I am doing some stress test of a simple spring cloud app backed by a zuul proxy service Sporadically we have a broken pipe exception due to a stale connection between Zuul and the service. I configure Zuul with diferent configuration options without success:
hystrix:
command:
default:
execution:
isolation:
strategy: THREAD
thread:
timeoutInMilliseconds: 61000
ribbon:
ReadTimeout: 60000
ConnectTimeout: 6000
zuul:
host:
socket-timeout-millis: 60000
connect-timeout-millis: 60000
The exception is as follows:
com.netflix.zuul.exception.ZuulException: Forwarding error at org.springframework.cloud.netflix.zuul.filters.route.RibbonRoutingFilter.forward(RibbonRoutingFilter.java:151) ~[router.jar!/:0.0.1] at org.springframework.cloud.netflix.zuul.filters.route.RibbonRoutingFilter.run(RibbonRoutingFilter.java:114) ~[router.jar!/:0.0.1] at com.netflix.zuul.ZuulFilter.runFilter(ZuulFilter.java:112) ~[zuul-core-1.0.28.jar!/:na]
Caused by: com.netflix.hystrix.exception.HystrixRuntimeException: query-serviceRibbonCommand failed and no fallback available. at com.netflix.hystrix.AbstractCommand$20.call(AbstractCommand.java:816) ~[hystrix-core-1.4.0-RC6.jar!/:na] at com.netflix.hystrix.AbstractCommand$20.call(AbstractCommand.java:798) ~[hystrix-core-1.4.0-RC6.jar!/:na] at rx.internal.operators.OperatorOnErrorResumeNextViaFunction$1.onError(OperatorOnErrorResumeNextViaFunction.java:77) ~[rxjava-1.0.4.jar!/:1.0.4] at rx.internal.operators.OperatorDoOnEach$1.onError(OperatorDoOnEach.java:70) ~[rxjava-1.0.4.jar!/:1.0.4]
Caused by: java.net.SocketException: Broken pipe at java.net.SocketOutputStream.socketWrite0(Native Method) ~[na:1.8.0_45] at java.net.SocketOutputStream.socketWrite(SocketOutputStream.java:109) ~[na:1.8.0_45] at java.net.SocketOutputStream.write(SocketOutputStream.java:153) ~[na:1.8.0_45] at org.apache.http.impl.io.AbstractSessionOutputBuffer.flushBuffer(AbstractSessionOutputBuffer.java:159) ~[httpcore-4.3.3.jar!/:4.3.3] ... at rx.internal.operators.OperatorMap$1.onNext(OperatorMap.java:55) ~[rxjava-1.0.4.jar!/:1.0.4] at com.netflix.loadbalancer.reactive.LoadBalancerCommand$1.call(LoadBalancerCommand.java:185) ~[ribbon-loadbalancer-2.0-RC13.jar!/:na] at com.netflix.loadbalancer.reactive.LoadBalancerCommand$1.call(LoadBalancerCommand.java:180
I had run out from the same problem: I have resolved it by adding zuul execution timeout to the application.yml of zuul server:
# Increase the Hystrix timeout to 60s (globally)
hystrix:
command:
default:
execution:
isolation:
thread:
timeoutInMilliseconds: 60000
Please refer to this thread on netflix issues: https://github.com/spring-cloud/spring-cloud-netflix/issues/321