Redis分布式锁实现指南

一、分布式锁的基本原理与需求

在分布式系统中，多个服务实例可能需要访问共享资源，此时需要分布式锁来保证操作的互斥性。Redis凭借高性能和原子操作特性，成为实现分布式锁的理想选择。

1. 分布式锁的核心需求

• 互斥性：同一时刻只能有一个客户端持有锁

• 安全性：锁只能被持有该锁的客户端释放，不能被其他客户端释放

• 容错性：在Redis节点宕机的情况下，锁仍能正常工作（可通过Redis集群解决）

• 可重入性：同一个客户端可以多次获取同一把锁（可选）

• 锁超时：防止死锁，锁在一定时间后自动释放

2. Redis实现分布式锁的基础

Redis实现分布式锁主要基于以下命令：

SET key value NX PX timeout

• NX：仅在键不存在时设置键值
• PX timeout：设置键的过期时间（毫秒）
• value：建议使用唯一值（如UUID），确保锁只能由持有者释放

二、Java实现Redis分布式锁

1. 基于Jedis客户端的简单实现

import redis.clients.jedis.Jedis;
import java.util.UUID;

public class RedisLock {
    private static final String LOCK_SUCCESS = "OK";
    private static final Long RELEASE_SUCCESS = 1L;
    private static final String SET_IF_NOT_EXIST = "NX";
    private static final String SET_WITH_EXPIRE_TIME = "PX";
    
    private Jedis jedis;
    private String lockKey;
    private String requestId;
    private int expireTime;
    
    public RedisLock(Jedis jedis, String lockKey, int expireTime) {
        this.jedis = jedis;
        this.lockKey = lockKey;
        this.expireTime = expireTime;
        this.requestId = UUID.randomUUID().toString();
    }
    
    /**
     * 获取锁
     * @return 是否成功获取锁
     */
    public boolean acquire() {
        String result = jedis.set(lockKey, requestId, SET_IF_NOT_EXIST, SET_WITH_EXPIRE_TIME, expireTime);
        return LOCK_SUCCESS.equals(result);
    }
    
    /**
     * 释放锁
     * @return 是否成功释放锁
     */
    public boolean release() {
        // 使用Lua脚本保证原子性
        String script = "if redis.call('get', KEYS[1]) == ARGV[1] then return redis.call('del', KEYS[1]) else return 0 end";
        Object result = jedis.eval(script, 1, lockKey, requestId);
        return RELEASE_SUCCESS.equals(result);
    }
    
    public String getRequestId() {
        return requestId;
    }
}

2. 使用示例

import redis.clients.jedis.Jedis;

public class RedisLockExample {
    public static void main(String[] args) {
        // 初始化Redis客户端
        Jedis jedis = new Jedis("localhost", 6379);
        
        // 创建锁实例
        RedisLock lock = new RedisLock(jedis, "my_distributed_lock", 30000);
        
        // 获取锁
        boolean acquired = lock.acquire();
        if (acquired) {
            try {
                // 执行临界区代码
                System.out.println("获取锁成功，执行临界区操作");
                Thread.sleep(5000);
            } catch (InterruptedException e) {
                e.printStackTrace();
            } finally {
                // 释放锁
                lock.release();
                System.out.println("释放锁");
            }
        } else {
            System.out.println("获取锁失败");
        }
        
        // 关闭Redis连接
        jedis.close();
    }
}

三、Redis分布式锁的进阶问题与解决方案

1. 锁续期问题

为避免业务执行时间超过锁的过期时间导致锁提前释放，可以使用"看门狗"机制自动续期：

import redis.clients.jedis.Jedis;
import java.util.UUID;
import java.util.concurrent.Executors;
import java.util.concurrent.ScheduledExecutorService;
import java.util.concurrent.TimeUnit;

public class RedisLockWithAutoRenew {
    private static final String LOCK_SUCCESS = "OK";
    private static final Long RELEASE_SUCCESS = 1L;
    private static final String SET_IF_NOT_EXIST = "NX";
    private static final String SET_WITH_EXPIRE_TIME = "PX";
    
    private Jedis jedis;
    private String lockKey;
    private String requestId;
    private int expireTime;
    private ScheduledExecutorService renewExecutor;
    private int renewInterval;
    private boolean isLocked = false;
    
    public RedisLockWithAutoRenew(Jedis jedis, String lockKey, int expireTime) {
        this.jedis = jedis;
        this.lockKey = lockKey;
        this.expireTime = expireTime;
        this.requestId = UUID.randomUUID().toString();
        this.renewExecutor = Executors.newSingleThreadScheduledExecutor();
        this.renewInterval = expireTime / 3; // 每三分之一的过期时间续期一次
    }
    
