One of the services offered by Amazon Web Services (AWS) is CloudFormation. CloudFormation allows one "to model a collection of AWS resources." This is done through a "template [which] describes [these] resources and their dependencies." This service is utilized to create the Umphbase stack which is responsible for doing daily updates of umphbase and backing up the database to a public endpoint.
The template.yaml file is used to create the stack. This template defines the following resources.
| Resource | Type | Description |
|---|---|---|
| RootRole | IAM::Role | Provides permissions to Lambda Functions |
| Secret | SecretsManager::Secret | Stores database access credentials |
| Umphbase | RDS::DBInstance | Database for storing setlist information |
| Bucket | S3::Bucket | Manages public endpoint for backup |
| Layer | Lambda::LayerVersion | Stores necessary Python dependencies |
| Update | Lambda::Function | Used to update the database |
| Backup | Lambda::Function | Used to backup the database |
Note: The update and backup directories contain the source code for each Lambda Function respectively. The layer directory contains the source code for the Lambda Layer.
The template for the CloudFormation stack requires multiple parameters which
are defined in the .env file. The .env.example file provides
an example setup. The UMPHBASE_BUCKET and DEPLOYMENT_BUCKET environment
variables define the names of the two S3 buckets
where the backup and Lambda Function source code is stored respectively. These
values must be changed as S3 Buckets in AWS must have globally unique names
and the default values are already in use. It is left as an exercise for the
reader to determine who is using them ;).
#.env.example -> .env
UMPHBASE_BUCKET=[something-else]
DEPLOYMENT_BUCKET=[something-else]
Other environment variables can be changed as well. Two recommended changes are to choose a new password and to choose an email to receive logging information when the Lambda Functions are run.
#.env.example -> .env
DB_PASSWORD=[your_password]
EMAIL=[[email protected]]
If an email is provided, the email make target can be used to verify the
email address. An email will be sent to the email address defined in the .env
file which provides a link to finish the verification.
Next, use the make target bucket to make the deployment bucket and populate
it with the source code for the Lambda Functions and Lambda Layer. After the
bucket has been created, use the target create-stack for creating the
Umphbase stack.
make bucket
make create-stack
Alternatively, you can create the stack using AWS Management Console or AWS Command Line Interface (CLI).
Lastly, the Update Lambda Function will only add the most recent shows to the
database. To initialize the AWS RDS database with the entire setlist history,
use the target init.
make init