Drift Monitoring Backend

A backend server to allow users to store drift jobs and information for monitoring.

Before starting the server, you need to generate the following secrets on /secrets folder:

• app_database_password: Mongo database password.

Development

A Flask API that handles drift run data and job status. Contains swagger documentation (openapi 3.1) for above APIs. Saves drift run data to a MongoDB database. Provides endpoints for saving job status and get data with multiple filters.

• Add sentry_sdk for error tracking: https://flask.palletsprojects.com/en/2.3.x/errorhandling/#error-logging-tools

Create a conda environment, make sure conda is installed (https://conda.io/docs/user-guide/install/):

$ conda create --name drift-run python=3.11
$ conda activate drift-run
$ pip install -r requirements.txt
$ pip install -r requirements-dev.txt
$ pip install -r requirements-test.txt

To run and debug the backend in local, you can use the following command:

export FLASK_APP=backend/autoapp:app
python -m flask run <args>

You will need to export to the environment the application required variables, see the file .env.sample for the required variables. Additionally you can create your .env file and launch the application using the vscode debugger configuration at .vscode/launch.json.

For testing you need a database running to run the tests, you can use the following commands to run bring up the database and then run the tests:

$ docker run --name drift-database -p 27017:27017 -d mongo:6.0.3

Testing

The repository is configure so vscode can locate the tests and run them using the debugger. To obtain the test plan you can use the following command:

python -m pytest --setup-plan tests

To run tests you can simply run:

$ python -m pytest -x -n=auto tests