Deploying our Node-Express-Mongoose application (our APIs) consists of 2 sets of steps. First, we'll set up a Mongo database that our Node-Express-Mongoose back-end can connect to, and then configure our Node-Express-Mongoose applications (our APIs) to connect to this new cloud-hosted database.
Clicking the link in the header above will take you to a set of instructions for setting up a cloud-hosted (via AWS) Mongo database you can use in your deployed Heroku application.
Following Heroku's Getting Started with Node.js guide takes care of many deployment issues most of you will encounter, and give you a sense of how to use Heroku's deployment environment and features. Take particular note of...
- Define a Procfile.
- Provision a database.
- If you've been storing API keys as environment variables locally, define config vars.
Speaking of config vars, using dotenv to store sensitive information in the environment is a fantastic practice, and accords with 12-factor principles.
> After reading documentation, make sure you add .env to your .gitignore file, since .env will contain the sensitive data you don't want under version control*.
More often that not, solving deployment issues requires a good deal of Googling. Don't expect to find a silver bullet -- often we must go through many different issues other users may have encountered to understand our own.
What should you Google?
- If you aren't able to deploy, Google the error that shows up in your terminal after trying to push your app.
- If you are able to deploy but your app doesn't load/function propertly, see what shows up after running
$ heroku logsin the Terminal.
If you notice somebody running into the same problem as you, try working together on debugging it!