本地搭建kubernetes（版本为1.17.0）集群的步骤

1 整体环境

本地win7环境，8c32g，安装了vmware，在vmware中安装了两台centos7.6 x64的机器（vm），分别如下：

编号	IP	配置	说明
1	192.168.117.132	2c2g	k8s master节点
2	192.168.117.133	8c16g	k8s node节点

注：这2台vm都已经安装了docker，详见上一篇文章

2 配置k8s master节点

2.1 环境准备：关闭防火墙、selinux、开启net.bridge.bridge-nf-call-iptables、关闭swap分区

systemctl stop firewalld.service
systemctl disable firewalld.service

setenforce 0
[root@192 ~]# cat /etc/selinux/config
#修改 SELINUX=permissive

sysctl -w net.bridge.bridge-nf-call-iptables=1
echo "net.bridge.bridge-nf-call-iptables=1" > /etc/sysctl.d/k8s.conf

swapoff -a && sed -i '/ swap / s/^/#/' /etc/fstab

2.2 配置kubernetes yum源，安装kubelet、kubeadm

#编辑文件/etc/yum.repos.d/kubernetes.repo ， 内容如下
[root@192 ~]# cat /etc/yum.repos.d/kubernetes.repo 
[kubernetes]
name=Kubernetes Repo
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64/
gpgcheck=1
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
enabled=1

yum list kubelet --showduplicates | sort -r
yum install kubelet-1.17.0-0
yum install kubeadm-1.17.0-0

#启用kubelet：其实不用，不管是运行kubeadm init还是kubeadm join都会自动启动kubelet服务
#systemctl start kubelet
#systemctl enable kubelet

2.3 运行kubeadm init 初始化k8s主节点

kubeadm init --image-repository registry.aliyuncs.com/google_containers

#看到类似下面的输出，就成功了
Your Kubernetes control-plane has initialized successfully!

To start using your cluster, you need to run the following as a regular user:

  mkdir -p $HOME/.kube
  sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
  sudo chown $(id -u):$(id -g) $HOME/.kube/config

You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
  https://kubernetes.io/docs/concepts/cluster-administration/addons/

Then you can join any number of worker nodes by running the following on each as root:

kubeadm join 192.168.117.132:6443 --token 6d0ov6.2q3ocbaz7z1x1590 \
    --discovery-token-ca-cert-hash sha256:759eab4d567cb211b3e2e2307e06fbe25d321f31746cf14c8b6d546034330de3

注：需要指定镜像仓库地址

2.4 在k8s主节点上验证一下

#kubectl配置自动补全
source <(kubectl completion bash)
echo "source <(kubectl completion bash)" >> ~/.bashrc

#按照kubeadm init最后的提示，配置$HOME/.kube/config

kubectl get po --all-namespaces

#可以看到主节点，且状态为NotReady
[root@192 ~]# kubectl get nodes 
NAME              STATUS     ROLES    AGE     VERSION
192.168.117.132   NotReady   master   5m30s   v1.17.0

3 配置k8s node节点

3.1 环境准备：关闭防火墙、selinux、开启net.bridge.bridge-nf-call-iptables、关闭swap分区(同master节点)

3.2 配置kubernetes yum源，安装kubelet、kubeadm(同master节点)

3.3 运行kubeadm join命令（按照kubeadm init最后的提示）

master节点上生成的join token默认一天后过期，参考：https://www.cnblogs.com/chenzhenqi/p/10695959.html

#看到类似下面的信息，就成功了
This node has joined the cluster:
* Certificate signing request was sent to apiserver and a response was received.
* The Kubelet was informed of the new secure connection details.

Run 'kubectl get nodes' on the control-plane to see this node join the cluster.

