前置
Java-Redis-Redisson配置基础上我们进行了改造,让锁的使用更加方便
基础设施
RedissonLock
import java.lang.annotation.ElementType;
import java.lang.annotation.Retention;
import java.lang.annotation.RetentionPolicy;
import java.lang.annotation.Target;
@Target(ElementType.METHOD)
@Retention(RetentionPolicy.RUNTIME)
public @interface RedissonLock {
int lockTime() default 3; //加锁的时间默认3秒, 如果任务在3秒内执行完毕那么自动释放锁,如果任务3秒内没有执行完毕也会释放锁, 所以内容执行时间过长适当加大锁的时间
String key() default "" ; //唯一标识,如果没有那么默认为token->sessionId
String doc() default "重复提交请求,请稍后再试";
boolean repeatLock() default false; //可重复加锁直到加锁成功,默认为false不能重复加锁
int repeatLockCount() default -1; //可重复加锁限制加锁的次数, 默认-1直到成功,设置10那么加锁10次都没成功就直接返回
int lockWaitTimeMs() default 100; //重复加锁默认的阻塞时间100毫秒,可以自己定义
}RepeatSubmitAspect
import com.application.Result;
import com.commonutils.NullUtils;
import com.redis.utils.DistributedRedisLock;
import org.aspectj.lang.ProceedingJoinPoint;
import org.aspectj.lang.annotation.Around;
import org.aspectj.lang.annotation.Aspect;
import org.aspectj.lang.annotation.Pointcut;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.beans.factory.annotation.Value;
import org.springframework.stereotype.Component;
import org.springframework.util.Assert;
import org.springframework.web.context.request.RequestContextHolder;
import org.springframework.web.context.request.ServletRequestAttributes;
import javax.servlet.http.HttpServletRequest;
@Aspect
@Component
public class RepeatSubmitAspect {
@Value("${spring.redis.redisson.tokenName}")
private String tokenName;
@Autowired
private DistributedRedisLock redisLock;
@Pointcut("@annotation(noRepeatSubmit)")
public void pointCut(RedissonLock noRepeatSubmit) {
}
public static HttpServletRequest getRequest() {
ServletRequestAttributes ra= (ServletRequestAttributes) RequestContextHolder.getRequestAttributes();
return ra.getRequest();
}
@Around("pointCut(noRepeatSubmit)")
public Object around(ProceedingJoinPoint pjp, RedissonLock noRepeatSubmit) throws Throwable {
int lockSeconds = noRepeatSubmit.lockTime();
String doc = noRepeatSubmit.doc();
String keyName = noRepeatSubmit.key();
boolean b = noRepeatSubmit.repeatLock();
int repeatLockCount = noRepeatSubmit.repeatLockCount();
int lockWaitTimeMs = noRepeatSubmit.lockWaitTimeMs();
HttpServletRequest request = getRequest();
Assert.notNull(request, "request can not null");
//如果没有唯一表示那么就使用token或者sessionID来唯一表示
if(!NullUtils.notEmpty(keyName)){
String token = request.getHeader(tokenName);
if(NullUtils.notEmpty(token)){
keyName=token;
}else{
//使用sessionID (注意保证分布式session共享)
keyName = request.getSession().getId();
}
System.out.println("tokenName:"+keyName);
}
String path = request.getServletPath();
String key = getKey(keyName, path);
//加锁
boolean isSuccess = redisLock.acquire(key, lockSeconds,b,repeatLockCount,lockWaitTimeMs);
if (isSuccess) {
// 获取锁成功
Object result;
try {
// 执行
result = pjp.proceed();
} finally {
// 解锁
redisLock.release(key);
}
return result;
} else {
// 获取锁失败，认为是重复提交的请求
return Result.Error(doc);
}
}
private String getKey(String token, String path) {
return token + path;
}
}DistributedRedisLock
import com.multithreading.utils.SleepTools;
import lombok.SneakyThrows;
import org.redisson.api.RLock;
import org.redisson.api.RedissonClient;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.stereotype.Component;
import java.util.concurrent.TimeUnit;
@Component
public class DistributedRedisLock {
private static final Logger logger = LoggerFactory.getLogger(DistributedRedisLock.class);
//从配置类中获取redisson对象
@Autowired
private RedissonClient redissonClient;
private final String LOCK_TITLE = "redisLock_";
private final ThreadLocal<Integer> count = new ThreadLocal<>();//计数
//加锁 , 线程加锁的时候发现,锁有人用了 ,那么就会进入自旋等待
@SneakyThrows
public boolean acquire(String lockKey, long seconds, boolean repeatLock, int repeatLockCount, int lockWaitTimeMs) {
