问题
在使用skywalking监控生产环境问题时,发现有一个请求的调用链路中微服务Platform的内部接口inner/userinfo
很奇怪的被重复调用了2次,如下图:
分析过程
相关版本:Spring Cloud版本为Dalston.SR4,Spring Boot版本为1.5.7.RELEASE
查看inner/userinfo
的服务方的2次调用分别耗时6043ms和8078ms,通过跟踪RetryableFeignLoadBalancer类的execute方法发现,feign的连接超时时间connectTimeout=2000(2秒),读超时时间readTimeout=5000(5秒),因此可以判断服务方的2次响应时间超过读超时时间阈值5秒了,因此调用方最终报了超时异常RetryableException。
我们知道,Spring Cloud中Feign整合了Ribbon,但Feign和Ribbon都有重试的功能,Spring Cloud为了统一两者的行为,将Feign的重试策略默认设置为 feign.Retryer#NEVER_RETRY
(即永不重试)。如要使用Feign的重试功能的话,只需使用Ribbon的重试配置即可。既然这样,那为什么会有以上现象呢?
进一步研究发现,对于Camden以及以后的版本,Feign的重试可使用如下属性进行配置:
ribbon:
# 同一实例最大重试次数,不包括首次调用。默认值为0
MaxAutoRetries: 0
# 同一个微服务其他实例的最大重试次数,不包括第一次调用的实例。默认值为1
MaxAutoRetriesNextServer: 1
# 是否所有操作(GET、POST等)都允许重试。默认值为false
OkToRetryOnAllOperations: false
我们的服务方Platform没有配置这些参数,因此应该是使用了默认值。因为我们Platform微服务只启动了一个实例,所以我重点关注MaxAutoRetries参数,实际跟踪发现它的值也是0,这跟我理解的有出入啊!
跟踪Spring Cloud的源码,发现Feign是在RetryTemplate的doExecute方法中进行重试的判断和调用的:
protected <T, E extends Throwable> T doExecute(...) {
...
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) {
...
}
...
}
...
}
其中canRetry方法用于判断当接口调用异常时是否需要进行重试。
跟进canRetry方法,终于找到了罪魁祸首——InterceptorRetryPolicy的canRetry方法:
org.springframework.cloud.client.loadbalancer.InterceptorRetryPolicy.java
@Override
public boolean canRetry(RetryContext context) {
LoadBalancedRetryContext lbContext = (LoadBalancedRetryContext)context;
if(lbContext.getRetryCount() == 0 && lbContext.getServiceInstance() == null) {
//We haven't even tried to make the request yet so return true so we do
lbContext.setServiceInstance(serviceInstanceChooser.choose(serviceName));
return true;
}
return policy.canRetryNextServer(lbContext);
}
留意最后一行:policy.canRetryNextServer(lbContext)。这一行意思是,根据策略判断是否重试服务的下一个实例。因为MaxAutoRetriesNextServer默认值为1,因此这里会返回true,所以inner/userinfo
就被调用了2次。
解决方案
找到原因就好办了,解决方案很简单,在调用方的yml配置MaxAutoRetriesNextServer的值为0即可:
ribbon:
# 同一实例最大重试次数,不包括首次调用。默认值为0
MaxAutoRetries: 0
# 同一个微服务其他实例的最大重试次数,不包括第一次调用的实例。默认值为1
MaxAutoRetriesNextServer: 0
# 是否所有操作(GET、POST等)都允许重试。默认值为false
OkToRetryOnAllOperations: false
配置后,在开发环境验证成功!
