K8s1.20 二进制部署&高可用扩容

MrZ • 2024-03-06 • 云原生, 默认分类 • 80 阅读

20230410 创建

20230412 修订：更新了部分images版本兼容问题，增加了重启主机测试cs的步骤，优化了部分sed指令

20230412 修订：修正了keepalived vip地址，修复ha dial tcp失败问题

20230609 修订：删除了nginx conf 80口暴露配置 以免nginx ingress controller 配置冲突

20230609 修订: 优化ssh免密脚本，节点改造为3master 5worker，lvs复用master节点

20230619 修订: 优化image离线批量导入，所有节点需要离线导入image

1 环境说明

1.1 主机规划

  • 3主5从，其中master与etcd复用
主机IP OS 主机名 角色
10.168.1.101 OracleLinux7.9 k8s-master01 kube-apiserver，kube-controller-manager，kube-scheduler，kubelet，kube-proxy，docker，nginx，keepalived，etcd
10.168.1.102 OracleLinux7.9 k8s-master02 kube-apiserver，kube-controller-manager，kube-scheduler，kubelet，kube-proxy，docker，nginx，keepalived，etcd
10.168.1.103 OracleLinux7.9 k8s-master03 kube-apiserver，kube-controller-manager，kube-scheduler，kubelet，kube-proxy，docker，nginx，keepalived，etcd
10.168.1.111 OracleLinux7.9 k8s-node01 kubelet，kube-proxy，docker
10.168.1.112 OracleLinux7.9 k8s-node02 kubelet，kube-proxy，docker
10.168.1.113 OracleLinux7.9 k8s-node03 kubelet，kube-proxy，docker
10.168.1.114 OracleLinux7.9 k8s-node04 kubelet，kube-proxy，docker
10.168.1.115 OracleLinux7.9 k8s-node05 kubelet，kube-proxy，docker
10.168.1.110 OracleLinux7.9 k8s-master-lb Nginx+Keepalived 不占硬件资源，在3台master上

1.2 IP规划

网络地址范围 角色 说明
10.96.0.0/16 pod 网段 kube-controller-manager中的--cluster-cidr字段
10.244.0.0/16 service 网段 后面搭建apiserver、kube-controller-manager的时候定义
10.244.0.1 ubernetes ClusterIP 集群service的第一个IP，自动分配
10.244.0.2 kube-dns ClusterIP 部署CoreDNS的时候需要把coredns.yaml中clusterIP字段修改成10.244.0.2
3000-32768 service pods ip pods IP地址的nodeport可用端口范围

部署之前先将deploy.zip的deploy文件夹上传到/opt/下，其中/root/k8s/deploy是tls生成目录，真实应用和ssl都在/opt目录下

2 部署流程一 : 3 节点master集群部署

2.1 master规划

k8s-master01      10.168.1.101     kube-apiserver，kube-controller-manager，kube-scheduler，etcd
k8s-master02      10.168.1.102     kube-apiserver，kube-controller-manager，kube-scheduler，etcd
k8s-master03      10.168.1.103     kube-apiserver，kube-controller-manager，kube-scheduler，etcd

2.2 修改hosts

修改主机名：

hostnamectl set-hostname k8s-master01 && bash  # master01执行
hostnamectl set-hostname k8s-master02 && bash # master02执行
hostnamectl set-hostname k8s-master03 && bash # master03执行
hostnamectl set-hostname k8s-node01  && bash  # node01执行
hostnamectl set-hostname k8s-node02  && bash  # node02执行
hostnamectl set-hostname k8s-node03  && bash  # node03执行
hostnamectl set-hostname k8s-node04  && bash  # node04执行
hostnamectl set-hostname k8s-node05  && bash  # node05执行

添加hosts（所有节点）:

cat >>  /etc/hosts << EOF
10.168.1.101 k8s-master01
10.168.1.102 k8s-master02
10.168.1.103 k8s-master03
10.168.1.110 k8s-master-lb
10.168.1.111 k8s-node01
10.168.1.112 k8s-node02
10.168.1.113 k8s-node03
10.168.1.114 k8s-node04
10.168.1.115 k8s-node05
EOF

2.3 OS优化

NTP

#如果是虚拟机同步主机时间则不需要配置
yum install ntp -y && ntpdate ntpdate ntp1.aliyun.com
systemctl stop firewalld && systemctl disable firewalld
sed -i 's/enforcing/disabled/' /etc/selinux/config
setenforce 0
sed -ri 's/.*swap.*/#&/' /etc/fstab
swapoff -a

开机加载br_netfilter:

modprobe br_netfilter       
cat > /etc/rc.sysinit << EOF
#!/bin/bash
for file in /etc/sysconfig/modules/*.modules ; do
[ -x $file ] && $file
done
EOF
cat > /etc/sysconfig/modules/br_netfilter.modules << EOF
modprobe br_netfilter
EOF
chmod 755 /etc/sysconfig/modules/br_netfilter.modules
lsmod |grep br_netfilter

将桥接的IPv4流量传递到iptables的链

cat > /etc/sysctl.d/k8s.conf << EOF 
net.bridge.bridge-nf-call-ip6tables = 1 
net.bridge.bridge-nf-call-iptables = 1 
EOF
cat >> /etc/sysctl.conf << EOF
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
net.bridge.bridge-nf-call-arptables = 1
net.ipv4.ip_forward = 1
EOF
sysctl --system && sysctl -p

2.4 免密登录

  • master01上执行
ssh-keygen -t rsa   
for i in k8s-master01 k8s-master02 k8s-master03 k8s-node01 k8s-node02;do ssh-copy-id -i ~/.ssh/id_rsa.pub $i;done
或者
yum install -y sshpass
ssh-keygen -t rsa 
export IP="k8s-master01 k8s-master02 k8s-master03 k8s-node01 k8s-node02 k8s-node03 k8s-node04 k8s-node05"
export SSHPASS=******
for HOST in $IP;do
     sshpass -e ssh-copy-id -o StrictHostKeyChecking=no $HOST
done

2.5 部署ETCD集群

  • Etcd 是一个分布式键值存储系统，Kubernetes使用Etcd进行数据存储，所以先准备一个Etcd数据库，为解决Etcd单点故障，应采用集群方式部署，这里使用3台组建集群，可容忍1台机器故障，当然，你也可以使用5台组建集群，可容忍2台机器故障。

ETCD数量通常为奇数个

etcd-01 10.168.1.101 k8s-master01
etcd-02 10.168.1.102 k8s-master02
etcd-03 10.168.1.103 k8s-master03

