本文将在前文的基础上介绍在kubernetes集群环境中配置dns服务，在k8s集群中，pod的生命周期是短暂的，pod重启后ip地址会产生变化，对于应用程序来说这是不可接受的，为解决这个问题，K8S集群巧妙的引入的dns服务来实现服务的发现，在k8s集群中dns总共需要使用4个组件，各组件分工如下： etcd：DNS存储 kube2sky：将Kubernetes Master中的service（服务）注册到etcd。 skyDNS：提供DNS域名解析服务。 healthz：提供对skydns服务的健康检查。
一、下载相关镜像文件，并纳入本地仓库统一管理
# docker pull docker.io/elcolio/etcd # docker pull docker.io/port/kubernetes-kube2sky # docker pull docker.io/skynetservices/skydns # docker pull docker.io/wu1boy/healthz
# docker tag docker.io/elcolio/etcd registry.fjhb.cn/etcd # docker tag docker.io/port/kubernetes-kube2sky registry.fjhb.cn/kubernetes-kube2sky # docker tag docker.io/skynetservices/skydns registry.fjhb.cn/skydns # docker tag docker.io/wu1boy/healthz registry.fjhb.cn/healthz
# docker push registry.fjhb.cn/etcd # docker push registry.fjhb.cn/kubernetes-kube2sky # docker push registry.fjhb.cn/skydns # docker push registry.fjhb.cn/healthz # docker images |grep fjhb
二、通过rc文件创建pod 这里面一个pod包含了4个组件，一个组件运行在一个docker容器中
# cat skydns-rc.yaml apiVersion: v1 kind: ReplicationController metadata: name: kube-dns namespace: default labels: k8s-app: kube-dns version: v12 kubernetes.io/cluster-service: "true" spec: replicas: 1 selector: k8s-app: kube-dns version: v12 template: metadata: labels: k8s-app: kube-dns version: v12 kubernetes.io/cluster-service: "true" spec: containers: - name: etcd image: registry.fjhb.cn/etcd resources: limits: cpu: 100m memory: 50Mi requests: cpu: 100m memory: 50Mi command: - /bin/etcd - --data-dir - /tmp/data - --listen-client-urls - http://127.0.0.1:2379,http://127.0.0.1:4001 - --advertise-client-urls - http://127.0.0.1:2379,http://127.0.0.1:4001 - --initial-cluster-token - skydns-etcd volumeMounts: - name: etcd-storage mountPath: /tmp/data - name: kube2sky image: registry.fjhb.cn/kubernetes-kube2sky resources: limits: cpu: 100m memory: 50Mi requests: cpu: 100m memory: 50Mi args: - -kube_master_url=http://192.168.115.5:8080 - -domain=cluster.local - name: skydns image: registry.fjhb.cn/skydns resources: limits: cpu: 100m memory: 50Mi requests: cpu: 100m memory: 50Mi args: - -machines=http://127.0.0.1:4001 - -addr=0.0.0.0:53 - -ns-rotate=false - -domain=cluster.local ports: - containerPort: 53 name: dns protocol: UDP - containerPort: 53 name: dns-tcp protocol: TCP - name: healthz image: registry.fjhb.cn/healthz resources: limits: cpu: 10m memory: 20Mi requests: cpu: 10m memory: 20Mi args: - -cmd=nslookup kubernetes.default.svc.cluster.local 127.0.0.1 >/dev/null - -port=8080 ports: - containerPort: 8080 protocol: TCP volumes: - name: etcd-storage emptyDir: {} dnsPolicy: Default
三、通过srv文件创建service
# cat skydns-svc.yaml apiVersion: v1 kind: Service metadata: name: kube-dns namespace: default labels: k8s-app: kube-dns kubernetes.io/cluster-service: "true" kubernetes.io/name: "KubeDNS" spec: selector: k8s-app: kube-dns clusterIP: 10.254.16.254 ports: - name: dns port: 53 protocol: UDP - name: dns-tcp port: 53 protocol: TCP# kubectl create -f skydns-rc.yaml # kubectl create -f skydns-svc.yaml # kubectl get rc # kubectl get pod # kubectl get svc
# kubectl describe svc kube-dns
# kubectl describe rc kube-dns
# kubectl describe pod kube-dns-9fllp Name: kube-dns-9fllp Namespace: default Node: 192.168.115.6/192.168.115.6 