Ultimate 2021 Guide to Deploying NodeJS (And DenoJS) Apps to Heroku

Heroku is a great platform for deploying full-stack applications (if your application is frontend-only then Netlify, Vercel or Render may be a better alternative). In this guide, I will be focusing on the different ways to deploy a node js app. If you are using DenoJS I’ll link a video below on deploying a Deno app.

• Using Docker to Deploy DenoJS

For today’s discussion, we will cover two ways to deploy to Heroku.

Before we get started please do two things:

• Open a Heroku Account
• Download the Heroku CLI

Setting Up

• Create a NodeJS Project and get it to where you’re happy with the project
• Create a git repo, the root of the git repo should where the package.json file is
• Heroku needs to know what command to run to start your projects (usually a script npm run start or npm run production). This will go in a file called the Procfile so create a file called Procfile in your project root (folder with package.json) and inside of it put something like this.

web: npm run production

The web: means Heroku is running a web process so it needs to assign the process an HTTP port (provided via the PORT environmental variable) and the npm run production is the command to start the application.

• make sure all your code is committed and now we may begin!

The Heroku CLI Method

• Assuming you’ve downloaded the HerokuCLI you must first log in to the Heroku CLI using the command heroku login which will open the browser to confirm login to your heroku account.
• After logging in we can generate a new Heroku project with the command heroku create projectName this will create a new Heroku project in your account with the name you passed (or a random name if you don’t specify a name). It will also add a new remote to your local git repository called heroku.
• So all you have to do to push your code to Heroku is push your code to the heroku remote git push heroku branchName and when this is done Heroku will immediately begin to deploy your code by downloading all dependencies in your package.json and then running the command specified in Procfile
• You may need to specify all of your environmental variables for your application to work, this can be done using the Heroku CLI with the command heroku config:set KEY=VALUE
• If you want to see all the current variables you’ve set heroku config these variables are normal environmental variables like using a .env file and will be accessible to anywhere you use the process.env object.

Github Method

• Go to the Heroku.com dashboard and create a new project
• switch to the deployment section of the project and select “github” deployment
• link your GitHub account
• select which repo on your account you want to deploy
• turn on the automatic deploys feature (this will redeploy your app anytime your GitHub is updated so you don’t have to make a separate push to Heroku)
• then trigger a manual deployment to get the site initially created
• to set environmental variables go to the settings section of the application dashboard and you’ll see a section called “config vars” and here you can add and remove variables.
• You can also use the CLI to add/remove variables but will need the --app=APP_NAME flag so it knows which app the command should be run for (usually it looks for a Heroku remote to determine this, but in this case that won’t exist).
• On the dashboard, you’ll see a button in the upper right area where you can access the logs and run terminal commands directed at your deployed app.

The Heroku CLI

• If you are using the CLI deployment method you can run the commands as is as long as you are in a folder within your repo.
• If you are using the GitHub deployment method you can run the commands below from anywhere but must include a flag specifying which app to run the command for --app=APP_NAME

• The heroku run commands are useful for running one-off scripts and tasks like migrating and seeding databases