Kubernetes二进制方式v1.13.2生产环境的安装与配置（HTTPS+RBAC）

一 背景

由于众所周知的原因，在国内无法直接访问Google的服务。二进制包由于其下载方便、灵活定制而深受广大kubernetes使用者喜爱，成为企业部署生产环境比较流行的方式之一，Kubernetes v1.13.2是目前的最新版本。安装部署过程可能比较复杂、繁琐，因此在安装过程中尽可能将操作步骤脚本话。文中涉及到的脚本已经通过本人测试。

二 环境及架构图

2.1 软件环境

OS（最小化安装版）：

cat /etc/centos-release

CentOS Linux release 7.6.1810 (Core)

Docker Engine：

docker version

Client: Version: 18.06.0-ce API version: 1.38 Go version: go1.10.3 Git commit: 0ffa825 Built: Wed Jul 18 19:08:18 2018 OS/Arch: linux/amd64 Experimental: false Server: Engine: Version: 18.06.0-ce API version: 1.38 (minimum version 1.12) Go version: go1.10.3 Git commit: 0ffa825 Built: Wed Jul 18 19:10:42 2018 OS/Arch: linux/amd64 Experimental: false

Kubenetes：

kubectl version

Client Version: version.Info{Major:"1", Minor:"13", GitVersion:"v1.13.2", GitCommit:"cff46ab41ff0bb44d8584413b598ad8360ec1def", GitTreeState:"clean", BuildDate:"2019-01-10T23:35:51Z", GoVersion:"go1.11.4", Compiler:"gc", Platform:"linux/amd64"} Server Version: version.Info{Major:"1", Minor:"13", GitVersion:"v1.13.2", GitCommit:"cff46ab41ff0bb44d8584413b598ad8360ec1def", GitTreeState:"clean", BuildDate:"2019-01-10T23:28:14Z", GoVersion:"go1.11.4", Compiler:"gc", Platform:"linux/amd64"}

ETCD:

etcd --version

etcd Version: 3.3.11 Git SHA: 2cf9e51d2 Go Version: go1.10.7 Go OS/Arch: linux/amd64

Flannel:

flanneld -version

v0.11.0

2.2 服务器规划

|IP|主机名（Hostname）|角色（Role）|组件（Component）|
|:-:|:-:|:-:|:-:|
|172.31.2.11|gysl-master|Master&Node|kube-apiserver，kube-controller-manager，kube-scheduler，etcd，（kubectl），kubelet，kube-proxy，docker，flannel|
|172.31.2.12|gysl-node1|Node|kubelet，kube-proxy，docker，flannel，etcd|
|172.31.2.13|gysl-node2|Node|kubelet，kube-proxy，docker，flannel，etcd|

注：加粗部分是Master节点必须安装的组件，etcd可以部署在其他节点，也可以部署在Master节点，kubectl是管理kubernetes的命令行工具。其余部分是Node节点必选组件。

2.3 节点或组件功能简介

Master节点：
Master节点上面主要由四个模块组成，apiserver，schedule，controller-manager，etcd。

apiserver: 负责对外提供RESTful的kubernetes API 的服务，它是系统管理指令的统一接口，任何对资源的增删该查都要交给apiserver处理后再交给etcd。kubectl(kubernetes提供的客户端工具，该工具内部是对kubernetes API的调用）是直接和apiserver交互的。

schedule: 负责调度Pod到合适的Node上，如果把scheduler看成一个黑匣子，那么它的输入是pod和由多个Node组成的列表，输出是Pod和一个Node的绑定。kubernetes目前提供了调度算法，同样也保留了接口。用户根据自己的需求定义自己的调度算法。

controller-manager: 如果apiserver做的是前台的工作的话，那么controller-manager就是负责后台的。每一个资源都对应一个控制器。而control manager就是负责管理这些控制器的，比如我们通过APIServer创建了一个Pod，当这个Pod创建成功后，apiserver的任务就算完成了。

etcd：etcd是一个高可用的键值存储系统，kubernetes使用它来存储各个资源的状态，从而实现了Restful的API。

Node节点：
每个Node节点主要由二个模块组成：kublet, kube-proxy。

kube-proxy: 该模块实现了kubernetes中的服务发现和反向代理功能。kube-proxy支持TCP和UDP连接转发，默认基Round Robin算法将客户端流量转发到与service对应的一组后端pod。服务发现方面，kube-proxy使用etcd的watch机制监控集群中service和endpoint对象数据的动态变化，并且维护一个service到endpoint的映射关系，从而保证了后端pod的IP变化不会对访问者造成影响，另外，kube-proxy还支持session affinity。

kublet：kublet是Master在每个Node节点上面的agent，是Node节点上面最重要的模块，它负责维护和管理该Node上的所有容器，但是如果容器不是通过kubernetes创建的，它并不会管理。本质上，它负责使Pod的运行状态与期望的状态一致。

2.4 Kubernetes架构图

2.5 Kubernetes工作流程图

三 操作步骤

3.1 针对性初始化设置

在所有主机上执行脚本KubernetesInstall-01.sh，以Master节点为例。

[root@gysl-master ~]# sh KubernetesInstall-01.sh

脚本内容如下：

#!/bin/bash # Initialize the machine. This needs to be executed on every machine. # Add host domain name. cat>>/etc/hosts<<EOF 172.31.2.11 gysl-master 172.31.2.12 gysl-node1 172.31.2.13 gysl-node2 EOF # Modify related kernel parameters. cat>/etc/sysctl.d/kubernetes.conf<<EOF net.ipv4.ip_forward = 1 net.bridge.bridge-nf-call-ip6tables = 1 net.bridge.bridge-nf-call-iptables = 1 EOF sysctl -p /etc/sysctl.d/kubernetes.conf>&/dev/null # Turn off and disable the firewalld. systemctl stop firewalld systemctl disable firewalld # Disable the SELinux. sed -i.bak 's/=enforcing/=disabled/' /etc/selinux/config # Disable the swap . sed -i.bak 's/^.*swap/#&/g' /etc/fstab # Reboot the machine. reboot

