Rewriting a simple tracker assignment as an AWS lambda solution for fun.

This started as a simple assignment.

The lambda function requires a couple of supporting libraries for database access. In particular the pymysql driver, but also the sqlalchemy orm package, because I started out with that in the original assignment and can't be bothered to refactorize out of the sqlalchemy dependency.

You need an environment with bash, Python >= 3.7 and corresponding pip.

Build with:

./build.sh

It installs the dependencies into a folder and zips the contents together with track.py into a package to be deployed into AWS.

I won't get into the details of setting up AWS, but you need an RDS mysql instance, an application load balancer, the lambda function and all the network topology and securit groups etc to hook these up to each other and the internet.

The lambda function takes one environment variable for the sqlalchemy database connection string. It should look something along the lines of this:

    LAMBDA_TRACKER_DB='mysql+pymysql://username:password@rds_dns_target_name/database'

https://docs.aws.amazon.com/lambda/latest/dg/services-rds-tutorial.html

https://docs.aws.amazon.com/lambda/latest/dg/services-alb.html