Yoga
On this page

Apollo Federation

GraphQL Yoga v3 documentation (superseded by v5): Learn how to use GraphQL Yoga with Apollo Federation to build a gateway server and a federated service.

Apollo Federation is a specification that applies microservice architecture through GraphQL APIs.

Thanks to Envelop’s Apollo Federation plugin, we can use GraphQL Yoga to build our gateway server.

Gateway

Installation for Gateway

npm i graphql-yoga
npm i @apollo/gateway
npm i @envelop/apollo-federation
npm i graphql@15

Example Gateway

import { createServer } from 'http'
import { createYoga } from 'graphql-yoga'
import { ApolloGateway } from '@apollo/gateway'
import { useApolloFederation } from '@envelop/apollo-federation'

// Initialize the gateway
const gateway = new ApolloGateway({
  serviceList: [
    { name: 'accounts', url: 'http://localhost:4001' },
    { name: 'products', url: 'http://localhost:4002' }
    // ...additional subgraphs...
  ]
})

// Make sure all services are loaded
await gateway.load()

const yoga = createYoga({
  plugins: [
    useApolloFederation({
      gateway
    })
  ]
})

// Start the server and explore http://localhost:4000/graphql
const server = createServer(yoga)
server.listen(4000, () => {
  console.info('Server is running on http://localhost:4000/graphql')
})

Handling Subgraph Errors

By default, GraphQL Yoga masks any unexpected GraphQL Errors. This is done to prevent leaking internal errors to the client. If you know that your subgraph is safe and you want to expose the errors to the client, you can customize the error masking bahviour.

Learn more about error masking

Expose subgraph errors to the client
import { createYoga, maskError } from 'graphql-yoga'
import { schema } from './schema.js'

const yoga = createYoga({
  schema,
  maskedErrors: {
    maskError(error, message, isDev) {
      if (error?.extensions?.code === 'DOWNSTREAM_SERVICE_ERROR') {
        return error
      }

      return maskError(error, message, isDev)
    }
  }
})

Federation Service

You don’t need any extra plugins for Yoga for Federation Service.

Installation

npm i graphql-yoga
npm i @apollo/subgraph
npm i graphql

Example User Service

const { parse } = require('graphql')
const { buildSubgraphSchema } = require('@apollo/subgraph')
const { createYoga } = require('graphql-yoga')
const { createServer } = require('http')

const typeDefs = parse(/* GraphQL */ `
  type Query {
    me: User
  }

  type User @key(fields: "id") {
    id: ID!
    username: String
  }
`)

const resolvers = {
  Query: {
    me() {
      return { id: '1', username: '@ava' }
    }
  },
  User: {
    __resolveReference(user, { fetchUserById }) {
      return fetchUserById(user.id)
    }
  }
}
const yoga = createYoga({
  schema: buildSubgraphSchema([{ typeDefs, resolvers }])
})

const server = createServer(yoga)

server.listen(4001, () => {
  console.log(`🚀 Server ready at http://localhost:4001`)
})

Federated tracing

Inject additional metrics for Apollo’s federated tracing.

You’ll need the @graphql-yoga/plugin-apollo-inline-trace Yoga plugin for this.

Installation

npm i graphql-yoga
npm i @graphql-yoga/plugin-apollo-inline-trace
npm i graphql

Example Federated tracing

import { createServer } from 'http'
import { createYoga } from 'graphql-yoga'
import { useApolloInlineTrace } from '@graphql-yoga/plugin-apollo-inline-trace'

const yoga = createYoga({
  plugins: [
    useApolloInlineTrace()
    // ...rest of your Apollo federation plugins
  ]
})

// Start the server and explore http://localhost:4000/graphql
const server = createServer(yoga)
server.listen(4000, () => {
  console.info('Server is running on http://localhost:4000/graphql')
})

Working Example

Check our working example to try it out.