3.2 安装Docker Engine并设置

在所有主机上执行脚本KubernetesInstall-02.sh，以Master节点为例。

[root@gysl-master ~]# sh KubernetesInstall-02.sh

脚本内容如下：

#!/bin/bash # Install the Docker engine. This needs to be executed on every machine. curl http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo -o /etc/yum.repos.d/docker-ce.repo>&/dev/null if [ $? -eq 0 ] ; then yum remove docker \ docker-client \ docker-client-latest \ docker-common \ docker-latest \ docker-latest-logrotate \ docker-logrotate \ docker-selinux \ docker-engine-selinux \ docker-engine>&/dev/null yum list docker-ce --showduplicates|grep "^doc"|sort -r yum -y install docker-ce-18.06.0.ce-3.el7 rm -f /etc/yum.repos.d/docker-ce.repo systemctl enable docker && systemctl start docker && systemctl status docker else echo "Install failed! Please try again! "; exit 110 fi

注意：以上步骤需要在每一个节点上执行。如果启用了swap，那么是需要禁用的（脚本KubernetesInstall-01.sh已有涉及），具体可以通过 free 命令查看详情。另外，还需要关注各个节点上的时间同步情况。

3.3 下载相关二进制包

在Master执行脚本KubernetesInstall-03.sh即可进行下载。

[root@gysl-master ~]# sh KubernetesInstall-03.sh

脚本内容如下：

#!/bin/bash # Download relevant softwares. Please verify sha512 yourself. while true; do echo "Downloading, please wait a moment." &&\ curl -L -C - -O https://dl.k8s.io/v1.13.2/kubernetes-server-linux-amd64.tar.gz && \ curl -L -C - -O https://github.com/etcd-io/etcd/releases/download/v3.2.26/etcd-v3.2.26-linux-amd64.tar.gz && \ curl -L -C - -O https://pkg.cfssl.org/R1.2/cfssl_linux-amd64 && \ curl -L -C - -O https://pkg.cfssl.org/R1.2/cfssljson_linux-amd64 && \ curl -L -C - -O https://pkg.cfssl.org/R1.2/cfssl-certinfo_linux-amd64 \ curl -L -C - -O https://github.com/coreos/flannel/releases/download/v0.11.0/flannel-v0.11.0-linux-amd64.tar.gz if [ $? -eq 0 ]; then echo "Congratulations! All software packages have been downloaded." break fi done

kubernetes-server-linux-amd64.tar.gz包括了kubernetes的主要组件，无需下载其他包。etcd-v3.2.26-linux-amd64.tar.gz是部署etcd需要用到的包。其余的是cfssl相关的软件，暂不深究。网络原因，只能写脚本来下载了，这个过程可能需要一会儿。

3.4 部署etcd集群

3.4.1 创建CA证书

在Master执行脚本KubernetesInstall-04.sh。

[root@gysl-master ~]# sh KubernetesInstall-04.sh 2019/01/28 16:29:47 [INFO] generating a new CA key and certificate from CSR 2019/01/28 16:29:47 [INFO] generate received request 2019/01/28 16:29:47 [INFO] received CSR 2019/01/28 16:29:47 [INFO] generating key: rsa-2048 2019/01/28 16:29:47 [INFO] encoded CSR 2019/01/28 16:29:47 [INFO] signed certificate with serial number 368034386524991671795323408390048460617296625670 2019/01/28 16:29:47 [INFO] generate received request 2019/01/28 16:29:47 [INFO] received CSR 2019/01/28 16:29:47 [INFO] generating key: rsa-2048 2019/01/28 16:29:48 [INFO] encoded CSR 2019/01/28 16:29:48 [INFO] signed certificate with serial number 714486490152688826461700674622674548864494534798 2019/01/28 16:29:48 [WARNING] This certificate lacks a "hosts" field. This makes it unsuitable for websites. For more information see the Baseline Requirements for the Issuance and Management of Publicly-Trusted Certificates, v.1.1.6, from the CA/Browser Forum (https://cabforum.org); specifically, section 10.2.3 ("Information Requirements"). /etc/etcd/ssl/ca-key.pem /etc/etcd/ssl/ca.pem /etc/etcd/ssl/server-key.pem /etc/etcd/ssl/server.pem

脚本内容如下：

#!/bin/bash mv cfssl* /usr/local/bin/ chmod +x /usr/local/bin/cfssl* ETCD_SSL=/etc/etcd/ssl mkdir -p $ETCD_SSL # Create some CA certificates for etcd cluster. cat<<EOF>$ETCD_SSL/ca-config.json { "signing": { "default": { "expiry": "87600h" }, "profiles": { "www": { "expiry": "87600h", "usages": [ "signing", "key encipherment", "server auth", "client auth" ] } } } } EOF cat<<EOF>$ETCD_SSL/ca-csr.json { "CN": "etcd CA", "key": { "algo": "rsa", "size": 2048 }, "names": [ { "C": "CN", "L": "Beijing", "ST": "Beijing" } ] } EOF cat<<EOF>$ETCD_SSL/server-csr.json { "CN": "etcd", "hosts": [ "172.31.2.11", "172.31.2.12", "172.31.2.13" ], "key": { "algo": "rsa", "size": 2048 }, "names": [ { "C": "CN", "L": "Beijing", "ST": "Beijing" } ] } EOF cd $ETCD_SSL cfssl_linux-amd64 gencert -initca ca-csr.json | cfssljson_linux-amd64 -bare ca - cfssl_linux-amd64 gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=www server-csr.json | cfssljson_linux-amd64 -bare server cd ~ # ca-key.pem ca.pem server-key.pem server.pem ls $ETCD_SSL/*.pem

