This project uses terraform to create a AWS Lambda environment which will run a Flask app.

The flask app is built using Docker and a Aws AMI image to match the lambda execution environment.

docker-compose up build_service

This command will use docker-compose to mount the source code using a volume. The entrypoint file for the build container will handle creating an archive from the Flask application including all python requirements. The out file generated from the build is flask-app.zip.

docker-compose up test_build

This command will start a docker container which will mount the current working dir into /app_build folder. It will then extract the flask-app.zip file into /app and run python3 run.py. The zip file contains all requirements for the app so there should be no need to install python requirements. If everything works you should be able to open http://localhost:5000/hello_lambda and you will get a response.

Inside of the aws folder, create a file .backend.tf with reference to a AWS S3 bucket for use with the Terraform s3 backend. And a terraform.tfvars file containing credentials you would like to use for postgres.

.backend.tf

bucket = "some-s3-bucket"
key = "path/to/terraform.tfstate"
region = "some region"

terraform.tfvars

postgres_pass = "somepgpass"
postgres_user = "somepguser"

Now initialize terraform using terraform init --backend-config .backend.tf. This should create a file storing the terraform state in the s3 bucket.

After terraform is initialized run terraform plan and then terraform apply.

If everything works correctly you should have a running flask app deployed to AWS Lambda.

This app uses environment variables for configuration. For local development, create a file .env and add the following lines.

FLASK_APP=run.py
FLASK_ENV=development
FLASK_DEBUG=1
SECRET_KEY=12345
DATABASE_URL=postgres://postgres:example@localhost:65432/postgres
SESSION_REDIS=redis://localhost:63798

The python package python-dotenv is used to load these environment variables from a .env file. Once this file is created you can run the flask commands as usual.

SQLAlchemy is used to create database model files. Migrations are generated from the model files using Flask-Migrate. Update your local database using flask db upgrade.