@nestjs/throttler

What is @nestjs/throttler?

@nestjs/throttler is a rate-limiting module for NestJS applications. It helps to control the rate of incoming requests to prevent abuse and ensure fair usage of resources. This package is particularly useful for APIs to avoid being overwhelmed by too many requests in a short period.

What are @nestjs/throttler's main functionalities?

Basic Rate Limiting

This feature allows you to set up basic rate limiting for your entire application. In this example, a maximum of 10 requests per minute is allowed.

```typescript
import { ThrottlerModule } from '@nestjs/throttler';

@Module({
  imports: [
    ThrottlerModule.forRoot({
      ttl: 60,
      limit: 10,
    }),
  ],
})
export class AppModule {}
```

Rate Limiting for Specific Routes

This feature allows you to apply rate limiting to specific routes. In this example, the 'limited' route is restricted to 5 requests per minute.

```typescript
import { Controller, Get, UseGuards } from '@nestjs/common';
import { ThrottlerGuard, Throttle } from '@nestjs/throttler';

@Controller('api')
@UseGuards(ThrottlerGuard)
export class ApiController {
  @Get('limited')
  @Throttle(5, 60)
  getLimited() {
    return 'This route is rate limited to 5 requests per minute';
  }
}
```

Customizing Rate Limiting

This feature allows you to customize the rate limiting storage mechanism. In this example, Redis is used as the storage backend for rate limiting.

```typescript
import { ThrottlerModule, ThrottlerGuard, ThrottlerStorageRedisService } from '@nestjs/throttler';
import { Module } from '@nestjs/common';
import { APP_GUARD } from '@nestjs/core';

@Module({
  imports: [
    ThrottlerModule.forRoot({
      ttl: 60,
      limit: 10,
      storage: new ThrottlerStorageRedisService(),
    }),
  ],
  providers: [
    {
      provide: APP_GUARD,
      useClass: ThrottlerGuard,
    },
  ],
})
export class AppModule {}
```

Other packages similar to @nestjs/throttler

express-rate-limit

express-rate-limit is a basic rate-limiting middleware for Express applications. It is simpler and more lightweight compared to @nestjs/throttler, but it lacks the deep integration with NestJS and some advanced features like custom storage backends.

rate-limiter-flexible

rate-limiter-flexible is a highly flexible rate-limiting library that supports various backends like Redis, MongoDB, and in-memory storage. It offers more customization options compared to @nestjs/throttler but requires more setup and integration effort.

koa-ratelimit

koa-ratelimit is a rate-limiting middleware for Koa applications. It provides similar functionalities to @nestjs/throttler but is designed specifically for Koa, making it less suitable for NestJS applications.

README

A progressive Node.js framework for building efficient and scalable server-side applications.

Description

A Rate-Limiter for NestJS, regardless of the context.

For an overview of the community storage providers, see Community Storage Providers.

This package comes with a couple of goodies that should be mentioned, first is the ThrottlerModule.

Table of Contents

• Description
• Table of Contents
• Usage
  • ThrottlerModule
  • Decorators
    • @Throttle()
    • @SkipThrottle()
  • Ignoring specific user agents
  • ThrottlerStorage
  • Working with Websockets
  • Working with GraphQL
• Community Storage Providers

Usage

ThrottlerModule

The ThrottleModule is the main entry point for this package, and can be used in a synchronous or asynchronous manner. All the needs to be passed is the ttl, the time to live in seconds for the request tracker, and the limit, or how many times an endpoint can be hit before returning a 429.

import { APP_GUARD } from '@nestjs/core';
import { ThrottlerGuard, ThrottlerModule } from 'nestjs-throttler';

@Module({
  imports: [
    ThrottlerModule.forRoot({
      ttl: 60,
      limit: 10,
    }),
  ],
  providers: [
    {
      provide: APP_GUARD,
      useClass: ThrottlerGuard,
    },
  ],
})
export class AppModule {}

The above would mean that 10 requests from the same IP can be made to a single endpoint in 1 minute.

@Module({
  imports: [
    ThrottlerModule.forRootAsync({
      imports: [ConfigModule],
      inject: [ConfigService],
      useFactory: (config: ConfigService) => ({
        ttl: config.get('THROTTLE_TTL'),
        limit: config.get('THROTTLE_LIMIT'),
      }),
    }),
  ],
  providers: [
    {
      provide: APP_GUARD,
      useClass: ThrottlerGuard,
    },
  ],
})
export class AppModule {}

The above is also a valid configuration for asynchronous registration of the module.

