說明
本文系搭建kubernetes v1.18.5 集群筆記，使用三台虛擬機作為 CentOS 測試機，安裝kubeadm、kubelet、kubectl均使用yum安裝，網絡組件選用的是 flannel
行文中難免出現錯誤，如果讀者有高見，請評論與我交流
如需轉載請注明原始出處 https://www.cnblogs.com/hellxz/p/use-kubeadm-init-kubernetes-cluster.html
環境准備
部署集群沒有特殊說明均使用root用戶執行命令
硬件信息
| ip | hostname | mem | disk | explain |
|---|---|---|---|---|
| 192.168.87.145 | kube-master | 4 GB | 20GB | k8s 控制平台節點 |
| 192.168.87.146 | kube-node1 | 4 GB | 20GB | k8s 執行節點1 |
| 192.168.87.147 | kube-node2 | 4 GB | 20GB | k8s 執行節點2 |
軟件信息
| software | version |
|---|---|
| CentOS | CentOS Linux release 7.7.1908 (Core) |
| Kubernetes | v1.18.5 |
| Docker | 19.03.12 |
保證環境正確性
| purpose | commands |
|---|---|
| 保證集群各節點互通 | ping -c 3 <ip> |
| 保證MAC地址唯一 | ip link 或 ifconfig -a |
| 保證集群內主機名唯一 | 查詢 hostnamectl status，修改 hostnamectl set-hostname <hostname> |
| 保證系統產品uuid唯一 | dmidecode -s system-uuid 或 sudo cat /sys/class/dmi/id/product_uuid |
修改MAC地址參考命令：
ifconfig eth0 down ifconfig eth0 hw ether 00:0C:18:EF:FF:ED ifconfig eth0 up如product_uuid不唯一，請考慮重裝CentOS系統
確保端口開放正常
kube-master節點端口檢查：
| Protocol | Direction | Port Range | Purpose |
|---|---|---|---|
| TCP | Inbound | 6443* | kube-api-server |
| TCP | Inbound | 2379-2380 | etcd API |
| TCP | Inbound | 10250 | Kubelet API |
| TCP | Inbound | 10251 | kube-scheduler |
| TCP | Inbound | 10252 | kube-controller-manager |
kube-node*節點端口檢查：
| Protocol | Direction | Port Range | Purpose |
|---|---|---|---|
| TCP | Inbound | 10250 | Kubelet API |
| TCP | Inbound | 30000-32767 | NodePort Services |
如果你對主機的防火牆配置不是很自信，可以關掉防火牆：
systemctl disable --now firewalld或者 清除iptables規則 （慎用）
iptables -F
配置主機互信
分別在各節點配置hosts映射：
cat >> /etc/hosts <<EOF 192.168.87.145 kube-master 192.168.87.146 kube-node1 192.168.87.147 kube-node2 EOF
kube-master生成ssh密鑰，分發公鑰到各節點：
#生成ssh密鑰，直接一路回車 ssh-keygen -t rsa #復制剛剛生成的密鑰到各節點可信列表中，需分別輸入各主機密碼 ssh-copy-id root@kube-master ssh-copy-id root@kube-node1 ssh-copy-id root@kube-node2
禁用swap
swap僅當內存不夠時會使用硬盤塊充當額外內存，硬盤的io較內存差距極大，禁用swap以提高性能
各節點均需執行：
swapoff -a sed -i 's/.*swap.*/#&/' /etc/fstab
關閉 SELinux
關閉 SELinux，否則 kubelet 掛載目錄時可能報錯 Permission denied，可以設置為permissive或disabled，permissive 會提示warn信息
各節點均需執行：
setenforce 0 sed -i 's/^SELINUX=.*/SELINUX=disabled/' /etc/selinux/config
設置系統時區、同步時間
timedatectl set-timezone Asia/Shanghai systemctl enable --now chronyd
查看同步狀態：
timedatectl status
輸出：
System clock synchronized: yes NTP service: active RTC in local TZ: no
System clock synchronized: yes，表示時鍾已同步；NTP service: active，表示開啟了時鍾同步服務；
# 將當前的 UTC 時間寫入硬件時鍾 timedatectl set-local-rtc 0 # 重啟依賴於系統時間的服務 systemctl restart rsyslog && systemctl restart crond
部署docker
所有節點均需安裝部署docker
添加docker yum源
#安裝必要依賴 yum install -y yum-utils device-mapper-persistent-data lvm2
#添加aliyun docker-ce yum源 yum-config-manager --add-repo http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
#重建yum緩存 yum makecache fast
安裝docker
#查看可用docker版本
