本地k8s环境minikube搭建过程

首先要安装docker这个环境是需要自己安装的。相关步骤如下：

1 2 3 4 5 6 7 8 9 10 11

yum install -y yum-utils device-mapper-persistent-data lvm2 yum-config-manager --add-repo https: //download .docker.com /linux/centos/docker-ce .repo 安装docker yum list docker-ce --showduplicates | sort -r 查看docker相关版本 #yum install docker-ce #由于repo中默认只开启stable仓库，故这里安装的是最新稳定版。 #yum install <FQPN> # 例如：yum install docker-ce-18.06.0.ce -y #以下是验证过的版本，建议安装 yum install docker-ce-18.06.0.ce -y systemctl start docker systemctl enable docker docker version（因为安装的是1.13.4版本的k8s,建议安装docker18.06）

然后使用阿里云修改好的minikube进行安装，否则在初始化minikube的时候会卡在墙上下不来

1 2 3

curl -Lo minikube http: //kubernetes .oss-cn-hangzhou.aliyuncs.com /minikube/releases/v0 .35.0 /minikube-linux-amd64 chmod +x minikube mv minikube /usr/bin/minikube

注意一点要关掉swap：关闭命令swapoff -a

加载阿里云k8s的官方源并且安装相关命令组件

1 2 3 4 5 6 7 8 9 10

cd /etc/yum .repos.d/ cat >>kubernetes.repo<<EOF [kubernetes] name=Kubernetes baseurl=https: //mirrors .aliyun.com /kubernetes/yum/repos/kubernetes-el7-x86_64/ enabled=1 gpgcheck=1 repo_gpgcheck=1 gpgkey=https: //mirrors .aliyun.com /kubernetes/yum/doc/yum-key .gpg https: //mirrors .aliyun.com /kubernetes/yum/doc/rpm-package-key .gpg EOF

1 2	yum install kubectl kubelet kubeadm -y systemctl start kubelet && systemctl enable kubelet

使用缺省VirtualBox驱动来创建Kubernetes本地环境

minikube start --registry-mirror=https: //registry .docker-cn.com

出现如下字样

- Verifying component health .....
+ kubectl is now configured to use "minikube"
= Done! Thank you for using minikube!

则本地的minikube安装完成。当然这个不能访问外网，单独装ingress或者端口转发即可

############################################################################

----------------------------------我是分割线------------------------------------割一下--------------------------------

############################################################################

ingress安装方法：

生成ingress：

创建depolyment.yaml:

apiVersion: v1 kind: Namespace metadata:    name: kube-system    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx --- kind: ConfigMap apiVersion: v1 metadata:    name: nginx-configuration    namespace: kube-system    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx --- kind: ConfigMap apiVersion: v1 metadata:    name: tcp-services    namespace: kube-system    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx --- kind: ConfigMap apiVersion: v1 metadata:    name: udp-services    namespace: kube-system    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx --- apiVersion: v1 kind: ServiceAccount metadata:    name: nginx-ingress-serviceaccount    namespace: kube-system    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx --- apiVersion: rbac.authorization.k8s.io /v1beta1 kind: ClusterRole metadata:    name: nginx-ingress-clusterrole    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx rules:    - apiGroups:        - ""      resources:        - configmaps        - endpoints        - nodes        - pods        - secrets      verbs:        - list        - watch    - apiGroups:        - ""      resources:        - nodes      verbs:        - get    - apiGroups:        - ""      resources:        - services      verbs:        - get        - list        - watch    - apiGroups:        - "extensions"      resources:        - ingresses      verbs:        - get        - list        - watch    - apiGroups:        - ""      resources:        - events      verbs:        - create        - patch    - apiGroups:        - "extensions"      resources:        - ingresses /status      verbs:        - update --- apiVersion: rbac.authorization.k8s.io /v1beta1 kind: Role metadata:    name: nginx-ingress-role    namespace: kube-system    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx rules:    - apiGroups:        - ""      resources:        - configmaps        - pods        - secrets        - namespaces      verbs:        - get    - apiGroups:        - ""      resources:        - configmaps      resourceNames:        # Defaults to "<election-id>-<ingress-class>"        # Here: "<ingress-controller-leader>-<nginx>"        # This has to be adapted if you change either parameter        # when launching the nginx-ingress-controller.        - "ingress-controller-leader-nginx"      verbs:        - get        - update    - apiGroups:        - ""      resources:        - configmaps      verbs:        - create    - apiGroups:        - ""      resources:        - endpoints      verbs:        - get --- apiVersion: rbac.authorization.k8s.io /v1beta1 kind: RoleBinding metadata:    name: nginx-ingress-role-nisa-binding    namespace: kube-system    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx roleRef:    apiGroup: rbac.authorization.k8s.io    kind: Role    name: nginx-ingress-role subjects:    - kind: ServiceAccount      name: nginx-ingress-serviceaccount      namespace: kube-system --- apiVersion: rbac.authorization.k8s.io /v1beta1 kind: ClusterRoleBinding metadata:    name: nginx-ingress-clusterrole-nisa-binding    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx roleRef:    apiGroup: rbac.authorization.k8s.io    kind: ClusterRole    name: nginx-ingress-clusterrole subjects:    - kind: ServiceAccount      name: nginx-ingress-serviceaccount      namespace: kube-system --- apiVersion: apps /v1 kind: Deployment metadata:    name: nginx-ingress-controller    namespace: kube-system    labels:      app.kubernetes.io /name : ingress-nginx      app.kubernetes.io /part-of : ingress-nginx spec:    replicas: 1    selector:      matchLabels:        app.kubernetes.io /name : ingress-nginx        app.kubernetes.io /part-of : ingress-nginx    template:      metadata:        labels:          app.kubernetes.io /name : ingress-nginx          app.kubernetes.io /part-of : ingress-nginx        annotations:          prometheus.io /port : "10254"          prometheus.io /scrape : "true"      spec:        serviceAccountName: nginx-ingress-serviceaccount        containers:          - name: nginx-ingress-controller            image: quay.io /kubernetes-ingress-controller/nginx-ingress-controller :0.23.0            args:              - /nginx-ingress-controller              - --configmap=$(POD_NAMESPACE) /nginx-configuration              - --tcp-services-configmap=$(POD_NAMESPACE) /tcp-services              - --udp-services-configmap=$(POD_NAMESPACE) /udp-services              - --publish-service=$(POD_NAMESPACE) /ingress-nginx              - --annotations-prefix=nginx.ingress.kubernetes.io            securityContext:              allowPrivilegeEscalation: true              capabilities:                drop:                  - ALL                add:                  - NET_BIND_SERVICE              # www-data -> 33              runAsUser: 33            env :              - name: POD_NAME                valueFrom:                  fieldRef:                    fieldPath: metadata.name              - name: POD_NAMESPACE                valueFrom:                  fieldRef:                    fieldPath: metadata.namespace            ports:              - name: http                containerPort: 80              - name: https                containerPort: 443            livenessProbe:              failureThreshold: 3              httpGet:                path: /healthz                port: 10254                scheme: HTTP              initialDelaySeconds: 10              periodSeconds: 10              successThreshold: 1              timeoutSeconds: 10            readinessProbe:              failureThreshold: 3              httpGet:                path: /healthz                port: 10254                scheme: HTTP              periodSeconds: 10              successThreshold: 1              timeoutSeconds: 10 ---