    /**
     * 获取锁
     * @return 是否成功获取锁
     */
    public boolean acquire() {
        String result = jedis.set(lockKey, requestId, SET_IF_NOT_EXIST, SET_WITH_EXPIRE_TIME, expireTime);
        if (LOCK_SUCCESS.equals(result)) {
            isLocked = true;
            startAutoRenew();
            return true;
        }
        return false;
    }
    
    /**
     * 释放锁
     * @return 是否成功释放锁
     */
    public boolean release() {
        isLocked = false;
        renewExecutor.shutdownNow();
        
        String script = "if redis.call('get', KEYS[1]) == ARGV[1] then return redis.call('del', KEYS[1]) else return 0 end";
        Object result = jedis.eval(script, 1, lockKey, requestId);
        return RELEASE_SUCCESS.equals(result);
    }
    
    /**
     * 启动自动续期
     */
    private void startAutoRenew() {
        renewExecutor.scheduleAtFixedRate(() -> {
            if (isLocked) {
                String script = "if redis.call('get', KEYS[1]) == ARGV[1] then return redis.call('pexpire', KEYS[1], ARGV[2]) else return 0 end";
                jedis.eval(script, 1, lockKey, requestId, String.valueOf(expireTime));
            }
        }, renewInterval, renewInterval, TimeUnit.MILLISECONDS);
    }
}

2. Redisson框架的使用

Redisson是一个基于Redis的Java驻内存数据网格（In-Memory Data Grid），提供了分布式锁的实现：

import org.redisson.Redisson;
import org.redisson.api.RLock;
import org.redisson.api.RedissonClient;
import org.redisson.config.Config;

public class RedissonLockExample {
    public static void main(String[] args) {
        // 配置Redisson
        Config config = new Config();
        config.useSingleServer().setAddress("redis://localhost:6379");
        
        // 创建Redisson客户端
        RedissonClient redisson = Redisson.create(config);
        
        // 获取锁实例
        RLock lock = redisson.getLock("myRedissonLock");
        
        try {
            // 尝试获取锁，最多等待100秒，锁持有时间为30秒
            boolean acquired = lock.tryLock(100, 30, java.util.concurrent.TimeUnit.SECONDS);
            
            if (acquired) {
                try {
                    // 执行临界区代码
                    System.out.println("获取Redisson锁成功，执行临界区操作");
                    Thread.sleep(5000);
                } finally {
                    // 释放锁
                    lock.unlock();
                    System.out.println("释放Redisson锁");
                }
            } else {
                System.out.println("获取Redisson锁失败");
            }
        } catch (InterruptedException e) {
            e.printStackTrace();
        } finally {
            // 关闭Redisson客户端
            redisson.shutdown();
        }
    }
}

四、Redis分布式锁的应用场景

1. 库存扣减

public class InventoryService {
    private Jedis jedis = new Jedis("localhost", 6379);
    
    public void deductInventory(String productId, int quantity) {
        RedisLock lock = new RedisLock(jedis, "inventory_lock:" + productId, 30000);
        
        try {
            if (lock.acquire()) {
                // 查询库存
                int stock = getInventoryFromDB(productId);
                
                if (stock >= quantity) {
                    // 扣减库存
                    boolean success = updateInventoryInDB(productId, stock - quantity);
                    if (success) {
                        System.out.println("库存扣减成功");
                    } else {
                        System.out.println("库存扣减失败");
                    }
                } else {
                    System.out.println("库存不足");
                }
            } else {
                System.out.println("获取锁失败，请稍后重试");
            }
        } finally {
            lock.release();
        }
    }
    
    private int getInventoryFromDB(String productId) {
        // 从数据库查询库存
        return 100; // 示例返回值
    }
    
    private boolean updateInventoryInDB(String productId, int newStock) {
        // 更新数据库中的库存
        return true; // 示例返回值
    }
}

