k8s、kubeadm安装

master（2C/4G，cpu核心数要求大于2）  192.168.86.11		docker、kubeadm、kubelet、kubectl、flannel
node01（2C/2G）	192.168.86.22		docker、kubeadm、kubelet、kubectl、flannel
node02（2C/2G）	192.168.86.55		docker、kubeadm、kubelet、kubectl、flannel
Harbor节点（hub.kgc.com）	 192.168.86.33		docker、docker-compose、harbor-offline-v1.2.21、在所有节点上安装Docker和kubeadm
2、部署Kubernetes Master
3、部署容器网络插件
4、部署 Kubernetes Node，将节点加入Kubernetes集群中
5、部署 Dashboard Web 页面，可视化查看Kubernetes资源
6、部署 Harbor 私有仓库，存放镜像资源

环境准备

#所有节点，关闭防火墙规则，关闭selinux，关闭swap交换
systemctl stop firewalld
systemctl disable firewalld
setenforce 0
sed -i 's/enforcing/disabled/' /etc/selinux/config
iptables -F && iptables -t nat -F && iptables -t mangle -F && iptables -X
swapoff -a						#交换分区必须要关闭
sed -ri 's/.*swap.*/#&/' /etc/fstab		#永久关闭swap分区，&符号在sed命令中代表上次匹配的结果#加载 ip_vs 模块
for i in $(ls /usr/lib/modules/$(uname -r)/kernel/net/netfilter/ipvs|grep -o "^[^.]*");do echo $i; /sbin/modinfo -F filename $i >/dev/null 2>&1 && /sbin/modprobe $i;done//修改主机名
hostnamectl set-hostname master01
hostnamectl set-hostname node01
hostnamectl set-hostname node02//所有节点修改hosts文件
vim /etc/hosts
192.168.86.11 master01
192.168.86.22 node01
192.168.86.55 node02//调整内核参数
cat > /etc/sysctl.d/kubernetes.conf << EOF
#开启网桥模式，可将网桥的流量传递给iptables链
net.bridge.bridge-nf-call-ip6tables=1
net.bridge.bridge-nf-call-iptables=1
#关闭ipv6协议
net.ipv6.conf.all.disable_ipv6=1
net.ipv4.ip_forward=1
EOF//生效参数
sysctl --system  

所有节点安装docker

yum install -y yum-utils device-mapper-persistent-data lvm2 
yum-config-manager --add-repo https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo 
yum install -y docker-ce docker-ce-cli containerd.iomkdir /etc/docker
cat > /etc/docker/daemon.json <<EOF
{"registry-mirrors":  ["https://xtx10u0l.mirror.aliyuncs.com"],"exec-opts": ["native.cgroupdriver=systemd"],"log-driver": "json-file","log-opts": {"max-size": "100m"}
}
EOF
#使用Systemd管理的Cgroup来进行资源控制与管理，因为相对Cgroupfs而言，Systemd限制CPU、内存等资源更加简单和成熟稳定。
#日志使用json-file格式类型存储，大小为100M，保存在/var/log/containers目录下，方便ELK等日志系统收集和管理日志。systemctl daemon-reload
systemctl restart docker.service
systemctl enable docker.service docker info | grep "Cgroup Driver"
Cgroup Driver: systemd

 所有节点安装kubeadm，kubelet和kubectl

#定义kubernetes源
cat > /etc/yum.repos.d/kubernetes.repo << EOF
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=0
repo_gpgcheck=0
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
EOFyum install -y kubelet-1.20.11 kubeadm-1.20.11 kubectl-1.20.11//开机自启kubelet
systemctl enable kubelet.service
#K8S通过kubeadm安装出来以后都是以Pod方式存在，即底层是以容器方式运行，所以kubelet必须设置开机自启

 部署K8S集群

#查看初始化需要的镜像
kubeadm config images list#在 master 节点上传 v1.20.11.zip 压缩包至 /opt 目录
mkdir /opt/k8s/
unzip v1.20.11.zip -d /opt/k8s
cd /opt/k8s/v1.20.11
for i in $(ls *.tar); do docker load -i $i; done#复制镜像和脚本到 node 节点，并在 node 节点上执行脚本加载镜像文件
scp -r /opt/k8s root@node01:/opt
scp -r /opt/k8s root@node02:/opt//初始化kubeadm
方法一：
kubeadm config print init-defaults > /opt/kubeadm-config.yamlcd /opt/
vim kubeadm-config.yaml
......
11行 localAPIEndpoint:
12行  advertiseAddress: 192.168.86.11		#指定master节点的IP地址
13行   bindPort: 6443
......
34行 kubernetesVersion: v1.20.11		#指定kubernetes版本号
35行 networking:
36行   dnsDomain: cluster.local
37行   podSubnet: "10.244.0.0/16"		#指定pod网段，10.244.0.0/16用于匹配flannel默认网段
38   serviceSubnet: 10.96.0.0/16	#指定service网段
39 scheduler: {}
#末尾再添加以下内容
--- 
apiVersion: kubeproxy.config.k8s.io/v1alpha1
kind: KubeProxyConfiguration
mode: ipvs									#把默认的kube-proxy调度方式改为ipvs模式