3.4.2 配置etcd服务

3.4.2.1 在Master节点上进行配置

在Master执行脚本KubernetesInstall-05.sh。

[root@gysl-master ~]# sh KubernetesInstall-05.sh

脚本内容如下：

#!/bin/bash # Deploy and configurate etcd service on the master node. ETCD_CONF=/etc/etcd/etcd.conf ETCD_SSL=/etc/etcd/ssl ETCD_SERVICE=/usr/lib/systemd/system/etcd.service tar -xzf etcd-v3.3.11-linux-amd64.tar.gz cp -p etcd-v3.3.11-linux-amd64/etc* /usr/local/bin/ # The etcd configuration file. cat>$ETCD_CONF<<EOF #[Member] ETCD_NAME="etcd-01" ETCD_DATA_DIR="/var/lib/etcd/default.etcd" ETCD_LISTEN_PEER_URLS="https://172.31.2.11:2380" ETCD_LISTEN_CLIENT_URLS="https://172.31.2.11:2379" #[Clustering] ETCD_INITIAL_ADVERTISE_PEER_URLS="https://172.31.2.11:2380" ETCD_ADVERTISE_CLIENT_URLS="https://172.31.2.11:2379" ETCD_INITIAL_CLUSTER="etcd-01=https://172.31.2.11:2380,etcd-02=https://172.31.2.12:2380,etcd-03=https://172.31.2.13:2380" ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster" ETCD_INITIAL_CLUSTER_STATE="new" EOF # The etcd servcie configuration file. cat>$ETCD_SERVICE<<EOF [Unit] Description=Etcd Server After=network.target After=network-online.target Wants=network-online.target [Service] Type=notify EnvironmentFile=$ETCD_CONF ExecStart=/usr/local/bin/etcd \ --name=\${ETCD_NAME} \ --data-dir=\${ETCD_DATA_DIR} \ --listen-peer-urls=\${ETCD_LISTEN_PEER_URLS} \ --listen-client-urls=\${ETCD_LISTEN_CLIENT_URLS},http://127.0.0.1:2379 \ --advertise-client-urls=\${ETCD_ADVERTISE_CLIENT_URLS} \ --initial-advertise-peer-urls=\${ETCD_INITIAL_ADVERTISE_PEER_URLS} \ --initial-cluster=\${ETCD_INITIAL_CLUSTER} \ --initial-cluster-token=\${ETCD_INITIAL_CLUSTER_TOKEN} \ --initial-cluster-state=new \ --cert-file=/etc/etcd/ssl/server.pem \ --key-file=/etc/etcd/ssl/server-key.pem \ --peer-cert-file=/etc/etcd/ssl/server.pem \ --peer-key-file=/etc/etcd/ssl/server-key.pem \ --trusted-ca-file=/etc/etcd/ssl/ca.pem \ --peer-trusted-ca-file=/etc/etcd/ssl/ca.pem Restart=on-failure LimitNOFILE=65536 [Install] WantedBy=multi-user.target EOF systemctl daemon-reload systemctl enable etcd.service --now systemctl status etcd

3.4.2.2 在Node1节点上进行配置

在Node1执行脚本KubernetesInstall-06.sh。

[root@gysl-master ~]# sh KubernetesInstall-06.sh

脚本内容如下：

#!/bin/bash # Deploy etcd on the node1. ETCD_SSL=/etc/etcd/ssl mkdir -p $ETCD_SSL scp gysl-master:~/etcd-v3.3.11-linux-amd64.tar.gz . scp gysl-master:$ETCD_SSL/{ca*pem,server*pem} $ETCD_SSL/ scp gysl-master:/etc/etcd/etcd.conf /etc/etcd/ scp gysl-master:/usr/lib/systemd/system/etcd.service /usr/lib/systemd/system/ tar -xvzf etcd-v3.3.11-linux-amd64.tar.gz mv ~/etcd-v3.3.11-linux-amd64/etcd* /usr/local/bin/ sed -i '/ETCD_NAME/{s/etcd-01/etcd-02/g}' /etc/etcd/etcd.conf sed -i '/ETCD_LISTEN_PEER_URLS/{s/2.11/2.12/g}' /etc/etcd/etcd.conf sed -i '/ETCD_LISTEN_CLIENT_URLS/{s/2.11/2.12/g}' /etc/etcd/etcd.conf sed -i '/ETCD_INITIAL_ADVERTISE_PEER_URLS/{s/2.11/2.12/g}' /etc/etcd/etcd.conf sed -i '/ETCD_ADVERTISE_CLIENT_URLS/{s/2.11/2.12/g}' /etc/etcd/etcd.conf rm -rf ~/etcd-v3.3.11-linux-amd64* systemctl daemon-reload systemctl enable etcd.service --now systemctl status etcd

3.4.2.3 在Node2节点上进行配置

在Node2执行脚本KubernetesInstall-07.sh。

[root@gysl-master ~]# sh KubernetesInstall-07.sh

脚本内容如下：

#!/bin/bash # Deploy etcd on the node2. ETCD_SSL=/etc/etcd/ssl mkdir -p $ETCD_SSL scp gysl-master:~/etcd-v3.3.11-linux-amd64.tar.gz . scp gysl-master:$ETCD_SSL/{ca*pem,server*pem} $ETCD_SSL/ scp gysl-master:/etc/etcd/etcd.conf /etc/etcd/ scp gysl-master:/usr/lib/systemd/system/etcd.service /usr/lib/systemd/system/ tar -xvzf etcd-v3.3.11-linux-amd64.tar.gz mv ~/etcd-v3.3.11-linux-amd64/etcd* /usr/local/bin/ sed -i '/ETCD_NAME/{s/etcd-01/etcd-03/g}' /etc/etcd/etcd.conf sed -i '/ETCD_LISTEN_PEER_URLS/{s/2.11/2.13/g}' /etc/etcd/etcd.conf sed -i '/ETCD_LISTEN_CLIENT_URLS/{s/2.11/2.13/g}' /etc/etcd/etcd.conf sed -i '/ETCD_INITIAL_ADVERTISE_PEER_URLS/{s/2.11/2.13/g}' /etc/etcd/etcd.conf sed -i '/ETCD_ADVERTISE_CLIENT_URLS/{s/2.11/2.13/g}' /etc/etcd/etcd.conf rm -rf ~/etcd-v3.3.11-linux-amd64* systemctl daemon-reload systemctl enable etcd.service --now systemctl status etcd

