Kubernetes(K8s)(五)——在K8s集群上搭建redis和docker的留言簿
(1).实验环境
需要三个镜像:web前端镜像php-frontend、redis-master和redis-slave。其中web前端通过JavaScript redis api实现与redis-master交互。
(2).扩展:flannel
flannel是K8s默认提供的网络插件。Flannel是由CoreOs团队开发社交的网络工具,CoreOS团队采用L3 Overlay模式设计flannel, 规定宿主机下各个Pod属于同一个子网,不同宿主机下的Pod属于不同的子网。
flannel会在每一个宿主机上运行名为flanneld代理,其负责为宿主机预先分配一个子网,并为Pod分配IP地址。Flannel使用Kubernetes或etcd来存储网络配置、分配的子网和主机公共IP等信息。数据包则通过VXLAN、UDP或host-gw这些类型的后端机制进行转发。
看一下flannel在Kubernetes中运行的整体过程:
(3).准备使用到的镜像
在两个node节点上下载docker.io/kubeguide/guestbook-php-frontend:latest、docker.io/kubeguide/redis-master:latest、docker.io/kubeguide/guestbook-redis-slave:latest。
# docker pull docker.io/kubeguide/guestbook-php-frontend Using default tag: latest Trying to pull repository docker.io/kubeguide/guestbook-php-frontend ... latest: Pulling from docker.io/kubeguide/guestbook-php-frontend 4d2e9ae40c41: Pull complete a3ed95caeb02: Pull complete 54a7f0f294dc: Pull complete a25fe735d07d: Pull complete 0c906d123edf: Pull complete 444908521f56: Pull complete 8c642d1244eb: Pull complete 9b221da943a2: Pull complete 88bb2a5d51e9: Pull complete d74e6c5bcbaf: Pull complete 54d6cf15da9c: Pull complete 55f6a4aed401: Pull complete 45539a0ff9ad: Pull complete 023f5f1f5f40: Pull complete dfdabf8691ea: Pull complete 09a6ba60b929: Pull complete 1b0035990bdd: Pull complete d455473d6864: Pull complete 7200260c49a3: Pull complete 400be507a72c: Pull complete 6d9e0a886b69: Pull complete 66e7bb4a52de: Pull complete Digest: sha256:195181e0263bcee4ae0c3e79352bbd3487224c0042f1b9ca8543b788962188ce Status: Downloaded newer image for docker.io/kubeguide/guestbook-php-frontend:latest # docker pull docker.io/kubeguide/redis-master Using default tag: latest Trying to pull repository docker.io/kubeguide/redis-master ... latest: Pulling from docker.io/kubeguide/redis-master a3ed95caeb02: Pull complete 9cf7e98c567b: Pull complete d2ff49536f4d: Pull complete f94adccdbb9c: Pull complete 6ead1ce47d52: Pull complete 62a3114e0031: Pull complete dcb903eb752c: Pull complete 0c2c7c7f11ef: Pull complete c7aa377bfeeb: Pull complete 3d7aab9ba591: Pull complete 85e54ffe4941: Pull complete 4756da354f92: Pull complete Digest: sha256:e11eae36476b02a195693689f88a325b30540f5c15adbf531caaecceb65f5b4d Status: Downloaded newer image for docker.io/kubeguide/redis-master:latest # docker pull docker.io/kubeguide/guestbook-redis-slave Using default tag: latest Trying to pull repository docker.io/kubeguide/guestbook-redis-slave ... latest: Pulling from docker.io/kubeguide/guestbook-redis-slave 70c964415e86: Pull complete a3ed95caeb02: Pull complete 3229dc640f23: Pull complete cc7958dd97b7: Pull complete 448093e93ac3: Pull complete 2bbcbb97d47b: Pull complete 87b4f6fd65c9: Pull complete 5b712ee108fe: Pull complete a1a961e320bc: Pull complete 6f37ec7ab693: Pull complete Digest: sha256:a36fec97659fe96b5b28750d88b5cfb84a45138bcf1397c8e237031b8855c58c Status: Downloaded newer image for docker.io/kubeguide/guestbook-redis-slave:latest
(4).创建配置文件
创建一个专门存放YAML配置文件的目录
[root@kube-master ~]# mkdir -p /etc/kubernetes/yaml
1)redis master deployment配置文件
[root@kube-master ~]# vim /etc/kubernetes/yaml/redis-master-deployment.yaml apiVersion: extensions/v1beta1 #api版本 kind: Deployment #类型 metadata: #deployment元数据(属性) name: redis-master #deployment名称