kubeadm init --config=kubeadm-config.yaml --upload-certs | tee kubeadm-init.log
#--experimental-upload-certs 参数可以在后续执行加入节点时自动分发证书文件，K8S V1.16版本开始替换为 --upload-certs#tee kubeadm-init.log 用以输出日志#查看 kubeadm-init 日志
less kubeadm-init.log#kubernetes配置文件目录
ls /etc/kubernetes/#存放ca等证书和密码的目录
ls /etc/kubernetes/pki方法二：
kubeadm init \
--apiserver-advertise-address=192.168.86.11 \
--image-repository registry.aliyuncs.com/google_containers \
--kubernetes-version=v1.20.11 \
--service-cidr=10.96.0.0/16 \
--pod-network-cidr=10.244.0.0/16 \
--token-ttl=0方法二初始化后需要修改 kube-proxy 的 configmap，开启 ipvs
kubectl edit cm kube-proxy -n=kube-system
修改mode: ipvs

#设定kubectl
kubectl需经由API server认证及授权后方能执行相应的管理操作，kubeadm 部署的集群为其生成了一个具有管理员权限的认证配置文件 /etc/kubernetes/admin.conf，它可由 kubectl 通过默认的 “$HOME/.kube/config” 的路径进行加载。mkdir -p $HOME/.kube
cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
chown $(id -u):$(id -g) $HOME/.kube/config#如果 kubectl get cs 发现集群不健康，更改以下两个文件
vim /etc/kubernetes/manifests/kube-scheduler.yaml 
vim /etc/kubernetes/manifests/kube-controller-manager.yaml#修改如下内容
把--bind-address=127.0.0.1变成--bind-address=192.168.86.11		#修改成k8s的控制节点master01的ip
把httpGet:字段下的hosts由127.0.0.1变成192.168.86.11（有两处）
#- --port=0		# 搜索port=0，把这一行注释掉systemctl restart kubelet

 

所有节点部署网络插件flannel
方法一：
//所有节点上传flannel镜像 flannel.tar 到 /opt 目录，master节点上传 kube-flannel.yml 文件
cd /opt
docker load < flannel.tar//在 master 节点创建 flannel 资源
kubectl apply -f kube-flannel.yml 方法二：
kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.ymlkubectl apply -f https://github.com/flannel-io/flannel/releases/latest/download/kube-flannel.yml//在 node 节点上执行 kubeadm join 命令加入群集
kubeadm join 192.168.86.11:6443 --token rc0kfs.a1sfe3gl4dvopck5 \--discovery-token-ca-cert-hash sha256:864fe553c812df2af262b406b707db68b0fd450dc08b34efb73dd5a4771d37a2

 

#在master节点查看节点状态
kubectl get nodeskubectl get pods -n kube-system
NAME                             READY   STATUS    RESTARTS   AGE
coredns-bccdc95cf-c9w6l          1/1     Running   0          71m
coredns-bccdc95cf-nql5j          1/1     Running   0          71m
etcd-master                      1/1     Running   0          71m
kube-apiserver-master            1/1     Running   0          70m
kube-controller-manager-master   1/1     Running   0          70m
kube-flannel-ds-amd64-kfhwf      1/1     Running   0          2m53s
kube-flannel-ds-amd64-qkdfh      1/1     Running   0          46m
kube-flannel-ds-amd64-vffxv      1/1     Running   0          2m56s
kube-proxy-558p8                 1/1     Running   0          2m53s
kube-proxy-nwd7g                 1/1     Running   0          2m56s
kube-proxy-qpz8t                 1/1     Running   0          71m
kube-scheduler-master            1/1     Running   0          70m#测试 pod 资源创建
kubectl create deployment nginx --image=nginxkubectl get pods -o wide
NAME                     READY   STATUS    RESTARTS   AGE   IP           NODE     NOMINATED NODE   READINESS GATES
nginx-554b9c67f9-zr2xs   1/1     Running   0          14m   10.244.1.2   node01   <none>           <none>#暴露端口提供服务
kubectl expose deployment nginx --port=80 --type=NodePortkubectl get svc
NAME         TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
kubernetes   ClusterIP   10.96.0.1      <none>        443/TCP        3h57m
myapp-ky20   NodePort    10.96.56.120   <none>        80:32404/TCP   3s#测试访问
curl http://node01:32404

 部署 Dashboard

#在 master01 节点上操作
#上传 recommended.yaml 文件到 /opt/k8s 目录中
cd /opt/k8s
vim recommended.yaml
#默认Dashboard只能集群内部访问，修改Service为NodePort类型，暴露到外部：
kind: Service
apiVersion: v1
metadata:labels:k8s-app: kubernetes-dashboardname: kubernetes-dashboardnamespace: kubernetes-dashboard
spec:ports:- port: 443targetPort: 8443nodePort: 30001     #添加type: NodePort          #添加selector:k8s-app: kubernetes-dashboardkubectl apply -f recommended.yaml#创建service account并绑定默认cluster-admin管理员集群角色
kubectl create serviceaccount dashboard-admin -n kube-system
kubectl create clusterrolebinding dashboard-admin --clusterrole=cluster-admin --serviceaccount=kube-system:dashboard-admin
kubectl describe secrets -n kube-system $(kubectl -n kube-system get secret | awk '/dashboard-admin/{print $1}')#使用输出的token登录Dashboard
https://NodeIP:30001

