4. Run your graph in production
Learn about deployment and essential developer tooling
Time to accomplish: 15 Minutes
Great job for making it this far! We've already learned how to build a GraphQL API with Apollo, connect it to REST and SQL data sources, and send GraphQL queries. Now that we've completed building our graph, it's finally time to deploy it! 🎉
An Apollo GraphQL API can be deployed to any cloud service, such as Heroku, AWS Lambda, or Netlify. In this tutorial, we'll deploy our graph API to Zeit Now. You will need to create a Now account in order to follow these steps. If you haven't already created an Apollo Engine account, you will need to sign up for one.
In a production application, you should set up this publishing script as part of your CI workflow. For now, we will run a script in our terminal that uses the Apollo CLI to publish our schema to Engine.
First, we need an Apollo Engine API key. Navigate to Apollo Engine, login, and click on New Service at the top. The prompt will instruct you to name your service. When you're finished, click Create Service. You'll see a key appear prefixed by
service:. Copy that key so we can save it as an environment variable.
Let's save our key as an environment variable. It's important to make sure we don't check our Engine API key into version control. Go ahead and make a copy of the
.env.example file located in
server/ and call it
.env. Add your Engine API key that you copied from the previous step to the file:
The entry should basically look like this:
Our key is now stored under the environment variable
It's time to publish our schema to Engine! First, start your server in one terminal window by running
npm start. In another terminal window, run:
npx apollo service:push --endpoint=http://localhost:4000
npx is a tool bundled with npm for easily running packages that are not installed globally.
This command publishes your schema to the Apollo registry. Once your schema is uploaded, you should be able to see your schema in the Apollo Engine explorer. In future steps, we will pull down our schema from Engine in order to power the Apollo VSCode extension.
For subsequent publishes, we may first want to check for any breaking changes in our new schema against the old version. In a terminal window, run:
npx apollo service:check --endpoint=http://localhost:4000
Publishing your schema to Apollo Engine unlocks many features necessary for running a graph API in production. Some of these features include:
- Schema explorer: With Engine's powerful schema registry, you can quickly explore all the types and fields in your schema with usage statistics on each field. This metric makes you understand the cost of a field. How expensive is a field? Is a certain field in so much demand?
- Schema history: Apollo Engine’s schema history allows developers to confidently iterate a graph's schema by validating the new schema against field-level usage data from the previous schema. This empowers developers to avoid breaking changes by providing insights into which clients will be broken by a new schema.
- Performance analytics: Fine-grained insights into every field, resolvers and operations of your graph's execution
- Client awareness: Report client identity (name and version) to your server for insights on client activity.
We also want to be transparent that the features we just described, such as viewing specific execution traces and validating schema changes against recent operations, are only available on a paid plan. Individual developers just getting started with GraphQL probably don't need these features, but they become incredibly valuable as you're working on a team. Additionally, layering these paid features on top of our free developer tools like Apollo VSCode makes them more intelligent over time.
We're committed to helping you succeed in building and running an Apollo graph API. This is why features such as publishing and downloading schemas from the registry, our open source offerings like Apollo Client and Apollo Server, and certain developer tools like Apollo VSCode and Apollo DevTools will always be free forever.
To deploy our app to Now, run the
now command from the
server directory of the app. The command may prompt you to login if you haven't already.
$ npx now
now command immediately deploys our graph API to the cloud and returns the hosted URL. Make sure you either copy the URL or run
npx now ls in your terminal to retrieve the URL, since we'll need it in the following section when we build our client.
Congrats on deploying your first Apollo graph API! 🚀 Let's move on to the second half of the tutorial where we connect the API we just built to a React app.