NOTE: If you add the ThrottlerGuard to your AppModule as a global guard then all the incoming requests will be throttled by default. This can also be omitted in favor of @UseGuards(ThrottlerGuard). The global guard check can be skipped using the @SkipThrottle() decorator mentioned later.

Example with @UseGuards(ThrottlerGuard):

// app.module.ts
@Module({
  imports: [
    ThrottlerModule.forRoot({
      ttl: 60,
      limit: 10,
    }),
  ],
})
export class AppModule {}

// app.controller.ts
@Controller()
export class AppController {
  @UseGuards(ThrottlerGuard)
  @Throttle(5, 30)
  normal() {}
}

Decorators

@Throttle()

@Throttle(limit: number = 30, ttl: number = 60)

This decorator will set THROTTLER_LIMIT and THROTTLER_TTL metadatas on the route, for retrieval from the Reflector class. Can be applied to controllers and routes.

@SkipThrottle()

@SkipThrottle(skip = true)

This decorator can be used to skip a route or a class or to negate the skipping of a route in a class that is skipped.

@SkipThrottle()
@Controller()
export class AppController {
  @SkipThrottle(false)
  dontSkip() {}

  doSkip() {}
}

In the above controller, dontSkip would be counted against and rate-limited while doSkip would not be limited in any way.

Ignoring specific user agents

You can use the ignoreUserAgents key to ignore specific user agents.

@Module({
  imports: [
    ThrottlerModule.forRoot({
      ttl: 60,
      limit: 10,
      ignoreUserAgents: [
        // Don't throttle request that have 'googlebot' defined in them.
        // Example user agent: Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)
        /googlebot/gi,

        // Don't throttle request that have 'bingbot' defined in them.
        // Example user agent: Mozilla/5.0 (compatible; Bingbot/2.0; +http://www.bing.com/bingbot.htm)
        new RegExp('bingbot', 'gi'),
      ],
    }),
  ],
})
export class AppModule {}

ThrottlerStorage

Interface to define the methods to handle the details when it comes to keeping track of the requests.

Currently the key is seen as an MD5 hash of the IP the ClassName and the MethodName, to ensure that no unsafe characters are used and to ensure that the package works for contexts that don't have explicit routes (like Websockets and GraphQL).

The interface looks like this:

export interface ThrottlerStorage {
  getRecord(key: string): Promise<number[]>;
  addRecord(key: string, ttl: number): Promise<void>;
}

So long as the Storage service implements this interface, it should be usable by the ThrottlerGuard.

Working with Websockets

To work with Websockets you can extend the ThrottlerGuard and override the handleRequest method with something like the following method

@Injectable()
export class WsThrottlerGuard extends ThrottlerGuard {
  async handleRequest(context: ExecutionContext, limit: number, ttl: number): Promise<boolean> {
    const client = context.switchToWs().getClient();
    // this is a generic method to switch between `ws` and `socket.io`. You can choose what is appropriate for you
    const ip = ['conn', '_socket']
      .map((key) => client[key])
      .filter((obj) => obj)
      .shift().remoteAddress;
    const key = this.generateKey(context, ip);
    const ttls = await this.storageService.getRecord(key);

    if (ttls.length >= limit) {
      throw new ThrottlerException();
    }

    await this.storageService.addRecord(key, ttl);
    return true;
  }
}

There are some things to take keep in mind when working with websockets:

• You cannot bind the guard with APP_GUARD or app.useGlobalGuards() due to how Nest binds global guards.
• When a limit is reached, Nest will emit an exception event, so make sure there is a listener ready for this.

Working with GraphQL

To get the ThrottlerModule to work with the GraphQL context, a couple of things must happen.

• You must use Express and apollo-server-express as your GraphQL server engine. This is the default for Nest, but the apollo-server-fastify package does not currently support passing res to the context, meaning headers cannot be properly set.
• When configuring your GraphQLModule, you need to pass an option for context in the form of ({ req, res}) => ({ req, res }). This will allow access to the Express Request and Response objects, allowing for the reading and writing of headers.
• You must add in some additional context switching to get the ExecutionContext to pass back values correctly (or you can override the method entirely)

@Injectable()
export class GqlThrottlerGuard extends ThrottlerGuard {
  getRequestResponse(context: ExecutionContext) {
    const gqlCtx = GqlExecutionContext.create(context);
    const ctx = gql.getContext();
    return { req, ctx.req, res: ctx.res }; // ctx.request and ctx.reply for fastify
  }
}

Community Storage Providers

• Redis

Feel free to submit a PR with your custom storage provider being added to this list.