rabbitmq

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

RabbitMQ Core Knowledge

RabbitMQ核心知识

Full Reference: See advanced.md for producer patterns (Node.js, Java, Python, Go), consumer patterns, security configuration, and high availability setup.

Deep Knowledge: Use
mcp__documentation__fetch_docs
with technology:
rabbitmq
for comprehensive documentation.

完整参考：查看advanced.md了解生产者模式（Node.js、Java、Python、Go）、消费者模式、安全配置和高可用部署方案。

深度知识：调用
mcp__documentation__fetch_docs
并指定技术为
rabbitmq
可获取完整文档。

Quick Start (Docker)

快速开始（Docker）

yaml

services:
  rabbitmq:
    image: rabbitmq:3-management
    ports:
      - "5672:5672"   # AMQP
      - "15672:15672" # Management UI
    environment:
      - RABBITMQ_DEFAULT_USER=admin
      - RABBITMQ_DEFAULT_PASS=admin

yaml

services:
  rabbitmq:
    image: rabbitmq:3-management
    ports:
      - "5672:5672"   # AMQP
      - "15672:15672" # Management UI
    environment:
      - RABBITMQ_DEFAULT_USER=admin
      - RABBITMQ_DEFAULT_PASS=admin

Core Concepts

核心概念

Concept	Description
Exchange	Routes messages to queues
Queue	Buffer that stores messages
Binding	Rule linking exchange to queue
Routing Key	Message attribute for routing

概念	描述
Exchange	将消息路由到队列
Queue	存储消息的缓冲区
Binding	关联交换机和队列的规则
Routing Key	用于路由的消息属性

Architecture

架构

Producer ──▶ Exchange ──binding──▶ Queue ──▶ Consumer
                │
                ├── direct   (exact routing key match)
                ├── topic    (pattern matching)
                ├── fanout   (broadcast to all)
                └── headers  (header-based routing)

生产者 ──▶ Exchange ──binding──▶ 队列 ──▶ 消费者
                │
                ├── direct   (路由键精确匹配)
                ├── topic    (模式匹配)
                ├── fanout   (广播到所有队列)
                └── headers  (基于消息头路由)

Exchange Types

交换机类型

Type	Routing	Use Case
direct	Exact key match	Task queues, RPC
topic	Pattern (*.#)	Event routing
fanout	Broadcast all	Notifications
headers	Header match	Complex routing

undefined

类型	路由规则	适用场景
direct	键精确匹配	任务队列、RPC
topic	模式匹配(*.#)	事件路由
fanout	全量广播	通知推送
headers	消息头匹配	复杂路由

undefined

Topic exchange patterns

Topic交换机匹配模式

order.* → matches order.created, order.updated order.# → matches order.created, order.item.added *.created → matches order.created, user.created

undefined

order.* → 匹配 order.created, order.updated order.# → 匹配 order.created, order.item.added *.created → 匹配 order.created, user.created

undefined

Queue Configuration

队列配置

Argument	Description
`x-message-ttl`	Message expiration (ms)
`x-max-length`	Max queue size
`x-dead-letter-exchange`	DLX for rejected messages
`x-dead-letter-routing-key`	DLQ routing key
`x-queue-type`	classic, quorum, stream

参数	描述
`x-message-ttl`	消息过期时间（毫秒）
`x-max-length`	队列最大长度
`x-dead-letter-exchange`	处理被拒绝消息的死信交换机
`x-dead-letter-routing-key`	死信队列路由键
`x-queue-type`	队列类型：classic、quorum、stream

When NOT to Use This Skill

本技能不适用场景

Event streaming with replay - Kafka provides persistent log
Ultra-high throughput (>100k msg/s) - Kafka or Pulsar scale better
AWS-native architecture - SQS integrates better
Lightweight messaging - NATS has simpler operations
JMS compliance required - ActiveMQ provides JMS API

需要回放的事件流场景：Kafka提供持久化日志能力更适配
超高吞吐量场景（>10万条消息/秒）：Kafka或Pulsar扩展性更好
AWS原生架构：SQS集成度更高
轻量级消息场景：NATS运维更简单
要求JMS兼容：ActiveMQ提供JMS API支持

Anti-Patterns

反模式

Anti-Pattern	Why It's Bad	Solution
No prefetch limit	Consumer overwhelmed	Set `prefetch` to 10-50
Classic queues in HA	Data loss on node failure	Use quorum queues
No DLX configured	Poison messages loop forever	Configure dead letter exchange
Automatic acks	Message loss on crash	Use manual acks
No publisher confirms	Silent message loss	Enable confirms for critical messages

反模式	弊端	解决方案
未设置prefetch限制	消费者负载过高	将 `prefetch` 设为10-50
高可用场景使用classic队列	节点故障时会丢失数据	使用quorum队列
未配置DLX	有毒消息会无限循环	配置死信交换机
自动ack	消费者崩溃时会丢失消息	使用手动ack
未开启生产者确认	消息丢失无感知	关键消息开启确认机制

Quick Troubleshooting

快速排查

Issue	Likely Cause	Fix
Messages piling up	Slow/dead consumer	Check consumer count and processing
High memory usage	Too many unacked messages	Reduce prefetch, add consumers
Messages not routed	Wrong routing key	Check exchange-queue bindings
Duplicate messages	Consumer crashed before ack	Implement idempotent processing
Publisher confirms timeout	Broker overloaded	Reduce publish rate

问题	可能原因	修复方案
消息堆积	消费者过慢或宕机	检查消费者数量和处理逻辑
内存占用过高	未ack消息过多	降低prefetch值，新增消费者
消息未被路由	路由键错误	检查交换机与队列的绑定关系
消息重复	消费者ack前崩溃	实现消费幂等逻辑
生产者确认超时	Broker负载过高	降低生产速率

Production Checklist

生产环境检查清单

Reference Documentation

参考文档

Available topics:

basics

exchanges

queues

consumers

clustering

production

可用主题：

basics

、

exchanges

、

queues

、

consumers

、

clustering

、

production