dd-trace-py RuntimeWorker daemon bug repro

app.py

from __future__ import print_function

import logging
import ddtrace

# disable "ERROR [ddtrace.internal.writer] [writer.py:270] - Failed to send traces to Datadog Agent at http://localhost:8126"
logging.getLogger("ddtrace.internal.writer").setLevel(logging.CRITICAL)

def main():
    with ddtrace.tracer.trace("app-main"):
        print("hello world")

if __name__ == "__main__":
    main()

docker-compose.yml

version: '3'

services:
  app:
    image: python:2
    volumes:
      - .:/project
    working_dir: /project
    environment:
      DD_AGENT_HOST: localhost  # doesn't actually exist
      DD_RUNTIME_METRICS_ENABLED: "True"
    command: >
      bash -c "
        pip install ddtrace==0.38.0 &&
        while true
        do
          ddtrace-run python app.py
        done
      "

error-captured.out

app_1      | Exception in thread RuntimeWorker (most likely raised during interpreter shutdown):
app_1      | Traceback (most recent call last):
app_1      |   File "/usr/local/lib/python2.7/threading.py", line 801, in __bootstrap_inner
app_1      |   File "/usr/local/lib/python2.7/threading.py", line 754, in run
app_1      |   File "/usr/local/lib/python2.7/site-packages/ddtrace/_worker.py", line 70, in _target
app_1      |   File "/usr/local/lib/python2.7/site-packages/ddtrace/_worker.py", line 79, in _on_shutdown
app_1      |   File "/usr/local/lib/python2.7/site-packages/ddtrace/internal/runtime/runtime_metrics.py", line 68, in flush
app_1      |   File "/usr/local/lib/python2.7/site-packages/ddtrace/vendor/dogstatsd/base.py", line 131, in __exit__
app_1      |   File "/usr/local/lib/python2.7/site-packages/ddtrace/vendor/dogstatsd/base.py", line 191, in close_buffer
app_1      |   File "/usr/local/lib/python2.7/site-packages/ddtrace/vendor/dogstatsd/base.py", line 344, in _flush_buffer
app_1      |   File "/usr/local/lib/python2.7/site-packages/ddtrace/vendor/dogstatsd/base.py", line 328, in _send_to_server
app_1      | <type 'exceptions.AttributeError'>: 'NoneType' object has no attribute 'timeout'

if you run docker-compose up for long enough you'll see an error like error-captured.out eventually. there are also some of these exceptions <type 'exceptions.ValueError'>: list.remove(x): x not in list but i'm not sure if that's the same issue or possibly caused by super short lifetime of app.py