Apollo Server plugin event reference

All plugin lifecycle methods are async, except for willResolveField and schemaDidLoadOrUpdate.

This reference describes the lifecycle events that your custom Apollo Server plugin can respond to.

Apollo Server fires two types of events that plugins can hook into: server lifecycle events and request lifecycle events.

Server lifecycle events are high-level events related to the lifecycle of Apollo Server itself (e.g., serverWillStart).
Request lifecycle events are associated with the lifecycle of a specific request.
- You define responses to these events within the response to a requestDidStart event, as described in Responding to request lifecycle events.

With two exceptions, all plugin methods in Apollo Server are async. The first exception is willResolveField, which is called much more frequently than other plugin methods. The second exception is schemaDidLoadOrUpdate, where making the method async would introduce unclear ordering semantics around method executions.

Server lifecycle events

`serverWillStart`

The serverWillStart event fires when Apollo Server is preparing to start serving GraphQL requests. The server doesn't start until this asynchronous method completes. If it throws (i.e., if the Promise it returns is rejected), startup fails and your server does not serve GraphQL operations. This helps you make sure all of your server's dependencies are available before attempting to begin serving requests.

This event is fired at different times depending on which Apollo Server integration you're using:

If you are using startStandaloneServer, it's fired when you invoke the startStandaloneServer function with your server instance.
In non-serverless integrations like expressMiddleware, it's fired from the start() method.
In serverless integrations, it's usually fired in response to the first incoming request.

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        console.log('Server starting!');
8
      },
9
    },
10
  ],
11
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        console.log('Server starting!');
8
      },
9
    },
10
  ],
11
});

`drainServer`

The drainServer event fires when Apollo Server is starting to shut down because ApolloServer.stop() has been invoked (either explicitly by your code, or by one of the termination signal handlers). While drainServer handlers run, GraphQL operations can still execute successfully. This hook is designed to allow you to stop accepting new connections and close existing connections. Apollo Server has a built-in plugin which uses this event to drain a Node http.Server.

You define your drainServer handler in the object returned by your serverWillStart handler, because the two handlers usually interact with the same data. Currently, drainServer handlers do not take arguments (this might change in the future).

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        return {
8
          async drainServer() {
9
            await myCustomServer.drain();
10
          },
11
        };
12
      },
13
    },
14
  ],
15
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        return {
8
          async drainServer() {
9
            await myCustomServer.drain();
10
          },
11
        };
12
      },
13
    },
14
  ],
15
});

`serverWillStop`

The serverWillStop event fires when Apollo Server is starting to shut down because ApolloServer.stop() has been invoked (either explicitly by your code, or by one of the termination signal handlers). If your plugin is running any background tasks, this is a good place to shut them down.

You define your serverWillStop handler in the object returned by your serverWillStart handler, because the two handlers usually interact with the same data. Currently, serverWillStop handlers do not take arguments (this might change in the future).

When your serverWillStop handler is called, Apollo Server is in a state where it will no longer start to execute new GraphQL operations, so it's a good place to flush observability data. If you are looking for a hook that runs while operations can still execute, try drainServer.

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        const interval = setInterval(doSomethingPeriodically, 1000);
8
        return {
9
          async serverWillStop() {
10
            clearInterval(interval);
11
          },
12
        };
13
      },
14
    },
15
  ],
16
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        const interval = setInterval(doSomethingPeriodically, 1000);
8
        return {
9
          async serverWillStop() {
10
            clearInterval(interval);
11
          },
12
        };
13
      },
14
    },
15
  ],
16
});

`renderLandingPage`

This event enables you to serve a custom landing page from Apollo Server's base URL. The event is fired once by Apollo Server after all serverWillStart events run. At most one installed plugin can define a renderLandingPage handler. Otherwise, Apollo Server throws an error on startup.