几个节点上的安装过程大同小异，唯一不同的是etcd配置文件中的服务器IP要写当前节点的IP。主要参数：

• ETCD_NAME：节点名称。
• ETCD_DATA_DIR：数据目录。
• ETCD_LISTEN_PEER_URLS：集群通信监听地址。
• ETCD_LISTEN_CLIENT_URLS：客户端访问监听地址。
• ETCD_INITIAL_ADVERTISE_PEER_URLS：集群通告地址。
• ETCD_ADVERTISE_CLIENT_URLS：客户端通告地址。
• ETCD_INITIAL_CLUSTER：集群节点地址。
• ETCD_INITIAL_CLUSTER_TOKEN：集群Token。
• ETCD_INITIAL_CLUSTER_STATE：加入集群的当前状态，new是新集群，existing表示加入已有集群。

3.4.3 验证etcd集群是否部署成功

执行以下命令：

[root@gysl-master ~]# etcdctl \ --ca-file=/etc/etcd/ssl/ca.pem \ --cert-file=/etc/etcd/ssl/server.pem \ --key-file=/etc/etcd/ssl/server-key.pem \ --endpoints="https://172.31.2.11:2379,https://172.31.2.12:2379,https://172.31.2.13:2379" cluster-health member 82184ce461853bed is healthy: got healthy result from https://172.31.2.12:2379 member d85d48cef1ccfeaf is healthy: got healthy result from https://172.31.2.13:2379 member fe6e7c664377ad3b is healthy: got healthy result from https://172.31.2.11:2379 cluster is healthy

"cluster is healthy"说明etcd集群部署成功！如果存在问题，那么首先看日志：/var/log/message 或 journalctl -u etcd，找到问题，逐一解决。命令看起来不是那么直观，可以直接复制下面的命令来进行检验：

etcdctl \ --ca-file=/etc/etcd/ssl/ca.pem \ --cert-file=/etc/etcd/ssl/server.pem \ --key-file=/etc/etcd/ssl/server-key.pem \ --endpoints="https://172.31.2.11:2379,https://172.31.2.12:2379,https://172.31.2.13:2379" cluster-health

3.5 部署Flannel网络

由于Flannel需要使用etcd存储自身的一个子网信息，所以要保证能成功连接Etcd，写入预定义子网段。写入的Pod网段${CLUSTER_CIDR}必须是/16段地址，必须与kube-controller-manager的–-cluster-cidr参数值一致。一般情况下，在每一个Node节点都需要进行配置，执行脚本KubernetesInstall-08.sh。

[root@gysl-master ~]# sh KubernetesInstall-08.sh

脚本内容如下：

#!/bin/bash KUBE_CONF=/etc/kubernetes FLANNEL_CONF=$KUBE_CONF/flannel.conf mkdir $KUBE_CONF tar -xvzf flannel-v0.11.0-linux-amd64.tar.gz mv {flanneld,mk-docker-opts.sh} /usr/local/bin/ # Check whether etcd cluster is healthy. etcdctl \ --ca-file=/etc/etcd/ssl/ca.pem \ --cert-file=/etc/etcd/ssl/server.pem \ --key-file=/etc/etcd/ssl/server-key.pem \ --endpoints="https://172.31.2.11:2379,\ https://172.31.2.12:2379,\ https://172.31.2.13:2379" cluster-health # Writing into a predetermined subnetwork. cd /etc/etcd/ssl etcdctl \ --ca-file=ca.pem --cert-file=server.pem --key-file=server-key.pem \ --endpoints="https://172.31.2.11:2379,https://172.31.2.12:2379,https://172.31.2.13:2379" \ set /coreos.com/network/config '{ "Network": "172.17.0.0/16", "Backend": {"Type": "vxlan"}}' cd ~ # Configuration the flannel service. cat>$FLANNEL_CONF<<EOF FLANNEL_OPTIONS="--etcd-endpoints=https://172.31.2.11:2379,https://172.31.2.12:2379,https://172.31.2.13:2379 -etcd-cafile=/etc/etcd/ssl/ca.pem -etcd-certfile=/etc/etcd/ssl/server.pem -etcd-keyfile=/etc/etcd/ssl/server-key.pem" EOF cat>/usr/lib/systemd/system/flanneld.service<<EOF [Unit] Description=Flanneld overlay address etcd agent After=network-online.target network.target Before=docker.service [Service] Type=notify EnvironmentFile=$FLANNEL_CONF ExecStart=/usr/local/bin/flanneld --ip-masq $FLANNEL_OPTIONS ExecStartPost=/usr/local/bin/mk-docker-opts.sh -k DOCKER_NETWORK_OPTIONS -d /run/flannel/subnet.env Restart=on-failure [Install] WantedBy=multi-user.target EOF # Modify the docker service. sed -i.bak -e '/ExecStart/i EnvironmentFile=\/run\/flannel\/subnet.env' -e 's/ExecStart=\/usr\/bin\/dockerd/ExecStart=\/usr\/bin\/dockerd $DOCKER_NETWORK_OPTIONS/g' /usr/lib/systemd/system/docker.service # Start or restart related services. systemctl daemon-reload systemctl enable flanneld --now systemctl restart docker systemctl status flanneld systemctl status docker ip address show