Start Time: Tue, 23 Jan 2018 10:55:19 -0500 Labels: k8s-app=kube-dns kubernetes.io/cluster-service=true version=v12 Status: Running IP: 172.16.37.5 Controllers: ReplicationController/kube-dns Containers: etcd: Container ID: docker://62ad76bfaca1797c5f43b0e9eebc04074169fce4cc15ef3ffc4cd19ffa9c8c19 Image: registry.fjhb.cn/etcd Image ID: docker-pullable://docker.io/elcolio/etcd@sha256:3b4dcd35a7eefea9ce2970c81dcdf0d0801a778d117735ee1d883222de8bbd9f Port: Command: /bin/etcd --data-dir /tmp/data --listen-client-urls http://127.0.0.1:2379,http://127.0.0.1:4001 --advertise-client-urls http://127.0.0.1:2379,http://127.0.0.1:4001 --initial-cluster-token skydns-etcd Limits: cpu: 100m memory: 50Mi Requests: cpu: 100m memory: 50Mi State: Running Started: Tue, 23 Jan 2018 10:55:23 -0500 Ready: True Restart Count: 0 Volume Mounts: /tmp/data from etcd-storage (rw) /var/run/secrets/kubernetes.io/serviceaccount from default-token-6pddn (ro) Environment Variables: <none> kube2sky: Container ID: docker://6b0bc6e8dce83e3eee5c7e654fbaca693730623fb7936a1fd9d73de1a1dd8152 Image: registry.fjhb.cn/kubernetes-kube2sky Image ID: docker-pullable://docker.io/port/kubernetes-kube2sky@sha256:0230d3fbb0aeb4ddcf903811441cf2911769dbe317a55187f58ca84c95107ff5 Port: Args: -kube_master_url=http://192.168.115.5:8080 -domain=cluster.local Limits: cpu: 100m memory: 50Mi Requests: cpu: 100m memory: 50Mi State: Running Started: Tue, 23 Jan 2018 10:55:25 -0500 Ready: True Restart Count: 0 Volume Mounts: /var/run/secrets/kubernetes.io/serviceaccount from default-token-6pddn (ro) Environment Variables: <none> skydns: Container ID: docker://ebc2aaaa54e2f922e370e454ec537665d813c69d37a21e3afd908e6dad056627 Image: registry.fjhb.cn/skydns Image ID: docker-pullable://docker.io/skynetservices/skydns@sha256:6f8a9cff0b946574bb59804016d3aacebc637581bace452db6a7515fa2df79ee Ports: 53/UDP, 53/TCP Args: -machines=http://127.0.0.1:4001 -addr=0.0.0.0:53 -ns-rotate=false -domain=cluster.local Limits: cpu: 100m memory: 50Mi Requests: cpu: 100m memory: 50Mi State: Running Started: Tue, 23 Jan 2018 10:55:27 -0500 Ready: True Restart Count: 0 Volume Mounts: /var/run/secrets/kubernetes.io/serviceaccount from default-token-6pddn (ro) Environment Variables: <none> healthz: Container ID: docker://f1de1189fa6b51281d414d7a739b86494b04c8271dc6bb5f20c51fac15ec9601 Image: registry.fjhb.cn/healthz Image ID: docker-pullable://docker.io/wu1boy/healthz@sha256:d6690c0a8cc4f810a5e691b6a9b8b035192cb967cb10e91c74824bb4c8eea796 Port: 8080/TCP Args: -cmd=nslookup kubernetes.default.svc.cluster.local 127.0.0.1 >/dev/null -port=8080 Limits: cpu: 10m memory: 20Mi Requests: cpu: 10m memory: 20Mi State: Running Started: Tue, 23 Jan 2018 10:55:29 -0500 Ready: True Restart Count: 0 Volume Mounts: /var/run/secrets/kubernetes.io/serviceaccount from default-token-6pddn (ro) Environment Variables: <none> Conditions: Type Status Initialized True Ready True PodScheduled True Volumes: etcd-storage: Type: EmptyDir (a temporary directory that shares a pod's lifetime) Medium: default-token-6pddn: Type: Secret (a volume populated by a Secret) SecretName: default-token-6pddn QoS Class: Guaranteed Tolerations: <none> Events: FirstSeen LastSeen Count From