3.4 验证

#在master节点上执行命令
[root@192 ~]# kubectl get nodes 
NAME              STATUS     ROLES    AGE     VERSION
192.168.117.132   NotReady   master   11m     v1.17.0
k8sslave1         NotReady   <none>   2m14s   v1.17.0

4 配置k8s集群网络环境

#在master节点上执行命令
curl -o net.yaml https://cloud.weave.works/k8s/net?k8s-version=$(kubectl version | base64 | tr -d '\n')
kubectl apply -f net.yaml

#等待3分钟，在master节点运行命令进行验证
[root@192 ~]# kubectl get nodes
NAME              STATUS   ROLES    AGE   VERSION
192.168.117.132   Ready    master   56m   v1.17.0
k8sslave1         Ready    <none>   46m   v1.17.0

5 本地学习的k8s集群搭建完成！，后续会在这个集群上运行一个完整的springcloud测试

[root@192 ~]# kubectl get pod --all-namespaces -o wide
NAMESPACE     NAME                                      READY   STATUS    RESTARTS   AGE   IP                NODE              NOMINATED NODE   READINESS GATES
kube-system   coredns-9d85f5447-h7dk6                   1/1     Running   0          58m   10.32.0.3         192.168.117.132   <none>           <none>
kube-system   coredns-9d85f5447-t622p                   1/1     Running   0          58m   10.32.0.2         192.168.117.132   <none>           <none>
kube-system   etcd-192.168.117.132                      1/1     Running   0          58m   192.168.117.132   192.168.117.132   <none>           <none>
kube-system   kube-apiserver-192.168.117.132            1/1     Running   0          58m   192.168.117.132   192.168.117.132   <none>           <none>
kube-system   kube-controller-manager-192.168.117.132   1/1     Running   0          58m   192.168.117.132   192.168.117.132   <none>           <none>
kube-system   kube-proxy-c78dt                          1/1     Running   0          48m   192.168.117.133   k8sslave1         <none>           <none>
kube-system   kube-proxy-rzrw5                          1/1     Running   0          58m   192.168.117.132   192.168.117.132   <none>           <none>
kube-system   kube-scheduler-192.168.117.132            1/1     Running   0          58m   192.168.117.132   192.168.117.132   <none>           <none>
kube-system   weave-net-jwq64                           2/2     Running   0          36m   192.168.117.133   k8sslave1         <none>           <none>
kube-system   weave-net-pwxjp                           2/2     Running   0          36m   192.168.117.132   192.168.117.132   <none>           <none>

6 安装kubernetes dashboard

参考： https://blog.51cto.com/shunzi115/2447601

7 部署一个简单的nodejs应用

#因为是从阿里云私服拉取镜像，所以需要先配置一个secret对象
#参考：https://www.cnblogs.com/unchch/p/11771005.html
#注意！！！其中的xxx需要替换为你自己的用户名、密码、邮箱
kubectl create secret docker-registry registry-secret --docker-server=registry.cn-hangzhou.aliyuncs.com --docker-username=xxx@163.com --docker-password=xxx --docker-email=xxx@163.com -n default


[root@192 ~]# cat kubia.yaml 
apiVersion: apps/v1
kind: Deployment
metadata:
  name: kubia-web-demo
spec:
  selector:
    matchLabels:
      app: kubia-web-demo
  replicas: 1
  template:
    metadata:
      labels:
        app: kubia-web-demo
    spec:
      containers:
      - name: kubia
        image: registry.cn-hangzhou.aliyuncs.com/jishusc/kubia
        ports:
        - containerPort: 8380
      imagePullSecrets:
      - name: registry-secret
---

apiVersion: v1
kind: Service
metadata:
  name: kubia-web-demo
spec:
  type: NodePort
  ports:
    - port: 8380
      targetPort: 8380
      nodePort: 30000
  selector:
    app: kubia-web-demo

8 部署镜像出错时的一些查看命令

#重点看 最后的Events
kubectl describe pod -n kubernetes-dashboard kubernetes-dashboard-6b86b44f87-qqtt5
kubectl logs -n kubernetes-dashboard kubernetes-dashboard-6b86b44f87-n8cnk --tail=20