You define your plugin's renderLandingPage handler in the object returned by your serverWillStart handler, which enables it to read values passed to serverWillStart:

index.ts

1
const server = new ApolloServer({
2
  typeDefs,
3
  resolvers,
4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        return {
8
          async renderLandingPage() {
9
            const html = `
10
              <!DOCTYPE html>
11
              <html>
12
                  <head>
13
                  </head>
14
                  <body>
15
                      <h1>Hello world!</h1>
16
                  </body>
17
              </html>`;
18
            return { html };
19
          },
20
        };
21
      },
22
    },
23
  ],
24
});

index.js

1
const server = new ApolloServer({
2
  typeDefs,
3
  resolvers,
4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        return {
8
          async renderLandingPage() {
9
            const html = `
10
              <!DOCTYPE html>
11
              <html>
12
                  <head>
13
                  </head>
14
                  <body>
15
                      <h1>Hello world!</h1>
16
                  </body>
17
              </html>`;
18
            return { html };
19
          },
20
        };
21
      },
22
    },
23
  ],
24
});

The handler should return an object with a string html property. The value of that property is served as HTML for any requests with accept: text/html headers. The html property can also be an async function that returns a string. This function is called for each landing page request.

For more landing page options, see Changing the landing page.

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        return {
8
          async renderLandingPage() {
9
            return {
10
              async html() {
11
                return `<html><body>Welcome to your server!</body></html>`;
12
              },
13
            };
14
          },
15
        };
16
      },
17
    },
18
  ],
19
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        return {
8
          async renderLandingPage() {
9
            return {
10
              async html() {
11
                return `<html><body>Welcome to your server!</body></html>`;
12
              },
13
            };
14
          },
15
        };
16
      },
17
    },
18
  ],
19
});

`requestDidStart`

New in Apollo Server 4: In Apollo Server 4, requestDidStart hooks are called in parallel rather than in series.

The requestDidStart event fires whenever Apollo Server begins fulfilling a GraphQL request.

1
requestDidStart?(
2
  requestContext: GraphQLRequestContext<TContext>,
3
): Promise<GraphQLRequestListener<TContext> | void>;

This function can optionally return an object that includes functions for responding to request lifecycle events that might follow requestDidStart.

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async requestDidStart(requestContext) {
7
        // Within this returned object, define functions that respond
8
        // to request-specific lifecycle events.
9
        return {
10
          // The `parsingDidStart` request lifecycle event fires
11
          // when parsing begins. The event is scoped within an
12
          // associated `requestDidStart` server lifecycle event.
13
          async parsingDidStart(requestContext) {
14
            console.log('Parsing started!');
15
          },
16
        };
17
      },
18
    },
19
  ],
20
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async requestDidStart(requestContext) {
7
        // Within this returned object, define functions that respond
8
        // to request-specific lifecycle events.
9
        return {
10
          // The `parsingDidStart` request lifecycle event fires
11
          // when parsing begins. The event is scoped within an
12
          // associated `requestDidStart` server lifecycle event.
13
          async parsingDidStart(requestContext) {
14
            console.log('Parsing started!');
15
          },
16
        };
17
      },
18
    },
19
  ],
20
});

If your plugin doesn't need to respond to any request lifecycle events, requestDidStart should not return a value.

`schemaDidLoadOrUpdate`

The schemaDidLoadOrUpdate event fires whenever Apollo Server initially loads the schema or updates the schema.

A schemaDidLoadOrUpdate handler is given the new API schema and optionally the new core schema (if using a gateway).

schemaDidLoadOrUpdate is a synchronous plugin API (i.e., it does not return a Promise).

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        return {
8
          schemaDidLoadOrUpdate({ apiSchema, coreSupergraphSdl }) {
9
            console.log(`The API schema is ${printSchema(apiSchema)}`);
10
            if (coreSupergraphSdl) {
11
              console.log(`The core schema is ${coreSupergraphSdl}`);
12
            }
13
          },
14
        };
15
      },
16
    },
17
  ],
18
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async serverWillStart() {
7
        return {
8
          schemaDidLoadOrUpdate({ apiSchema, coreSupergraphSdl }) {
9
            console.log(`The API schema is ${printSchema(apiSchema)}`);
10
            if (coreSupergraphSdl) {
11
              console.log(`The core schema is ${coreSupergraphSdl}`);
12
            }
13
          },
14
        };
15
      },
16
    },
17
  ],
18
});

`startupDidFail`

The startupDidFail hook is triggered if your server fails to start. This can occur if the schema fails to load or a serverWillStart or renderLandingPage hook throws an error. This hook receives the thrown error, which is the same error that await server.start() throws.

