How Springboot implements Redis distributed reentrant lock source code analysis based on Redisson

1. Preface

We are using Redis to implement distributed locks. At first, we generally use SET resource-name anystring NX EX max-lock-timeTo lock, use Lua script to ensure atomicity to release the lock. This manual implementation is troublesome. The Redis official website also clearly states that the Java version uses Redisson to implement it. The editor also looked at the official website and slowly figured it out, and took a close-up to record it. From the official website to integrating Springboot to source code interpretation, take a single node as an example.

2. Why use Redisson

1. We open the official website

redis Chinese official website

2. We can see that the official allows us to use other

3. Open the official recommendation

4. Find the document

Redisson address

5. Redisson structure

3. Springboot integrates Redisson

1. Import dependencies

<dependency>
    <groupId>org.springframework.boot</groupId>
    <artifactId>spring-boot-starter-data-redis</artifactId>
</dependency>
<dependency>
    <groupId>redis.clients</groupId>
    <artifactId>jedis</artifactId>
</dependency>
<!--redis分布式锁-->
<dependency>
    <groupId>org.redisson</groupId>
    <artifactId>redisson</artifactId>
    <version>3.12.0</version>
</dependency>

2. Take the official website as an example to see how to configure

3. Write the configuration class

import org.redisson.Redisson;
import org.redisson.api.RedissonClient;
import org.redisson.config.Config;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;

/**
 * @author wangzhenjun
 * @date 2022/2/9 9:57
 */
@Configuration
public class MyRedissonConfig {

    /**
     * 所有对redisson的使用都是通过RedissonClient来操作的
     * @return
     */
    @Bean(destroyMethod="shutdown")
    public RedissonClient redisson(){
        // 1. 创建配置
        Config config = new Config();
        // 一定要加redis://
        config.useSingleServer().setAddress("redis://192.168.17.130:6379");
        // 2. 根据config创建出redissonClient实例
        RedissonClient redissonClient = Redisson.create(config);
        return redissonClient;
    }
}

4. Official website test locking example

5. Write according to the simple Controller interface on the official website

@ResponseBody
@GetMapping("/hello")
public String hello(){
    // 1.获取一把锁，只要锁名字一样，就是同一把锁
    RLock lock = redisson.getLock("my-lock");
    // 2. 加锁
    lock.lock();// 阻塞试等待  默认加的都是30s
    // 带参数情况
    // lock.lock(10, TimeUnit.SECONDS);// 10s自动解锁，自动解锁时间一定要大于业务的执行时间。
    try {
        System.out.println("加锁成功" + Thread.currentThread().getId());
        Thread.sleep(30000);
    } catch (InterruptedException e) {
        e.printStackTrace();
    } finally {
        // 3. 解锁
        System.out.println("解锁成功：" + Thread.currentThread().getId());
        lock.unlock();
    }
    return "hello";
}

6. Test

##4. lock.lock() Source code analysis

1. Open the RedissonLock implementation class

2. Find the implementation method

@Override
public void lock() {
    try {
    	// 我们发现不穿过期时间源码默认过期时间为-1
        lock(-1, null, false);
    } catch (InterruptedException e) {
        throw new IllegalStateException();
    }
}

3. Press Ctrl to enter the lock method

private void lock(long leaseTime, TimeUnit unit, boolean interruptibly) throws InterruptedException {
	// 获取线程的id，占有锁的时候field的值为UUID:线程号id
    long threadId = Thread.currentThread().getId();
    // 尝试获得锁
    Long ttl = tryAcquire(leaseTime, unit, threadId);
    // lock acquired 获得锁，返回
    if (ttl == null) {
        return;
    }
	// 这里说明获取锁失败，就通过线程id订阅这个锁
    RFuture<RedissonLockEntry> future = subscribe(threadId);
    if (interruptibly) {
        commandExecutor.syncSubscriptionInterrupted(future);
    } else {
        commandExecutor.syncSubscription(future);
    }

    try {
    	// 这里进行自旋，不断尝试获取锁
        while (true) {
        	// 继续尝试获取锁
            ttl = tryAcquire(leaseTime, unit, threadId);
            // lock acquired 获取成功
            if (ttl == null) {
            	// 直接返回，挑出自旋
                break;
            }

            // waiting for message 继续等待获得锁
            if (ttl >= 0) {
                try {
                    future.getNow().getLatch().tryAcquire(ttl, TimeUnit.MILLISECONDS);
                } catch (InterruptedException e) {
                    if (interruptibly) {
                        throw e;
                    }
                    future.getNow().getLatch().tryAcquire(ttl, TimeUnit.MILLISECONDS);
                }
            } else {
                if (interruptibly) {
                    future.getNow().getLatch().acquire();
                } else {
                    future.getNow().getLatch().acquireUninterruptibly();
                }
            }
        }
    } finally {
     	// 取消订阅
        unsubscribe(future, threadId);
    }
//        get(lockAsync(leaseTime, unit));
}

