软件编程
位置:首页>> 软件编程>> java编程>> Spring Cloud 请求重试机制核心代码分析

Spring Cloud 请求重试机制核心代码分析

作者:每当变幻时  发布时间:2023-04-11 04:00:23 

标签:Spring,Cloud,请求,重试

场景

发布微服务的操作一般都是打完新代码的包,kill掉在跑的应用,替换新的包,启动。

spring cloud 中使用eureka为注册中心,它是允许服务列表数据的延迟性的,就是说即使应用已经不在服务列表了,客户端在一段时间内依然会请求这个地址。那么就会出现请求正在发布的地址,而导致失败。

我们会优化服务列表的刷新时间,以提高服务列表信息的时效性。但是无论怎样,都无法避免有那么一段时间是数据不一致的。

所以我们想到一个办法就是重试机制,当a机子在重启时,同个集群的b是可以正常提供服务的,如果有重试机制就可以在上面这个场景里进行重试到b而不影响正确响应。

操作

需要进行如下的操作:


ribbon:
ReadTimeout: 10000
ConnectTimeout: 10000
MaxAutoRetries: 0
MaxAutoRetriesNextServer: 1
OkToRetryOnAllOperations: false

引入spring-retry包


<dependency>
 <groupId>org.springframework.retry</groupId>
 <artifactId>spring-retry</artifactId>
</dependency>

以zuul为例子还需要配置开启重试:


zuul.retryable=true

遇到了问题

然而万事总没那么一帆风顺,通过测试重试机制生效了,但是并没有我想象的去请求另一台健康的机子,于是被迫去吧开源码看一看,最终发现是源码的bug,不过已经修复,升级版本即可。

代码分析

使用的版本是

spring-cloud-netflix-core:1.3.6.RELEASE

spring-retry:1.2.1.RELEASE

spring cloud 依赖版本:


<dependencyManagement>
   <dependencies>
     <dependency>
       <groupId>org.springframework.cloud</groupId>
       <artifactId>spring-cloud-dependencies</artifactId>
       <version>${spring-cloud.version}</version>
       <type>pom</type>
       <scope>import</scope>
     </dependency>
   </dependencies>
 </dependencyManagement>

因为启用了重试,所以请求应用时会执行RetryableRibbonLoadBalancingHttpClient.execute方法:


public RibbonApacheHttpResponse execute(final RibbonApacheHttpRequest request, final IClientConfig configOverride) throws Exception {
   final RequestConfig.Builder builder = RequestConfig.custom();
   IClientConfig config = configOverride != null ? configOverride : this.config;
   builder.setConnectTimeout(config.get(
       CommonClientConfigKey.ConnectTimeout, this.connectTimeout));
   builder.setSocketTimeout(config.get(
       CommonClientConfigKey.ReadTimeout, this.readTimeout));
   builder.setRedirectsEnabled(config.get(
       CommonClientConfigKey.FollowRedirects, this.followRedirects));

final RequestConfig requestConfig = builder.build();
   final LoadBalancedRetryPolicy retryPolicy = loadBalancedRetryPolicyFactory.create(this.getClientName(), this);
   RetryCallback retryCallback = new RetryCallback() {
     @Override
     public RibbonApacheHttpResponse doWithRetry(RetryContext context) throws Exception {
       //on retries the policy will choose the server and set it in the context
       //extract the server and update the request being made
       RibbonApacheHttpRequest newRequest = request;
       if(context instanceof LoadBalancedRetryContext) {
         ServiceInstance service = ((LoadBalancedRetryContext)context).getServiceInstance();
         if(service != null) {
           //Reconstruct the request URI using the host and port set in the retry context
           newRequest = newRequest.withNewUri(new URI(service.getUri().getScheme(),
               newRequest.getURI().getUserInfo(), service.getHost(), service.getPort(),
               newRequest.getURI().getPath(), newRequest.getURI().getQuery(),
               newRequest.getURI().getFragment()));
         }
       }
       newRequest = getSecureRequest(request, configOverride);
       HttpUriRequest httpUriRequest = newRequest.toRequest(requestConfig);
       final HttpResponse httpResponse = RetryableRibbonLoadBalancingHttpClient.this.delegate.execute(httpUriRequest);
       if(retryPolicy.retryableStatusCode(httpResponse.getStatusLine().getStatusCode())) {
         if(CloseableHttpResponse.class.isInstance(httpResponse)) {
           ((CloseableHttpResponse)httpResponse).close();
         }
         throw new RetryableStatusCodeException(RetryableRibbonLoadBalancingHttpClient.this.clientName,
             httpResponse.getStatusLine().getStatusCode());
       }
       return new RibbonApacheHttpResponse(httpResponse, httpUriRequest.getURI());
     }
   };
   return this.executeWithRetry(request, retryPolicy, retryCallback);
 }

我们发现先new 一个RetryCallback,然后执行this.executeWithRetry(request, retryPolicy, retryCallback);

而这个RetryCallback.doWithRetry的代码我们清楚看到是实际请求的代码,也就是说this.executeWithRetry方法最终还是会调用RetryCallback.doWithRetry


protected <T, E extends Throwable> T doExecute(RetryCallback<T, E> retryCallback,
     RecoveryCallback<T> recoveryCallback, RetryState state)
     throws E, ExhaustedRetryException {

RetryPolicy retryPolicy = this.retryPolicy;
   BackOffPolicy backOffPolicy = this.backOffPolicy;

// Allow the retry policy to initialise itself...
   RetryContext context = open(retryPolicy, state);
   if (this.logger.isTraceEnabled()) {
     this.logger.trace("RetryContext retrieved: " + context);
   }