注：为了节省机器，这里与K8s节点机器复用。也可以独立于k8s集群之外部署，只要apiserver能连接到就行。

2.5.1 安装cfssl

cd /opt/deploy/tls/cfssl
chmod +x cfssl_linux-amd64 cfssljson_linux-amd64 cfssl-certinfo_linux-amd64
cp cfssl_linux-amd64 /usr/local/bin/cfssl
cp cfssljson_linux-amd64 /usr/local/bin/cfssljson
cp cfssl-certinfo_linux-amd64 /usr/local/bin/cfssl-certinfo

2.5.2 etcd ssl

mkdir -p /root/k8s/deploy/tls/{etcd,k8s} && cd /root/k8s/deploy/tls/etcd

自签CA

cat > ca-config.json << EOF
{
  "signing": {
    "default": {
      "expiry": "87600h"
    },
    "profiles": {
      "www": {
         "expiry": "87600h",
         "usages": [
            "signing",
            "key encipherment",
            "server auth",
            "client auth"
        ]
      }
    }
  }
}
EOF
cat > ca-csr.json << EOF
{
    "CN": "etcd CA",
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "names": [
        {
            "C": "CN",
            "L": "SuZou",
            "ST": "SuZou"
        }
    ]
}
EOF

生成证书：会生成ca.pem和ca-key.pem文件

cfssl gencert -initca ca-csr.json | cfssljson -bare ca -

2.5.3 etcd https ssl

cd /root/k8s/deploy/tls/etcd
cat > server-csr.json << EOF
{
    "CN": "etcd",
    "hosts": [
    "10.168.1.100",
    "10.168.1.101",
    "10.168.1.102",
    "10.168.1.103",
    "10.168.1.104",
    "10.168.1.105",
    "10.168.1.106",
    "10.168.1.107",
    "10.168.1.108",
    "10.168.1.109",
    "10.168.1.110",
    "10.168.1.111",
    "10.168.1.112",
    "10.168.1.113",
    "10.168.1.114",
    "10.168.1.115",
    "10.168.1.116",
    "10.168.1.117",
    "10.168.1.118",
    "10.168.1.119",
    "10.168.1.120"
    ],
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "names": [
        {
            "C": "CN",
            "L": "SuZou",
            "ST": "SuZou"
        }
    ]
}
EOF

注：上述文件hosts字段中IP为所有etcd节点的集群内部通信IP，一个都不能少！为了方便后期扩容可以多写几个预留的IP，我把另外两台master节点也加了进去，后面做扩展用。

生成证书，会生成server.pem和server-key.pem文件

cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=www server-csr.json | cfssljson -bare server

2.5.4 etcd集群应用部署

  • master01操作

创建工作目录

cd /opt/deploy/package
mkdir -p /data/etcd/{bin,cfg,ssl}
tar -zxvf etcd-v3.4.9-linux-amd64.tar.gz
mv etcd-v3.4.9-linux-amd64/{etcd,etcdctl} /data/etcd/bin/

创建etcd配置文件

cat > /data/etcd/cfg/etcd.conf << EOF
#[Member]
ETCD_NAME="etcd-01"
ETCD_DATA_DIR="/var/lib/etcd/default.etcd"
ETCD_LISTEN_PEER_URLS="https://10.168.1.101:2380"
ETCD_LISTEN_CLIENT_URLS="https://10.168.1.101:2379"

#[Clustering]
ETCD_INITIAL_ADVERTISE_PEER_URLS="https://10.168.1.101:2380"
ETCD_ADVERTISE_CLIENT_URLS="https://10.168.1.101:2379"
ETCD_INITIAL_CLUSTER="etcd-01=https://10.168.1.101:2380,etcd-02=https://10.168.1.102:2380,etcd-03=https://10.168.1.103:2380"
ETCD_INITIAL_CLUSTER_TOKEN="etcd-cluster"
ETCD_INITIAL_CLUSTER_STATE="new"
EOF

配置文件说明：

ETCD_NAME：节点名称，集群中唯一 
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_PEERURLS：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_INITIALCLUSTER_TOKEN：集群Token 
ETCD_INITIALCLUSTER_STATE：加入集群的当前状态，new是新集群，existing表示加入已有集群

systemd管理etcd

cat > /usr/lib/systemd/system/etcd.service << EOF
[Unit]
Description=Etcd Server
After=network.target
After=network-online.target
Wants=network-online.target
[Service]
Type=notify
EnvironmentFile=/data/etcd/cfg/etcd.conf
ExecStart=/data/etcd/bin/etcd \
--cert-file=/data/etcd/ssl/server.pem \
--key-file=/data/etcd/ssl/server-key.pem \
--peer-cert-file=/data/etcd/ssl/server.pem \
--peer-key-file=/data/etcd/ssl/server-key.pem \
--trusted-ca-file=/data/etcd/ssl/ca.pem \
--peer-trusted-ca-file=/data/etcd/ssl/ca.pem \
--logger=zap
Restart=on-failure
LimitNOFILE=65536
[Install]
WantedBy=multi-user.target
EOF

拷贝生成的证书至指定位置

cp /root/k8s/deploy/tls/etcd/ca*pem /root/k8s/deploy/tls/etcd/server*pem /data/etcd/ssl/

启动并设置开机启动

systemctl daemon-reload && systemctl start etcd && systemctl enable etcd && systemctl status etcd

注意：此时启动一台etcd会显示hang住，没有处于Running状态，暂时忽略，是因为其他两个节点并没有启动，可以查看日志/var/log/messages

将上面mster节点1所有生成的文件拷贝到master节点2和master节点3 (在此)

ssh  k8s-master02 "mkdir  /data"
scp -r /data/etcd/ k8s-master02:/data/
scp /usr/lib/systemd/system/etcd.service k8s-master02:/usr/lib/systemd/system/

ssh  k8s-master03 "mkdir  /data"
scp -r /data/etcd/ k8s-master03:/data/
scp /usr/lib/systemd/system/etcd.service k8s-master03:/usr/lib/systemd/system/

在master01、master02节点上分别修改 /data/etcd/cfg/etcd.conf 配置文件中的节点名称和当前服务器IP

修改字段：

vim /data/etcd/cfg/etcd.conf        # master01、master02节点操作
ETCD_NAME                           # 修改此处，节点2改为etcd-02，节点3改为etcd-03
ETCD_LISTEN_PEER_URL                # 修改此处为当前服务器IP
ETCD_LISTEN_CLIENT_URLS             # 修改此处为当前服务器IP
ETCD_INITIAL_ADVERTISE_PEER_URLS    # 修改此处为当前服务器IP
ETCD_ADVERTISE_CLIENT_URLS          # 修改此处为当前服务器IP