yum list docker-ce.x86_64 --showduplicates | sort -r
#安裝指定版本docker
yum install -y docker-ce-19.03.12-3.el7
這里以安裝19.03.12版本舉例，注意版本號不包含
:與之前的數字
確保網絡模塊開機自動加載
lsmod | grep overlay
lsmod | grep br_netfilter
若上面命令無返回值輸出或提示文件不存在，需執行以下命令：
cat > /etc/modules-load.d/docker.conf <<EOF
overlay
br_netfilter
EOF
modprobe overlay
modprobe br_netfilter
使橋接流量對iptables可見
各節點均需執行：
cat > /etc/sysctl.d/k8s.conf <<EOF net.bridge.bridge-nf-call-ip6tables = 1 net.bridge.bridge-nf-call-iptables = 1 EOF sysctl --system
驗證是否生效，均返回 1 即正確
sysctl -n net.bridge.bridge-nf-call-iptables
sysctl -n net.bridge.bridge-nf-call-ip6tables
配置docker
$ mkdir /etc/docker #修改cgroup驅動為systemd[k8s官方推薦]、限制容器日志量、修改存儲類型，最后的docker家目錄可修改 $ cat > /etc/docker/daemon.json <<EOF { "exec-opts": ["native.cgroupdriver=systemd"], "log-driver": "json-file", "log-opts": { "max-size": "100m" }, "storage-driver": "overlay2", "storage-opts": [ "overlay2.override_kernel_check=true" ], "registry-mirrors": ["https://7uuu3esz.mirror.aliyuncs.com"], "data-root": "/data/docker" } EOF #添加開機自啟，立即啟動 $ systemctl enable --now docker
驗證docker是否正常
#查看docker信息，判斷是否與配置一致 docker info
#hello-docker測試 docker run --rm hello-world
#刪除測試image docker rmi hello-world
添加用戶到docker組
非root用戶，無需sudo即可使用docker命令
#添加用戶到docker組 usermod -aG docker <USERNAME>
#當前會話立即更新docker組 newgrp docker
部署kubernetes集群
未特殊說明，各節點均需執行如下步驟
添加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=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
#重建yum緩存，輸入y添加證書認證
yum makecache fast
安裝kubeadm、kubelet、kubectl
各節點均需安裝kubeadm、kubelet，kubectl僅kube-master節點需安裝（作為worker節點，kubectl無法使用，可以不裝）
yum install -y kubelet kubeadm kubectl --disableexcludes=kubernetes
systemctl enable --now kubelet
配置自動補全命令
#安裝bash自動補全插件 yum install bash-completion -y
#設置kubectl與kubeadm命令補全，下次login生效 kubectl completion bash >/etc/bash_completion.d/kubectl kubeadm completion bash > /etc/bash_completion.d/kubeadm
預拉取kubernetes鏡像
由於國內網絡因素，kubernetes鏡像需要從mirrors站點或通過dockerhub用戶推送的鏡像拉取
#查看指定k8s版本需要哪些鏡像 kubeadm config images list --kubernetes-version v1.18.5
另因阿里雲的鏡像暫時還沒更新到v1.18.5版本，所以通過dockerhub上拉取，目前阿里雲最新同步版本是v1.18.3，想通過v1.18.3版本拉取鏡像請參考 <https://www.cnblogs.com/hellxz/p/13204093.html
在 /root/k8s 目錄下，新建腳本get-k8s-images.sh，內容如下：
#!/bin/bash # Script For Quick Pull K8S Docker Images # by Hellxz Zhang <hellxz001@foxmail.com> KUBE_VERSION=v1.18.5 PAUSE_VERSION=3.2 CORE_DNS_VERSION=1.6.7 ETCD_VERSION=3.4.3-0 # pull kubernetes images from hub.docker.com docker pull kubeimage/kube-proxy-amd64:$KUBE_VERSION docker pull kubeimage/kube-controller-manager-amd64:$KUBE_VERSION docker pull kubeimage/kube-apiserver-amd64:$KUBE_VERSION docker pull kubeimage/kube-scheduler-amd64:$KUBE_VERSION # pull aliyuncs mirror docker images docker pull registry.cn-hangzhou.aliyuncs.com/google_containers/pause:$PAUSE_VERSION docker pull registry.cn-hangzhou.aliyuncs.com/google_containers/coredns:$CORE_DNS_VERSION docker pull