附录
相关节点的异常:
1、Hystrix/IUserService#getUserInfo()/Execution节点的异常信息为:
Read timed out executing GET http://prong-cloud-server-platform/inner/userinfo
feign.RetryableException: Read timed out executing GET http://prong-cloud-server-platform/inner/userinfo
at feign.FeignException.errorExecuting(FeignException.java:67)
at feign.SynchronousMethodHandler.executeAndDecode(SynchronousMethodHandler.java:104)
at feign.SynchronousMethodHandler.invoke(SynchronousMethodHandler.java:76)
at feign.hystrix.HystrixInvocationHandler$1.run$original$guV1ukWA(HystrixInvocationHandler.java:108)
at feign.hystrix.HystrixInvocationHandler$1.run$original$guV1ukWA$accessor$1jiMEUrR(HystrixInvocationHandler.java)
at feign.hystrix.HystrixInvocationHandler$1$auxiliary$sEUgyNk3.call(Unknown Source)
at org.apache.skywalking.apm.agent.core.plugin.interceptor.enhance.InstMethodsInter.intercept(InstMethodsInter.java:93)
at feign.hystrix.HystrixInvocationHandler$1.run(HystrixInvocationHandler.java)
at com.netflix.hystrix.HystrixCommand$2.call(HystrixCommand.java:302)
at com.netflix.hystrix.HystrixCommand$2.call(HystrixCommand.java:298)
at rx.internal.operators.OnSubscribeDefer.call(OnSubscribeDefer.java:46)
at rx.internal.operators.OnSubscribeDefer.call(OnSubscribeDefer.java:35)
at rx.internal.operators.OnSubscribeLift.call(OnSubscribeLift.java:48)
at rx.internal.operators.OnSubscribeLift.call(OnSubscribeLift.java:30)
at rx.internal.operators.OnSubscribeLift.call(OnSubscribeLift.java:48)
at rx.internal.operators.OnSubscribeLift.call(OnSubscribeLift.java:30)
at rx.internal.operators.OnSubscribeLift.call(OnSubscribeLift.java:48)
at rx.internal.operators.OnSubscribeLift.call(OnSubscribeLift.java:30)
at rx.Observable.unsafeSubscribe(Observable.java:10211)
at rx.internal.operators.OnSubscribeDefer.call(OnSubscribeDefer.java:51)
at rx.internal.operators.OnSubscribeDefer.call(OnSubscribeDefer.java:35)
at rx.Observable.unsafeSubscribe(Observable.java:10211)
at rx.internal.operators.OnSubscribeDoOnEach.call(OnSubscribeDoOnEach.java:41)
at rx.internal.operators.OnSubscribeDoOnEach.call(OnSubscribeDoOnEach.java:30)
at rx.internal.operators.OnSubscribeLift.call(OnSubscribeLift.java:48)
at rx.internal.operators.OnSubscribeLift.call(OnSubscribeLift.java:30)
at rx.Observable.unsafeSubscribe(Observable.java:10211)
at rx.internal.operators.OperatorSubscribeOn$1.call(OperatorSubscribeOn.java:94)
at com.netflix.hystrix.strategy.concurrency.HystrixContexSchedulerAction$1.call(HystrixContexSchedulerAction.java:56)
at com.netflix.hystrix.strategy.concurrency.HystrixContexSchedulerAction$1.call(HystrixContexSchedulerAction.java:47)
at io.prong.autoconfigure.auth.service.ProngHystrixConcurrencyStrategy$WrappedCallable.call(ProngHystrixConcurrencyStrategy.java:121)
at com.netflix.hystrix.strategy.concurrency.HystrixContexSchedulerAction.call(HystrixContexSchedulerAction.java:69)
at rx.internal.schedulers.ScheduledAction.run(ScheduledAction.java:55)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)
java.net.SocketTimeoutException: Read timed out
at feign.FeignException.errorExecuting(FeignException.java:67)
at feign.SynchronousMethodHandler.executeAndDecode(SynchronousMethodHandler.java:104)
at feign.SynchronousMethodHandler.invoke(SynchronousMethodHandler.java:76)
at feign.hystrix.HystrixInvocationHandler$1.run$original$guV1ukWA(HystrixInvocationHandler.java:108)
at feign.hystrix.HystrixInvocationHandler$1.run$original$guV1ukWA$accessor$1jiMEUrR(HystrixInvocationHandler.java)
at feign.hystrix.HystrixInvocationHandler$1$auxiliary$sEUgyNk3.call(Unknown Source)
at org.apache.skywalking.apm.agent.core.plugin.interceptor.enhance.InstMethodsInter.intercept(InstMethodsInter.java:93)
at feign.hystrix.HystrixInvocationHandler$1.run(HystrixInvocationHandler.java)
at com.netflix.hystrix.HystrixCommand$2.call(HystrixCommand.java:302)
at com.netflix.hystrix.HystrixCommand$2.call(HystrixCommand.java:298)
2、/inner/userinfo节点的异常信息为:
java.net.SocketTimeoutException: Read timed out