# 如果没有设置标签,这些标签可以从Pod模板中应用(获取?) # these labels can be applied automatically from the labels in the pod template if not set # labels: # app: redis # role: master # tier: backend spec: #deployment详细定义
# 此副本值是默认值,请根据需要修改 # this replicas value is default modify it according to your case replicas: 1 #副本期望值
# 如果没有设置选择器,选择器可以从Pod模板中的标签中应用(获取?) # selector can be applied automatically from the labels in the pod template if not set # selector: # matchLabels: # app: guestbook # role: master # tier: backend template: #模板 metadata: #Pod元数据(属性) labels: #自定义标签 app: redis role: master tier: backend spec: #Pod详细定义 containers: #容器 - name: master #容器名称的一部分 image: docker.io/kubeguide/redis-master:latest #采用镜像 imagePullPolicy: IfNotPresent #镜像拉取策略 resources: #容器运行资源限制 requests: #运行下限 cpu: 100m #0.1核 memory: 100Mi #100Mb(1024单位) ports: #对外开放端口列表 - containerPort: 6379 #容器对外开放端口
2)redis master service配置文件
[root@kube-master ~]# vim /etc/kubernetes/yaml/redis-master-service.yaml apiVersion: v1 #api版本 kind: Service #类型 metadata: #service元数据(属性) name: redis-master #service名称 labels: #自定义标签 app: redis role: master tier: backend spec: #service详细定义 ports: #服务对外开放端口列表 # the port that this service should serve on - port: 6379 #提供给内部Pod访问使用的端口 targetPort: 6379 #Pod内服务的端口 selector: #service作用范围,此处是同时具备以下标签时,采用当前service app: redis role: master tier: backend
3)redis slave deployment配置文件
[root@kube-master ~]# vim /etc/kubernetes/yaml/redis-slave-deployment.yaml apiVersion: extensions/v1beta1 #api版本 kind: Deployment #类型 metadata: #deployment元数据(属性) name: redis-slave #deployment名称 # 如果没有设置标签,这些标签可以从Pod模板中应用(获取?) # these labels can be applied automatically from the labels in the pod template if not set # labels: # app: redis # role: slave # tier: backend spec: #deployment详细定义 # 此副本值是默认值,请根据需要修改 # this replicas value is default modify it according to your case replicas: 2 #副本期望值 # 如果没有设置选择器,选择器可以从Pod模板中的标签中应用(获取?) # selector can be applied automatically from the labels in the pod template if not set # selector: # matchLabels: # app: guestbook # role: slave # tier: backend template: #模板 metadata: #Pod元数据(属性) labels: #自定义标签 app: redis role: slave tier: backend spec: #Pod详细定义 containers: #容器 - name: slave #容器名称的一部分 image: docker.io/kubeguide/guestbook-redis-slave:latest #采用镜像 imagePullPolicy: IfNotPresent #镜像拉取策略 resources: #容器运行资源限制 requests: #运行下限 cpu: 100m #0.1核 memory: 100Mi #100Mb(1024单位) env: #环境变量设置 - name: GET_HOSTS_FROM #环境变量名称 value: env #环境变量值 # 如果集群配置不包括DNS服务,则要访问环境变量来查找主服务的主机,请注释掉上面的'value: dns',并取消注释下面的行 # If your cluster config does not include a dns service, then to # instead access an environment variable to find the master # service's host, comment out the 'value: dns' line above, and # uncomment the line below. # value: env ports: #对外开放端口列表 - containerPort: 6379 #容器对外开放端口
4)redis slave service配置文件
[root@kube-master ~]# vim /etc/kubernetes/yaml/redis-slave-service.yaml apiVersion: v1 #api版本 kind: Service #类型 metadata: #service元数据(属性) name: redis-slave #service名称 labels: #自定义标签 app: redis role: slave tier: backend spec: #service详细定义 ports: #服务对外开放端口列表 # the port that this service should serve on - port: 6379 #提供给内部Pod访问使用的端口 selector: #service作用范围,此处同时具备以下标签时,采用当前service app: redis role: slave tier: backend
5)php frontend deployment配置文件
[root@kube-master ~]# vim /etc/kubernetes/yaml/frontend-deployment.yaml apiVersion: extensions/v1beta1 #api版本 kind: Deployment #类型 metadata: #deployment元数据(属性) name: frontend #deployment名称 # 如果没有设置标签,这些标签可以从Pod模板中应用(获取?) # these labels can be applied automatically from the labels in the pod template if not set # labels: # app: guestbook # tier: frontend spec: #deployment详细定义 # 此副本值是默认值,请根据需要修改 # this replicas value is default modify it according to your case replicas: 3 #副本期望值 # 如果没有设置选择器,选择器可以从Pod模板中的标签中应用(获取?) # selector can be applied automatically from the labels in the pod template if not set # selector: # matchLabels: # app: guestbook # tier: frontend template: #模板 metadata: #Pod元数据(属性) labels: #自定义标签 app: guestbook tier: frontend spec: #Pod详细定义 containers: #容器 - name: php-redis #容器名称的一部分 image: docker.io/kubeguide/guestbook-php-frontend:latest #采用镜像 imagePullPolicy: IfNotPresent #镜像拉取策略 resources: #容器运行资源限制 requests: #下限 cpu: 100m #0.1核 memory: 100Mi #100Mb(1024单位) env: #环境变量设置 - name: GET_HOSTS_FROM #环境变量名称 value: env #环境变量值 # 如果集群配置不包含DNS服务,则要访问环境变量来查找主服务的主机,请注释掉上面的'value: dns',并取消注释下面的行 # If your cluster config does not include a dns service, then to # instead access environment variables to find service host # info, comment out the 'value: dns' line above, and uncomment the # line below. # value: env ports: #对外开放端口列表 - containerPort: 80 #容器对外开放端口
