1. 架构设计和环境设计

1.1. 架构设计

  1. 部署 Haproxy 为 Kubernetes 提供 Endpoint 访问入口
  2. 使用 Keepalived 将 Endpoint 入口地址设置为 Virtual IP 并通过部署多台节点的方式实现冗余
  3. 使用 kubeadm 部署高可用 Kubernetes 集群, 指定 Endpoint IP 为 Keepalived 生成的 Virtual IP
  4. 使用 prometheus 作为 Kubernetes 的集群监控系统, 使用 grafana 作为图表监控图表展示系统, 使用 alertmanager 作为报警系统
  5. 使用 jenkins + gitlab + harbor 构建 CI/CD 系统
  6. 使用单独的域名在 Kubernetes 集群内进行通信, 在内网搭建 DNS 服务用于解析域名
    Kubernetes 系列第二篇: Kubernetes 架构设计和部署_第1张图片

1.2. 环境设计

主机名 IP 角色
kube-master-01.sk8s.io-01.sk8s.io 192.168.0.201 k8s master, haprxoy + keepalived(虚拟IP: 192.168.0.250)
kube-master-01.sk8s.io-02.sk8s.io 192.168.0.202 k8s master, haprxoy + keepalived(虚拟IP: 192.168.0.250)
kube-master-01.sk8s.io-03.sk8s.io 192.168.0.203 k8s master, DNS, Storage, GitLab, Harbor
kube-node-01.sk8s.io 192.168.0.204 node
kube-node-02.sk8s.io 192.168.0.205 node

2. 操作系统初始化设置

2.1. 关闭 SELINUX

[root@localhost ~]# setenforce 0
[root@localhost ~]# sed -i 's#^SELINUX=.*#SELINUX=disabled#' /etc/sysconfig/selinux
[root@localhost ~]# sed -i 's#^SELINUX=.*#SELINUX=disabled#' /etc/selinux/config

2.2. 关闭无用服务

[root@localhost ~]# systemctl disable firewalld postfix auditd kdump NetworkManager

2.3. 升级系统内核

[root@master ~]# rpm --import https://www.elrepo.org/RPM-GPG-KEY-elrepo.org
[root@master ~]# rpm -Uvh http://www.elrepo.org/elrepo-release-7.0-3.el7.elrepo.noarch.rpm 
[root@master ~]# yum -y --disablerepo=\* --enablerepo=elrepo-kernel install kernel-lt.x86_64 kernel-lt-devel.x86_64 kernel-lt-headers.x86_64
[root@master ~]# yum -y remove kernel-tools-libs.x86_64 kernel-tools.x86_64
[root@master ~]# yum -y --disablerepo=\* --enablerepo=elrepo-kernel install kernel-lt-tools.x86_64
[root@master ~]# cat < /etc/default/grub
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
GRUB_DEFAULT=0
GRUB_DISABLE_SUBMENU=true
GRUB_TERMINAL_OUTPUT="console"
GRUB_CMDLINE_LINUX="crashkernel=auto console=ttyS0 console=tty0 panic=5"
GRUB_DISABLE_RECOVERY="true"
GRUB_TERMINAL="serial console"
GRUB_TERMINAL_OUTPUT="serial console"
GRUB_SERIAL_COMMAND="serial --speed=9600 --unit=0 --word=8 --parity=no --stop=1"
EOF
[root@localhost ~]# grub2-mkconfig -o /boot/grub2/grub.cfg

2.4. 统一网卡名称

[root@localhost ~]# grub_cinfig='GRUB_CMDLINE_LINUX="crashkernel=auto ipv6.disable=1 net.ifnames=0 rd.lvm.lv=centos/root rd.lvm.lv=centos/swap rhgb quiet"'
[root@localhost ~]# sed -i "s#GRUB_CMDLINE_LINUX.*#${grub_cinfig}#" /etc/default/grub
[root@localhost ~]# grub2-mkconfig -o /boot/grub2/grub.cfg

# ATTR{address} 为网卡 MAC 地址, NAME 为修改后的地址
[root@localhost ~]# cat /etc/udev/rules.d/70-persistent-net.rules 
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="ec:xx:yy:cc:b6:xx", NAME="eth0"
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="ec:xx:yy:cc:b6:xx", NAME="eth1"

# 重启之前先修改 /etc/sysconfig/network-scripts/ 下面的网卡配置文件
[root@localhost ~]# reboot

2.5. 其他配置

[root@localhost ~]# yum -y install vim net-tools lrzsz lbzip2 bzip2 ntpdate curl wget psmisc
[root@localhost ~]# timedatectl set-timezone Asia/Shanghai
[root@localhost ~]# echo "nameserver 223.5.5.5" > /etc/resolv.conf
[root@localhost ~]# echo "nameserver 114.114.114.114" >> /etc/resolv.conf
[root@localhost ~]# echo 'LANG="en_US.UTF-8"' > /etc/locale.conf
[root@localhost ~]# echo 'export LANG="en_US.UTF-8"' >> /etc/profile.d/custom.sh