SubObjectPath Type Reason Message --------- -------- ----- ---- ------------- -------- ------ ------- 7m 7m 1 {default-scheduler } Normal Scheduled Successfully assigned kube-dns-9fllp to 192.168.115.6 7m 7m 1 {kubelet 192.168.115.6} spec.containers{etcd} Normal Pulling pulling image "registry.fjhb.cn/etcd" 7m 7m 1 {kubelet 192.168.115.6} spec.containers{etcd} Normal Pulled Successfully pulled image "registry.fjhb.cn/etcd" 7m 7m 1 {kubelet 192.168.115.6} spec.containers{etcd} Normal Created Created container with docker id 62ad76bfaca1; Security:[seccomp=unconfined] 7m 7m 1 {kubelet 192.168.115.6} spec.containers{kube2sky} Normal Pulled Successfully pulled image "registry.fjhb.cn/kubernetes-kube2sky" 7m 7m 1 {kubelet 192.168.115.6} spec.containers{etcd} Normal Started Started container with docker id 62ad76bfaca1 7m 7m 1 {kubelet 192.168.115.6} spec.containers{kube2sky} Normal Pulling pulling image "registry.fjhb.cn/kubernetes-kube2sky" 7m 7m 1 {kubelet 192.168.115.6} spec.containers{kube2sky} Normal Created Created container with docker id 6b0bc6e8dce8; Security:[seccomp=unconfined] 7m 7m 1 {kubelet 192.168.115.6} spec.containers{skydns} Normal Pulled Successfully pulled image "registry.fjhb.cn/skydns" 7m 7m 1 {kubelet 192.168.115.6} spec.containers{skydns} Normal Pulling pulling image "registry.fjhb.cn/skydns" 7m 7m 1 {kubelet 192.168.115.6} spec.containers{kube2sky} Normal Started Started container with docker id 6b0bc6e8dce8 7m 7m 1 {kubelet 192.168.115.6} spec.containers{skydns} Normal Created Created container with docker id ebc2aaaa54e2; Security:[seccomp=unconfined] 7m 7m 1 {kubelet 192.168.115.6} spec.containers{skydns} Normal Started Started container with docker id ebc2aaaa54e2 7m 7m 1 {kubelet 192.168.115.6} spec.containers{healthz} Normal Pulling pulling image "registry.fjhb.cn/healthz" 7m 7m 1 {kubelet 192.168.115.6} spec.containers{healthz} Normal Pulled Successfully pulled image "registry.fjhb.cn/healthz" 7m 7m 1 {kubelet 192.168.115.6} spec.containers{healthz} Normal Created Created container with docker id f1de1189fa6b; Security:[seccomp=unconfined] 7m 7m 1 {kubelet 192.168.115.6} spec.containers{healthz} Normal Started Started container with docker id f1de1189fa6b
四、修改kubelet配置文件并重启服务
注意:
--cluster-dns参数要和前面svc文件中的clusterIP参数一致
--cluster-domain参数要和前面rc文件中的-domain参数一致
集群内所有的kubelet节点都需要修改
# grep 'KUBELET_ADDRESS' /etc/kubernetes/kubelet KUBELET_ADDRESS="--address=192.168.115.5 --cluster-dns=10.254.16.254 --cluster-domain=cluster.local" # systemctl restart kubelet
五、运行一个busybox和curl进行测试
# cat busybox.yaml apiVersion: v1 kind: Pod metadata: name: busybox spec: containers: - name: busybox image: docker.io/busybox command: - sleep - "3600"# cat curl.yaml apiVersion: v1 kind: Pod metadata: name: curl spec: containers: - name: curl image: docker.io/webwurst/curl-utils command: - sleep - "3600"# kubectl create -f busybox.yaml # kubectl create -f curl.yaml
通过busybox容器对kubernetes的service进行解析，发现service被自动解析成了对应的集群ip地址，而并不是172.16网段的docker地址
# kubectl get svc # kubectl exec busybox -- nslookup frontend # kubectl exec busybox -- nslookup Redis-master # kubectl exec busybox -- nslookup redis-slave
通过curl容器访问前面创建的php留言板
# kubectl exec curl -- curl frontend
