1. Configure Docker & docker-compose (if you don't have it yet) & understand what this project consists of.
2. Bring up everything with docker-compose up. This will start maindb (source database), stabledb (target database) and inserter (script that simulates frequent insertion of dummy events into maindb).
3. Create a new application (it can be added as an extra service in docker-compose.yaml or executed seperately outside of Docker, as you wish) which will periodically migrate data from maindb to stabledb (let's say, every minute). The script should be easily configurable (DB credentials, migration interval etc) via command-line arguments or environment variables.
4. Add reporting to the script (stdout, text file, web interface - as you wish). Report must contain information about the amount of data that was migrated and time taken to process everything.
5. Bonus points: implement a simple web-server that will start data migration when a POST request is received.
6. Bonus points: edit inserter/main.py, decrease the insertion interval to 0 (line 28) and see if your script still works.
7. Bonus points: make the ETL process configurable so that if events table changes, code change will not be required (e. g. define ETL mapping via some seperate YAML config file.)
8. Bonus points: fork this repo, implement your changes in your fork and create a pull request to this repo.

Note: application must be written in Python.

Hint: you can add comments to describe your code. You can also provide multiple solutions if there is more than one good way to solve something.