Queue & Messaging Recipes

spoonfeed provides 3 queue and messaging recipes covering message brokers, job queues, and dead letter handling. Use these recipes for asynchronous processing, event-driven architectures, and background job execution. Choose RabbitMQ for message routing and pub/sub, or BullMQ for Redis-backed job queues with scheduling and retries.

Recommended combinations

• Reliable async processing: rabbitmq + dead-letter-queue + transactional-outbox
• Background jobs with retries: bullmq + redis-cache + dead-letter-queue

---

RabbitMQ

RabbitMQ message broker integration via @nestjs/microservices.

ID	rabbitmq
Dependencies	@nestjs/microservices amqplib
Dev dependencies	@types/amqplib
Compatible with	HTTP API, AWS Lambda, Microservice, Scheduled Worker, Monorepo, Full-Stack

Environment variables:

Variable	Default	Description
RABBITMQ_URL	amqp://guest:guest@localhost:5672	RabbitMQ connection URL

Usage:

• Use @MessagePattern() for request/response (RPC) patterns
• Use @EventPattern() for fire-and-forget event patterns
• Configure queues in the microservice module registration
• Use durable queues in production
• Define message contracts as typed interfaces
• Always acknowledge messages after successful processing

Requires a running RabbitMQ instance

RabbitMQ must be running locally or via Docker. The docker-compose-dev recipe can include a RabbitMQ service.

Tip

Pair with the dead-letter-queue recipe for handling permanently failed messages.

Pairs well with: dead-letter-queue, transactional-outbox, correlation-id

---

BullMQ

Redis-backed job queue with BullMQ for background job processing.

ID	bullmq
Dependencies	@nestjs/bullmq bullmq
Compatible with	HTTP API, AWS Lambda, Microservice, Scheduled Worker, Monorepo, Full-Stack

Environment variables:

Variable	Default	Description
REDIS_HOST	localhost	Redis host for BullMQ
REDIS_PORT	6379	Redis port for BullMQ

Usage:

• Define processors by extending WorkerHost with @Processor() decorator
• Inject Queue to add jobs: queue.add('job-name', data)
• Use named jobs for routing to specific handlers
• Handle failed jobs with @OnWorkerEvent() decorator
• Configure concurrency and rate limiting per queue

@Processor('email')
export class EmailProcessor extends WorkerHost {
  async process(job: Job<EmailPayload>) {
    // Process the job
  }
}

Requires a running Redis instance

BullMQ uses Redis as its backing store. Ensure Redis is running locally or via Docker.

Pairs well with: redis-cache, dead-letter-queue

---

Dead Letter Queue

Handle permanently failed messages with DLQ routing and replay.

ID	dead-letter-queue
Compatible with	HTTP API, Microservice, Scheduled Worker, Monorepo, Full-Stack

Usage:

• DeadLetterQueueService stores permanently failed messages
• Call dlqService.add() in queue error handlers when a message cannot be processed after retries
• Inspect dead letters with getAll() or getByQueue()
• Remove processed dead letters with remove()

Integrates with BullMQ or RabbitMQ by calling dlqService.add() in your consumer error handlers.

try {
  await processMessage(message);
} catch (error) {
  if (retryCount >= maxRetries) {
    dlqService.add({
      queue: 'orders',
      payload: message,
      error: error.message,
      attempts: retryCount,
    });
  }
}