4. Go in and try to acquire the lock method

private Long tryAcquire(long leaseTime, TimeUnit unit, long threadId) {
	// 直接进入异步方法
    return get(tryAcquireAsync(leaseTime, unit, threadId));
}

private <T> RFuture<Long> tryAcquireAsync(long leaseTime, TimeUnit unit, long threadId) {
    // 这里进行判断如果没有设置参数leaseTime = -1
    if (leaseTime != -1) {
        return tryLockInnerAsync(leaseTime, unit, threadId, RedisCommands.EVAL_LONG);
    }
    // 此方法进行获得锁，过期时间为看门狗的默认时间
    // private long lockWatchdogTimeout = 30 * 1000;看门狗默认过期时间为30s
    // 加锁和过期时间要保证原子性，这个方法后面肯定调用执行了Lua脚本，我们下面在看
    RFuture<Long> ttlRemainingFuture = tryLockInnerAsync(commandExecutor.getConnectionManager().getCfg().getLockWatchdogTimeout(), TimeUnit.MILLISECONDS, threadId, RedisCommands.EVAL_LONG);
    // 开启一个定时任务进行不断刷新过期时间
    ttlRemainingFuture.onComplete((ttlRemaining, e) -> {
        if (e != null) {
            return;
        }
        // lock acquired 获得锁
        if (ttlRemaining == null) {
        	// 刷新过期时间方法，我们下一步详细说一下
            scheduleExpirationRenewal(threadId);
    });
    return ttlRemainingFuture;

5. Check out the tryLockInnerAsync() method

<T> RFuture<T> tryLockInnerAsync(long leaseTime, TimeUnit unit, long threadId, RedisStrictCommand<T> command) {
    internalLockLeaseTime = unit.toMillis(leaseTime);

    return commandExecutor.evalWriteAsync(getName(), LongCodec.INSTANCE, command,
    		  // 首先判断锁是否存在
              "if (redis.call(&#39;exists&#39;, KEYS[1]) == 0) then " +
              		// 存在则获取锁
                  "redis.call(&#39;hset&#39;, KEYS[1], ARGV[2], 1); " +
                  // 然后设置过期时间
                  "redis.call(&#39;pexpire&#39;, KEYS[1], ARGV[1]); " +
                  "return nil; " +
              "end; " +
              // hexists查看哈希表的指定字段是否存在，存在锁并且是当前线程持有锁
              "if (redis.call(&#39;hexists&#39;, KEYS[1], ARGV[2]) == 1) then " +
              		// hincrby自增一
                  "redis.call(&#39;hincrby&#39;, KEYS[1], ARGV[2], 1); " +
                  	// 锁的值大于1，说明是可重入锁，重置过期时间
                  "redis.call(&#39;pexpire&#39;, KEYS[1], ARGV[1]); " +
                  "return nil; " +
              "end; " +
              // 锁已存在，且不是本线程，则返回过期时间ttl
              "return redis.call(&#39;pttl&#39;, KEYS[1]);",
                Collections.<Object>singletonList(getName()), internalLockLeaseTime, getLockName(threadId));
}

6. Enter what is left in 4 ScheduleExpirationRenewal() method of scheduled tasks

Go to the source code step by step: scheduleExpirationRenewal --->renewExpiration

According to the source code below, the refresh time of scheduled tasks is: internalLockLeaseTime / 3, which is a watchdog 1/3, which is refreshed every 10 seconds

private void renewExpiration() {
    ExpirationEntry ee = EXPIRATION_RENEWAL_MAP.get(getEntryName());
    if (ee == null) {
        return;
    }
    
    Timeout task = commandExecutor.getConnectionManager().newTimeout(new TimerTask() {
        @Override
        public void run(Timeout timeout) throws Exception {
            ExpirationEntry ent = EXPIRATION_RENEWAL_MAP.get(getEntryName());
            if (ent == null) {
                return;
            }
            Long threadId = ent.getFirstThreadId();
            if (threadId == null) {
                return;
            }
            
            RFuture<Boolean> future = renewExpirationAsync(threadId);
            future.onComplete((res, e) -> {
                if (e != null) {
                    log.error("Can&#39;t update lock " + getName() + " expiration", e);
                    return;
                }
                
                if (res) {
                    // reschedule itself
                    renewExpiration();
                }
            });
        }
    }, internalLockLeaseTime / 3, TimeUnit.MILLISECONDS);
    
    ee.setTimeout(task);
}

5. Lock.lock(10, TimeUnit.SECONDS) source code analysis

1. Open the implementation class

@Override
public void lock(long leaseTime, TimeUnit unit) {
    try {
    	// 这里的过期时间为我们输入的10
        lock(leaseTime, unit, false);
    } catch (InterruptedException e) {
        throw new IllegalStateException();
    }
}

2 . Method

lock() implements display, same as 3.3 source code

3. Go directly to try to obtain the lock

tryAcquireAsync()method

private <T> RFuture<Long> tryAcquireAsync(long leaseTime, TimeUnit unit, long threadId) {
    // 这里进行判断如果没有设置参数leaseTime = -1,此时我们为10
    if (leaseTime != -1) {
    	// 来到此方法
        return tryLockInnerAsync(leaseTime, unit, threadId, RedisCommands.EVAL_LONG);
    }
    // 此处省略后面内容，前面以详细说明。。。。
}

4. Open the

tryLockInnerAsync()method

It is not difficult to find that it is the same as the method without passing the expiration time, except that the value of leaseTime has changed.