[root@localhost ~]# cat >> /etc/security/limits.conf < /etc/sysctl.d/k8s.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
vm.overcommit_memory=1
vm.panic_on_oom=0
fs.inotify.max_user_instances=8192
fs.inotify.max_user_watches=1048576
fs.file-max=52706963
fs.nr_open=52706963
net.ipv4.tcp_sack = 0
EOF

[root@localhost ~]# sysctl -p /etc/sysctl.d/k8s.conf

2.6. 配置 Hosts Deny

[root@localhost ~]# echo "sshd:192.168.0." > /etc/hosts.allow
[root@localhost ~]# echo "sshd:ALL" > /etc/hosts.deny

2.7. ssh 配置

# 创建管理员用户, 并生成 ssh key(将私钥下载下来, 禁止在服务器留存; 将公钥复制到其他服务器上的 ~/.ssh/authorized_keys)
[root@localhost ~]# useradd huyuan
[root@localhost ~]# echo "sycx123" | passwd --stdin huyuan
[root@localhost ~]# su - huyuan
[root@localhost ~]# ssh-keygen -b 4096
[root@localhost ~]# mv ~/.ssh/id_rsa.pub ~/.ssh/authorized_keys

# 回到 root 用户
[root@localhost ~]# exit

# 禁止 DNS 反解, 优化 SSH 连接速度
[root@localhost ~]# sed -i 's/^#UseDNS.*/UseDNS no/' /etc/ssh/sshd_config 

# 禁用密码认证
[root@localhost ~]# sed -i 's/^PasswordAuthentication.*/PasswordAuthentication no/' /etc/ssh/sshd_config

# 禁止 root 用户登录
[root@localhost ~]# sed -i 's/#PermitRootLogin.*/PermitRootLogin no/' /etc/ssh/sshd_config

# 只允许 huyuan 登录服务器, 多个用户使用空格进行分隔
[root@localhost ~]# echo "AllowUsers huyuan" >> /etc/ssh/sshd_config

# 重启服务
[root@localhost ~]# systemctl restart sshd

2.7. 设置统一 root 密码

[root@localhost ~]# echo "xxxxx" | passwd --stdin root 

2.8. 设置主机名

[root@localhost ~]# hostnamectl set-hostname kube-master-01.sk8s.io-01.sk8s.io
[root@localhost ~]# echo "192.168.0.201 kube-master-01.sk8s.io-01.sk8s.io" >> /etc/hosts
[root@localhost ~]# echo "192.168.0.202 kube-master-01.sk8s.io-02.sk8s.io" >> /etc/hosts
[root@localhost ~]# echo "192.168.0.203 kube-master-01.sk8s.io-03.sk8s.io" >> /etc/hosts
[root@localhost ~]# echo "192.168.0.204 kube-node-01.sk8s.io" >> /etc/hosts
[root@localhost ~]# echo "192.168.0.205 kube-node-02.sk8s.io" >> /etc/hosts

3. 初始化 Kubernetes 集群

3.1. 安装和配置 docker (所有节点)

[[email protected] ~]# yum install -y yum-utils device-mapper-persistent-data lvm2
[[email protected] ~]# yum-config-manager --add-repo http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
[[email protected] ~]# yum -y install docker-ce-18.09.6 docker-ce-cli-18.09.6
[[email protected] ~]# cat /etc/docker/daemon.json 
{
"registry-mirrors": ["https://c7i79lkw.mirror.aliyuncs.com"],
"insecure-registries": ["122.228.208.72:9000"],
"exec-opts": ["native.cgroupdriver=systemd"],
"log-driver": "json-file",
"graph": "/opt/docker",
"log-opts": {
    "max-size": "100m"
},
"storage-driver": "overlay2"
}

[[email protected] ~]# systemctl enable docker
[[email protected] ~]# systemctl start docker

3.2. 配置 haproxy 作为 ApiServer 代理

# 在 kube-master-01.sk8s.io01 和 kube-master-01.sk8s.io02 主机上安装和配置
[[email protected] ~]# yum -y install haproxy
[[email protected] ~]# cat > /etc/haproxy/haproxy.cfg << EOF
global
    log 127.0.0.1  local2 notice
    chroot /var/lib/haproxy
    stats socket /var/run/haproxy.sock mode 660 level admin
    stats timeout 30s
    user haproxy
    group haproxy
    daemon
    nbproc 1
defaults
    log     global
    timeout connect 5000
    timeout client  10m
    timeout server  10m
listen  admin_stats
    bind 0.0.0.0:8000
    mode http
    stats refresh 30s
    stats uri /status
    stats realm welcome login\ Haproxy
    stats auth admin:tuitui99
    stats hide-version
    stats admin if TRUE
listen kube-master-01.sk8s.io
    bind 0.0.0.0:8443
    mode tcp
    option tcplog
    balance source
    server 192.168.0.201 192.168.0.201:6443 check inter 2000 fall 2 rise 2 weight 1
    server 192.168.0.202 192.168.0.202:6443 check inter 2000 fall 2 rise 2 weight 1
    server 192.168.0.203 192.168.0.203:6443 check inter 2000 fall 2 rise 2 weight 1
EOF