count.set(0); //初始化值
//获取锁对象
RLock mylock = redissonClient.getLock((LOCK_TITLE + lockKey));
do {
//尝试加锁
boolean b = mylock.tryLock(0, seconds, TimeUnit.SECONDS);;
if (b) {
logger.info("获取锁成功");
return true;
}
logger.info("尝试获取锁" + Thread.currentThread().getName());
//获取加锁的次数,如果是-1那么持续加锁,如果满足加锁次数那么结束加锁
if (repeatLockCount != -1 && count.get().equals(repeatLockCount)) {
logger.warn(Thread.currentThread().getName() + "尝试加锁失败以尝试了:" + count.get() + "次");
return false;
}
SleepTools.ms(lockWaitTimeMs);
//加锁次数增加
count.set(count.get() + 1);
} while (repeatLock);
logger.warn("重复提交");
return false;
}
//手动锁的释放
public void release(String lockKey) {
//获取所对象
RLock mylock = redissonClient.getLock((LOCK_TITLE + lockKey));
// 这里判断下当前key是否上锁，不然业务执行时间大于锁自动释放时间后，解锁报异常
if (mylock.isLocked()&&mylock.isHeldByCurrentThread()) { // 是否还是锁定状态并且锁是当前线程的
mylock.unlock(); // 释放锁
logger.info("解锁:" + lockKey);
}
}
}
功能使用和介绍
- 支持配置锁的时间
- 支持配置锁的key(同一个key的请求会被锁住)
- 支持重复加锁
- 支持重复加锁的次数
- 支持重复加锁的间隔时间
通过以上功能的的组合能做到,幂等性,分布式悲观锁, 超时丢弃 ,不止在Controller里使用,而是在任何基于Spring容器管理的Bean都支持,当然如果特殊的场景我们可以直接使用DistributedRedisLock类,注意在任何时候都要释放锁
@RedissonLock
@GetMapping(value = "/updateAgeAsyncLock" )
public Result updateAgeAsyncLock() {
userService.updateAgeAsyncLock();
return Result.Ok();
}
@RedissonLock(doc = "加锁失败")
@GetMapping(value = "/updateAgeAsyncLockDoc" )
public Result updateAgeAsyncLockDoc() {
userService.updateAgeAsyncLock();
return Result.Ok();
}
@RedissonLock(lockTime = 15)
@GetMapping(value = "/updateAgeAsyncLock0" )
public Result updateAgeAsyncLock0() {
userService.updateAgeAsyncLock();
return Result.Ok();
}
@RedissonLock(lockTime = 15,key = "updateAgeAsyncLock")
@GetMapping(value = "/updateAgeAsyncLock1" )
public Result updateAgeAsyncLock1() {
userService.updateAgeAsyncLock();
return Result.Ok();
}
@RedissonLock(lockTime = 15,key = "updateAgeAsyncLock",repeatLock = true)
@GetMapping(value = "/updateAgeAsyncLock2" )
public Result updateAgeAsyncLock2() {
userService.updateAgeAsyncLock();
return Result.Ok();
}
@RedissonLock(lockTime = 15,key = "updateAgeAsyncLock",repeatLock = true,repeatLockCount = 10)
@GetMapping(value = "/updateAgeAsyncLock3" )
public Result updateAgeAsyncLock3() {
userService.updateAgeAsyncLock();
return Result.Ok();
}
@RedissonLock(lockTime = 15,key = "updateAgeAsyncLock",repeatLock = true,repeatLockCount = 10,lockWaitTimeMs = 500)
@GetMapping(value = "/updateAgeAsyncLock4" )
public Result updateAgeAsyncLock4() {
userService.updateAgeAsyncLock();
return Result.Ok();
}
@RedissonLock(lockTime = 15,key = "updateAgeAsyncLock",repeatLock = true,lockWaitTimeMs = 500)
@GetMapping(value = "/updateAgeAsyncLock4" )
public Result updateAgeAsyncLock4() {
userService.updateAgeAsyncLock();
return Result.Ok();
}
验证方式: 使用jmeter多线程10000请求 ,注意使用jmeter的时候不要使用默认锁key的方式,因为jmeter每次请sessionID都不同的,想要验证效果我们需要手动加key或者使用token
其他悲观锁的实现方式
public class DistributedRedisLock {
//从配置类中获取redisson对象
@Autowired
private RedissonClient redissonClient;
private final String LOCK_TITLE = "redisLock_";
//加锁 , 线程加锁的时候发现,锁有人用了 ,那么就会进入自旋等待
public boolean acquire(String lockName){
//声明key对象
String key = LOCK_TITLE + lockName;
//获取锁对象
RLock mylock = redissonClient.getLock(key);
//一直等待直到加锁成功后，并且设置锁过期时间，防止死锁的产生
mylock.lock(10, TimeUnit.SECONDS);
System.err.println("======lock======"+Thread.currentThread().getName());
//加锁成功
return true;
}
//锁的释放
public void release(String lockName){
//必须是和加锁时的同一个key
String key = LOCK_TITLE + lockName;
//获取所对象
RLock mylock = redissonClient.getLock(key);
// 这里判断下当前key是否上锁，不然业务执行时间大于锁自动释放时间后，解锁报异常
if(mylock.isLocked()){ // 是否还是锁定状态
if(mylock.isHeldByCurrentThread()){ // 时候是当前执行线程的锁
mylock.unlock(); // 释放锁
System.err.println("======unlock======"+Thread.currentThread().getName());
}
}
}
}
到此这篇关于Java-Redis-Redisson-分布式锁的文章就介绍到这了,更多相关Java-Redis-Redisson分布式锁内容请搜索编程网以前的文章或继续浏览下面的相关文章希望大家以后多多支持编程网！