启动etcd并设置开机启动

在master02和master03节点上操作

systemctl daemon-reload && systemctl start etcd && systemctl enable etcd && systemctl status etcd

再把master01上的etcd-1重启下：

systemctl daemon-reload && systemctl restart etcd && systemctl status etcd

注意etcd的启动顺序是 主1 主2 主3  再重置主1 如果由于配置问题导致主1没起来，可以停止3节点的etcd服务删除所有数据目录再重启，如果已经有数据则可以修改集群状态值来处理数据同步问题。

查看集群状态V3版本

master01操作

ETCDCTL_API=3 /data/etcd/bin/etcdctl --cacert=/data/etcd/ssl/ca.pem --cert=/data/etcd/ssl/server.pem --key=/data/etcd/ssl/server-key.pem --endpoints="https://10.168.1.101:2379,https://10.168.1.102:2379,https://10.168.1.103:2379" endpoint health --write-out=table
+---------------------------+--------+-------------+-------+
|         ENDPOINT          | HEALTH |    TOOK     | ERROR |
+---------------------------+--------+-------------+-------+
| https://10.168.1.101:2379 |   true | 10.503607ms |       |
| https://10.168.1.103:2379 |   true | 10.851736ms |       |
| https://10.168.1.102:2379 |   true | 10.801035ms |       |
+---------------------------+--------+-------------+-------+

2.6 部署docker ce

  • 所有节点,建议部署19版本以后
yum install -y yum-utils device-mapper-persistent-data lvm2
yum install docker-ce -y
cat >> /etc/docker/daemon.json  << EOF  
{
      "insecure-registries":["lzyum.luxsan-ict.com"],
      "bip":"192.168.252.1/24",
      "debug":true,
      "log-opts": {"max-size":"50m", "max-file":"3"},
      "default-address-pools":[
      {
              "base":"192.168.240.0/20",
              "size":24
      }
      ]
}
EOF
systemctl daemon-reload && systemctl restart docker &&  systemctl enable docker   && \
systemctl status docker

2.7 部署Master集群(在此)

2.7.1 ssl布局

  • 在master01上操作

自签证书签发机构（CA）

cd /root/k8s/deploy/tls/k8s
cat > ca-config.json << EOF
{
  "signing": {
    "default": {
      "expiry": "87600h"
    },
    "profiles": {
      "kubernetes": {
         "expiry": "87600h",
         "usages": [
            "signing",
            "key encipherment",
            "server auth",
            "client auth"
        ]
      }
    }
  }
}
EOF
cat > ca-csr.json << EOF
{
    "CN": "kubernetes",
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "names": [
        {
            "C": "CN",
            "L": "SuZou",
            "ST": "SuZou",
            "O": "k8s",
            "OU": "System"
        }
    ]
}
EOF

生成证书：生成ca.pem和ca-key.pem文件

cfssl gencert -initca ca-csr.json | cfssljson -bare ca -

使用自签CA签发kube-apiserver HTTPS证书

创建证书请求文件

cat > server-csr.json << EOF
{
    "CN": "kubernetes",
    "hosts": [
      "10.244.0.1",
      "127.0.0.1",
      "10.168.1.100",
      "10.168.1.101",
      "10.168.1.102",
      "10.168.1.103",
      "10.168.1.104",
      "10.168.1.105",
      "10.168.1.106",
      "10.168.1.107",
      "10.168.1.108",
      "10.168.1.109",
      "10.168.1.110",
      "10.168.1.111",
      "10.168.1.112",
      "10.168.1.113",
      "10.168.1.114",
      "10.168.1.115",
      "10.168.1.116",
      "10.168.1.117",
      "10.168.1.118",
      "10.168.1.119",
      "10.168.1.120",
      "kubernetes",
      "kubernetes.default",
      "kubernetes.default.svc",
      "kubernetes.default.svc.cluster",
      "kubernetes.default.svc.cluster.local"
    ],
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "names": [
        {
            "C": "CN",
            "L": "SuZou",
            "ST": "SuZou",
            "O": "k8s",
            "OU": "System"
        }
    ]
}
EOF

注意：上述文件hosts字段中IP为所有Master/LB/VIP IP，一个都不能少！为了方便后期扩容可以多写几个预留的IP。

注： 如果 hosts 字段不为空则需要指定授权使用该证书的 IP 或域名列表。 由于该证书后续被 kubernetes master 集群使用，需要将master节点的IP都填上，同时还需要填写 service 网络的首个IP。(一般是 kube-apiserver 指定的 service-cluster-ip-range 网段的第一个IP，如 10.244.0.1)

      "10.244.0.1",   #servicer IP
      "127.0.0.1",
      "10.168.1.101", #master01
      "10.168.102",   #master02，预留
      "10.168.103",   #master03，预留
      "10.42.221.136",    #vip
      "10.42.221.137",    #预留
      "10.42.221.138",    #预留
      "10.42.221.139",    #预留
      "10.168.110"        #预留

生成证书，生成server.pem和server-key.pem

cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes server-csr.json | cfssljson -bare server

2.7.2 部署api-server （V1.20.15）

创建工作空间

cd /opt/deploy/package
mkdir -p /data/kubernetes/{bin,cfg,ssl,logs} && \
tar -zxvf kubernetes-server-linux-amd64.tar.gz && \
cd kubernetes/server/bin && \
cp kube-apiserver kube-scheduler kube-controller-manager kubectl /data/kubernetes/bin && \
cp kubectl /usr/bin/