[[email protected] ~]# systemctl enable haproxy
[[email protected] ~]# systemctl start haproxy

3.3. 配置 keepalived 为 haproxy 做主从备份

# 在 kube-master-01.sk8s.io01 和 kube-master-01.sk8s.io02 主机上安装和配置
[[email protected] ~]# yum -y install keepalived
[[email protected] ~]# cp /etc/keepalived/keepalived.conf{,.bak}
[[email protected] ~]# cat > /etc/keepalived/keepalived.conf << EOF
global_defs {
    # 唯一表示 192.168.0.201 节点 ip
    router_id master-192.168.0.201
    # 执行 notify_master notify_backup notify_fault 等脚本的用户
    script_user root
}
vrrp_script check-haproxy {
    # 检测进程是否存在
    script "/bin/killall -0 haproxy &>/dev/null"
    interval 5
    weight -30
    user root
}
vrrp_instance k8s {
    state MASTER
    # 设置优先级, 主服务器为 120, 从服务器为 100
    priority 120
    dont_track_primary
    interface eth0
    virtual_router_id 80
    advert_int 3
    track_script {
        check-haproxy
    }
    authentication {
        auth_type PASS
        auth_pass tuitui99
    }
    virtual_ipaddress {
        # 设置虚拟 ip
        192.168.0.254
    }
    # 脚本参考 https://blog.51cto.com/hongchen99/2298896
    notify_master "/bin/python /etc/keepalived/notify_keepalived.py master"
    notify_backup "/bin/python /etc/keepalived/notify_keepalived.py backup"
    notify_fault "/bin/python /etc/keepalived/notify_keepalived.py fault"
}
EOF

[[email protected] ~]# chmod +x /etc/keepalived/notify_keepalived.py
[[email protected] ~]# systemctl enable keepalived
[[email protected] ~]# systemctl start keepalived

3.4. 配置 haproxy 和 keepalived 日志

# 配置 haproxy 日志
[[email protected] ~]# echo "local2.*  /var/log/haproxy.log" >> /etc/rsyslog.conf

# 配置 keepalived 日志
[[email protected] ~]# cp /etc/sysconfig/keepalived{,.bak}
[[email protected] ~]# echo KEEPALIVED_OPTIONS="-D -d -S 0" > /etc/sysconfig/keepalived
[[email protected] ~]# echo "local0.*   /var/log/keepalived.log" >> /etc/rsyslog.conf

# 由于 haproxy 日志通过 udp 传输, 需要打开 rsyslog 的 udp 端口, 在 rsyslog 里面, 去掉下面两个变量的注释
[[email protected] ~]# cat /etc/rsyslog.conf
$ModLoad imudp
$UDPServerRun 514

[[email protected] ~]# systemctl restart rsyslog
[[email protected] ~]# systemctl restart haproxy
[[email protected] ~]# systemctl restart keepalived

3.5. 安装 kubelet kubeadm 和 kubectl

[[email protected] ~]# 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

[[email protected] ~]# yum install -y kubelet-1.14.0 kubeadm-1.14.0 kubectl-1.14.0
[[email protected] ~]# systemctl enable kubelet

3.6. 初始化 kubernetes 集群

# 加载 ipvs 模块
[[email protected] ~]# cat > /etc/sysconfig/modules/ipvs.modules << EOF
#!/bin/bash
modprobe -- ip_vs
modprobe -- ip_vs_rr
modprobe -- ip_vs_wrr
modprobe -- ip_vs_sh
modprobe -- nf_conntrack_ipv4
EOF

[[email protected] ~]# chmod +x /etc/sysconfig/modules/ipvs.modules
[[email protected] ~]# sh /etc/sysconfig/modules/ipvs.modules
[[email protected] ~]# lsmod | grep ip_vs

