注释：此次操作使用VMware Workstation Pro17虚拟机进行
本次使用单master节点，双worker节点组成的最小化单主节点的学习环境
1.K8S所有节点环境准备
xshell发送所用会话，包括harbor仓库
虚拟机操作系统环境准备
参考链接:
https://kubernetes.io/zh/docs/setup/production-environment/tools/kubeadm/install-kubeadm/
2.关闭swap分区
临时关闭
swapoff -a && sysctl -w vm.swappiness=0
swapoff -a
基于配置文件关闭
sed -ri '/^[^#]*swap/s@^@#@' /etc/fstab
3.确保各个节点MAC地址或product_uuid唯一
ifconfig eth0 | grep ether | awk '{print $2}'
cat /sys/class/dmi/id/product_uuid
温馨提示:
一般来讲，硬件设备会拥有唯一的地址，但是有些虚拟机的地址可能会重复。
Kubernetes使用这些值来唯一确定集群中的节点。 如果这些值在每个节点上不唯一，可能会导致安装失败。
4.检查网络环境是否互通
可以使用ping命令来测试。
5.允许使用iptable检查桥接流量
cat <<EOF | tee /etc/modules-load.d/k8s.conf
br_netfilter
EOF
cat <<EOF | tee /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
EOF
sysctl --system
6.检查端口是否被占用
ss -ntl
参考链接: https://kubernetes.io/zh-cn/docs/reference/networking/ports-and-protocols/
7.检查docker环境
安装docker和docker-compose，也可参考docker官网
8.禁用防火墙（内网环境）
systemctl disable --now firewalld
9.禁用selinux
sed -i 's/^SELINUX=enforcing$/SELINUX=disabled/' /etc/selinux/config
grep ^SELINUX= /etc/selinux/config
10.配置host解析
cat >> /etc/hosts <<'EOF'
192.168.52.231 master231
192.168.52.232 worker232
192.168.52.233 worker233
192.168.52.250 harbor.lzh.com
EOF
cat /etc/hosts
# 注意windows也要配置解析
11.所有节点创建自定义证书目录
mkdir -pv /etc/docker/certs.d/harbor.lzh.com
12.检查端口是否被占用
https://kubernetes.io/zh-cn/docs/reference/networking/ports-and-protocols/
13.安装harbor
13.1.下载harbor软件
可以去GitHub上下载其他版本，我这里先用以前下载的版本
13.2.解压harbor软件包
[root@harbor250 ~]# tar xf harbor.tar.gz -C /test-harbor/softwares/
13.3.安装harbor
[root@harbor250 ~]# cd /test-harbor/softwares/harbor/
[root@harbor250 harbor]#
[root@harbor250 harbor]# ./install.sh
13.4.将客户端证书推送到所有的k8s集群
[root@harbor250 harbor]# scp certs/custom/client/* master231:/etc/docker/certs.d/harbor.lzh.com/
[root@harbor250 harbor]#
[root@harbor250 harbor]# scp certs/custom/client/* worker232:/etc/docker/certs.d/harbor.lzh.com/
[root@harbor250 harbor]#
[root@harbor250 harbor]# scp certs/custom/client/* worker233:/etc/docker/certs.d/harbor.lzh.com/
13.5.挑选任意K8S节点测试harbor能否正常访问
[root@master231 ~]# docker login -u admin -p 1 harbor.lzh.com
.....
