Redisson概述
Redisson是一个在Redis的基础上实现的Java驻内存数据网格(In-Memory Data Grid)。它不仅提供了一系列的分布式的Java常用对象,还提供了许多分布式服务。其中包括(BitSet, Set, Multimap, SortedSet, Map, List, Queue, BlockingQueue, Deque, BlockingDeque, Semaphore, Lock, AtomicLong, CountDownLatch, Publish / Subscribe, Bloom filter, Remote service, Spring cache, Executor service, Live Object service, Scheduler service) Redisson提供了使用Redis的最简单和最便捷的方法。Redisson的宗旨是促进使用者对Redis的关注分离(Separation of Concern),从而让使用者能够将精力更集中地放在处理业务逻辑上。
Redisson底层采用的是Netty 框架。支持Redis 2.8以上版本,支持Java1.6+以上版本。
关于Redisson更多详细介绍,可参考Redssion概述
Redisson提供的分布式锁
可重入锁
Redisson的分布式可重入锁RLock Java对象实现了java.util.concurrent.locks.Lock接口,同时还支持自动过期解锁。下面是RLock的基本使用方法:
RLock lock = redisson.getLock("anyLock"); // 最常见的使用方法 lock.lock(); // 支持过期解锁功能 // 10秒钟以后自动解锁 // 无需调用unlock方法手动解锁 lock.lock(10, TimeUnit.SECONDS); // 尝试加锁,最多等待100秒,上锁以后10秒自动解锁 boolean res = lock.tryLock(100, 10, TimeUnit.SECONDS); ... lock.unlock();
Redisson同时还为分布式锁提供了异步执行的相关方法:
RLock lock = redisson.getLock("anyLock"); lock.lockAsync(); lock.lockAsync(10, TimeUnit.SECONDS); Future<Boolean> res = lock.tryLockAsync(100, 10, TimeUnit.SECONDS);
公平锁
Redisson分布式可重入公平锁也是实现了java.util.concurrent.locks.Lock接口的一种RLock对象。在提供了自动过期解锁功能的同时,保证了当多个Redisson客户端线程同时请求加锁时,优先分配给先发出请求的线程。
RLock fairLock = redisson.getFairLock("anyLock"); // 最常见的使用方法 fairLock.lock(); // 支持过期解锁功能 // 10秒钟以后自动解锁 // 无需调用unlock方法手动解锁 fairLock.lock(10, TimeUnit.SECONDS); // 尝试加锁,最多等待100秒,上锁以后10秒自动解锁 boolean res = fairLock.tryLock(100, 10, TimeUnit.SECONDS); ... fairLock.unlock();
其他锁
Redisson还提供了其他机制的锁,如联锁(MultiLock)、红锁(RedLock)等。详细可参考:分布式锁和同步器
使用Redisson实现分布式锁
- 定义回调接口
/** * 分布式锁回调接口 */ public interface DistributedLockCallback<T> { /** * 调用者必须在此方法中实现需要加分布式锁的业务逻辑 * * @return */ public T process(); /** * 得到分布式锁名称 * * @return */ public String getLockName(); }
- 定义分布式锁模板
/** * 分布式锁操作模板 */ public interface DistributedLockTemplate { long DEFAULT_WAIT_TIME = 30; long DEFAULT_TIMEOUT = 5; TimeUnit DEFAULT_TIME_UNIT = TimeUnit.SECONDS; /** * 使用分布式锁,使用锁默认超时时间。 * @param callback * @param fairLock 是否使用公平锁 * @return */ <T> T lock(DistributedLockCallback<T> callback, boolean fairLock); /** * 使用分布式锁。自定义锁的超时时间 * * @param callback * @param leaseTime 锁超时时间。超时后自动释放锁。 * @param timeUnit * @param fairLock 是否使用公平锁 * @param <T> * @return */ <T> T lock(DistributedLockCallback<T> callback, long leaseTime, TimeUnit timeUnit, boolean fairLock); /** * 尝试分布式锁,使用锁默认等待时间、超时时间。 * @param callback * @param fairLock 是否使用公平锁 * @param <T> * @return */ <T> T tryLock(DistributedLockCallback<T> callback, boolean fairLock); /** * 尝试分布式锁,自定义等待时间、超时时间。 * @param callback * @param waitTime 获取锁最长等待时间 * @param leaseTime 锁超时时间。超时后自动释放锁。 * @param timeUnit * @param fairLock 是否使用公平锁 * @param <T> * @return */ <T> T tryLock(DistributedLockCallback<T> callback, long waitTime, long leaseTime, TimeUnit timeUnit, boolean fairLock); }
- 实现分布式锁模板
public class SingleDistributedLockTemplate implements DistributedLockTemplate { private RedissonClient redisson; public SingleDistributedLockTemplate() { } public SingleDistributedLockTemplate(RedissonClient redisson) { this.redisson = redisson; } @Override public <T> T lock(DistributedLockCallback<T> callback, boolean fairLock) { return lock(callback, DEFAULT_TIMEOUT, DEFAULT_TIME_UNIT, fairLock); } @Override public <T> T lock(DistributedLockCallback<T> callback, long leaseTime, TimeUnit timeUnit, boolean fairLock) { RLock lock = getLock(redisson, callback.getLockName(), fairLock); try { lock.lock(leaseTime, timeUnit); return callback.process(); } finally { if (lock != null) { lock.unlock(); } } } @Override public <T> T tryLock(DistributedLockCallback<T> callback, boolean fairLock) { return tryLock(callback, DEFAULT_WAIT_TIME, DEFAULT_TIMEOUT, DEFAULT_TIME_UNIT, fairLock); } @Override public <T> T tryLock(DistributedLockCallback<T> callback, long waitTime, long leaseTime, TimeUnit timeUnit, boolean fairLock) { RLock lock = getLock(redisson, callback.getLockName(), fairLock); try { if (lock.tryLock(waitTime, leaseTime, timeUnit)) { return callback.process(); } } catch (InterruptedException e) { } finally { if (lock != null) { lock.unlock(); } } return null; } private RLock getLock(RedissonClient redisson, String lockName, boolean fairLock) { RLock lock; if (fairLock) { lock = redisson.getFairLock(lockName); } else { lock = redisson.getLock(lockName); } return lock; } public void setRedisson(RedissonClient redisson) { this.redisson = redisson; } }
