在阿里云VPC环境下，部署了k8sV1.11.1高可用集群，kubelet一直有报一个错误
Oct 18 10:03:20 worker001 kubelet[18956]: E1018 10:03:20.625638 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91373; INTERNAL_ERROR
Oct 18 10:04:03 worker001 kubelet[18956]: E1018 10:04:03.711397 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91511; INTERNAL_ERROR
Oct 18 10:04:38 worker001 kubelet[18956]: E1018 10:04:38.792179 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91529; INTERNAL_ERROR
Oct 18 10:05:03 worker001 kubelet[18956]: E1018 10:05:03.812098 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91643; INTERNAL_ERROR
Oct 18 10:05:48 worker001 kubelet[18956]: E1018 10:05:48.638481 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91681; INTERNAL_ERROR
Oct 18 10:06:03 worker001 kubelet[18956]: E1018 10:06:03.896857 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91781; INTERNAL_ERROR
Oct 18 10:06:53 worker001 kubelet[18956]: E1018 10:06:53.674940 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91819; INTERNAL_ERROR
Oct 18 10:07:03 worker001 kubelet[18956]: E1018 10:07:03.975904 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91853; INTERNAL_ERROR
Oct 18 10:07:54 worker001 kubelet[18956]: E1018 10:07:54.042442 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91907; INTERNAL_ERROR
Oct 18 10:08:04 worker001 kubelet[18956]: E1018 10:08:04.055452 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91933; INTERNAL_ERROR
Oct 18 10:08:54 worker001 kubelet[18956]: E1018 10:08:54.129991 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 91995; INTERNAL_ERROR
Oct 18 10:09:04 worker001 kubelet[18956]: E1018 10:09:04.144145 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 92005; INTERNAL_ERROR
Oct 18 10:09:54 worker001 kubelet[18956]: E1018 10:09:54.218016 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 92071; INTERNAL_ERROR
Oct 18 10:10:04 worker001 kubelet[18956]: E1018 10:10:04.232644 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 92085; INTERNAL_ERROR
Oct 18 10:10:54 worker001 kubelet[18956]: E1018 10:10:54.302705 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 92147; INTERNAL_ERROR
Oct 18 10:11:04 worker001 kubelet[18956]: E1018 10:11:04.316685 18956 streamwatcher.go:109] Unable to decode an event from the watch stream: stream error: stream ID 92161; INTERNAL_ERROR
目前集群运行正常，但是通过kubectl get nodes -w会发现偶尔节点会NotReady状态，大约40s左右偶尔一下，大部分时间都是正常running状态。这个导致的结果是我通过外部访问集群的服务，偶尔出现：
java.net.ConnectException: 拒绝连接 (Connection refused)
不知道是我的服务部署问题，还是集群本身的问题。
写回答
-
NotReady的问题已经找到，是因为配置了controller-manager的node-monitor-grace-period: 10s 和 pod-eviction-timeout: 10s导致controller获取不到node的状态，影响了node上的服务，只需要改成默认值即可，现在集群中的服务正常了。
但是在kube-proxy中还是有很多下面的错误，不知道什么原因：
Unable to decode an event from the watch stream: stream error: stream ID 92071; INTERNAL_ERROR2019-07-17 23:09:04赞同 1 展开评论 打赏 -
版权声明：本文内容由阿里云实名注册用户自发贡献，版权归原作者所有，阿里云开发者社区不拥有其著作权，亦不承担相应法律责任。具体规则请查看《阿里云开发者社区用户服务协议》和《阿里云开发者社区知识产权保护指引》。如果您发现本社区中有涉嫌抄袭的内容，填写侵权投诉表单进行举报，一经查实，本社区将立刻删除涉嫌侵权内容。
