AWESOME-FORK-XD / MT-DB-Backend

Proof of concept project for MT product database

Summary

Last Revision by Derek Gau on 1/11/2023

This is the API Backend environment used by both Medten Product DB and Medten Product Catalog applications.

It includes:

• CRUD APIs for virtually all of the database tables in the Medten Product database.
• Authentication and Authorization to support logins.

AWS Architecture

This application is deployed as a "serverless" application on AWS. Each application has its own AWS template file, as defined in the sam-backend-[environment].yaml files in the root folder. The AWS services used are:

• API Gateway - provides the URLs exposed as the API
• Lambda - for NodeJS functions for handling calls received throught API gateway
• CloudWatch Logs - Lambda functions produce logs in CloudWatch

Prerequisites for Development

It is highly recommended to use VSCode or an IDE that supports the AWS command line environment.

1. You should have installed a version of NodeJS 14.x (updated January 2023)
2. The AWS CLI must be installed on your machine.
3. You should run aws configure to log in to AWS as the correct IAM user and region before running any AWS CLI operations. (These credentials will be granted by your manager).

How to Package and Deploy the Application on AWS

Prerequisite: .env files.

Included in the root folder is an example.env file. This specifies the environment-specific variables needed to deploy and run code. If you were not not provided .env files, copy the example.env file and rename it following the convention .env.[environment].local where environment is one of development, test, or production. Note that .env.*.local files MUST NOT BE SAVED under version control, because they typically contain credentials and/or sensitive information. Keep them in a secure, password-protected, encrypted storage (such as a password manager).

For example, prior to deploying to a development, test or production environment you must have .env.development.local, .env.test.local, .env.production.local files respectively available in the root of the project.

Package and Deploy the application code.

Before packaging and deploying the code, it is important to understand two variables that are used in the following commands. These are the env and profile variables.

The env variable is mostly synonymous with the NODE_ENV environment variable. Indeed, it specifies the environment-specific settings (usually stored in a .env file) that are used by the app. It is only relevant during the packaging process.

The profile variable indicates the named AWS profile. You should have configured this already in your ~/.aws directory. You may or may not have a dedicated AWS environment for TEST vs PRODUCTION. To accomodate these differences, this variable provides the needed flexibility to direct the code to the correct environment.

First, package the application code.

Example: packaging for NODE_ENV=test and an medten-test AWS profile...

npm run package --env=uiteam --profile=medten-test

Example: packaging for NODE_ENV=production and an medten AWS profile...

npm run package --env=production --profile=medten

Once the code is packaged, you can deploy it. Take care to use the correct --env and --profile variables. They should match what you choose for packaging.

This script uses AWS SAM guided deployment. This allows you to make choices (and save them for future use, if desired, about how you want to deploy the code). Typically all the defaults are used, but make sure to set the AWS Region to us-west-1, as the suggested default may not be correct.

Example: deploying using the medten AWS profile...

npm run deploy --env=production --profile=medten

Note, if you choose to save the configuration after all the guided deployment prompts, it will be saved in a samconfig.toml file by default. Since this file could be developer-specific, it should NOT be added to version control. To use this file for subsequent deployments, you can use the command by specifying the --config-file parameter with the file and the --config-env parameter set to the environment name whose settings you want to use (assuming you saved them in the samconfig.toml file).

Example: deploy a pkg-backend.yaml packaged application using samconfig.toml for the test environment settings.

sam deploy -t pkg-backend.yaml --config-file samconfig.toml --config-env test