创建配置文件

注意修改etcd-server IP、apiserver IP、service IP段

cat > /data/kubernetes/cfg/kube-apiserver.conf << EOF
KUBE_APISERVER_OPTS="--logtostderr=false \\
--v=2 \\
--log-dir=/data/kubernetes/logs \\
--etcd-servers=https://10.168.1.101:2379,https://10.168.1.102:2379,https://10.168.1.103:2379 \\
--bind-address=10.168.1.101 \\
--secure-port=6443 \\
--advertise-address=10.168.1.101 \\
--allow-privileged=true \\
--service-cluster-ip-range=10.244.0.0/16 \\
--enable-admission-plugins=NamespaceLifecycle,LimitRanger,ServiceAccount,ResourceQuota,NodeRestriction \\
--authorization-mode=RBAC,Node \\
--enable-bootstrap-token-auth=true \\
--token-auth-file=/data/kubernetes/cfg/token.csv \\
--service-node-port-range=3000-32767 \\
--kubelet-client-certificate=/data/kubernetes/ssl/server.pem \\
--kubelet-client-key=/data/kubernetes/ssl/server-key.pem \\
--tls-cert-file=/data/kubernetes/ssl/server.pem  \\
--tls-private-key-file=/data/kubernetes/ssl/server-key.pem \\
--client-ca-file=/data/kubernetes/ssl/ca.pem \\
--service-account-key-file=/data/kubernetes/ssl/ca-key.pem \\
--service-account-issuer=api \\
--service-account-signing-key-file=/data/kubernetes/ssl/server-key.pem \\
--etcd-cafile=/data/etcd/ssl/ca.pem \\
--etcd-certfile=/data/etcd/ssl/server.pem \\
--etcd-keyfile=/data/etcd/ssl/server-key.pem \\
--requestheader-client-ca-file=/data/kubernetes/ssl/ca.pem \\
--proxy-client-cert-file=/data/kubernetes/ssl/server.pem \\
--proxy-client-key-file=/data/kubernetes/ssl/server-key.pem \\
--requestheader-allowed-names=kubernetes \\
--requestheader-extra-headers-prefix=X-Remote-Extra- \\
--requestheader-group-headers=X-Remote-Group \\
--requestheader-username-headers=X-Remote-User \\
--enable-aggregator-routing=true \\
--audit-log-maxage=30 \\
--audit-log-maxbackup=3 \\
--audit-log-maxsize=100 \\
--audit-log-path=/data/kubernetes/logs/k8s-audit.log"
EOF

注：上面两个\ \ 第一个是转义符，第二个是换行符，使用转义符是为了使用EOF保留换行符。

参数说明：

--logtostderr：启用日志

--v：日志等级

--log-dir：日志目录

--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机制

--token-auth-file：bootstrap token文件

--service-node-port-range：Service nodeport类型默认分配端口范围

--kubelet-client-xxx：apiserver访问kubelet客户端证书

--tls-xxx-file：apiserver https证书

1.20版本必须加的参数：--service-account-issuer，--service-account-signing-key-file

--etcd-xxxfile：连接Etcd集群证书

--audit-log-xxx：审计日志

启动聚合层相关配置：--requestheader-client-ca-file，--proxy-client-cert-file，--proxy-client-key-file，--requestheader-allowed-names，--requestheader-extra-headers-prefix，--requestheader-group-headers，--requestheader-username-headers，--enable-aggregator-routing

拷贝生成的证书

cd /root/k8s/deploy/tls/k8s
cp /root/k8s/deploy/tls/k8s/ca*pem /root/k8s/deploy/tls/k8s/server*pem /data/kubernetes/ssl/

启用 TLS Bootstrapping 机制

TLS Bootstraping：Master apiserver启用TLS认证后，Node节点kubelet和kube-proxy要与kube-apiserver进行通信，必须使用CA签发的有效证书才可以，当Node节点很多时，这种客户端证书颁发需要大量工作，同样也会增加集群扩展复杂度。为了简化流程，Kubernetes引入了TLS bootstraping机制来自动颁发客户端证书，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。所以强烈建议在Node上使用这种方式，目前主要用于kubelet，kube-proxy还是由我们统一颁发一个证书。

创建token文件

格式：token，用户名，UID，用户组

生成token:

[root@k8s-master01 k8s]# head -c 16 /dev/urandom | od -An -t x | tr -d ' '
f766e563143dd013a2988279c1871831

cat > /data/kubernetes/cfg/token.csv << EOF
f766e563143dd013a2988279c1871831,kubelet-bootstrap,10001,"system:node-bootstrapper"
EOF

systemd管理apiserver

cat > /usr/lib/systemd/system/kube-apiserver.service << EOF
[Unit]
Description=Kubernetes API Server
Documentation=https://github.com/kubernetes/kubernetes
[Service]
EnvironmentFile=/data/kubernetes/cfg/kube-apiserver.conf
ExecStart=/data/kubernetes/bin/kube-apiserver \$KUBE_APISERVER_OPTS
Restart=on-failure
[Install]
WantedBy=multi-user.target
EOF

启动并设置开机启动

systemctl daemon-reload && \
systemctl start kube-apiserver && \
systemctl enable kube-apiserver && \
systemctl status kube-apiserver

测试

curl --insecure https://10.168.1.101:6443/

有返回说明启动正常,虽然此时是403但是表明api功能好的 。

2.7.3 部署kube-controller-manager

  • 在master01上操作

创建配置文件

cat > /data/kubernetes/cfg/kube-controller-manager.conf << EOF
KUBE_CONTROLLER_MANAGER_OPTS="--logtostderr=false \\
--v=2 \\
--log-dir=/data/kubernetes/logs \\
--leader-elect=true \\
--kubeconfig=/data/kubernetes/cfg/kube-controller-manager.kubeconfig \\
--bind-address=127.0.0.1 \\
--allocate-node-cidrs=true \\
--cluster-cidr=10.96.0.0/16 \\
--service-cluster-ip-range=10.244.0.0/16 \\
--cluster-signing-cert-file=/data/kubernetes/ssl/ca.pem \\
--cluster-signing-key-file=/data/kubernetes/ssl/ca-key.pem  \\
--root-ca-file=/data/kubernetes/ssl/ca.pem \\
--service-account-private-key-file=/data/kubernetes/ssl/ca-key.pem \\
--cluster-signing-duration=87600h0m0s"
EOF

释义：

--cluster-cidr  #pod IP段，掩码需要是16位
--service-cluster-ip-range  #service IP段
--kubeconfig：连接apiserver配置文件
--leader-elect：当该组件启动多个时，自动选举（HA）
--cluster-signing-cert-file/--cluster-signing-key-file：自动为kubelet颁发证书的CA，与apiserver保持一致

生成kubeconfig文件

生成kube-controller-manager证书：

cd /root/k8s/deploy/tls/k8s

创建证书请求文件：

cat > kube-controller-manager-csr.json << EOF
{
    "CN": "system:kube-controller-manager",
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "hosts": [
      "127.0.0.1",
      "10.168.1.100",
      "10.168.1.101",
      "10.168.1.102",
      "10.168.1.103",
      "10.168.1.104",
      "10.168.1.105",
      "10.168.1.106",
      "10.168.1.107",
      "10.168.1.108",
      "10.168.1.109",
      "10.168.1.110",
      "10.168.1.111",
      "10.168.1.112",
      "10.168.1.113",
      "10.168.1.114",
      "10.168.1.115",
      "10.168.1.116",
      "10.168.1.117",
      "10.168.1.118",
      "10.168.1.119",
      "10.168.1.120"
    ],
    "names": [
      {
        "C": "CN",
        "ST": "SuZou",
        "L": "SuZou",
        "O": "system:kube-controller-manager",
        "OU": "system"
      }
    ]
}
EOF

