k8s部署elastic相关问题

Question

在AzureVM中部署了k8s集群,在k8s中部署了elastic服务,elastic服务启动了3个pod,成功run,但是服务不可用,查看pod日志发现不能解析其他两个pod的dns,只能解析当前pod的dns;已经成功部署headlessService

Answer 1

问题详细描述
docker版本:

docker --version

Docker version 26.0.0, build 2ae903e

kubelet、kubeadm、kubectl version一致:

Client Version: version.Info{Major:"1", Minor:"22", GitVersion:"v1.22.2", GitCommit:"8b5a19147530eaac9476b0ab82980b4088bbc1b2", GitTreeState:"clean", BuildDate:"2021-09-15T21:38:50Z", GoVersion:"go1.16.8", Compiler:"gc", Platform:"linux/amd64"}

Server Version: version.Info{Major:"1", Minor:"22", GitVersion:"v1.22.2", GitCommit:"8b5a19147530eaac9476b0ab82980b4088bbc1b2", GitTreeState:"clean", BuildDate:"2021-09-15T21:32:41Z", GoVersion:"go1.16.8", Compiler:"gc", Platform:"linux/amd64"}

calico版本:3.25.0

现在k8s已经部署完成,节点发现已经配置完成

现在在当前k8s集群中部署elastic的时候,日志提示:failed to resolve host [elastic-cluster-master-0.es7-headless.elastic-worker.svc.cluster.local]",

"stacktrace": ["java.net.UnknownHostException: elastic-cluster-master-0.es7-headless.elastic-worker.svc.cluster.local"

 

k8s中elastic的组件:pvc、pv、service、statefulset、serviceaccount、secrite等

不能解析的域名分别为: elastic-cluster-master-0.es7-headless.elastic-worker.svc.cluster.local、elastic-cluster-data-0.es7-headless.elastic-worker.svc.cluster.local、elastic-cluster-ctl-0.es7-headless.elastic-worker.svc.cluster.local