Deploying AWS Lambda Functions in Node.js

This recipe guides you through setting up AWS Lambda functions with Nx.

Getting set up

Depending on your current situation, you can either

  • create a new project with the goal of primarily developing and publishing AWS Lambda functions
  • add AWS Lambda functions to an existing Node.js project in an Nx workspace

Starting a New Project

To create a new project, run

npx create-nx-workspace@latest my-functions --preset=@nx/aws-lambda
Nx 15 and lower use @nrwl/ instead of @nx/

Configure Existing Projects

First, make sure you have @nx/aws-lambda installed.

Next, use the corresponding Nx generator to add the AWS Lambda configuration to an existing project:

nx g @nx/aws-lambda:setup-serverless
Nx 15 and lower use @nrwl/ instead of @nx/

This will setup your project to use AWS Lambda functions:

  1. Creates a new AWS lambda function in directory functions/hello-world.
  2. Adds samconfig.toml and template.yaml in the root of the project.
  3. Updates your project.json to have 2 new targets serve-functions & deploy-functions.

Serve and Develop Your Functions Locally

The project.json should have a new target serve-functions:

project.json
{ "name": "my-functions", ... "targets": { ... "serve-functions": { "command": "sam build && sam local start-api" }, ... } }

This allows to just run nx serve-functions to start a local server that serves your functions. As you can see it leverages the SAM CLI underneath.

Configure Your AWS Lambda Deploy Settings

Prerequiste

You need to configure your AWS credentials inside AWS before attempting to deploy.

Deployment

The following requirements need to be met in order to run the AWS Lambda function deployment:

  • SAM installed on your machine
  • esbuild available in your PATH (SAM need this). Example: npm install -g esbuild.

Your samconfig.toml stores default parameters for the SAM CLI. On the other hand, if you want to configure your lambda function settings such as the AWS region, runtime, and handler function, update your template.yaml.

The Nx project.json already contains a deploy-functions target we can invoke to trigger the deployment:

project.json
{ "name": "my-functions", ... "targets": { ... "deploy-functions": { "command": "sam build && sam deploy --guided" } } }

Just run:

nx deploy-functions

That's it! For monitoring or further permission settings, please refer to the AWS Lambda console.