释义：

注：
hosts 列表包含所有 kube-controller-manager 节点 IP；我这里填写3台master的IP，另外的为预留IP

CN 为 system:kube-controller-manager、O 为 system:kube-controller-manager，kubernetes 内置的 ClusterRoleBindings system:kube-controller-manager 赋予 kube-controller-manager 工作所需的权限

生成证书：

cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes kube-controller-manager-csr.json | cfssljson -bare kube-controller-manager

生成kubeconfig文件（以下是linux命令，直接全部复制到终端执行）：

cd /root/k8s/deploy/tls/k8s
KUBE_CONFIG="/data/kubernetes/cfg/kube-controller-manager.kubeconfig"
KUBE_APISERVER="https://10.168.1.101:6443"

kubectl config set-cluster kubernetes \
  --certificate-authority=/data/kubernetes/ssl/ca.pem \
  --embed-certs=true \
  --server=${KUBE_APISERVER} \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-credentials kube-controller-manager \
  --client-certificate=./kube-controller-manager.pem \
  --client-key=./kube-controller-manager-key.pem \
  --embed-certs=true \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-context default \
  --cluster=kubernetes \
  --user=kube-controller-manager \
  --kubeconfig=${KUBE_CONFIG}
kubectl config use-context default --kubeconfig=${KUBE_CONFIG}

systemd管理controller-manager

cat > /usr/lib/systemd/system/kube-controller-manager.service << EOF
[Unit]
Description=Kubernetes Controller Manager
Documentation=https://github.com/kubernetes/kubernetes
[Service]
EnvironmentFile=/data/kubernetes/cfg/kube-controller-manager.conf
ExecStart=/data/kubernetes/bin/kube-controller-manager \$KUBE_CONTROLLER_MANAGER_OPTS
Restart=on-failure
[Install]
WantedBy=multi-user.target
EOF
systemctl daemon-reload && \
systemctl start kube-controller-manager && \
systemctl enable kube-controller-manager && \
systemctl status kube-controller-manager

2.7.4 部署kube-scheduler和kubectl

  • master01

创建配置文件

cat > /data/kubernetes/cfg/kube-scheduler.conf << EOF
KUBE_SCHEDULER_OPTS="--logtostderr=false \\
--v=2 \\
--log-dir=/data/kubernetes/logs \\
--leader-elect \\
--kubeconfig=/data/kubernetes/cfg/kube-scheduler.kubeconfig \\
--bind-address=127.0.0.1"
EOF

释义：

--kubeconfig：连接apiserver配置文件
--leader-elect：当该组件启动多个时，自动选举（HA）

生成kubeconfig文件

生成kube-scheduler证书：

cd /root/k8s/deploy/tls/k8s

创建证书请求文件

cat > kube-scheduler-csr.json << EOF
{
    "CN": "system:kube-scheduler",
    "hosts": [
      "127.0.0.1",
      "10.168.1.100",
      "10.168.1.101",
      "10.168.1.102",
      "10.168.1.103",
      "10.168.1.104",
      "10.168.1.105",
      "10.168.1.106",
      "10.168.1.107",
      "10.168.1.108",
      "10.168.1.109",
      "10.168.1.110",
      "10.168.1.111",
      "10.168.1.112",
      "10.168.1.113",
      "10.168.1.114",
      "10.168.1.115",
      "10.168.1.116",
      "10.168.1.117",
      "10.168.1.118",
      "10.168.1.119",
      "10.168.1.120"
    ],
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "names": [
      {
        "C": "CN",
        "ST": "SuZou",
        "L": "SuZou",
        "O": "system:kube-scheduler",
        "OU": "system"
      }
    ]
}
EOF

注：

hosts 列表包含所有 kube-scheduler 节点 IP；我这里填的3台master IP，顺便预留了一些 。

CN 为 system:kube-scheduler、O 为 system:kube-scheduler，kubernetes 内置的 ClusterRoleBindings system:kube-scheduler 将赋予 kube-scheduler 工作所需的权限。

生成证书：

cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes kube-scheduler-csr.json | cfssljson -bare kube-scheduler

生成kubeconfig文件（以下是shell命令，直接在终端执行）：

记得修改下KUBE_APISERVER的地址为master01地址

KUBE_CONFIG="/data/kubernetes/cfg/kube-scheduler.kubeconfig"
KUBE_APISERVER="https://10.168.1.101:6443"

kubectl config set-cluster kubernetes \
  --certificate-authority=/data/kubernetes/ssl/ca.pem \
  --embed-certs=true \
  --server=${KUBE_APISERVER} \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-credentials kube-scheduler \
  --client-certificate=./kube-scheduler.pem \
  --client-key=./kube-scheduler-key.pem \
  --embed-certs=true \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-context default \
  --cluster=kubernetes \
  --user=kube-scheduler \
  --kubeconfig=${KUBE_CONFIG}
kubectl config use-context default --kubeconfig=${KUBE_CONFIG}

systemd管理scheduler

cat > /usr/lib/systemd/system/kube-scheduler.service << EOF
[Unit]
Description=Kubernetes Scheduler
Documentation=https://github.com/kubernetes/kubernetes
[Service]
EnvironmentFile=/data/kubernetes/cfg/kube-scheduler.conf
ExecStart=/data/kubernetes/bin/kube-scheduler \$KUBE_SCHEDULER_OPTS
Restart=on-failure
[Install]
WantedBy=multi-user.target
EOF
systemctl daemon-reload && \
systemctl start kube-scheduler && \
systemctl enable kube-scheduler && \
systemctl status kube-scheduler

查看集群状态

生成kubectl连接集群的证书：

cd /root/k8s/deploy/tls/k8s

创建证书请求文件:

cat > admin-csr.json <<EOF
{
  "CN": "admin",
   "hosts": [
      "127.0.0.1",
      "10.168.1.100",
      "10.168.1.101",
      "10.168.1.102",
      "10.168.1.103",
      "10.168.1.104",
      "10.168.1.105",
      "10.168.1.106",
      "10.168.1.107",
      "10.168.1.108",
      "10.168.1.109",
      "10.168.1.110",
      "10.168.1.111",
      "10.168.1.112",
      "10.168.1.113",
      "10.168.1.114",
      "10.168.1.115",
      "10.168.1.116",
      "10.168.1.117",
      "10.168.1.118",
      "10.168.1.119",
      "10.168.1.120"
    ],
  "key": {
    "algo": "rsa",
    "size": 2048
  },
  "names": [
    {
      "C": "CN",
      "L": "SuZou",
      "ST": "SuZou",
      "O": "system:masters",
      "OU": "System"
    }
  ]
}
EOF

