Summary

Operators should use consistent logging mechanisms to ensure log file availability when debugging and troubleshooting is required. The commonly accepted method for this is described in Kubernetes documentation for "Cluster level logging architectures":
https://kubernetes.io/docs/concepts/cluster-administration/logging/#cluster-level-logging-architectures

Specifically, we will implement the sidecar logging architecture:

https://kubernetes.io/docs/concepts/cluster-administration/logging/#sidecar-container-with-logging-agent

Requirements

We need to implement this in each of the pods we manage, Apache containers running WSGI services should log to their own files, and then a sidecar container will stream those logs to stdout. 

For the heat-engine service, it will log to the traditional /var/log/heat/heat-engine.log file and the sidecar container will stream the contents of this file to stdout.