. )

• provided in permalink, but screenshot attached.

Your Environment
[TIP]: # ( Include as many relevant details about your environment as possible including the running version of New Relic software and any relevant configurations. )

• added in the permalink
  Additional context
  [TIP]: # ( Add any other context about the problem here. For example, relevant community posts or support tickets. )

[hmstepanek]

If main loops forever and never exits than you need to move the background task decorator into something that has a finite start and end time. We will not record a distributed trace until the thing the background task is wrapping has exited. You may also need to set the startup timeout, startup_timeout = 30 so that the first background task is captured.