- 使用SingleDistributedLockTemplate
DistributedLockTemplate lockTemplate = ...; final String lockName = ...; lockTemplate.lock(new DistributedLockCallback<Object>() { @Override public Object process() { //do some business return null; } @Override public String getLockName() { return lockName; } }, false);
但是每次使用分布式锁都要写类似上面的重复代码,有没有什么方法可以只关注核心业务逻辑代码的编写,即上面的"do some business"。下面介绍如何使用Spring AOP来实现这一目标。
使用Spring AOP简化分布式锁
- 定义注解@DistributedLock
@Target({ElementType.METHOD}) @Retention(RetentionPolicy.RUNTIME) @Documented public @interface DistributedLock { String lockName() default ""; String lockNamePre() default ""; //lockName后缀 String lockNamePost() default ""; //lockName后缀 String param() default ""; //获取注解的方法第一个参数对象的某个属性值来作为lockName。因为有时候lockName是不固定的。 boolean fairLock() default false; //是否使用公平锁。 boolean tryLock() default false; //是否使用尝试锁。 long waitTime() default 30L; long leaseTime() default 5L; TimeUnit timeUnit() default TimeUnit.SECONDS; }
- 定义切面织入的代码
@Aspect @Component public class DistributedLockAspect { @Autowired private DistributedLockTemplate lockTemplate; @Pointcut("@annotation(cn.sprinkle.study.distributedlock.common.annotation.DistributedLock)") public void DistributedLockAspect() {} @Around(value = "DistributedLockAspect()") public Object doAround(ProceedingJoinPoint pjp) throws Throwable { //切点所在的类名 String targetName = pjp.getTarget().getClass().getName(); //使用了注解的方法 String methodName = pjp.getSignature().getName(); Class targetClass = Class.forName(targetName); Method[] methods = targetClass.getMethods(); Object[] arguments = pjp.getArgs(); //得到使用注解的方法。可使用Method.getAnnotation(Class<T> annotationClass)获取指定的注解,然后可获得注解的属性 Optional<Method> optional = Arrays.stream(methods) .parallel() .filter(method -> method.getName().equals(methodName)) .findAny(); if (optional.isPresent()) { Method m = optional.get(); final String lockName = getLockName(m, arguments); return lock(pjp, m, lockName); } return null; } public String getLockName(Method method, Object[] args) throws Throwable { DistributedLock annotation = method.getAnnotation(DistributedLock.class); String lockName = annotation.lockName(), param = annotation.param(); if (StringUtils.isEmpty(lockName)) { if (!StringUtils.isEmpty(param)) { if (args.length > 0) { Object arg = args[0]; lockName = String.valueOf(getParam(arg, param)); String preLockName = annotation.lockNamePre(), postLockName = annotation.lockNamePost(); lockName = preLockName + lockName + postLockName; return lockName; } } } else { return lockName; } throw new IllegalArgumentException("lockName can‘t be empty!"); } /** * 从方法参数获取数据 * * @param param * @param arg 方法的参数数组 * @return */ public Object getParam(Object arg, String param) throws Throwable { if (!StringUtils.isEmpty(param) && arg != null) { Object result = PropertyUtils.getProperty(arg, param); return result; } return null; } public Object lock(ProceedingJoinPoint pjp, Method method, final String lockName) { DistributedLock annotation = method.getAnnotation(DistributedLock.class); boolean fairLock = annotation.fairLock(); boolean tryLock = annotation.tryLock(); if (tryLock) { return tryLock(pjp, annotation, lockName, fairLock); } else { return lock(pjp,lockName, fairLock); } } public Object lock(ProceedingJoinPoint pjp, final String lockName, boolean fairLock) { return lockTemplate.lock(new DistributedLockCallback<Object>() { @Override public Object process() { return proceed(pjp); } @Override public String getLockName() { return lockName; } }, fairLock); } public Object tryLock(ProceedingJoinPoint pjp, DistributedLock annotation, final String lockName, boolean fairLock) { long waitTime = annotation.waitTime(), leaseTime = annotation.leaseTime(); TimeUnit timeUnit = annotation.timeUnit(); return lockTemplate.tryLock(new DistributedLockCallback<Object>() { @Override public Object process() { return proceed(pjp); } @Override public String getLockName() { return lockName; } }, waitTime, leaseTime, timeUnit, fairLock); } public Object proceed(ProceedingJoinPoint pjp) { try { return pjp.proceed(); } catch (Throwable throwable) { throwable.printStackTrace(); } return null; } }