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async startupDidFail({ error }) {
7
        console.log(`Startup failed: ${error}`);
8
      },
9
    },
10
  ],
11
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async startupDidFail({ error }) {
7
        console.log(`Startup failed: ${error}`);
8
      },
9
    },
10
  ],
11
});

Request lifecycle events

If you're using TypeScript to create your plugin, implement the GraphQLRequestListener interface to define functions for request lifecycle events.

When Apollo Server processes a request, these events fire in the order listed (with the exception of didEncounterErrors, which might fire in one of a few places depending on when errors occur). See the flow diagram

Note that not every event fires for every request (for example, parsingDidStart doesn't fire for an operation that Apollo Server has cached and doesn't need to parse again).

`didResolveSource`

The didResolveSource event is invoked after Apollo Server has determined the String-representation of the incoming operation that it will act upon. In the event that this String was not directly passed in from the client, this may be retrieved from a cache store (e.g., Automated Persisted Queries).

At this stage, there is not a guarantee that the operation is not malformed.

1
didResolveSource?(
2
  requestContext: WithRequired<
3
    GraphQLRequestContext<TContext>, 'source' | 'queryHash'>,
4
  >,
5
): Promise<void>;

`parsingDidStart`

The parsingDidStart event fires whenever Apollo Server will parse a GraphQL request to create its associated document AST.

If Apollo Server receives a request with a query string that matches a previous request, the associated document might already be available in Apollo Server's cache. In this case, parsingDidStart is not called for the request, because parsing does not occur.

1
parsingDidStart?(
2
  requestContext: WithRequired<
3
    GraphQLRequestContext<TContext>, 'source' | 'queryHash'
4
  >,
5
): Promise<void | (err?: Error) => Promise<void>>;

`validationDidStart`

The validationDidStart event fires whenever Apollo Server will validate a request's document AST against your GraphQL schema.

Like parsingDidStart, this event does not fire if a request's document is already available in Apollo Server's cache (only successfully validated documents are cached by Apollo Server).

The document AST is guaranteed to be available at this stage, because parsing must succeed for validation to occur.

1
validationDidStart?(
2
  requestContext: WithRequired<
3
    GraphQLRequestContext<TContext>,
4
    'source' | 'queryHash' | 'document'
5
  >,
6
): Promise<void | (err?: ReadonlyArray<Error>) => Promise<void>>;

`didResolveOperation`

The didResolveOperation event fires after the graphql library successfully determines the operation to execute from a request's document AST. At this stage, both the operationName string and operation AST are available.

This event is not associated with your GraphQL server's resolvers. When this event fires, your resolvers have not yet executed (they execute after executionDidStart).

If the operation is anonymous (i.e., the operation is query { ... } instead of query NamedQuery { ... }), then operationName is null.

If a didResolveOperation hook throws a GraphQLError, that error is serialized and returned to the client with an HTTP status code of 500 unless it specifies a different status code.

The didResolveOperation hook is a great spot to perform extra validation because it has access to the parsed and validated operation and the request-specific context (i.e., contextValue). Multiple plugins can run the didResolveOperation in parallel, but if more than one plugin throws, the client only receives a single error.

1
didResolveOperation?(
2
  requestContext: WithRequired<
3
    GraphQLRequestContext<TContext>,
4
    'source' | 'queryHash' | 'document' | 'operationName'
5
  >,
6
): Promise<void>;

`responseForOperation`

The responseForOperation event is fired immediately before GraphQL execution would take place. If its return value resolves to a non-null GraphQLResponse, that result is used instead of executing the query. Hooks from different plugins are invoked in series, and the first non-null response is used.