# 安装 ipvsadm 管理 ipvs
[[email protected] ~]# yum -y install ipvsadm

# 编写初始化配置文件
[[email protected] ~]# cat > kubeadm-init.yaml << EOF
apiVersion: kubeadm.k8s.io/v1beta1
kind: ClusterConfiguration
kubernetesVersion: v1.14.0
# 192.168.0.254 为虚拟 IP, 8443 为 haproxy 监听的端口
controlPlaneEndpoint: "192.168.0.254:8443"
# 设置拉取初始化镜像地址
imageRepository: registry.cn-hangzhou.aliyuncs.com/google_containers
networking:
  podSubnet: 10.244.0.0/16

apiServer:
  certSANs:
  - "kube-master-01.sk8s.io01"
  - "kube-master-01.sk8s.io02"
  - "kube-master-01.sk8s.io03"
  - "192.168.0.201"
  - "192.168.0.202"
  - "192.168.0.203"
  - "192.168.0.254"
  - "127.0.0.1"

---
apiVersion: kubeproxy.config.k8s.io/v1alpha1
kind: KubeProxyConfiguration
mode: "ipvs"
EOF

# 初始化 kubernetes 集群 --experimental-upload-certs 共享证书
[[email protected] ~]# kubeadm init --config=kubeadm-init.yaml --experimental-upload-certs

# 创建 kubernetes 集群管理用户
[[email protected] ~]# groupadd -g 5000 kubelet
[[email protected] ~]# useradd -c "kubernetes-admin-user" -G docker -u 5000 -g 5000 kubelet
[[email protected] ~]# echo "kubelet" | passwd --stdin kubelet

# 复制 kubernetes 集群配置文件到管理用户
[[email protected] ~]# mkdir /home/kubelet/.kube
[[email protected] ~]# cp -i /etc/kubernetes/admin.conf /home/kubelet/.kube/config
[[email protected] ~]# chown -R kubelet:kubelet /home/kubelet/.kube

3.7. 更新 coredns

[[email protected] ~]# su - kubelet
[[email protected] ~]$ cat > coredns.yaml << EOF
apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    k8s-app: kube-dns
  name: coredns
  namespace: kube-system
spec:
  replicas: 3
  selector:
    matchLabels:
      k8s-app: kube-dns
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 1
    type: RollingUpdate
  template:
    metadata:
      labels:
        k8s-app: kube-dns
    spec:
      affinity:
        podAntiAffinity:
          preferredDuringSchedulingIgnoredDuringExecution:
          - weight: 100
            podAffinityTerm:
              labelSelector:
                matchExpressions:
                - key: k8s-app
                  operator: In
                  values:
                  - kube-dns
              topologyKey: kubernetes.io/hostname
      containers:
      - args:
        - -conf
        - /etc/coredns/Corefile
        image: coredns/coredns:1.5.0
        imagePullPolicy: IfNotPresent
        livenessProbe:
          failureThreshold: 5
          httpGet:
            path: /health
            port: 8080
            scheme: HTTP
          initialDelaySeconds: 60
          periodSeconds: 10
          successThreshold: 1
          timeoutSeconds: 5
        name: coredns
        ports:
        - containerPort: 53
          name: dns
          protocol: UDP
        - containerPort: 53
          name: dns-tcp
          protocol: TCP
        - containerPort: 9153
          name: metrics
          protocol: TCP
        resources:
          limits:
            memory: 170Mi
          requests:
            cpu: 100m
            memory: 70Mi
        securityContext:
          allowPrivilegeEscalation: false
          capabilities:
            add:
            - NET_BIND_SERVICE
            drop:
            - all
          readOnlyRootFilesystem: true
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
        volumeMounts:
        - mountPath: /etc/coredns
          name: config-volume
          readOnly: true
      dnsPolicy: Default
      restartPolicy: Always
      schedulerName: default-scheduler
      securityContext: {}
      serviceAccount: coredns
      serviceAccountName: coredns
      terminationGracePeriodSeconds: 30
      tolerations:
      - key: CriticalAddonsOnly
        operator: Exists
      - effect: NoSchedule
        key: node-role.kubernetes.io/master
      - effect: NoSchedule
        key: node.kubernetes.io/not-ready
      volumes:
      - configMap:
          defaultMode: 420
          items:
          - key: Corefile
            path: Corefile
          name: coredns
        name: config-volume
EOF

[[email protected] ~]$ kubectl apply -f coredns.yaml

3.8. 其他节点加入集群

# 其他 master 加入集群
[[email protected] ~]$ kubeadm join 192.168.0.254:8443 --token h4n7uy.5qibssxu27vveko5 \
--discovery-token-ca-cert-hash sha256:a27738a4457d57ee611dd1c0281aeaabd32bc834797fe307980b95755b052e41 \
--experimental-control-plane --certificate-key eb37e5810fe300a42c5b610117ad57acf682a92da928cf94435a135aa338bc12

# 其他 node 加入集群
[[email protected] ~]$ kubeadm join 192.168.0.254:8443 --token h4n7uy.5qibssxu27vveko5 \
--discovery-token-ca-cert-hash sha256:a27738a4457d58ee611dd1c0281aeaabd34bc834797fe307980b95755b052e41