Login Succeeded
[root@master231 ~]#
14.所有节点安装kubeadm，kubelet，kubectl
14.1.配置软件源 （阿里云的镜像站）
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
EOF
14.2.查看kubeadm的版本（将来你要安装的K8S时请所有组件版本均保持一致!）
yum -y list kubeadm --showduplicates | sort -r
yum list kubelet.x86_64 --showduplicates | sort -r
yum list kubectl.x86_64 --showduplicates | sort -r
14.3.安装kubeadm，kubelet，kubectl软件包 #（所有节点）
yum -y install kubeadm-1.23.17-0 kubelet-1.23.17-0 kubectl-1.23.17-0
14.4.启动kubelet服务
(若服务启动失败时正常现象，其会自动重启，因为缺失配置文件，初始化集群后恢复！此步骤可跳过！推荐设置开机自启动)
systemctl enable --now kubelet
systemctl status kubelet
参考链接:
https://kubernetes.io/zh/docs/tasks/tools/install-kubectl-linux/
15.初始化master节点
15.1.使用kubeadm初始化master节点
[root@master231 ~]# kubeadm init --kubernetes-version=v1.23.17 --image-repository registry.aliyuncs.com/google_containers --apiserver-advertise-address=192.168.52.11 --pod-network-cidr=10.96.0.0/16 --service-cidr=172.31.0.0/16
kubeadm init --image-repository registry.cn-hangzhou.aliyuncs.com/lfy_k8s_images --kubernetes-version v1.23.17 --service-cidr=10.200.0.0/16 --pod-network-cidr=10.100.0.0/16
--kubernetes-version=v1.23.17: 指定 Kubernetes 的版本为 v1.23.17。
--image-repository registry.aliyuncs.com/google_containers: 指定使用阿里云的镜像仓库 repo，并拉取谷歌容器镜像。
--pod-network-cidr=10.100.0.0/16: 指定 Pod 网络的地址段为 10.100.0.0/16。
--service-cidr=10.200.0.0/16: 指定SVC的网段 指定 Service 的地址段为 10.200.0.0/16。
--service-dns-domain=lzh.com: 指定 Service 的 DNS 域名为 lzh.com。若不指定，默认为"cluster.local"。
15.2.拷贝授权文件，用于管理kubernetes集群
[root@master231 ~]# mkdir -p $HOME/.kube
[root@master231 ~]# sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
[root@master231 ~]# sudo chown $(id -u):$(id -g) $HOME/.kube/config
15.3.查看master组件
[root@master231 ~]# kubectl get componentstatuses
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME STATUS MESSAGE ERROR
scheduler Healthy ok
controller-manager Healthy ok
etcd-0 Healthy {"health":"true","reason":""}
[root@master231 ~]#
[root@master231 ~]# kubectl get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME STATUS MESSAGE ERROR
controller-manager Healthy ok
etcd-0 Healthy {"health":"true","reason":""}
scheduler Healthy ok
[root@master231 ~]#
16.配置所有节点加入k8s集群
16.1.加入集群，注意TOKEN
[root@worker232 ~]# docker load -i worker-node.tar.gz 不导入镜像就直接下载，网络好的话无所谓。
[root@worker232 ~]#kubeadm join 192.168.52.231:6443 --token uzg2le.ll3ay03eyfdupiw9 \
> --discovery-token-ca-cert-hash sha256:3e5b1a7125e6a1139478f702a9762f1857fb70bcdb52e335c46c3081ce7dc2b9
kubeadm join 192.168.52.11:6443 --token xovu1o.3vtmh02hrrsokxe7 \
--discovery-token-ca-cert-hash sha256:543b5163dc1f547f7d67993878264c3f1f75acc0eb60ebcb68e719cb0834676c
16.2.master查看集群节点数量
[root@master231 ~]# kubectl get nodes
NAME STATUS ROLES AGE VERSION
master231 NotReady control-plane,master 22m v1.23.17
worker232 NotReady <none> 6m14s v1.23.17
worker233 NotReady <none> 6m10s v1.23.17
[root@master231 ~]#
[root@master231 ~]#
[root@master231 ~]# kubectl get no
NAME STATUS ROLES AGE VERSION
master231 NotReady control-plane,master 22m v1.23.17
worker232 NotReady <none> 6m15s v1.23.17
worker233 NotReady <none> 6m11s v1.23.17
[root@master231 ~]#
17.配置自动补全功能
[root@master231 ~]# echo "source <(kubectl completion bash)" >> ~/.bashrc && source ~/.bashrc
[root@master231 ~]#
[root@master231 ~]# kubectl
alpha auth cordon diff get patch run version
annotate autoscale cp drain help plugin scale wait
api-resources certificate create edit kustomize port-forward set
api-versions cluster-info debug exec label proxy taint
apply completion delete explain logs replace top
attach config describe expose options rollout uncordon
[root@master231 ~]# kubectl
18. 安装cni插件
18.1.下载插件的配置文件
[root@master231 ~]# mkdir -p /manifests/cni
[root@master231 ~]#
[root@master231 ~]# wget https://github.com/flannel-io/flannel/releases/latest/download/kube-flannel.yml -O /manifests/cni/kube-flannel.yml
18.2.修改配置文件
[root@master231 ~]# vim /manifests/cni/kube-flannel.yml
...