1
responseForOperation?(
2
  requestContext: WithRequired<
3
    GraphQLRequestContext<TContext>,
4
    'source' | 'queryHash' | 'document' | 'operationName' | 'operation'
5
): Promise<GraphQLResponse | null>;

`executionDidStart`

The executionDidStart event fires whenever Apollo Server begins executing the GraphQL operation specified by a request's document AST.

1
executionDidStart?(
2
  requestContext: WithRequired<
3
    GraphQLRequestContext<TContext>,
4
    'source' | 'queryHash' | 'document' | 'operationName' | 'operation'
5
  >,
6
): Promise<GraphQLRequestExecutionListener | void>;

executionDidStart may return an object with one or both of the methods executionDidEnd and willResolveField. executionDidEnd is treated like an end hook: it is called after execution with any errors that occurred. (If the operation uses incremental delivery directives such as @defer, executionDidEnd is called when the fields required to fill the initial payload have finished executing; you can use willSendSubsequentPayload to hook into the end of execution for each subsequent payload.) willResolveField is documented in the next section.

`willResolveField`

The willResolveField event fires whenever Apollo Server is about to resolve a single field during the execution of an operation. The handler is passed an object with four fields (source, args, contextValue, and info) that correspond to the four positional arguments passed to resolvers. Note that source corresponds to the argument often called parent in our docs.

You provide your willResolveField handler in the object returned by your executionDidStart handler.

Your willResolveField handler can optionally return an "end hook" function that's invoked with the resolver's result (or the error that it throws). The end hook is called when your resolver has fully resolved (e.g., if the resolver returns a Promise, the hook is called with the Promise's eventual resolved result).

willResolveField and its end hook are synchronous plugin APIs (i.e., they do not return Promises).

willResolveField only fires when a field is resolved inside the Apollo Server itself; it does not fire at all if the server is a Gateway.

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async requestDidStart(initialRequestContext) {
7
        return {
8
          async executionDidStart(executionRequestContext) {
9
            return {
10
              willResolveField({ source, args, contextValue, info }) {
11
                const start = process.hrtime.bigint();
12
                return (error, result) => {
13
                  const end = process.hrtime.bigint();
14
                  console.log(`Field ${info.parentType.name}.${info.fieldName} took ${end - start}ns`);
15
                  if (error) {
16
                    console.log(`It failed with ${error}`);
17
                  } else {
18
                    console.log(`It returned ${result}`);
19
                  }
20
                };
21
              },
22
            };
23
          },
24
        };
25
      },
26
    },
27
  ],
28
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async requestDidStart(initialRequestContext) {
7
        return {
8
          async executionDidStart(executionRequestContext) {
9
            return {
10
              willResolveField({ source, args, contextValue, info }) {
11
                const start = process.hrtime.bigint();
12
                return (error, result) => {
13
                  const end = process.hrtime.bigint();
14
                  console.log(`Field ${info.parentType.name}.${info.fieldName} took ${end - start}ns`);
15
                  if (error) {
16
                    console.log(`It failed with ${error}`);
17
                  } else {
18
                    console.log(`It returned ${result}`);
19
                  }
20
                };
21
              },
22
            };
23
          },
24
        };
25
      },
26
    },
27
  ],
28
});

`didEncounterErrors`

The didEncounterErrors event fires when Apollo Server encounters errors while parsing, validating, or executing a GraphQL operation. The errors are available on requestContext.errors.

(If the operation uses incremental delivery directives such as @defer, didEncounterErrors is only called when errors that will be sent in the initial payload are encountered; you can use didEncounterSubsequentErrors to find out if more errors are found later.)

1
didEncounterErrors?(
2
  requestContext: WithRequired<
3
    GraphQLRequestContext<TContext>, 'errors'
4
  >,
5
): Promise<void>;

`didEncounterSubsequentErrors`

The didEncounterSubsequentErrors event only fires for operations that use incremental delivery directives such as @defer. This hook is called when any execution errors are encountered after the initial payload is sent; didEncounterErrors is not called in this case. The errors in question are provided as the second argument to the hook (not as requestContext.errors, which will continue to be the list of errors sent in the initial payload).

