Consolidate server logging

[eboileau]

Aims/objectives.

For production.

A clear and concise description of todo items.

• To "harmonize" logging throughout the app, e.g. some modules/classes have no logging at all, logging levels are not used consistently i.e. some info should actually be debug, and vice versa, etc.
• Some functions are defined in utils but are not used. Currently, logging is configured in config.py
• What is the difference between from flask import current_app, current_app.logger and import logging, logging.getLogger(__name__)?
• For production, logs are accessed e.g. with podman-compose logs. I believe they are somewhere under /scimodom_data/containers/, but where exactly I don't know. Is this "configurable"? e.g how long are logs kept? What if we need to go back a few days? Do we need a more flexible solution e.g. app-only logging, rotating logs?

Re format, see also injecting-request-information.

[HaraldWilhelmi]

The command "podman-compose logs" just grants access to the stdout/stderr of the applications running in the container. One possible approach is this:

• Log the level WARNING and up to stderr so that they are visible with "podman-compose logs".
• Write INFO and above to some log file.
• Write DEBUG too to the log file if a DEBUG option is given.

We could also make the log file accessible from outside the container by adding a bind volume to the docker-compose.yml.