- 使用注解@DistributedLock实现分布式锁
有了上面两段代码,以后需要用到分布式锁,只需在核心业务逻辑方法添加注解@DistributedLock,并设置LockName、fairLock等即可。
@Service public class DistributionService { @Autowired private RedissonClient redissonClient; @DistributedLock(lockName = "lock", lockNamePost = ".lock") public Integer aspect() { RMap<String, Integer> map = redissonClient.getMap("distributionTest"); Integer count = map.get("count"); if (count > 0) { count = count - 1; map.put("count", count); } return count; } @DistributedLock(param = "id", lockNamePost = ".lock") public Integer aspect(Person person) { RMap<String, Integer> map = redissonClient.getMap("distributionTest"); Integer count = map.get("count"); if (count > 0) { count = count - 1; map.put("count", count); } return count; } }
- 测试
定义一个Worker类:
public class Worker implements Runnable { private final CountDownLatch startSignal; private final CountDownLatch doneSignal; private final DistributionService service; public Worker(CountDownLatch startSignal, CountDownLatch doneSignal, DistributionService service) { this.startSignal = startSignal; this.doneSignal = doneSignal; this.service = service; } @Override public void run() { try { startSignal.await(); System.out.println(Thread.currentThread().getName() + " start"); //Integer count = service.aspect(); Integer count = service.aspect(new Person(1, "张三")); System.out.println(Thread.currentThread().getName() + ": count = " + count); doneSignal.countDown(); } catch (InterruptedException ex) { System.out.println(ex); } } }
定义Controller类:
@RestController @RequestMapping("/distributedLockTest") public class DistributedLockTestController { private int count = 10; @Autowired private RedissonClient redissonClient; @Autowired private DistributionService service; @RequestMapping(method = RequestMethod.GET) public String distributedLockTest() throws Exception { RMap<String, Integer> map = redissonClient.getMap("distributionTest"); map.put("count", 8); CountDownLatch startSignal = new CountDownLatch(1); CountDownLatch doneSignal = new CountDownLatch(count); for (int i = 0; i < count; ++i) { // create and start threads new Thread(new Worker(startSignal, doneSignal, service)).start(); } startSignal.countDown(); // let all threads proceed doneSignal.await(); System.out.println("All processors done. Shutdown connection"); return "finish"; } }
Redisson基本配置:
singleServerConfig: idleConnectionTimeout: 10000 pingTimeout: 1000 connectTimeout: 10000 timeout: 3000 retryAttempts: 3 retryInterval: 1500 reconnectionTimeout: 3000 failedAttempts: 3 password: subscriptionsPerConnection: 5 clientName: null address: "redis://127.0.0.1:6379" subscriptionConnectionMinimumIdleSize: 1 subscriptionConnectionPoolSize: 50 connectionMinimumIdleSize: 10 connectionPoolSize: 64 database: 0 dnsMonitoring: false dnsMonitoringInterval: 5000 threads: 0 nettyThreads: 0 codec: !<org.redisson.codec.JsonJacksonCodec> {} useLinuxNativeEpoll: false
工程中需要注入的对象:
@Value("classpath:/redisson-conf.yml") Resource configFile; @Bean(destroyMethod = "shutdown") RedissonClient redisson() throws IOException { Config config = Config.fromYAML(configFile.getInputStream()); return Redisson.create(config); } @Bean DistributedLockTemplate distributedLockTemplate(RedissonClient redissonClient) { return new SingleDistributedLockTemplate(redissonClient); }
需要引入的依赖:
<dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-aop</artifactId> </dependency> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-web</artifactId> </dependency> <dependency> <groupId>org.redisson</groupId> <artifactId>redisson</artifactId> <version>3.5.3</version> </dependency> <dependency> <groupId>commons-beanutils</groupId> <artifactId>commons-beanutils</artifactId> <version>1.8.3</version> </dependency>
最后启动工程,然后访问localhost:8080/distributedLockTest,可以看到如下结果:
观察结果,可以看出,10个线程中只有8个线程能执行count减1操作,而且多个线程是依次执行的。也就是说分布式锁起作用了。
至此,使用Redisson实现分布式锁,然后使用Spring AOP简化分布式锁介绍完毕。
若有什么地方有错误的或需要改进的,欢迎留言一起讨论交流。