背景
在浏览器访问服务器时,网速非常慢的情况下。为了获取结果,用户常常会进行重复点击操作。这将会使得一些非幂等性操作的操作结果变得非常不可靠。
举例而言,用户进行付款操作就是一个非幂等性操作。
非幂等性,简单而言,就是一个操作是不可重复的。
方案
在用户浏览器cookie中,加上idempotent_token ,然后在各个微服务中使用拦截器拦截,并且使用分布式锁 进行全局锁定。
由于微服务是分布式的,那么将会出现一种情况是,在某种负载均衡的策略,用户在访问仓库微服务(1),并且同时访问仓库微服务(2),并且同时对库存进行修改。这种情景是合乎情理的,并且他们将会带上同一个idempotent_token进行仓库的微服务操作。这个时候是必须要使用分布式锁进行加锁操作。
原理与实现
拦截器
/**
* 幂等的拦截器,用于处理非幂等性操作。
* 幂等性将不予处理,直接放行
*/
public class IdempotentTokenInterceptor extends HandlerInterceptorAdapter {
private static final Logger log = LoggerFactory.getLogger(IdempotentTokenInterceptor.class);
public static final String IDEMPOTENT_TOKEN = "idempotent_token";
@Resource
private IdempotentDb<IdempotentRo> defaultIdempotentDb;
@Value("${spring.cloud.consul.host}")
private String consulHost;
@Value("${spring.cloud.consul.port}")
private int consulPort;
/**
* 返回幂等错误的信息
*
* @param response http的响应
* @param message 返回的http message
* @return true -> 继续向下执行;false -> 不继续向下执行,将被拦截
*/
private boolean with(HttpServletResponse response, String message) {
response.setStatus(HttpStatus.UNAUTHORIZED.value());
response.setContentType("application/json");
response.setCharacterEncoding("utf-8");
CookieUtil.addCookie(response, IDEMPOTENT_TOKEN, UUID.randomUUID().toString().replaceAll("-", ""), 3600 * 2);
try (PrintWriter writer = response.getWriter()) {
writer.append(new Gson().toJson(Result.ResultBuilder.errorWith(message).build()));
writer.flush();
} catch (IOException e) {
e.printStackTrace();
log.error("cannot close response print writer");
e.printStackTrace();
}
return false;
}
@Override
public boolean preHandle(HttpServletRequest request, HttpServletResponse response, Object handler) throws Exception {
return Optional.ofNullable(request.getCookies())
.map(cookies -> Stream.of(cookies).filter(x -> IDEMPOTENT_TOKEN.equalsIgnoreCase(x.getName()))
.map(x -> {
String timeStamp = UUID.randomUUID().toString().replaceAll("-", "");
// 如果已经找到幂等token,判断是否幂等的值为空
return Optional.ofNullable(x.getValue()).map(v -> {
List<IdempotentRo> list = defaultIdempotentDb.findByAuthKey(v);
if (CollectionUtils.isEmpty(list))
list = new ArrayList<>();
// 查找该url是否已经存在与幂等键值对之中
boolean hasRequested = list.stream().anyMatch(ir -> request.getMethod().equals(ir.getMethod())
&& request.getRequestURI().equals(ir.getAuthUrl()));
if (hasRequested) {
log.error("already requested with idempotent token from the URL of {} by {} method", request.getRequestURI(), request.getMethod());
return with(response, "Please do not repeat the submission");
} else {
defaultIdempotentDb.insert(IdempotentRo.IdempotentRoBuilder.build().set(v, request.getRequestURI(), request.getMethod()).create());
CookieUtil.addCookie(response, IDEMPOTENT_TOKEN, UUID.randomUUID().toString().replaceAll("-", ""), 3600 * 2);
return true;
}
}).orElseGet(() -> {
log.error("cannot find value of idempotent token from the URL of {} by {} method", request.getRequestURI(), request.getMethod());
return with(response, "Please do not fake the idempotent token");
});
}).reduce((x, y) -> x && y).orElseGet(() -> {
log.error("cannot find idempotent token from the URL of {} by {} method", request.getRequestURI(), request.getMethod());
return with(response, "Please do request with idempotent token");
})).orElseGet(() -> {
log.error("cannot find cookies from the URL of {} by {} method", request.getRequestURI(), request.getMethod());
return with(response, "Please do not fake the request...");
});
}
@Override
public void postHandle(HttpServletRequest request, HttpServletResponse response, Object handler, ModelAndView modelAndView) throws Exception {
super.postHandle(request, response, handler, modelAndView);
}
@Override
public void afterCompletion(HttpServletRequest request, HttpServletResponse response, Object handler, Exception ex) throws Exception {
super.afterCompletion(request, response, handler, ex);
}
}
分布式锁
public class IdempotentDistributedLock implements DistributedLock {
private final Consul consul;
private final Session value;
private final SessionClient sessionClient;
private final SessionCreatedResponse session;
public static final String KEY = "consul_key";
public static final Logger log= LoggerFactory.getLogger(IdempotentDistributedLock.class);
public IdempotentDistributedLock(Consul consul, String sessionId) {
this.consul = consul;
// 获取摘要,作为session id,并创建会话
this.value = ImmutableSession.builder().name(sessionId).build();
this.sessionClient = consul.sessionClient();
this.session = sessionClient.createSession(value);
}
@Override
public void lock() {
// 进行获取锁的操作,获取不到则将线程进行放入到缓冲队列中
KeyValueClient keyValueClient = consul.keyValueClient();
boolean hasAcquired=keyValueClient.acquireLock(KEY,this.value.getName().get(), this.session.getId());
if(!hasAcquired)
throw new AlreadyLockedException();
}
@Override
@Deprecated
public void lockInterruptibly() throws InterruptedException {
throw new UnsupportedOperationException();
}
@Override
public boolean tryLock() {
return false;
}
@Override
public boolean tryLock(long time, TimeUnit unit) throws InterruptedException {
return false;
}
@Override
public void unlock() {
KeyValueClient keyValueClient = consul.keyValueClient();
keyValueClient.deleteKey(KEY);
sessionClient.destroySession(session.getId());
}
@Override
@Deprecated
public Condition newCondition() {
throw new UnsupportedOperationException();
}
}
性能测试
在Chrome的slow 3g下,用户访问同个操作,主要的延迟来自于业务的处理。
好处
- 能够有效的防止用户重复点击
- 分布式锁实现JVM的Lock接口,用户可无学习难度的使用,并且作为分布式锁进行资源锁定
- 在以consul的作为一致性的基础服务情况下,用户也可以有效的进行调试排查,直接将所有会话列表查询出来
局限性
- 用户对于单一资源的锁定将会出现有时难以决断
- 用户只能进行一次性操作,对于其他想要进行资源的操作的话,将会直接熔断,不再进行等待
未来的趋势
- 将会解决分布式锁中,复杂多资源的锁定
引用
有疑问加站长微信联系(非本文作者)