 安装Harbor私有仓库（192.168.86.33）

#修改主机名
hostnamectl set-hostname hub.kgc.com#所有节点加上主机名映射
echo '192.168.10.23 hub.kgc.com' >> /etc/hosts#安装 docker
yum install -y yum-utils device-mapper-persistent-data lvm2 
yum-config-manager --add-repo https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo 
yum install -y docker-ce docker-ce-cli containerd.iomkdir /etc/docker
cat > /etc/docker/daemon.json <<EOF
{"registry-mirrors":  ["https://xtx10u0l.mirror.aliyuncs.com"],"exec-opts": ["native.cgroupdriver=systemd"],"log-driver": "json-file","log-opts": {"max-size": "100m"},"insecure-registries": ["https://hub.kgc.com"]
}
EOFsystemctl start docker
systemctl enable docker#所有 node 节点都修改 docker 配置文件，加上私有仓库配置
cat > /etc/docker/daemon.json <<EOF
{"registry-mirrors":  ["https://xtx10u0l.mirror.aliyuncs.com"],"exec-opts": ["native.cgroupdriver=systemd"],"log-driver": "json-file","log-opts": {"max-size": "100m"},"insecure-registries": ["https://hub.kgc.com"]
}
EOFsystemctl daemon-reload
systemctl restart docker#安装 Harbor
#上传 harbor-offline-installer-v1.2.2.tgz 和 docker-compose 文件到 /opt 目录
cd /opt
cp docker-compose /usr/local/bin/
chmod +x /usr/local/bin/docker-composetar zxvf harbor-offline-installer-v1.2.2.tgz
cd harbor/
vim harbor.cfg
5行  hostname = hub.kgc.com
9行  ui_url_protocol = https
24行 ssl_cert = /data/cert/server.crt
25行 ssl_cert_key = /data/cert/server.key
59行 harbor_admin_password = Harbor12345生成证书
mkdir -p /data/cert
cd /data/cert
#生成私钥
openssl genrsa -des3 -out server.key 2048
输入两遍密码：123456#生成证书签名请求文件
openssl req -new -key server.key -out server.csr
输入私钥密码：123456
输入国家名：CN
输入省名：BJ
输入市名：BJ
输入组织名：KGC
输入机构名：KGC
输入域名：hub.kgc.com
输入管理员邮箱：[e-mail]
其它全部直接回车#备份私钥
cp server.key server.key.org#清除私钥密码
openssl rsa -in server.key.org -out server.key
输入私钥密码：123456#签名证书
openssl x509 -req -days 1000 -in server.csr -signkey server.key -out server.crtchmod +x /data/cert/*cd /opt/harbor/
./install.sh在本地使用火狐浏览器访问：https://hub.kgc.com
添加例外 -> 确认安全例外
用户名：admin
密码：Harbor12345

#在一个node节点上登录harbor
docker login -u admin -p Harbor12345 https://hub.kgc.com#上传镜像
docker tag nginx:latest hub.kgc.com/library/nginx:v1
docker push hub.kgc.com/library/nginx:v1#在master节点上删除之前创建的nginx资源
kubectl delete deployment nginxkubectl create deployment nginx-deployment --image=hub.kgc.com/library/nginx:v1 --port=80 --replicas=3kubectl expose deployment nginx-deployment --port=30000 --target-port=80kubectl get svc,pods
NAME                       TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)     AGE
service/kubernetes         ClusterIP   10.96.0.1       <none>        443/TCP     10m
service/nginx-deployment   ClusterIP   10.96.222.161   <none>        30000/TCP   3m15sNAME                                    READY   STATUS    RESTARTS   AGE
pod/nginx-deployment-77bcbfbfdc-bv5bz   1/1     Running   0          16s
pod/nginx-deployment-77bcbfbfdc-fq8wr   1/1     Running   0          16s
pod/nginx-deployment-77bcbfbfdc-xrg45   1/1     Running   0          3m39syum install ipvsadm -y
ipvsadm -Lncurl 10.96.222.161:30000kubectl edit svc nginx-deployment
25   type: NodePort						#把调度策略改成NodePortkubectl get svc
NAME                       TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)           AGE
service/kubernetes         ClusterIP   10.96.0.1       <none>        443/TCP           29m
service/nginx-deployment   NodePort    10.96.222.161   <none>        30000:32034/TCP   22m浏览器访问：
192.168.86.11:32034
192.168.86.22:32034
192.168.86.55:32034#将cluster-admin角色权限授予用户system:anonymous
kubectl create clusterrolebinding cluster-system-anonymous --clusterrole=cluster-admin --user=system:anonymous