目录
一、环境配置
1、机器部署
2、部署大致流程
二、实验环境配置
1、所有节点关闭防火墙核心防护以及关闭swap交换
2、所有节点安装docker
3、所有节点安装kubeadm，kubelet和kubectl
4、部署K8s集群
5、设定kubectl
6、所有节点部署网络插件flannel
7、部署Dashboard
三、安装harbor私有仓库
一、环境配置
1、机器部署
master（2C/4G） 192.168.170.111 docker、kubeadm、kubelet、kubectl、flannel
node01（2C/2G） 192.168.170.113 docker、kubeadm、kubelet、kubectl、flannel
node02（2C/2G） 192.168.170.114 docker、kubeadm、kubelet、kubectl、flannel
Harbor节点（hub.rmh.com）192.168.170.115 docker、docker-compose、harbor-offline-v1.2.2
2、部署大致流程
①在所有节点安装docker和kubeadm
②部署kubernetes Master
③部署容器网络插件
④部署kubernets 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
二、实验环境配置
1、所有节点关闭防火墙核心防护以及关闭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
192.168.170.111 master01
192.168.170.113 node01
192.168.170.114 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 //时间同步
yum install ntpdate -y
ntpdate time.windows.com
2、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://6ijb8ubo.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
3、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-1.22.2 kubeadm-1.22.2 kubectl-1.22.2//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
4、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
unzip v1.20.11.zip -d /opt/k8s
cd /opt/k8s/v1.20.11
for i in $(ls *.tar); do docker load -i $i; done//复制镜像和脚本到 node 节点，并在 node 节点上执行脚本加载镜像文件
scp -r /opt/k8s root@node01:/opt
scp -r /opt/k8s root@node02:/opt//初始化kubeadm
方法一：
kubeadm config print init-defaults > /opt/kubeadm-config.yamlcd /opt/
vim kubeadm-config.yaml
......
11 localAPIEndpoint:
12 advertiseAddress: 192.168.170.111 #指定master节点的IP地址
13 bindPort: 6443
......
34 kubernetesVersion: v1.20.11 #指定kubernetes版本号
35 networking:
36 dnsDomain: cluster.local
37 podSubnet: "10.244.0.0/16" #指定pod网段，10.244.0.0/16用于匹配flannel默认网段
38 serviceSubnet: 10.96.0.0/16 #指定service网段
39 scheduler: {}
#末尾再添加以下内容
---
apiVersion: kubeproxy.config.k8s.io/v1alpha1
kind: KubeProxyConfiguration
mode: ipvs #把默认的kube-proxy调度方式改为ipvs模式kubeadm init --config=kubeadm-config.yaml --upload-certs | tee kubeadm-init.log
#--experimental-upload-certs 参数可以在后续执行加入节点时自动分发证书文件，K8S V1.16版本开始替换为 --upload-certs
#tee kubeadm-init.log 用以输出日志
//查看 kubeadm-init 日志
less kubeadm-init.log//kubernetes配置文件目录
ls /etc/kubernetes///存放ca等证书和密码的目录
ls /etc/kubernetes/pki 方法二：
kubeadm init \
--apiserver-advertise-address=192.168.10.19 \
--image-repository registry.aliyuncs.com/google_containers \
--kubernetes-version=v1.20.11 \
--service-cidr=10.96.0.0/16 \
--pod-network-cidr=10.244.0.0/16 \
--token-ttl=0
查看kubeadm-init日志
kubernetes配置文件目录
存放ca等证书和密码的目录
5、设定kubectl
//设定kubectl
kubectl需经由API server认证及授权后方能执行相应的管理操作，kubeadm 部署的集群为其生成了一个具有管理员权限的认证配置文件 /etc/kubernetes/admin.conf，它可由 kubectl 通过默认的 “$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=192.168.170.111 #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192.168.170.111（有两处）
#- --port=0 # 搜索port=0，把这一行注释掉systemctl restart kubelet
有两处host需要修改
6、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 192.168.170.111:6443 --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//扩展3个副本
kubectl scale deployment nginx --replicas=3
kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx-554b9c67f9-9kh4s 1/1 Running 0 66s 10.244.1.3 node01 <none> <none>
nginx-554b9c67f9-rv77q 1/1 Running 0 66s 10.244.2.2 node02 <none> <none>
nginx-554b9c67f9-zr2xs 1/1 Running 0 17m 10.244.1.2 node01 <none> <none>
在node节点执行kubeadm.join命令加入集群
复制此内容去节点执行加入集群
7、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
三、安装harbor私有仓库
192.168.170.115
//修改主机名
hostnamectl set-hostname hub.benet.com//所有节点加上主机名映射
echo '192.168.170.115 hub.rmh.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://6ijb8ubo.mirror.aliyuncs.com"],"exec-opts": ["native.cgroupdriver=systemd"],"log-driver": "json-file","log-opts": {"max-size": "100m"},"insecure-registries": ["https://hub.benet.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://6ijb8ubo.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 hostname = hub.benet.com
9 ui_url_protocol = https
24 ssl_cert = /data/cert/server.crt
25 ssl_cert_key = /data/cert/server.key
59 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：BENET
输入机构名：BENET
输入域名：hub.benet.com
输入管理员邮箱：admin@benet.com
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.benet.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.benet.com//上传镜像
docker tag nginx:latest hub.benet.com/library/nginx:v1
docker push hub.benet.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.benet.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> 30000:32340/TCP 22m浏览器访问：
192.168.10.19:32340
192.168.10.20:32340
192.168.10.21:32340#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########### 内核参数优化方案 ##########
cat > /etc/sysctl.d/kubernetes.conf <<EOF
net.bridge.bridge-nf-call-iptables=1
net.bridge.bridge-nf-call-ip6tables=1
net.ipv4.ip_forward=1
net.ipv4.tcp_tw_recycle=0
vm.swappiness=0 #禁止使用 swap 空间，只有当系统内存不足（OOM）时才允许使用它
vm.overcommit_memory=1 #不检查物理内存是否够用
vm.panic_on_oom=0 #开启 OOM
fs.inotify.max_user_instances=8192
fs.inotify.max_user_watches=1048576
fs.file-max=52706963 #指定最大文件句柄数
fs.nr_open=52706963 #仅4.4以上版本支持
net.ipv6.conf.all.disable_ipv6=1
net.netfilter.nf_conntrack_max=2310720
EOF