在脚本执行之前需要把Flannel安装包拷贝到用户的HOME目录。脚本执行完毕之后需要检查各服务的状态，确保docker0和flannel.1在同一网段。

3.6 部署Master节点

3.6.1 创建CA证书

这一步中创建了kube-apiserver和kube-proxy相关的CA证书，在Master节点执行脚本KubernetesInstall-09.sh。

[root@gysl-master ~]# sh KubernetesInstall-09.sh

脚本内容如下：

#!/bin/bash # Deploy the master node. KUBE_SSL=/etc/kubernetes/ssl mkdir $KUBE_SSL # Create CA. cat>$KUBE_SSL/ca-config.json<<EOF { "signing": { "default": { "expiry": "87600h" }, "profiles": { "kubernetes": { "expiry": "87600h", "usages": [ "signing", "key encipherment", "server auth", "client auth" ] } } } } EOF cat>$KUBE_SSL/ca-csr.json<<EOF { "CN": "kubernetes", "key": { "algo": "rsa", "size": 2048 }, "names": [ { "C": "CN", "L": "Beijing", "ST": "Beijing", "O": "k8s", "OU": "System" } ] } EOF cat>$KUBE_SSL/server-csr.json<<EOF { "CN": "kubernetes", "hosts": [ "10.0.0.1", "127.0.0.1", "172.31.2.11", "kubernetes", "kubernetes.default", "kubernetes.default.svc", "kubernetes.default.svc.cluster", "kubernetes.default.svc.cluster.local" ], "key": { "algo": "rsa", "size": 2048 }, "names": [ { "C": "CN", "L": "Beijing", "ST": "Beijing", "O": "k8s", "OU": "System" } ] } EOF cd $KUBE_SSL cfssl_linux-amd64 gencert -initca ca-csr.json | cfssljson_linux-amd64 -bare ca - cfssl_linux-amd64 gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes server-csr.json | cfssljson_linux-amd64 -bare server # Create kube-proxy CA. cat>$KUBE_SSL/kube-proxy-csr.json<<EOF { "CN": "system:kube-proxy", "hosts": [], "key": { "algo": "rsa", "size": 2048 }, "names": [ { "C": "CN", "L": "BeiJing", "ST": "BeiJing", "O": "k8s", "OU": "System" } ] } EOF cfssl_linux-amd64 gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes kube-proxy-csr.json | cfssljson_linux-amd64 -bare kube-proxy ls *.pem cd ~

执行完毕之后应该看到以下文件：
/etc/kubernetes/ssl/ca-key.pem /etc/kubernetes/ssl/kube-proxy-key.pem /etc/kubernetes/ssl/server-key.pem
/etc/kubernetes/ssl/ca.pem /etc/kubernetes/ssl/kube-proxy.pem /etc/kubernetes/ssl/server.pem

3.6.2 安装配置kube-apiserver服务

将备好的安装包解压，并移动到相关目录，进行相关配置，执行脚本KubernetesInstall-10.sh。

[root@gysl-master ~]# sh KubernetesInstall-10.sh

脚本内容如下：

#!/bin/bash KUBE_ETC=/etc/kubernetes KUBE_API_CONF=/etc/kubernetes/apiserver.conf tar -xvzf kubernetes-server-linux-amd64.tar.gz mv kubernetes/server/bin/{kube-apiserver,kube-scheduler,kube-controller-manager} /usr/local/bin/ # Create a token file. cat>$KUBE_ETC/token.csv<<EOF $(head -c 16 /dev/urandom | od -An -t x | tr -d ' '),kubelet-bootstrap,10001,"system:kubelet-bootstrap" EOF # Create a kube-apiserver configuration file. cat >$KUBE_API_CONF<<EOF KUBE_APISERVER_OPTS="--logtostderr=true \ --v=4 \ --etcd-servers=https://172.31.2.11:2379,https://172.31.2.12:2379,https://172.31.2.13:2379 \ --bind-address=172.31.2.11 \ --secure-port=6443 \ --advertise-address=172.31.2.11 \ --allow-privileged=true \ --service-cluster-ip-range=10.0.0.0/24 \ --enable-admission-plugins=NamespaceLifecycle,LimitRanger,SecurityContextDeny,ServiceAccount,ResourceQuota,NodeRestriction \ --authorization-mode=RBAC,Node \ --enable-bootstrap-token-auth \ --token-auth-file=$KUBE_ETC/token.csv \ --service-node-port-range=30000-50000 \ --tls-cert-file=$KUBE_ETC/ssl/server.pem \ --tls-private-key-file=$KUBE_ETC/ssl/server-key.pem \ --client-ca-file=$KUBE_ETC/ssl/ca.pem \ --service-account-key-file=$KUBE_ETC/ssl/ca-key.pem \ --etcd-cafile=/etc/etcd/ssl/ca.pem \ --etcd-certfile=/etc/etcd/ssl/server.pem \ --etcd-keyfile=/etc/etcd/ssl/server-key.pem" EOF # Create the kube-apiserver service. cat>/usr/lib/systemd/system/kube-apiserver.service<<EOF [Unit] Description=Kubernetes API Server Documentation=https://github.com/kubernetes/kubernetes After=etcd.service Wants=etcd.service [Service] EnvironmentFile=-$KUBE_API_CONF ExecStart=/usr/local/bin/kube-apiserver \$KUBE_APISERVER_OPTS Restart=on-failure [Install] WantedBy=multi-user.target EOF systemctl daemon-reload systemctl enable kube-apiserver.service --now systemctl status kube-apiserver.service

参数说明：

• --logtostderr：启用日志。
• --v：日志等级。
• --etcd-servers：etcd集群地址。
• --bind-address：监听地址。
• --secure-port：https安全端口。
• --advertise-address：集群通告地址。
• --allow-privileged：启用授权。
• --service-cluster-ip-range：Service虚拟IP地址段。
• --enable-admission-plugins：准入控制模块。
• --authorization-mode：认证授权，启用RBAC授权和节点自管理。
• --enable-bootstrap-token-auth：启用TLS bootstrap功能。
• --token-auth-file：token文件。
• --service-node-port-range：Service Node类型默认分配端口范围。