registry.cn-hangzhou.aliyuncs.com/google_containers/etcd:$ETCD_VERSION # retag to k8s.gcr.io prefix docker tag kubeimage/kube-proxy-amd64:$KUBE_VERSION k8s.gcr.io/kube-proxy:$KUBE_VERSION docker tag kubeimage/kube-controller-manager-amd64:$KUBE_VERSION k8s.gcr.io/kube-controller-manager:$KUBE_VERSION docker tag kubeimage/kube-apiserver-amd64:$KUBE_VERSION k8s.gcr.io/kube-apiserver:$KUBE_VERSION docker tag kubeimage/kube-scheduler-amd64:$KUBE_VERSION k8s.gcr.io/kube-scheduler:$KUBE_VERSION docker tag registry.cn-hangzhou.aliyuncs.com/google_containers/pause:$PAUSE_VERSION k8s.gcr.io/pause:$PAUSE_VERSION docker tag registry.cn-hangzhou.aliyuncs.com/google_containers/coredns:$CORE_DNS_VERSION k8s.gcr.io/coredns:$CORE_DNS_VERSION docker tag registry.cn-hangzhou.aliyuncs.com/google_containers/etcd:$ETCD_VERSION k8s.gcr.io/etcd:$ETCD_VERSION # untag origin tag, the images won't be delete. docker rmi kubeimage/kube-proxy-amd64:$KUBE_VERSION docker rmi kubeimage/kube-controller-manager-amd64:$KUBE_VERSION docker rmi kubeimage/kube-apiserver-amd64:$KUBE_VERSION docker rmi kubeimage/kube-scheduler-amd64:$KUBE_VERSION docker rmi registry.cn-hangzhou.aliyuncs.com/google_containers/pause:$PAUSE_VERSION docker rmi registry.cn-hangzhou.aliyuncs.com/google_containers/coredns:$CORE_DNS_VERSION docker rmi registry.cn-hangzhou.aliyuncs.com/google_containers/etcd:$ETCD_VERSION
腳本添加可執行權限，執行腳本拉取鏡像：
chmod +x get-k8s-images.sh ./get-k8s-images.sh
拉取完成，執行 docker images 查看鏡像
初始化kube-master
僅 kube-master 節點需要執行此步驟
修改kubelet配置默認cgroup driver
cat > /var/lib/kubelet/config.yaml <<EOF apiVersion: kubelet.config.k8s.io/v1beta1 kind: KubeletConfiguration cgroupDriver: systemd EOF systemctl restart kubelet
生成kubeadm初始化配置文件 [可選] 僅當需自定義初始化配置時用
kubeadm config print init-defaults > init.default.yaml
測試環境是否正常（WARNING是正常的）
kubeadm init phase preflight [--config kubeadm-init.yaml]
上圖提示Warning是正常的，校驗不了k8s信息是因為連不上被ban的網站，最后一個提示是因我本地未關閉防火牆，請我看清楚必要放行的端口號是否暢通
初始化master 10.244.0.0/16是flannel固定使用的IP段，設置取決於網絡組件要求
kubeadm init --pod-network-cidr=10.244.0.0/16 --kubernetes-version=v1.18.5 [--config kubeadm-init.yaml]
輸出如下：
[root@kube-master k8s]# kubeadm init --pod-network-cidr=10.244.0.0/16 --kubernetes-version=v1.18.5 W0703 18:49:19.076654 16469 configset.go:202] WARNING: kubeadm cannot validate component configs for API groups [kubelet.config.k8s.io kubeproxy.config.k8s.io] [init] Using Kubernetes version: v1.18.5 [preflight] Running pre-flight checks [WARNING Firewalld]: firewalld is active, please ensure ports [6443 10250] are open or your cluster may not function correctly [preflight] Pulling images required for setting up a Kubernetes cluster [preflight] This might take a minute or two, depending on the speed of your internet connection [preflight] You can also perform this action in beforehand using 'kubeadm config images