0

I am running a docker container which has a python application that spawns other processes. I am using signals to try to gracefully shut down the python application and all of the spawned processes. When I send SIGTERM using the template from How to process SIGTERM signal gracefully? the entire container terminates. If I remove the logic of managing spawned processes, then the application closes normally and the container remains running (but my other processes are still running). I have tried this with other signals as well, but the result was the same.

Entering loop to listen for sigterm
^CLeaving loop
Killing 188 devices
/bin/sh: line 1:    31 Killed                  /bin/bash
user2680142
  • 59
  • 1
  • 10
  • Please post the output you're receiving from the containerized process, including any stack traces and debug output from the Docker daemon. Really *any* output that you receive will be useful. – L0j1k Dec 20 '17 at 02:59
  • Added basic output. Will try to get more verbose log. – user2680142 Dec 20 '17 at 22:25

0 Answers0