#hosts 列表包含所有 节点 IP,包括node节点以及预留的IP

10.42.221.150 - 10.42.221.160为 work nodes 节点预留IP

说明： 后续 kube-apiserver 使用 RBAC 对客户端(如 kubelet、kube-proxy、Pod)请求进行授权； kube-apiserver 预定义了一些 RBAC 使用的 RoleBindings，如 cluster-admin 将 Group system:masters 与 Role cluster-admin 绑定，该 Role 授予了调用kube-apiserver 的所有 API的权限；

O指定该证书的 Group 为 system:masters，kubelet 使用该证书访问 kube-apiserver 时 ，由于证书被 CA 签名，所以认证通过，同时由于证书用户组为经过预授权的 system:masters，所以被授予访问所有 API 的权限；

注： 这个admin 证书，是将来生成管理员用的kube config 配置文件用的，现在我们一般建议使用RBAC 来对kubernetes 进行角色权限控制， kubernetes 将证书中的CN 字段 作为User， O 字段作为 Group； “O”: “system:masters”, 必须是system:masters，否则后面kubectl create clusterrolebinding报错。

创建kubeconfig配置文件 kubeconfig 为 kubectl 的配置文件，包含访问 apiserver 的所有信息，如 apiserver 地址、CA 证书和自身使用的证书。

生成证书：

cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes admin-csr.json | cfssljson -bare admin

创建kubeconfig配置文件 kubeconfig 为 kubectl 的配置文件，包含访问 apiserver 的所有信息，如 apiserver 地址、CA 证书和自身使用的证书

生成kubeconfig文件：

mkdir /root/.kube

KUBE_APISERVER IP修改成master01的地址

KUBE_CONFIG="/root/.kube/config"
KUBE_APISERVER="https://10.168.1.101:6443"

kubectl config set-cluster kubernetes \
  --certificate-authority=/data/kubernetes/ssl/ca.pem \
  --embed-certs=true \
  --server=${KUBE_APISERVER} \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-credentials cluster-admin \
  --client-certificate=./admin.pem \
  --client-key=./admin-key.pem \
  --embed-certs=true \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-context default \
  --cluster=kubernetes \
  --user=cluster-admin \
  --kubeconfig=${KUBE_CONFIG}
kubectl config use-context default --kubeconfig=${KUBE_CONFIG}

授权kubernetes证书访问kubelet api权限:

kubectl create clusterrolebinding kube-apiserver:kubelet-apis --clusterrole=system:kubelet-api-admin --user kubernetes

测试集群：

kubectl cluster-info    #会获取一些kubernetes信息
Kubernetes control plane is running at https://10.168.1.101:6443
To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.

kubectl get cs  #查看各组件健康状态
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME                 STATUS    MESSAGE             ERROR
scheduler            Healthy   ok                  
controller-manager   Healthy   ok                  
etcd-0               Healthy   {"health":"true"}   
etcd-1               Healthy   {"health":"true"}   
etcd-2               Healthy   {"health":"true"}  

kubectl get all --all-namespaces    #查看集群内所有资源

配置kubectl命令自动补全：

yum install -y bash-completion && \
source /usr/share/bash-completion/bash_completion && \
source <(kubectl completion bash) && \
kubectl completion bash > ~/.kube/completion.bash.inc && \
source '/root/.kube/completion.bash.inc' && \
source $HOME/.bash_profile && \
echo "source <(kubectl completion bash)" >> ~/.bashrc && \
source ~/.bashrc

授权kubelet-bootstrap用户允许请求证书

kubectl create clusterrolebinding kubelet-bootstrap \
--clusterrole=system:node-bootstrapper \
--user=kubelet-bootstrap

查看：

kubectl get clusterrolebinding | grep -i bootstrap

到此，单节点master部署完成，并且已做好了接收node节点注册的准备。

2.8 node角色部署

  • 下面还是在master01节点上操作，即同时作为Worker Node复用。

2.8.1 创建工作目录并拷贝二进制文件

在master01节点拷贝kubernetes-server安装包到/opt/deploy/package目录
ssh k8s-node01 "mkdir -p /opt/deploy/package"; \
ssh k8s-node02 "mkdir -p /opt/deploy/package"; \
ssh k8s-node03 "mkdir -p /opt/deploy/package"; \
ssh k8s-node04 "mkdir -p /opt/deploy/package"; \
ssh k8s-node05 "mkdir -p /opt/deploy/package"

cd /opt/deploy/package && \
scp kubernetes-server-linux-amd64.tar.gz k8s-node01:/opt/deploy/package && \
scp kubernetes-server-linux-amd64.tar.gz k8s-node02:/opt/deploy/package && \
scp kubernetes-server-linux-amd64.tar.gz k8s-node03:/opt/deploy/package && \
scp kubernetes-server-linux-amd64.tar.gz k8s-node04:/opt/deploy/package && \
scp kubernetes-server-linux-amd64.tar.gz k8s-node05:/opt/deploy/package

然后分别到node01-05执行以下操作
mkdir -p /data/kubernetes/{bin,cfg,ssl,logs} && \
cd /opt/deploy/package && \
tar -xzvf  kubernetes-server-linux-amd64.tar.gz && \
cd kubernetes/server/bin && \
cp kubelet kube-proxy /data/kubernetes/bin

2.8.2 部署kubelet

master01操作

先注意导入image pause-amd64:v3.0

cd /opt/deploy/images
for i in `ls ./*`;do docker load <$i;done

创建配置文件

cat > /data/kubernetes/cfg/kubelet.conf << EOF
KUBELET_OPTS="--logtostderr=false \\
--v=2 \\
--log-dir=/data/kubernetes/logs \\
--hostname-override=k8s-master01 \\
--network-plugin=cni \\
--kubeconfig=/data/kubernetes/cfg/kubelet.kubeconfig \\
--bootstrap-kubeconfig=/data/kubernetes/cfg/bootstrap.kubeconfig \\
--config=/data/kubernetes/cfg/kubelet-config.yml \\
--cert-dir=/data/kubernetes/ssl \\
--pod-infra-container-image=pause-amd64:v3.0"
EOF

参数说明： 

--hostname-override：显示名称，集群中唯一 
--network-plugin：启用CNI 
--kubeconfig：空路径，会自动生成，后面用于连接apiserver 
--bootstrap-kubeconfig：首次启动向apiserver申请证书 
--config：配置参数文件 
--cert-dir：kubelet证书生成目录 
--pod-infra-container-image：管理Pod网络容器的镜像