at java.net.SocketInputStream.socketRead0(Native Method)
at java.net.SocketInputStream.socketRead(SocketInputStream.java:116)
at java.net.SocketInputStream.read(SocketInputStream.java:171)
at java.net.SocketInputStream.read(SocketInputStream.java:141)
at java.io.BufferedInputStream.fill(BufferedInputStream.java:246)
at java.io.BufferedInputStream.read1(BufferedInputStream.java:286)
at java.io.BufferedInputStream.read(BufferedInputStream.java:345)
at sun.net.www.http.HttpClient.parseHTTPHeader(HttpClient.java:735)
at sun.net.www.http.HttpClient.parseHTTP(HttpClient.java:678)
at sun.net.www.protocol.http.HttpURLConnection.getInputStream0(HttpURLConnection.java:1587)
at sun.net.www.protocol.http.HttpURLConnection.getInputStream(HttpURLConnection.java:1492)
at java.net.HttpURLConnection.getResponseCode(HttpURLConnection.java:480)
at feign.Client$Default.convertResponse(Client.java:152)
at feign.Client$Default.execute$original$Z12I5rH3(Client.java:74)
at feign.Client$Default.execute$original$Z12I5rH3$accessor$MeGe8flP(Client.java)
at feign.Client$Default$auxiliary$uXNNBULq.call(Unknown Source)
at org.apache.skywalking.apm.agent.core.plugin.interceptor.enhance.InstMethodsInter.intercept(InstMethodsInter.java:93)
at feign.Client$Default.execute(Client.java)
at org.springframework.cloud.netflix.feign.ribbon.RetryableFeignLoadBalancer$1.doWithRetry(RetryableFeignLoadBalancer.java:92)
at org.springframework.cloud.netflix.feign.ribbon.RetryableFeignLoadBalancer$1.doWithRetry(RetryableFeignLoadBalancer.java:77)
at org.springframework.retry.support.RetryTemplate.doExecute(RetryTemplate.java:287)
at org.springframework.retry.support.RetryTemplate.execute(RetryTemplate.java:164)
at org.springframework.cloud.netflix.feign.ribbon.RetryableFeignLoadBalancer.execute(RetryableFeignLoadBalancer.java:77)
at org.springframework.cloud.netflix.feign.ribbon.RetryableFeignLoadBalancer.execute(RetryableFeignLoadBalancer.java:48)
at com.netflix.client.AbstractLoadBalancerAwareClient$1.call(AbstractLoadBalancerAwareClient.java:109)
at com.netflix.loadbalancer.reactive.LoadBalancerCommand$3$1.call(LoadBalancerCommand.java:303)
at com.netflix.loadbalancer.reactive.LoadBalancerCommand$3$1.call(LoadBalancerCommand.java:287)
at rx.internal.util.ScalarSynchronousObservable$3.call(ScalarSynchronousObservable.java:231)
at rx.internal.util.ScalarSynchronousObservable$3.call(ScalarSynchronousObservable.java:228)
at rx.Observable.unsafeSubscribe(Observable.java:10211)
at rx.internal.operators.OnSubscribeConcatMap$ConcatMapSubscriber.drain(OnSubscribeConcatMap.java:286)
at rx.internal.operators.OnSubscribeConcatMap$ConcatMapSubscriber.onNext(OnSubscribeConcatMap.java:144)
at com.netflix.loadbalancer.reactive.LoadBalancerCommand$1.call(LoadBalancerCommand.java:185)
at com.netflix.loadbalancer.reactive.LoadBalancerCommand$1.call(LoadBalancerCommand.java:180)
at rx.Observable.unsafeSubscribe(Observable.java:10211)
at rx.internal.operators.OnSubscribeConcatMap.call(OnSubscribeConcatMap.java:94)
at rx.internal.operators.OnSubscribeConcatMap.call(OnSubscribeConcatMap.java:42)
at rx.Observable.unsafeSubscribe(Observable.java:10211)
at rx.internal.operators.OperatorRetryWithPredicate$SourceSubscriber$1.call(OperatorRetryWithPredicate.java:127)
at rx.internal.schedulers.TrampolineScheduler$InnerCurrentThreadScheduler.enqueue(TrampolineScheduler.java:73)
at rx.internal.schedulers.TrampolineScheduler$InnerCurrentThreadScheduler.schedule(TrampolineScheduler.java:52)
at rx.internal.operators.OperatorRetryWithPredicate$SourceSubscriber.onNext(OperatorRetryWithPredicate.java:79)
at rx.internal.operators.OperatorRetryWithPredicate$SourceSubscriber.onNext(OperatorRetryWithPredicate.java:45)
at rx.internal.util.ScalarSynchronousObservable$WeakSingleProducer.request(ScalarSynchronousObservable.java:276)
at rx.Subscriber.setProducer(Subscriber.java:209)
at rx.internal.util.ScalarSynchronousObservable$JustOnSubscribe.call(ScalarSynchronousObservable.java:138)
网友评论