采用redis token,分布式锁的接口幂等性实现

2019独角兽企业重金招聘Python工程师标准>>>

每一次进行幂等校验之前先获取token，因为token的时效性只有1次，我们每次获得的token在幂等操作后就无效了，所以一个token不需要长期保存在redis中。

@RestController
public class TokenController {
    @Autowired
    private RedisService redisService;
    @GetMapping("/users-anon/gettoken")
    public Map getToken(@RequestParam("url") String url) {
        Map<String,String> tokenMap = new HashMap();
        String tokenValue = UUID.randomUUID().toString();
        tokenMap.put(url + tokenValue, tokenValue);
        redisService.set(url + tokenValue, tokenValue);
        return tokenMap;
    }
}

获取token后，访问该url的接口，此时我们使用拦截器进行拦截(/add/**可表示为所有有新增操作的接口)

@SpringBootConfiguration
public class TokenInterceptorConfig extends WebMvcConfigurerAdapter {
    @Autowired
    private TokenInterceptor tokenInterceptor;

    @Override
    public void addInterceptors(InterceptorRegistry registry) {
        registry.addInterceptor(tokenInterceptor).addPathPatterns("/add/**");
    }
}

拦截器的具体内容为

@Slf4j
@Component
public class TokenInterceptor implements HandlerInterceptor {
    @Autowired
    private RedisService redisService;
    @Override
    public boolean preHandle(HttpServletRequest request, HttpServletResponse response, Object handler) throws Exception {
        String tokenName = request.getRequestURI() + request.getParameter("token_value");
        String tokenValue = request.getParameter("token_value");
        if (tokenValue != null && !tokenValue.equals("")) {
            log.info("tokenName:{},tokenValue:{}",tokenName,tokenValue);
            return handleToken(request,response,handler);
        }
        return false;
    }

    @Override
    public void postHandle(HttpServletRequest request, HttpServletResponse response, Object handler, @Nullable ModelAndView modelAndView) throws Exception {
        if (redisService.exists(request.getParameter("token_value"))) {
            RedisTool.releaseDistributedLock(redisService, request.getParameter("token_value"), request.getParameter("token_value"));
        }
    }

    @Override
    public void afterCompletion(HttpServletRequest request, HttpServletResponse response, Object handler, @Nullable Exception ex) throws Exception {

    }

    /**
     * 分布式锁处理
     * @param request
     * @param response
     * @param handler
     * @return
     * @throws Exception
     */
    private boolean handleToken(HttpServletRequest request, HttpServletResponse response, Object handler) throws Exception {
        //当大量高并发下所有带token参数的请求进来时，进行分布式锁定,允许某一台服务器的一个线程进入，锁定时间3分钟
        if (RedisTool.tryGetDistributedLock(redisService,request.getParameter("token_value"),request.getParameter("token_value"),180)) {
            if (redisService.exists(request.getRequestURI() + request.getParameter("token_value"))) {
                //当请求的url与token与redis中的存储相同时
                if (redisService.get(request.getRequestURI() + request.getParameter("token_value")).equals(request.getParameter("token_value"))) {
                    //放行的该线程删除redis中存储的token
                    redisService.del(request.getRequestURI() + request.getParameter("token_value"));
                    //放行
                    return true;
                }
            }
            //当请求的url与token与redis中的存储不相同时，解除锁定
            RedisTool.releaseDistributedLock(redisService,request.getParameter("token_value"),request.getParameter("token_value"));
            //进行拦截
            return false;
        }
        return false;
    }

直到后续controller操作执行完毕后释放分布式锁,见postHandle。

分布式锁的具体实现为

package com.cloud.user.config.redis;

/**
 * Created by Administrator on 2018-08-05.
 */
public class RedisTool {
    private static final String LOCK_SUCCESS = "OK";
    private static final Long RELEASE_SUCCESS = 1L;
    /**
     * 尝试获取分布式锁
     * @param lockKey 锁
     * @param requestId 请求标识
     * @param expireTime 超期时间
     * @return 是否获取成功
     */
    public static boolean tryGetDistributedLock(RedisService redisService, String lockKey, String requestId, int expireTime) {

        String result = redisService.set(lockKey, requestId, expireTime);

        if (LOCK_SUCCESS.equals(result)) {
            return true;
        }
        return false;
    }
    /**
     * 释放分布式锁
     * @param lockKey 锁
     * @param requestId 请求标识
     * @return 是否释放成功
     */
    public static boolean releaseDistributedLock(RedisService redisService, String lockKey, String requestId) {
        Object result = redisService.eval(lockKey,requestId);

        if (RELEASE_SUCCESS.equals(result)) {
            return true;
        }
        return false;

    }
}

在RedisServiceImpl实现类中，以上set跟eval的具体实现为

private static final String SET_IF_NOT_EXIST = "NX";
private static final String SET_WITH_EXPIRE_TIME = "PX";
@Autowired
private JedisPool jedisPool;

public <T> T execute(RedisFunction<T, Jedis> fun) {
    Jedis jedis = null;
    try {
        jedis = jedisPool.getResource();
        return (T)fun.callback(jedis);
    }catch (Exception e) {
        logger.error(e.getMessage());
        return null;
    }finally {
        if (jedis != null) {
            jedis.close();
        }
    }
}

@Override
public String set(String lockKey, String requestId, int expireTime) {
    return execute(new RedisFunction<String, Jedis>() {

        @Override
        public String callback(Jedis jedis) {
            return jedis.set(lockKey,requestId,SET_IF_NOT_EXIST,SET_WITH_EXPIRE_TIME,expireTime);
        }

    });
}

@Override
public Object eval(String lockKey, String requestId) {
    return execute(new RedisFunction<String, Jedis>() {
        @Override
        public Object callback(Jedis jedis) {
            String script = "if redis.call('get', KEYS[1]) == ARGV[1] then return redis.call('del', KEYS[1]) else return 0 end";
            return jedis.eval(script, Collections.singletonList(lockKey),Collections.singletonList(requestId));
        }
    });
}

public interface RedisFunction<T, E> {

    Object callback(E jedis);
}

最后解释一下为什么在分布式锁中不直接使用setnx+expire来设置分布式锁，而使用set(key,value,"NX","PX",过期时间)来做分布式锁，因为你鬼知道它在执行setnx的时候是不是刚好系统崩溃了，这样expire就未执行，结果大家都清楚这个分布式锁就永远锁定了。

转载于:https://my.oschina.net/u/3768341/blog/1922950