---

---

再创建svc,yaml：

Service:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

apiVersion: v1 kind: Service metadata:    annotations:      #service.beta.kubernetes.io/alicloud-loadbalancer-id: "lb-wz9du18pa4e7f93vetzww"    labels:      app: nginx-ingress    name: nginx-ingress    namespace: kube-system spec:    ports:    - name: http      nodePort: 30468      port: 80      protocol: TCP      targetPort: 80    - name: https      nodePort: 30471      port: 443      protocol: TCP      targetPort: 443    selector:      #app: ingress-nginx      app.kubernetes.io /name : ingress-nginx    #type: LoadBalancer    type : NodePort status:    loadBalancer:      ingress:      - ip: 39.108.26.119（此处更改成自己本机ip）

以上yaml创建pod的命令是：

kubectl apply -f xxxx.yaml

业务镜像可以拉取gitlab的，这里没做cofigmap，需要自己配。生成业务编排需自己编写yaml

以下是简单安装脚本。

#!/bin/bash
#安装docker相关，用以拉取本地所需镜像,版本采用docker-ce 18.06版，支持1.13版kubernetes
#检测网卡是否是固定ip
grep -rE "dhcp" /etc/sysconfig/network-scripts/ifcfg-*
if [ $? -eq 0 ];
then
    echo "网卡为DHCP模式请更改为规定ip"
    exit
else
    echo "网卡正常。"
fi
yum clean all && yum repolist
yum install -y yum-utils device-mapper-persistent-data lvm2
yum-config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo
yum install docker-ce-18.06.0.ce -y
systemctl start docker
systemctl enable docker
VERSION=`docker version`
if [ $? -eq 0 ];
then
    echo "输出docker版本信息：$VERSION"
else
    echo "docker安装出错，请检查错误日志"
    exit
fi
echo "1" >/proc/sys/net/bridge/bridge-nf-call-iptables #此步是保证iptables正确转发获取镜像，否则会报dns解析错误
########获取minikube二进制文件并且添加系统命令########
cd /data
curl -Lo minikube http://kubernetes.oss-cn-hangzhou.aliyuncs.com/minikube/releases/v0.35.0/minikube-linux-amd64
chmod +x minikube
mv minikube /usr/bin/minikube
swapoff -a #强制关闭swap不然初始化的时候会提示错误
cd /etc/yum.repos.d/
cat>>kubernetes.repo<<EOF
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64/
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
EOF
yum install kubectl kubelet kubeadm -y
systemctl start kubelet && systemctl enable kubelet
########启动minikube########
minikube start --vm-driver=none
if [ $? -eq 0 ];
then
    echo "minikube初始化成功"
else
    echo "minikube初始化失败,请检查报错输出，重新执行初始化命令minikube start --vm-driver=none 命令，如果仍有报错，请执行清理集群命令minikube delete，并重新执行初始化命令！"
    minikube delete
    exit
fi
#缺省Minikube使用VirtualBox驱动来创建Kubernetes本地环境
#minikube start --registry-mirror=https://registry.docker-cn.com
STATUS=`kubectl get node | awk '{print$2}' | sed -n '2p'`
if [ $STATUS = "Ready" ];
then
    echo "输出集群状态$STATUS"
else
    echo "输出状态不是Ready，请联系运维."
fi
#echo "输出集群状态$STATUS"
#echo "输出状态不是Ready，请联系运维."