配置参数文件

记得修改clusterDNS IP，该IP为service IP段第二个IP

cat > /data/kubernetes/cfg/kubelet-config.yml << EOF
kind: KubeletConfiguration
apiVersion: kubelet.config.k8s.io/v1beta1
address: 0.0.0.0
port: 10250
readOnlyPort: 10255
cgroupDriver: cgroupfs
clusterDNS:
- 10.244.0.2
clusterDomain: cluster.local 
failSwapOn: false
authentication:
  anonymous:
    enabled: false
  webhook:
    cacheTTL: 2m0s
    enabled: true
  x509:
    clientCAFile: /data/kubernetes/ssl/ca.pem 
authorization:
  mode: Webhook
  webhook:
    cacheAuthorizedTTL: 5m0s
    cacheUnauthorizedTTL: 30s
evictionHard:
  imagefs.available: 15%
  memory.available: 100Mi
  nodefs.available: 10%
  nodefs.inodesFree: 5%
maxOpenFiles: 1000000
maxPods: 110
EOF

生成kubelet初次加入集群引导kubeconfig文件

KUBE_APISERVER为master01 IP地址

TOKEN为先前生成的/data/kubernetes/cfg/token.csv #两者一定要相同

KUBE_CONFIG="/data/kubernetes/cfg/bootstrap.kubeconfig"
KUBE_APISERVER="https://10.168.1.101:6443"
TOKEN="f766e563143dd013a2988279c1871831"

kubectl config set-cluster kubernetes \
  --certificate-authority=/data/kubernetes/ssl/ca.pem \
  --embed-certs=true \
  --server=${KUBE_APISERVER} \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-credentials "kubelet-bootstrap" \
  --token=${TOKEN} \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-context default \
  --cluster=kubernetes \
  --user="kubelet-bootstrap" \
  --kubeconfig=${KUBE_CONFIG}
kubectl config use-context default --kubeconfig=${KUBE_CONFIG}

注意如果这步出错，比如token值拷贝错误会导致csr没有请求，可以删除config重新生成，需要重启kubelet重新发起注册请求。

systemd管理kubelet

先把kubelet执行文件拷贝到/data/kubernetes/bin/

cd /opt/deploy/package && \
cp kubernetes/server/bin/kubelet /data/kubernetes/bin/
cat > /usr/lib/systemd/system/kubelet.service << EOF
[Unit]
Description=Kubernetes Kubelet
After=docker.service
[Service]
EnvironmentFile=/data/kubernetes/cfg/kubelet.conf
ExecStart=/data/kubernetes/bin/kubelet \$KUBELET_OPTS
Restart=on-failure
LimitNOFILE=65536
[Install]
WantedBy=multi-user.target
EOF
systemctl daemon-reload && \
systemctl start kubelet && \
systemctl enable kubelet && \
systemctl status kubelet

批准kubelet证书申请并加入集群

# 查看kubelet证书请求
[root@k8s-master01 package]# kubectl get csr
NAME                                                   AGE   SIGNERNAME                                    REQUESTOR           CONDITION
node-csr-AWalJ6dniBjNF-_K5wL5JKAWtns-AwI_a8aqEV6pg4g   46s   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Pending

# 批准申请
kubectl certificate approve node-csr-AWalJ6dniBjNF-_K5wL5JKAWtns-AwI_a8aqEV6pg4g

# 查看节点（由于网络插件还没有部署，节点显示准备就绪 NotReady，暂时先忽略。）
kubectl get node   #这里有问题，没有获取到资源,查看日志显示system:kube-controller-manager有权限问题，解决办法：

kubectl create clusterrolebinding  controller-node-clusterrolebing --clusterrole=system:controller:node-controller --user=system:kube-controller-manager
kubectl create clusterrolebinding kube-controller-manager --clusterrole=cluster-admin --user=system:kube-controller-manager

重启所有组件:
systemctl restart etcd.service && \
systemctl status etcd.service && \
systemctl restart kube-apiserver.service && \
systemctl status kube-apiserver.service && \
systemctl restart kube-controller-manager.service && \
systemctl status kube-controller-manager.service && \
systemctl restart kube-scheduler.service && \
systemctl status kube-scheduler.service && \
systemctl restart kubelet.service && \
systemctl status kubelet.service

等待30s左右
[root@k8s-master01 package]# kubectl get  nodes 
NAME           STATUS     ROLES    AGE   VERSION
k8s-master01   NotReady   <none>   15s   v1.20.15

2.8.3 部署kube-proxy

master01操作

创建配置文件

cat > /data/kubernetes/cfg/kube-proxy.conf << EOF
KUBE_PROXY_OPTS="--logtostderr=false \\
--v=2 \\
--log-dir=/data/kubernetes/logs \\
--config=/data/kubernetes/cfg/kube-proxy-config.yml"
EOF

配置参数文件

注意clusterCIDR为pod网段

hostnameOverride为master01节点主机名，别写错了

cat > /data/kubernetes/cfg/kube-proxy-config.yml << EOF
kind: KubeProxyConfiguration
apiVersion: kubeproxy.config.k8s.io/v1alpha1
bindAddress: 0.0.0.0
metricsBindAddress: 0.0.0.0:10249
clientConnection:
  kubeconfig: /data/kubernetes/cfg/kube-proxy.kubeconfig
hostnameOverride: k8s-master01
clusterCIDR: 10.96.0.0/16
EOF

生成kube-proxy.kubeconfig文件

生成kube-proxy证书：

创建证书请求文件：

cd /root/k8s/deploy/tls/k8s
cat > kube-proxy-csr.json << EOF
{
  "CN": "system:kube-proxy",
  "hosts": [
      "127.0.0.1",
      "10.168.1.100",
      "10.168.1.101",
      "10.168.1.102",
      "10.168.1.103",
      "10.168.1.104",
      "10.168.1.105",
      "10.168.1.106",
      "10.168.1.107",
      "10.168.1.108",
      "10.168.1.109",
      "10.168.1.110",
      "10.168.1.111",
      "10.168.1.112",
      "10.168.1.113",
      "10.168.1.114",
      "10.168.1.115",
      "10.168.1.116",
      "10.168.1.117",
      "10.168.1.118",
      "10.168.1.119",
      "10.168.1.120"
  ],
  "key": {
    "algo": "rsa",
    "size": 2048
  },
  "names": [
    {
      "C": "CN",
      "L": "SuZou",
      "ST": "SuZou",
      "O": "k8s",
      "OU": "System"
    }
  ]
}
EOF