6)php frontend service配置文件
[root@kube-master ~]# vim /etc/kubernetes/yaml/frontend-service.yaml apiVersion: v1 #api版本 kind: Service #类型 metadata: #service元数据(属性) name: frontend #service名称 labels: #自定义标签 app: guestbook tier: frontend spec: #service详细定义 # if your cluster supports it, uncomment the following to automatically creat e # an external load-balanced IP for the frontend service. # type: LoadBalancer type: NodePort #服务访问方式 ports: #服务对外开放端口列表 # the port that this service should serve on - port: 80 #提供给内部Pod访问使用的端口 nodePort: 30001 #提供给外部访问服务的节点上的端口 selector: #service作用范围,此处同时具备以下标签时,采用当前service app: guestbook tier: frontend
(5).通过YAML文件启动
启动deployment和service
[root@kube-master ~]# kubectl create -f /etc/kubernetes/yaml/redis-master-deployment.yaml deployment "redis-master" created [root@kube-master ~]# kubectl create -f /etc/kubernetes/yaml/redis-master-service.yaml service "redis-master" created [root@kube-master ~]# kubectl create -f /etc/kubernetes/yaml/redis-slave-deployment.yaml deployment "redis-slave" created [root@kube-master ~]# kubectl create -f /etc/kubernetes/yaml/redis-slave-service.yaml service "redis-slave" created [root@kube-master ~]# kubectl create -f /etc/kubernetes/yaml/frontend-deployment.yaml deployment "frontend" created [root@kube-master ~]# kubectl create -f /etc/kubernetes/yaml/frontend-service.yaml service "frontend" created
查看deployment和service,下方数据为清理掉非相关内容后的数据。
[root@kube-master ~]# kubectl get deployment -o wide NAME DESIRED CURRENT UP-TO-DATE AVAILABLE AGE frontend 3 3 3 3 6m redis-master 1 1 1 1 6m redis-slave 2 2 2 2 6m [root@kube-master ~]# kubectl get replicaset -o wide NAME DESIRED CURRENT READY AGE CONTAINER(S) IMAGE(S) SELECTOR frontend-1186687533 3 3 3 6m php-redis docker.io/kubeguide/guestbook-php-frontend:latest app=guestbook,pod-template-hash=1186687533,tier=frontend redis-master-3671804942 1 1 1 7m master docker.io/kubeguide/redis-master:latest app=redis,pod-template-hash=3671804942,role=master,tier=backend redis-slave-2377017994 2 2 2 7m slave docker.io/kubeguide/guestbook-redis-slave:latest app=redis,pod-template-hash=2377017994,role=slave,tier=backend [root@kube-master ~]# kubectl get pod -o wide NAME READY STATUS RESTARTS AGE IP NODE frontend-1186687533-4jns2 1/1 Running 0 7m 10.255.31.8 kube-node2 frontend-1186687533-dqcdj 1/1 Running 0 7m 10.255.39.4 kube-node1 frontend-1186687533-vg8fw 1/1 Running 0 7m 10.255.39.5 kube-node1 redis-master-3671804942-8tq84 1/1 Running 0 8m 10.255.39.2 kube-node1 redis-slave-2377017994-1zctb 1/1 Running 0 8m 10.255.31.7 kube-node2 redis-slave-2377017994-rqr4m 1/1 Running 0 8m 10.255.39.3 kube-node1 [root@kube-master ~]# kubectl get service -o wide NAME CLUSTER-IP EXTERNAL-IP PORT(S) AGE SELECTOR frontend 10.254.186.13 <nodes> 80:30001/TCP 8m app=guestbook,tier=frontend kubernetes 10.254.0.1 <none> 443/TCP 42d <none> redis-master 10.254.137.74 <none> 6379/TCP 8m app=redis,role=master,tier=backend redis-slave 10.254.72.23 <none> 6379/TCP 8m app=redis,role=slave,tier=backend
(6).开启node节点上的路由转发功能
临时开启方法一:
# iptables -P FORWARD ACCEPT
临时开启方法二:
# echo "1" > /proc/sys/net/ipv4/ip_forward
永久开启
# vim /etc/sysctl.conf //添加或修改如下行 net.ipv4.ip_forward = 1 # sysctl -p //刷新使参数生效 net.ipv4.ip_forward = 1
(7).通过浏览器访问