1
didEncounterSubsequentErrors?(
2
  requestContext: GraphQLRequestContextDidEncounterSubsequentErrors<TContext>,
3
  errors: ReadonlyArray<GraphQLError>,
4
): Promise<void>;

`willSendResponse`

The willSendResponse event fires whenever Apollo Server is about to send a response for a GraphQL operation. This event fires (and Apollo Server sends a response) even if the GraphQL operation encounters one or more errors.

(If the operation uses incremental delivery directives such as @defer, willSendResponse is called before the initial payload is sent; you can use willSendSubsequentPayload to find out when more payloads will be sent.)

1
willSendResponse?(
2
  requestContext: WithRequired<
3
    GraphQLRequestContext<TContext>, 'source' | 'queryHash'
4
  >,
5
): Promise<void>;

`willSendSubsequentPayload`

The willSendSubsequentPayload event only fires for operations that use incremental delivery directives such as @defer. This hook is called before each payload after the initial one is sent, similarly to willSendResponse. The payload in question is provided as the second argument to the hook (not on requestContext). If this is the last payload, payload.hasNext will be false. Note that the precise format of payload is determined by the graphql-js project, and incremental delivery support has not yet (as of September 2022) been released in an official release of graphql-js. When the official release (expected to be graphql@17) is released, the format of this argument may potentially change; in this case, Apollo Server may change the precise details of this hook in a backward-incompatible way in a minor release of Apollo Server. (For now, this hook can only be called if you install a pre-release of graphql@17.)

1
willSendSubsequentPayload?(
2
  requestContext: GraphQLRequestContextWillSendSubsequentPayload<TContext>,
3
  payload: GraphQLExperimentalFormattedSubsequentIncrementalExecutionResult,
4
): Promise<void>;

`contextCreationDidFail`

The contextCreationDidFail event fires if the user-provided context function throws an error. Because the request context is incomplete, this hook only receives the error (not a full GraphQLRequestContext). If you want more information about the request that triggered this event, consider wrapping your context function in a try/catch block and adding the appropriate information to the thrown error.

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async contextCreationDidFail({ error }) {
7
        console.log(`Context creation failed: ${error}`);
8
      },
9
    },
10
  ],
11
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async contextCreationDidFail({ error }) {
7
        console.log(`Context creation failed: ${error}`);
8
      },
9
    },
10
  ],
11
});

`invalidRequestWasReceived`

The invalidRequestWasReceived event fires any time a "Bad Request" error is thrown during request execution. This includes CSRF prevention and malformed requests (e.g., incorrect headers, invalid JSON body, or invalid search params for GET), but doesn't include malformed GraphQL. Because the request context is incomplete, this hook only receives the error (not a full GraphQLRequestContext). If you want more information about the request that triggered this event, consider wrapping your context function in a try/catch block and adding the appropriate information to the thrown error.

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async invalidRequestWasReceived({ error }) {
7
        console.log(`Bad request: ${error}`);
8
      },
9
    },
10
  ],
11
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async invalidRequestWasReceived({ error }) {
7
        console.log(`Bad request: ${error}`);
8
      },
9
    },
10
  ],
11
});

`unexpectedErrorProcessingRequest`

The unexpectedErrorProcessingRequest event fires whenever an "unexpected" error is thrown during request execution. "Unexpected" errors don't include common client data errors such as validation, parsing, or GraphQL execution errors. Instead, an unexpected error indicates a programming error, such as a plugin hook throwing unexpectedly or Apollo Server encountering a bug. No matter the cause, Apollo Server masks this type of error from a client.

Note this hook is on the top level of the plugin rather than on the object returned from requestDidStart, because this hook triggers if requestDidStart throws.

Example

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async unexpectedErrorProcessingRequest({ requestContext, error }) {
7
        console.log(`Something went wrong: ${error}`);
8
      },
9
    },
10
  ],
11
});

1
const server = new ApolloServer({
2
  /* ... other necessary configuration ... */
3

4
  plugins: [
5
    {
6
      async unexpectedErrorProcessingRequest({ requestContext, error }) {
7
        console.log(`Something went wrong: ${error}`);
8
      },
9
    },
10
  ],
11
});

Creating plugins

Setup