servicecomb-java-chassis icon indicating copy to clipboard operation
servicecomb-java-chassis copied to clipboard

ServiceComb Java Chassis is a Software Development Kit (SDK) for rapid development of microservices in Java, providing service registration, service discovery, dynamic routing, and service management...

Results 233 servicecomb-java-chassis issues
Sort by recently updated
recently updated
newest added

通过实现ServerListFilterExt#getFilteredListOfServers来做一个lb的功能,想确认下,入参中的List servers,实例的status是否准确实时? 想知道客户端获得的服务端实例信息是怎么获取到的?缓存机制是什么?

本地暂时规避方案: 覆盖了 org.apache.servicecomb.common.rest.filter.inner.ServerRestArgsFilter catch了 `org.apache.servicecomb.common.rest.codec.RestCodec#restToArgs` 抛出的异常 然后`return processException(invocation, e);` ``` protected Response processException(SwaggerInvocation invocation, Throwable e) { if (e instanceof InvocationTargetException) { e = ((InvocationTargetException) e).getTargetException(); } return ExceptionFactory.convertExceptionToResponse(invocation, e);...

enhancement
need discussion

2021-12-22 14:50:51.794 | [Visual_SDK-7] | INFO | o.a.s.registry.consumer.MicroserviceVersions | create MicroserviceVersions, appId=hivoice, microserviceName=service C. 2021-12-22 14:50:51.795 | [Visual_SDK-7] | INFO | o.a.s.s.r.cache.RefreshableMicroserviceCache | find instances[1] from service center success. service=hivoice/service...

可能在metrics日志上看到的CPU使用率很低, 但实际上容器CPU使用情况已达到配额限制, 容易产生误导. 麻烦分析一下有什么解决办法呢?

前端调用后端接口是将参数用gzip方式压缩,cse后端接口解析时无法解析出全部参数。

日志如下: 2021-12-10 15:13:37.576 [thread-1] ERROR - [xx.java:reload:86] - [] - Reload xx data failed, with message[null]: java.lang.NullPointerException: null at org.apache.servicecomb.core.filter.OperationInstancesDiscoveryFilter.groupByVersion(OperationInstancesDiscoveryFilter.java:115) ~[java-chassis-core-1.3.6.jar!/:1.3.6] at org.apache.servicecomb.core.filter.OperationInstancesDiscoveryFilter.init(OperationInstancesDiscoveryFilter.java:72) ~[java-chassis-core-1.3.6.jar!/:1.3.6] at org.apache.servicecomb.serviceregistry.discovery.AbstractDiscoveryFilter.discovery(AbstractDiscoveryFilter.java:33) ~[service-registry-1.3.6.jar!/:?] at org.apache.servicecomb.serviceregistry.discovery.DiscoveryTree.doDiscovery(DiscoveryTree.java:166) ~[service-registry-1.3.6.jar!/:?]...

![image](https://user-images.githubusercontent.com/10349459/93186921-c3689e80-f771-11ea-937d-0f63db5a9e5d.png) --- 上图为metric日志报错截图。 定位过程: 1、观察到客户端请求失败,查看interfacelog,发现错误码为490;客户端运行无明显运行异常日志 2、查看服务端运行日志,无任何异常信息,服务端未收到请求,时间点前后无任何异常日志打印。 3、查看metric日志,见上面截图,RestClientInvocation:]|Failed to send request, local:not connected, remote:/10.33.8.46:8082. io.netty.channel.ConnectTimeoutException: connection timed out: /10.33.8.46:8082 4、观察服务端出现问题时间点前后日志,服务可以正常收到来自上述客户端的请求,并能处理成功。 以上,请专家帮忙指点可能是哪里问题?观察过自测环境和现网,均有低概率发生。十分感谢~

performance
connection&timeout

我们服务计划将现网一个微服务接入CSE注册中心的方式改为HTTPS,但该注册中心其他微服务依然沿用HTTP接入。不知道这样是否可行呢? ![image](https://user-images.githubusercontent.com/1498592/144342282-b3dc81e0-30e4-49b0-9542-f48566b37db6.png)