3.6.3 安装配置kube-scheduler服务

之前已经将kube-scheduler相关的二进制文件移动到了相关目录，直接执行脚本KubernetesInstall-11.sh。

[root@gysl-master ~]# sh KubernetesInstall-11.sh

脚本内容如下：

#!/bin/bash # Deploy the scheduler service. KUBE_ETC=/etc/kubernetes KUBE_SCHEDULER_CONF=$KUBE_ETC/kube-scheduler.conf cat>$KUBE_SCHEDULER_CONF<<EOF KUBE_SCHEDULER_OPTS="--logtostderr=true \ --v=4 \ --master=127.0.0.1:8080 \ --leader-elect" EOF cat>/usr/lib/systemd/system/kube-scheduler.service<<EOF [Unit] Description=Kubernetes Scheduler Documentation=https://github.com/kubernetes/kubernetes [Service] EnvironmentFile=-$KUBE_SCHEDULER_CONF ExecStart=/usr/local/bin/kube-scheduler \$KUBE_SCHEDULER_OPTS Restart=on-failure [Install] WantedBy=multi-user.target EOF systemctl daemon-reload systemctl enable kube-scheduler.service --now sleep 20 systemctl status kube-scheduler.service

参数说明：

• --master：连接本地apiserver。
• --leader-elect：当该组件启动多个时，自动选举（HA），被选为 leader的节点负责处理工作，其它节点为阻塞状态。

3.6.4 安装配置kube-controller服务

之前已经将kube-scheduler相关的二进制文件移动到了相关目录，直接执行脚本KubernetesInstall-12.sh。

[root@gysl-master ~]# sh KubernetesInstall-12.sh

脚本内容如下：

#!/bin/bash # Deploy the controller-manager service. KUBE_CONTROLLER_CONF=/etc/kubernetes/kube-controller-manager.conf cat>$KUBE_CONTROLLER_CONF<<EOF KUBE_CONTROLLER_MANAGER_OPTS="--logtostderr=true \ --v=4 \ --master=127.0.0.1:8080 \ --leader-elect=true \ --address=127.0.0.1 \ --service-cluster-ip-range=10.0.0.0/24 \ --cluster-name=kubernetes \ --cluster-signing-cert-file=/etc/kubernetes/ssl/ca.pem \ --cluster-signing-key-file=/etc/kubernetes/ssl/ca-key.pem \ --root-ca-file=/etc/kubernetes/ssl/ca.pem \ --service-account-private-key-file=/etc/kubernetes/ssl/ca-key.pem" EOF cat>/usr/lib/systemd/system/kube-controller-manager.service<<EOF [Unit] Description=Kubernetes Controller Manager Documentation=https://github.com/kubernetes/kubernetes [Service] EnvironmentFile=-$KUBE_CONTROLLER_CONF ExecStart=/usr/local/bin/kube-controller-manager \$KUBE_CONTROLLER_MANAGER_OPTS Restart=on-failure [Install] WantedBy=multi-user.target EOF systemctl daemon-reload systemctl enable kube-controller-manager.service --now sleep 20 systemctl status kube-controller-manager.service

3.6.5 查看集群状态

直接执行脚本KubernetesInstall-13.sh。

[root@gysl-master ~]# sh KubernetesInstall-13.sh

脚本内容如下：

#!/bin/bash # Check the service. mv kubernetes/server/bin/kubectl /usr/local/bin/ kubectl get cs

如果部署成功的话，将看到如下结果：

[root@gysl-master ~]# kubectl get cs NAME STATUS MESSAGE ERROR controller-manager Healthy ok scheduler Healthy ok etcd-2 Healthy {"health":"true"} etcd-0 Healthy {"health":"true"} etcd-1 Healthy {"health":"true"}

3.7 部署Node节点

3.7.1 创建bootstrap和kube-proxy的kubeconfig文件

Master apiserver启用TLS认证后，Node节点kubelet组件想要加入集群，必须使用CA签发的有效证书才能与apiserver通信，当Node节点很多时，签署证书是一件很繁琐的事情，因此有了TLS Bootstrapping机制，kubelet会以一个低权限用户自动向apiserver申请证书，kubelet的证书由apiserver动态签署。在前面创建的token文件在这一步派上了用场，在Master节点上执行脚本KubernetesInstall-14.sh创建bootstrap.kubeconfig和kube-proxy.kubeconfig。

[root@gysl-master ~]# sh KubernetesInstall-14.sh

脚本内容如下：

#!/bin/bash BOOTSTRAP_TOKEN=$(awk -F "," '{print $1}' /etc/kubernetes/token.csv) KUBE_SSL=/etc/kubernetes/ssl/ KUBE_APISERVER="https://172.31.2.11:6443" cd $KUBE_SSL # Set cluster parameters. kubectl config set-cluster kubernetes \ --certificate-authority=./ca.pem \ --embed-certs=true \ --server=${KUBE_APISERVER} \ --kubeconfig=bootstrap.kubeconfig # Set client parameters. kubectl config set-credentials kubelet-bootstrap \ --token=${BOOTSTRAP_TOKEN} \ --kubeconfig=bootstrap.kubeconfig # Set context parameters. kubectl config set-context default \ --cluster=kubernetes \ --user=kubelet-bootstrap \ --kubeconfig=bootstrap.kubeconfig # Set context. kubectl config use-context default --kubeconfig=bootstrap.kubeconfig # Create kube-proxy kubeconfig file. kubectl config set-cluster kubernetes \ --certificate-authority=./ca.pem \ --embed-certs=true \ --server=${KUBE_APISERVER} \ --kubeconfig=kube-proxy.kubeconfig kubectl config set-credentials kube-proxy \ --client-certificate=./kube-proxy.pem \ --client-key=./kube-proxy-key.pem \ --embed-certs=true \ --kubeconfig=kube-proxy.kubeconfig kubectl config set-context default \ --cluster=kubernetes \ --user=kube-proxy \ --kubeconfig=kube-proxy.kubeconfig kubectl config use-context default --kubeconfig=kube-proxy.kubeconfig cd ~ # Bind kubelet-bootstrap user to system cluster roles. kubectl create clusterrolebinding kubelet-bootstrap \ --clusterrole=system:node-bootstrapper \ --user=kubelet-bootstrap

3.7.2 配置kube-proxy和kubelet服务

因为kubernetes-server-linux-amd64.tar.gz已经在Master节点的HOME目录解压，所以可以在各节点上执行脚本KubernetesInstall-15.sh。