将
"Network": "10.244.0.0/16",
修改为:
"Network": "10.100.0.0/16",
根据自己的时实际情况修改，不需要和我的一样
18.3.安装flannel插件
[root@master231 ~]# kubectl apply -f /manifests/cni/kube-flannel.yml
18.4.验证网络插件是否部署成功
[root@master231 ~]# kubectl -n kube-flannel get pods
NAME READY STATUS RESTARTS AGE
kube-flannel-ds-btrqw 1/1 Running 0 5m14s
kube-flannel-ds-krq6g 1/1 Running 0 5m14s
kube-flannel-ds-mh2q7 1/1 Running 0 5m14s
[root@master231 ~]#
[root@master231 ~]# kubectl get nodes
NAME STATUS ROLES AGE VERSION
master231 Ready control-plane,master 91m v1.23.17
worker232 Ready <none> 75m v1.23.17
worker233 Ready <none> 74m v1.23.17
[root@master231 ~]#
[root@master231 ~]# kubectl -n kube-flannel get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
kube-flannel-ds-btrqw 1/1 Running 0 18m 192.168.52.231 master231 <none> <none>
kube-flannel-ds-krq6g 1/1 Running 0 18m 192.168.52.232 worker232 <none> <none>
kube-flannel-ds-mh2q7 1/1 Running 0 18m 192.168.52.233 worker233 <none> <none>
[root@master231 ~]#
18.5.推送测试镜像到harbor
[root@master231 ~]# docker login -u admin -p 1 harbor.lzh.com
[root@master231 ~]#
[root@master231 ~]# docker tag alpine harbor.lzh.com/test-linux/alpine
[root@master231 ~]#
[root@master231 ~]# docker push harbor.lzh.com/test-linux/alpine
Using default tag: latest
The push refers to repository [harbor.lzh.com/test-linux/alpine]
8d3ac3489996: Pushed
latest: digest: sha256:e7d88de73db3d3fd9b2d63aa7f447a10fd0220b7cbf39803c803f2af9ba256b3 size: 528
[root@master231 ~]#
18.6.启动pod测试
1.[root@master231 ~]# mkdir /manifests/pod
[root@master231 ~]#
2.[root@master231 ~]# cat /manifests/pod/01-flannel-test.yaml
# 指定apiserver版本号
apiVersion: v1
# 指定资源的类型
kind: Pod
# 定义源数据信息
metadata:
# Pod的名称
name: pod-c1
# 用户定义资源期望运行的状态
spec:
# 指定在worker232的工作节点运行
nodeName: worker232
# 在Pod内运行的容器定义
containers:
# 容器的名称
- name: c1
# 镜像名称
image: harbor.lzh.com/test-linux/alpine:latest
# 相当于Dockerfile的ENTRYPOINT指令，指定容器运行的命令
command: ["tail","-f","/etc/hosts"]
---
apiVersion: v1
kind: Pod
metadata:
name: pod-c2
spec:
nodeName: worker233
containers:
- name: c2
image: harbor.lzh.com/test-linux/alpine:latest
command: ["sleep","3600"]
[root@master231 ~]#
3.[root@master231 ~]# kubectl apply -f /manifests/pod/01-flannel-test.yaml
pod/pod-c1 created
pod/pod-c2 created
[root@master231 ~]#
4.[root@master231 ~]# kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
pod-c1 1/1 Running 0 8s 10.100.1.2 worker232 <none> <none>
pod-c2 1/1 Running 0 8s 10.100.2.2 worker233 <none> <none>
[root@master231 ~]#