分布式锁是控制分布式系统之间同步访问共享资源的一种方式

锁接口定义

定义一个锁通用接口，对外提供锁服务

import java.util.concurrent.TimeUnit;public interface LockService {    /**      * 尝试获取锁     * @param 锁名     */    public boolean tryLock(String name);    /**      * 在指定时间内尝试获取锁     * @param 锁名     * @param 尝试时间     */    public boolean tryLock(String name, long timeout, TimeUnit unit);    /**      * 获取锁，阻塞方法，直到获取锁成功为止     * @param 锁名     */    public void lock(String name) throws Exception;    /**      * 释放锁     * @param 锁名     */    public void unlock(String name);}

redis锁的实现：

public class RedisLockService implements LockService {    private static final String NAMESPACE = "lock-service:";    private static final Logger log = LogManager.getLogger(RedisLockService.class);    @Autowired    private RedisTemplate
    
      stringRedisTemplate;    /** 锁默认超时时间，单位：秒 */    @Value("${lockService.timeout:300}")    private int timeout;          /** 服务ID，用于区分释放锁权限。 */    @Value("${instance.id:locks}")    private String instanceId;    @Override    public boolean tryLock(String name) {        String key = NAMESPACE + name;        String value = getValue();        boolean success = stringRedisTemplate.opsForValue().setIfAbsent(key, value);    //如果不存在key，则set        if (success)            stringRedisTemplate.expire(key, this.timeout, TimeUnit.SECONDS);            //设置默认锁有效时间        return success;    }    @Override    public boolean tryLock(String name, long timeout, TimeUnit unit) {        if (timeout <= 0)            return tryLock(name);        String key = NAMESPACE + name;        String value = getValue();        boolean success = stringRedisTemplate.opsForValue().setIfAbsent(key, value);        long millisTimeout = unit.toMillis(timeout);        long start = System.currentTimeMillis();        while (!success) {          //若获取锁不成功，则在限定时间内不断尝试，直到获取成功，或超时            try {                Thread.sleep(100);            } catch (InterruptedException e) {                return false;            }            if ((System.currentTimeMillis() - start) >= millisTimeout)                break;            success = stringRedisTemplate.opsForValue().setIfAbsent(key, value);        }        if (success)            stringRedisTemplate.expire(key, this.timeout, TimeUnit.SECONDS);        return success;    }    @Override    public void lock(String name) {        String key = NAMESPACE + name;        String value = getValue();        boolean success = stringRedisTemplate.opsForValue().setIfAbsent(key, value);        while (!success) {            try {                Thread.sleep(100);          //每间隔100毫秒，再次尝试获取锁            } catch (InterruptedException e) {                e.printStackTrace();            }            success = stringRedisTemplate.opsForValue().setIfAbsent(key, value);            if (success)                stringRedisTemplate.expire(key, this.timeout, TimeUnit.SECONDS);        }    }    @Override    public void unlock(String name) {        String key = NAMESPACE + name;        String value = getValue();        //一行lua脚本，如果KEY的值等于VALUE，则删除KEY，否则返回0        String str = "if redis.call(\"get\",KEYS[1]) == ARGV[1] then return redis.call(\"del\",KEYS[1]) else return 0 end";        RedisScript
     
       script = new DefaultRedisScript
      
       (str, Long.class);        Long ret = stringRedisTemplate.execute(script, Collections.singletonList(key), value);        if (ret == 0)            log.warn("Lock [{}] is not hold by instance [{}]", name, value);    }    /**     * 生成由服务ID+线程名组成的线程唯一value。用于保证只能被拥有锁的线程解锁     */    private String getValue(){        return instanceId + Thread.currentThread().getName();    }}