[root@gysl-node1 ~]# sh KubernetesInstall-15.sh

脚本内容如下：

#!/bin/bash KUBE_CONF=/etc/kubernetes KUBE_SSL=$KUBE_CONF/ssl IP=172.31.2.13 mkdir $KUBE_SSL scp gysl-master:~/kubernetes/server/bin/{kube-proxy,kubelet} /usr/local/bin/ scp gysl-master:$KUBE_CONF/ssl/{bootstrap.kubeconfig,kube-proxy.kubeconfig} $KUBE_CONF cat>$KUBE_CONF/kube-proxy.conf<<EOF KUBE_PROXY_OPTS="--logtostderr=true \ --v=4 \ --hostname-override=$IP \ --cluster-cidr=10.0.0.0/24 \ --kubeconfig=$KUBE_CONF/kube-proxy.kubeconfig" EOF cat>/usr/lib/systemd/system/kube-proxy.service<<EOF [Unit] Description=Kubernetes Proxy After=network.target [Service] EnvironmentFile=-$KUBE_CONF/kube-proxy.conf ExecStart=/usr/local/bin/kube-proxy \$KUBE_PROXY_OPTS Restart=on-failure [Install] WantedBy=multi-user.target EOF systemctl daemon-reload systemctl enable kube-proxy.service --now sleep 20 systemctl status kube-proxy.service -l cat>$KUBE_CONF/kubelet.yaml<<EOF kind: KubeletConfiguration apiVersion: kubelet.config.k8s.io/v1beta1 address: $IP port: 10250 readOnlyPort: 10255 cgroupDriver: cgroupfs clusterDNS: ["10.0.0.2"] clusterDomain: cluster.local. failSwapOn: false authentication: anonymous: enabled: true EOF cat>$KUBE_CONF/kubelet.conf<<EOF KUBELET_OPTS="--logtostderr=true \ --v=4 \ --hostname-override=$IP \ --kubeconfig=$KUBE_CONF/kubelet.kubeconfig \ --bootstrap-kubeconfig=$KUBE_CONF/bootstrap.kubeconfig \ --config=$KUBE_CONF/kubelet.yaml \ --cert-dir=$KUBE_SSL \ --pod-infra-container-image=registry.cn-hangzhou.aliyuncs.com/google-containers/pause-amd64:3.0" EOF cat>/usr/lib/systemd/system/kubelet.service<<EOF [Unit] Description=Kubernetes Kubelet After=docker.service Requires=docker.service [Service] EnvironmentFile=$KUBE_CONF/kubelet.conf ExecStart=/usr/local/bin/kubelet \$KUBELET_OPTS Restart=on-failure KillMode=process [Install] WantedBy=multi-user.target EOF systemctl daemon-reload systemctl enable kubelet.service --now sleep 20 systemctl status kubelet.service -l

以上脚本有多少个Node节点就在相应的Node节点上执行多少次，每次执行只需修改IP的值即可。

参数说明：

• --hostname-override：在集群中显示的主机名。
• --kubeconfig：指定kubeconfig文件位置，会自动生成。
• --bootstrap-kubeconfig：指定刚才生成的bootstrap.kubeconfig文件。
• --cert-dir：颁发证书存放位置。
• --pod-infra-container-image：管理Pod网络的镜像。

3.7.3 Approve kubelet CSR请求

可以手动或自动approve CSR请求。推荐使用自动的方式，因为从 v1.8 版本开始，可以自动轮转approve csr后生成的证书。未approve之前如下：

[root@gysl-master ~]# kubectl get csr NAME AGE REQUESTOR CONDITION node-csr-FpTP2sCI0SiYDCxaIHa1SRukS_5u9BQN10BsTd6RU1Y 20m kubelet-bootstrap Pending node-csr-YYfnPwAws2LxJzV-OgYjJ22zy_z9XQM8PT0MnqZN910 24m kubelet-bootstrap Pending

在Master节点上执行脚本KubernetesInstall-15.sh。

[root@gysl-master ~]# sh KubernetesInstall-15.sh certificatesigningrequest.certificates.k8s.io/node-csr-FpTP2sCI0SiYDCxaIHa1SRukS_5u9BQN10BsTd6RU1Y approved certificatesigningrequest.certificates.k8s.io/node-csr-YYfnPwAws2LxJzV-OgYjJ22zy_z9XQM8PT0MnqZN910 approved

脚本内容如下：

#!/bin/bash CSRS=$(kubectl get csr | awk '{if(NR>1) print $1}') for csr in $CSRS; do kubectl certificate approve $csr; done

3.7.4 查看集群状态

[root@gysl-master ~]# kubectl get cs NAME STATUS MESSAGE ERROR controller-manager Healthy ok scheduler Healthy ok etcd-2 Healthy {"health":"true"} etcd-1 Healthy {"health":"true"} etcd-0 Healthy {"health":"true"} [root@gysl-master ~]# kubectl get node NAME STATUS ROLES AGE VERSION 172.31.2.12 Ready <none> 11m v1.13.2 172.31.2.13 Ready <none> 11m v1.13.2

3.8 运行一个测试

[root@gysl-master ~]# kubectl run nginx --image=nginx --replicas=3 kubectl run --generator=deployment/apps.v1 is DEPRECATED and will be removed in a future version. Use kubectl run --generator=run-pod/v1 or kubectl create instead. deployment.apps/nginx created [root@gysl-master ~]# kubectl expose deployment nginx --port=88 --target-port=80 --type=NodePort service/nginx exposed [root@gysl-master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE nginx-7cdbd8cdc9-7h946 0/1 ContainerCreating 0 33s nginx-7cdbd8cdc9-vtkqf 0/1 ContainerCreating 0 33s nginx-7cdbd8cdc9-wdjtj 0/1 ContainerCreating 0 33s [root@gysl-master ~]# kubectl get svc NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE kubernetes ClusterIP 10.0.0.1 <none> 443/TCP 8h nginx NodePort 10.0.0.2 <none> 88:46705/TCP 28s [root@gysl-master ~]# kubectl get pods NAME READY STATUS RESTARTS AGE nginx-7cdbd8cdc9-7h946 1/1 Running 0 2m4s nginx-7cdbd8cdc9-vtkqf 1/1 Running 0 2m4s nginx-7cdbd8cdc9-wdjtj 1/1 Running 0 2m4s