pull' [kubelet-start] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env" [kubelet-start] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml" [kubelet-start] Starting the kubelet [certs] Using certificateDir folder "/etc/kubernetes/pki" [certs] Generating "ca" certificate and key [certs] Generating "apiserver" certificate and key [certs] apiserver serving cert is signed for DNS names [kube-master kubernetes kubernetes.default kubernetes.default.svc kubernetes.default.svc.cluster.local] and IPs [10.96.0.1 192.168.87.145] [certs] Generating "apiserver-kubelet-client" certificate and key [certs] Generating "front-proxy-ca" certificate and key [certs] Generating "front-proxy-client" certificate and key [certs] Generating "etcd/ca" certificate and key [certs] Generating "etcd/server" certificate and key [certs] etcd/server serving cert is signed for DNS names [kube-master localhost] and IPs [192.168.87.145 127.0.0.1 ::1] [certs] Generating "etcd/peer" certificate and key [certs] etcd/peer serving cert is signed for DNS names [kube-master localhost] and IPs [192.168.87.145 127.0.0.1 ::1] [certs] Generating "etcd/healthcheck-client" certificate and key [certs] Generating "apiserver-etcd-client" certificate and key [certs] Generating "sa" key and public key [kubeconfig] Using kubeconfig folder "/etc/kubernetes" [kubeconfig] Writing "admin.conf" kubeconfig file [kubeconfig] Writing "kubelet.conf" kubeconfig file [kubeconfig] Writing "controller-manager.conf" kubeconfig file [kubeconfig] Writing "scheduler.conf" kubeconfig file [control-plane] Using manifest folder "/etc/kubernetes/manifests" [control-plane] Creating static Pod manifest for "kube-apiserver" [control-plane] Creating static Pod manifest for "kube-controller-manager" W0703 18:49:23.039913 16469 manifests.go:225] the default kube-apiserver authorization-mode is "Node,RBAC"; using "Node,RBAC" [control-plane] Creating static Pod manifest for "kube-scheduler" W0703 18:49:23.040907 16469 manifests.go:225] the default kube-apiserver authorization-mode is "Node,RBAC"; using "Node,RBAC" [etcd] Creating static Pod manifest for local etcd in "/etc/kubernetes/manifests" [wait-control-plane] Waiting for the kubelet to boot up the control plane as static Pods from directory "/etc/kubernetes/manifests". This can take up to 4m0s [apiclient] All control plane components are healthy after 21.505101 seconds [upload-config] Storing the configuration used in ConfigMap "kubeadm-config" in the "kube-system" Namespace [kubelet] Creating a ConfigMap "kubelet-config-1.18" in namespace kube-system with the configuration for the kubelets in the cluster [upload-certs] Skipping phase. Please see --upload-certs [mark-control-plane] Marking the node kube-master as control-plane by adding the label "node-role.kubernetes.io/master=''" [mark-control-plane] Marking the node kube-master as control-plane by adding the taints [node-role.kubernetes.io/master:NoSchedule] [bootstrap-token] Using token: 2b7cfv.6bhz4z3a3vzyg498 [bootstrap-token] Configuring bootstrap tokens, cluster-info ConfigMap, RBAC Roles [bootstrap-token] configured RBAC rules to allow Node Bootstrap tokens to get nodes [bootstrap-token] configured RBAC rules to allow Node Bootstrap tokens to post CSRs in order for nodes to get long term certificate credentials [bootstrap-token] configured RBAC rules to allow the csrapprover controller automatically approve CSRs from a Node Bootstrap Token [bootstrap-token] configured RBAC rules to allow certificate rotation for all node client certificates in the cluster [bootstrap-token] Creating the "cluster-info" ConfigMap in the "kube-public" namespace [kubelet-finalize] Updating "/etc/kubernetes/kubelet.conf" to point to a rotatable kubelet client certificate and key [addons] Applied essential addon: CoreDNS [addons] Applied essential addon: kube-proxy 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 192.168.87.145:6443 --token 2b7cfv.6bhz4z3a3vzyg498 \ --discovery-token-ca-cert-hash sha256:79bd63d82634f9953cc9d6b5a923fa87c973f0c3fd9ed7270167052dd834c026
為日常使用集群的用戶添加kubectl使用權限
su hellxz mkdir -p $HOME/.kube sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/admin.conf sudo chown $(id -u):$(id -g) $HOME/.kube/admin.conf echo "export KUBECONFIG=$HOME/.kube/admin.conf" >> ~/.bashrc exit
配置master認證
echo 'export KUBECONFIG=/etc/kubernetes/admin.conf' >> /etc/profile . /etc/profile
如果不配置這個，會提示如下輸出：
The connection to the server localhost:8080 was refused - did you specify the right host or port?此時master節點已經初始化成功，但是還未完裝網絡組件，還無法與其他節點通訊
安裝網絡組件，以flannel為例
cd ~/k8s yum install -y wget
#下載flannel最新配置文件 wget https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml kubectl apply -f kube-flannel.yml
查看kube-master節點狀態
kubectl get nodes
如果STATUS提示
NotReady，可以通過kubectl describe node kube-master查看具體的描述信息，性能差的服務器到達Ready狀態時間會長些
備份鏡像供其他節點使用
在kube-master節點將鏡像備份出來，便於后續傳輸給其他node節點，當然有鏡像倉庫更好
docker save k8s.gcr.io/kube-proxy:v1.18.5 \ k8s.gcr.io/kube-apiserver:v1.18.5 \ k8s.gcr.io/kube-controller-manager:v1.18.5 \ k8s.gcr.io/kube-scheduler:v1.18.5 \ k8s.gcr.io/pause:3.2 \ k8s.gcr.io/coredns:1.6.7 \ k8s.gcr.io/etcd:3.4.3-0 > k8s-imagesV1.18.5.tar
初始化kube-node*節點並加入集群
拷貝鏡像到node節點，以kube-node1舉例，node2不再累述
#此時命令在kube-node*節點上執行 mkdir ~/k8s scp root@kube-master:/root/k8s/k8s-imagesV1.18.5.tar ~/k8s
獲取加入kubernetes命令，未忘可不選
剛才在初始化kube-master節點時，有在最后輸出其加入集群的命令，假如我沒記下來，那怎么辦呢？
訪問kube-master輸入創建新token命令，同時輸出加入集群的命令：
kubeadm token create --print-join-command
在kube-node*節點上執行加入集群命令
kubeadm join 192.168.87.145:6443 --token jdyzyq.icwlpkm36kgs6nqh --discovery-token-ca-cert-hash sha256:24f9b05fa10307ef6fff4132e0ec3c8b54917d4ff440b36108908aca588d8be7
查看集群節點狀態
kubectl get nodes
參考
- 《Kubernetes權威指南》第4版
- 官方文檔 https://kubernetes.io/docs/setup/production-environment/tools/kubeadm/install-kubeadm/