// Make sure the context is available globally for clients who need
   // it...
   RetrySynchronizationManager.register(context);

Throwable lastException = null;

boolean exhausted = false;
   try {

// Give clients a chance to enhance the context...
     boolean running = doOpenInterceptors(retryCallback, context);

if (!running) {
       throw new TerminatedRetryException(
           "Retry terminated abnormally by interceptor before first attempt");
     }

// Get or Start the backoff context...
     BackOffContext backOffContext = null;
     Object resource = context.getAttribute("backOffContext");

if (resource instanceof BackOffContext) {
       backOffContext = (BackOffContext) resource;
     }

if (backOffContext == null) {
       backOffContext = backOffPolicy.start(context);
       if (backOffContext != null) {
         context.setAttribute("backOffContext", backOffContext);
       }
     }

/*
      * We allow the whole loop to be skipped if the policy or context already
      * forbid the first try. This is used in the case of external retry to allow a
      * recovery in handleRetryExhausted without the callback processing (which
      * would throw an exception).
      */
     while (canRetry(retryPolicy, context) && !context.isExhaustedOnly()) {

try {
         if (this.logger.isDebugEnabled()) {
           this.logger.debug("Retry: count=" + context.getRetryCount());
         }
         // Reset the last exception, so if we are successful
         // the close interceptors will not think we failed...
         lastException = null;
         return retryCallback.doWithRetry(context);
       }
       catch (Throwable e) {

lastException = e;

try {
           registerThrowable(retryPolicy, state, context, e);
         }
         catch (Exception ex) {
           throw new TerminatedRetryException("Could not register throwable",
               ex);
         }
         finally {
           doOnErrorInterceptors(retryCallback, context, e);
         }

if (canRetry(retryPolicy, context) && !context.isExhaustedOnly()) {
           try {
             backOffPolicy.backOff(backOffContext);
           }
           catch (BackOffInterruptedException ex) {
             lastException = e;
             // back off was prevented by another thread - fail the retry
             if (this.logger.isDebugEnabled()) {
               this.logger
                   .debug("Abort retry because interrupted: count="
                       + context.getRetryCount());
             }
             throw ex;
           }
         }

if (this.logger.isDebugEnabled()) {
           this.logger.debug(
               "Checking for rethrow: count=" + context.getRetryCount());
         }

if (shouldRethrow(retryPolicy, context, state)) {
           if (this.logger.isDebugEnabled()) {
             this.logger.debug("Rethrow in retry for policy: count="
                 + context.getRetryCount());
           }
           throw RetryTemplate.<E>wrapIfNecessary(e);
         }

}

/*
        * A stateful attempt that can retry may rethrow the exception before now,
        * but if we get this far in a stateful retry there's a reason for it,
        * like a circuit breaker or a rollback classifier.
        */
       if (state != null && context.hasAttribute(GLOBAL_STATE)) {
         break;
       }
     }

if (state == null && this.logger.isDebugEnabled()) {
       this.logger.debug(
           "Retry failed last attempt: count=" + context.getRetryCount());
     }

exhausted = true;
     return handleRetryExhausted(recoveryCallback, context, state);

}
   catch (Throwable e) {
     throw RetryTemplate.<E>wrapIfNecessary(e);
   }
   finally {
     close(retryPolicy, context, state, lastException == null || exhausted);
     doCloseInterceptors(retryCallback, context, lastException);
     RetrySynchronizationManager.clear();
   }
 }

在一个while循环里实现重试机制,当执行retryCallback.doWithRetry(context)出现异常的时候,就会catch异常,然后用 retryPolicy判断是否进行重试,特别注意registerThrowable(retryPolicy, state, context, e);方法,不但判断了是否重试,在重试情况下会新选出一个机子放入context,然后再去执行retryCallback.doWithRetry(context)时带入,如此就实现了换机子重试了。

但是我的配置怎么会没有换机子呢?调试代码发现registerThrowable(retryPolicy, state, context, e);选出来的机子没问题,就是新的健康的机子,但是在执行retryCallback.doWithRetry(context)代码的时候依然请求的是那台挂掉的机子。

所以我们再仔细看一下retryCallback.doWithRetry(context)的代码:

我们发现了这行代码:


newRequest = getSecureRequest(request, configOverride);
protected RibbonApacheHttpRequest getSecureRequest(RibbonApacheHttpRequest request, IClientConfig configOverride) {
   if (isSecure(configOverride)) {
     final URI secureUri = UriComponentsBuilder.fromUri(request.getUri())
         .scheme("https").build(true).toUri();
     return request.withNewUri(secureUri);
   }
   return request;
 }

newRequest在前面已经使用context构建完毕,request是上一次请求的数据,只要执行这个代码就会发现newRequest永远都会被request覆盖。看到这里我们才发现原来是一个源码bug。

issue地址:https://github.com/spring-cloud/spring-cloud-netflix/issues/2667

总结

这是一次很普通的查问题过程,在这个过程中当我发现配置没有达到我的预期时,我先查看了配置的含义,尝试多次无果,于是进行断点调试发现异常中断点后,因为场景需要一台机子健康一台机子下线,我模拟了数百次,最终才定位到了这行代码。开源项目即使是优秀的项目必然也会有bug存在,不迷信,不盲目。另一方面,阅读源码能力也是一个解决问题的重要能力,像我在找源码入口,定位代码时耗费了很多的时间。

来源:http://www.cnblogs.com/killbug/p/9150067.html

0
投稿

猜你喜欢

手机版 软件编程 asp之家 www.aspxhome.com