<T> RFuture<T> tryLockInnerAsync(long leaseTime, TimeUnit unit, long threadId, RedisStrictCommand<T> command) {
    internalLockLeaseTime = unit.toMillis(leaseTime);

    return commandExecutor.evalWriteAsync(getName(), LongCodec.INSTANCE, command,
    		  // 首先判断锁是否存在
              "if (redis.call(&#39;exists&#39;, KEYS[1]) == 0) then " +
              		// 存在则获取锁
                  "redis.call(&#39;hset&#39;, KEYS[1], ARGV[2], 1); " +
                  // 然后设置过期时间
                  "redis.call(&#39;pexpire&#39;, KEYS[1], ARGV[1]); " +
                  "return nil; " +
              "end; " +
              // hexists查看哈希表的指定字段是否存在，存在锁并且是当前线程持有锁
              "if (redis.call(&#39;hexists&#39;, KEYS[1], ARGV[2]) == 1) then " +
              		// hincrby自增一
                  "redis.call(&#39;hincrby&#39;, KEYS[1], ARGV[2], 1); " +
                  	// 锁的值大于1，说明是可重入锁，重置过期时间
                  "redis.call(&#39;pexpire&#39;, KEYS[1], ARGV[1]); " +
                  "return nil; " +
              "end; " +
              // 锁已存在，且不是本线程，则返回过期时间ttl
              "return redis.call(&#39;pttl&#39;, KEYS[1]);",
                Collections.<Object>singletonList(getName()), internalLockLeaseTime, getLockName(threadId));
}

6. Lock.unlock() source code analysis

1. Open method implementation

@Override
public void unlock() {
    try {
    	// 点击进入释放锁方法
        get(unlockAsync(Thread.currentThread().getId()));
    } catch (RedisException e) {
        if (e.getCause() instanceof IllegalMonitorStateException) {
            throw (IllegalMonitorStateException) e.getCause();
        } else {
            throw e;
        }
    }
    
//        Future<Void> future = unlockAsync();
//        future.awaitUninterruptibly();
//        if (future.isSuccess()) {
//            return;
//        }
//        if (future.cause() instanceof IllegalMonitorStateException) {
//            throw (IllegalMonitorStateException)future.cause();
//        }
//        throw commandExecutor.convertException(future);
}

2. Open

unlockAsync()method

@Override
public RFuture<Void> unlockAsync(long threadId) {
    RPromise<Void> result = new RedissonPromise<Void>();
    // 解锁方法，后面展开说
    RFuture<Boolean> future = unlockInnerAsync(threadId);
	// 完成
    future.onComplete((opStatus, e) -> {
        if (e != null) {
        	// 取消到期续订
            cancelExpirationRenewal(threadId);
            // 将这个未来标记为失败并通知所有人
            result.tryFailure(e);
            return;
        }
		// 状态为空，说明解锁的线程和当前锁不是同一个线程
        if (opStatus == null) {
            IllegalMonitorStateException cause = new IllegalMonitorStateException("attempt to unlock lock, not locked by current thread by node id: "
                    + id + " thread-id: " + threadId);
            result.tryFailure(cause);
            return;
        }
        
        cancelExpirationRenewal(threadId);
        result.trySuccess(null);
    });

    return result;
}

3. Open

unlockInnerAsync()method

protected RFuture<Boolean> unlockInnerAsync(long threadId) {
    return commandExecutor.evalWriteAsync(getName(), LongCodec.INSTANCE, RedisCommands.EVAL_BOOLEAN,
    		// 判断释放锁的线程和已存在锁的线程是不是同一个线程，不是返回空
            "if (redis.call(&#39;hexists&#39;, KEYS[1], ARGV[3]) == 0) then " +
                "return nil;" +
            "end; " +
            // 释放锁后，加锁次数减一
            "local counter = redis.call(&#39;hincrby&#39;, KEYS[1], ARGV[3], -1); " +
            // 判断剩余数量是否大于0
            "if (counter > 0) then " +
            	// 大于0 ，则刷新过期时间
                "redis.call(&#39;pexpire&#39;, KEYS[1], ARGV[2]); " +
                "return 0; " +
            "else " +
            	// 释放锁，删除key并发布锁释放的消息
                "redis.call(&#39;del&#39;, KEYS[1]); " +
                "redis.call(&#39;publish&#39;, KEYS[2], ARGV[1]); " +
                "return 1; "+
            "end; " +
            "return nil;",
            Arrays.<Object>asList(getName(), getChannelName()), LockPubSub.UNLOCK_MESSAGE, internalLockLeaseTime, getLockName(threadId));

}

The above is the detailed content of How Springboot implements Redis distributed reentrant lock source code analysis based on Redisson. For more information, please follow other related articles on the PHP Chinese website!