[root@gysl-node1 ~]# curl http://10.0.0.2:88 <!DOCTYPE html> <html> <head> <title>Welcome to nginx!</title> <style> body { width: 35em; margin: 0 auto; font-family: Tahoma, Verdana, Arial, sans-serif; } </style> </head> <body> <h1>Welcome to nginx!</h1> <p>If you see this page, the nginx web server is successfully installed and working. Further configuration is required.</p> <p>For online documentation and support please refer to <a href="http://nginx.org/">nginx.org</a>.<br/> Commercial support is available at <a href="http://nginx.com/">nginx.com</a>.</p> <p><em>Thank you for using nginx.</em></p> </body> </html>

如果此时在浏览器输入：http://10.0.0.2:88 ，那么将出现nginx的默认页面。

3.9 在Master节点部署Node节点的相关组件

资源比较充裕的情况下，Master节点仅仅做为服务接口、调度、控制节点，必须部署的组件有：kube-apiserver、kube-controller-manager、kube-scheduler、kubectl、etcd。除此之外，一般还需要做HA等相关部署。如果Master节点资源比较充裕，有些实验也要求至少有三个节点在运行，那么也可以将Master节点部署设置为一般Node节点来使用。为此，直接执行脚本KubernetesInstall-17.sh。

[root@gysl-master ~]# KubernetesInstall-17.sh

脚本内容如下：

#!/bin/bash KUBE_CONF=/etc/kubernetes KUBE_SSL=$KUBE_CONF/ssl IP=172.31.2.11 mkdir $KUBE_SSL cp ~/kubernetes/server/bin/{kube-proxy,kubelet} /usr/local/bin/ cp $KUBE_CONF/ssl/{bootstrap.kubeconfig,kube-proxy.kubeconfig} $KUBE_CONF cat>$KUBE_CONF/kube-proxy.conf<<EOF KUBE_PROXY_OPTS="--logtostderr=true \ --v=4 \ --hostname-override=$IP \ --cluster-cidr=10.0.0.0/24 \ --kubeconfig=$KUBE_CONF/kube-proxy.kubeconfig" EOF cat>/usr/lib/systemd/system/kube-proxy.service<<EOF [Unit] Description=Kubernetes Proxy After=network.target [Service] EnvironmentFile=-$KUBE_CONF/kube-proxy.conf ExecStart=/usr/local/bin/kube-proxy \$KUBE_PROXY_OPTS Restart=on-failure [Install] WantedBy=multi-user.target EOF systemctl daemon-reload systemctl enable kube-proxy.service --now sleep 20 systemctl status kube-proxy.service -l cat>$KUBE_CONF/kubelet.yaml<<EOF kind: KubeletConfiguration apiVersion: kubelet.config.k8s.io/v1beta1 address: $IP port: 10250 readOnlyPort: 10255 cgroupDriver: cgroupfs clusterDNS: ["10.0.0.2"] clusterDomain: cluster.local. failSwapOn: false authentication: anonymous: enabled: true EOF cat>$KUBE_CONF/kubelet.conf<<EOF KUBELET_OPTS="--logtostderr=true \ --v=4 \ --hostname-override=$IP \ --kubeconfig=$KUBE_CONF/kubelet.kubeconfig \ --bootstrap-kubeconfig=$KUBE_CONF/bootstrap.kubeconfig \ --config=$KUBE_CONF/kubelet.yaml \ --cert-dir=$KUBE_SSL \ --pod-infra-container-image=registry.cn-hangzhou.aliyuncs.com/google-containers/pause-amd64:3.0" EOF cat>/usr/lib/systemd/system/kubelet.service<<EOF [Unit] Description=Kubernetes Kubelet After=docker.service Requires=docker.service [Service] EnvironmentFile=$KUBE_CONF/kubelet.conf ExecStart=/usr/local/bin/kubelet \$KUBELET_OPTS Restart=on-failure KillMode=process [Install] WantedBy=multi-user.target EOF systemctl daemon-reload systemctl enable kubelet.service --now sleep 20 systemctl status kubelet.service -l kubectl certificate approve $(kubectl get csr | awk '{if(NR>1) print $1}') kubectl get csr kubectl label node 172.31.2.11 node-role.kubernetes.io/master='master' kubectl label node 172.31.2.12 node-role.kubernetes.io/node='node' kubectl label node 172.31.2.13 node-role.kubernetes.io/node='node' kubectl get nodes

部署成功之后，将出现以下内容：

NAME STATUS ROLES AGE VERSION 172.31.2.11 Ready master 22m v1.13.2 172.31.2.12 Ready node 11h v1.13.2 172.31.2.13 Ready node 11h v1.13.2

四 总结

4.1 Kubernetes的二进制安装部署是一个比较复杂的过程，其中涉及到的步骤比较多，需要理解清楚各节点及组件之间的关系，逐步进行，每一个步骤成功了再进行下一步，切不可急躁。

4.2 在安装部署的过程中，日志及帮助信息是十分重要的，journalctl命令较为常用，--help也会起到柳暗花明又一村的效果。

4.3 把执行步骤脚本化，显得清晰有效，在后续的工作、学习过程中要继续保持。

4.4 由于时间仓促，安装部署中的很多个性化配置并未配置，在后续过程中会根据实际使用情况进行完善。比如：每一个服务或组件并未将日志单独保存。

4.5 其他不尽如人意的地方正在完善。

4.6 文中的两张图片来源于互联网，如有侵权，请联系删除。

五 参考资料

5.1 认证相关

5.2 证书相关

5.3 cfssl官方资料

5.4 Systemd相关资料)

5.5 Kubernetes基本概念

5.6 本文涉及到的脚本及配置文件