2. 定时任务排重

public class ScheduledTask {
    private RedissonClient redisson;
    
    public ScheduledTask() {
        Config config = new Config();
        config.useSingleServer().setAddress("redis://localhost:6379");
        redisson = Redisson.create(config);
    }
    
    public void executeDailyReport() {
        RLock lock = redisson.getLock("daily_report_task_lock");
        
        try {
            // 尝试获取锁，5秒内获取不到则放弃
            boolean acquired = lock.tryLock(5, 30, TimeUnit.SECONDS);
            
            if (acquired) {
                try {
                    // 执行每日报告生成任务
                    generateDailyReport();
                } finally {
                    lock.unlock();
                }
            } else {
                System.out.println("另一个实例正在执行每日报告任务");
            }
        } catch (InterruptedException e) {
            Thread.currentThread().interrupt();
        }
    }
    
    private void generateDailyReport() {
        System.out.println("开始生成每日报告...");
        // 执行报告生成逻辑
    }
}

五、分布式锁的监控与告警

import java.util.concurrent.atomic.AtomicInteger;

public class MonitorableRedisLock extends RedisLock {
    private static final AtomicInteger totalAcquireCount = new AtomicInteger(0);
    private static final AtomicInteger totalReleaseCount = new AtomicInteger(0);
    private static final AtomicInteger currentLockedCount = new AtomicInteger(0);
    
    private long acquireTime;
    
    public MonitorableRedisLock(Jedis jedis, String lockKey, int expireTime) {
        super(jedis, lockKey, expireTime);
    }
    
    @Override
    public boolean acquire() {
        long startTime = System.currentTimeMillis();
        boolean result = super.acquire();
        if (result) {
            acquireTime = System.currentTimeMillis();
            totalAcquireCount.incrementAndGet();
            currentLockedCount.incrementAndGet();
            
            // 记录获取锁耗时
            long acquireTime = System.currentTimeMillis() - startTime;
            if (acquireTime > 1000) { // 超过1秒的获取操作
                System.err.println("警告: 获取锁耗时过长 - " + acquireTime + "ms, 锁: " + getLockKey());
            }
        }
        return result;
    }
    
    @Override
    public boolean release() {
        boolean result = super.release();
        if (result) {
            totalReleaseCount.incrementAndGet();
            currentLockedCount.decrementAndGet();
            
            // 记录锁持有时间
            long holdTime = System.currentTimeMillis() - acquireTime;
            if (holdTime > getExpireTime() * 0.8) { // 接近过期时间
                System.err.println("警告: 锁持有时间过长 - " + holdTime + "ms, 锁: " + getLockKey());
            }
        }
        return result;
    }
    
    // 监控指标获取方法
    public static int getTotalAcquireCount() {
        return totalAcquireCount.get();
    }
    
    public static int getTotalReleaseCount() {
        return totalReleaseCount.get();
    }
    
    public static int getCurrentLockedCount() {
        return currentLockedCount.get();
    }
}

六、Redis分布式锁的最佳实践

1. 设置合理的过期时间：根据业务执行时间设置锁的过期时间，避免死锁

2. 使用唯一标识：锁的value使用唯一标识，确保锁只能由持有者释放

3. 保证解锁原子性：使用Lua脚本保证解锁操作的原子性

4. 考虑锁续期：对于长时间运行的任务，考虑使用"看门狗"机制自动续期

5. 高可用部署：在生产环境中使用Redis集群或哨兵模式保证可用性

6. 监控与告警：监控锁的持有时间和竞争情况，及时发现问题

7. 选择合适的锁类型：根据业务需求选择普通锁、红锁、读写锁等不同类型

通过合理使用Redis实现分布式锁，可以有效解决分布式系统中的资源竞争问题，提高系统的可靠性和稳定性。