生成证书：

cfssl gencert -ca=ca.pem -ca-key=ca-key.pem -config=ca-config.json -profile=kubernetes kube-proxy-csr.json | cfssljson -bare kube-proxy

生成kubeconfig文件：

注意修改KUBE_APISERVER IP地址为master01地址

KUBE_CONFIG="/data/kubernetes/cfg/kube-proxy.kubeconfig"
KUBE_APISERVER="https://10.168.1.101:6443"

kubectl config set-cluster kubernetes \
  --certificate-authority=/data/kubernetes/ssl/ca.pem \
  --embed-certs=true \
  --server=${KUBE_APISERVER} \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-credentials kube-proxy \
  --client-certificate=./kube-proxy.pem \
  --client-key=./kube-proxy-key.pem \
  --embed-certs=true \
  --kubeconfig=${KUBE_CONFIG}
kubectl config set-context default \
  --cluster=kubernetes \
  --user=kube-proxy \
  --kubeconfig=${KUBE_CONFIG}
kubectl config use-context default --kubeconfig=${KUBE_CONFIG}

systemd管理kube-proxy

cp /opt/deploy/package/kubernetes/server/bin/kube-proxy /data/kubernetes/bin
cat > /usr/lib/systemd/system/kube-proxy.service << EOF
[Unit]
Description=Kubernetes Proxy
After=network.target

[Service]
EnvironmentFile=/data/kubernetes/cfg/kube-proxy.conf
ExecStart=/data/kubernetes/bin/kube-proxy \$KUBE_PROXY_OPTS
Restart=on-failure
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target
EOF
systemctl daemon-reload && \
systemctl start kube-proxy && \
systemctl enable kube-proxy && \
systemctl status kube-proxy

2.8.4 部署网络组件Calico

Calico是一个纯三层的数据中心网络方案，是目前Kubernetes主流的网络方案。

官网：https://projectcalico.docs.tigera.io/about/about-calico

部署Calico：

注意：yaml中的images都是本地的，版本如下,具体容器包，在个人阿里云K8S部署-->二进制部署-->V1.20文件夹中

calico/cni v3.14.2 calico/cni:v3.14.2
calico/pod2daemon-flexvol v3.14.2 calico/pod2daemon-flexvol:v3.14.2
calico/node v3.14.2 calico/node:v3.14.2
calico/kube-controllers v3.14.2 calico/kube-controllers:v3.14.2 
cd /opt/deploy/package
kubectl apply -f calico.yaml

查看calico状态：

注意之前的pause pod镜像必须正常运行 否则calico-node pod 会pending
kubectl get po,svc,deploy -A -o wide

NAMESPACE     NAME                                           READY   STATUS    RESTARTS   AGE     IP              NODE           NOMINATED NODE   READINESS GATES
kube-system   pod/calico-kube-controllers-7d8686bbf8-ssc7r   1/1     Running   0          7m29s   172.16.32.129   k8s-master01   <none>           <none>
kube-system   pod/calico-node-s2762                          1/1     Running   0          4m15s   10.168.1.101   k8s-master01   <none>           <none>

2.9 授权apiserver访问kubelet

  • master01操作
cd /opt/deploy/package
cat > apiserver-to-kubelet-rbac.yaml << EOF
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  annotations:
    rbac.authorization.kubernetes.io/autoupdate: "true"
  labels:
    kubernetes.io/bootstrapping: rbac-defaults
  name: system:kube-apiserver-to-kubelet
rules:
  - apiGroups:
      - ""
    resources:
      - nodes/proxy
      - nodes/stats
      - nodes/log
      - nodes/spec
      - nodes/metrics
      - pods/log
    verbs:
      - "*"
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: system:kube-apiserver
  namespace: ""
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: system:kube-apiserver-to-kubelet
subjects:
  - apiGroup: rbac.authorization.k8s.io
    kind: User
    name: kubernetes
EOF

执行：

kubectl apply -f apiserver-to-kubelet-rbac.yaml

至此，一个单master单work node复用就部署完成了（包含了所有部署功能步骤），之后扩容可以完成cp上面配置文件来实现。

按照之前的规划：

k8s-master01    10.168.1.101     kube-apiserver，kube-controller-manager，kube-scheduler，etcd
k8s-node01      10.168.1.111     kubelet，kube-proxy，docker，etcd
k8s-node02      10.168.1.112     kubelet，kube-proxy，docker，etcd
k8s-node03      10.168.1.113     kubelet，kube-proxy，docker，etcd
k8s-node04      10.168.1.114     kubelet，kube-proxy，docker，etcd
k8s-node05      10.168.1.115     kubelet，kube-proxy，docker，etcd

下面将node01-05加到master01中。

3 部署流程二:扩容Worker Nodes节点

  • master01

3.1 新增Worker Node

1）拷贝已部署好的Node相关文件到新节点

在master01节点将Worker Node涉及文件拷贝到新节点node1-5

注意如果不是使用公有image则会出现其他node拉取calico node image失败的情况，节点未显示未就绪，需要提前导入所有image（默认calico yaml拉取策略是Never）

 在master01上
scp -r  /opt/deploy/images/ root@k8s-node01:/opt/deploy
scp -r  /opt/deploy/images/ root@k8s-node02:/opt/deploy
scp -r  /opt/deploy/images/ root@k8s-node03:/opt/deploy
scp -r  /opt/deploy/images/ root@k8s-node04:/opt/deploy
scp -r  /opt/deploy/images/ root@k8s-node05:/opt/deploy 
在所有node01-5节点对应的目录下导入c操作
cd /opt/deploy/images/ &&  for i in  `ls ./*`;do docker load <$i;done
此时node节点才会引入calico-node pod成功
[root@k8s-master01 deploy]#  kubectl get pod  -o wide  -nkube-system
NAME                                       READY   STATUS    RESTARTS   AGE   IP              NODE           NOMINATED NODE   READINESS GATES
calico-kube-controllers-54b8449c7d-kfcqk   1/1     Running   0          31m   172.16.32.129   k8s-master01   <none>           <none>
calico-node-nnjlr                          1/1     Running   0          17m   10.168.1.111    k8s-node01     <none>           <none>
calico-node-xqsdb                          1/1     Running   0          31m   10.168.1.101    k8s-master01   <none>           <none>

拷贝到node01-05节点：

scp -r /data/kubernetes root@k8s-node01:/data/ && \
scp -r /usr/lib/systemd/system/{kubelet,kube-proxy}.service root@k8s-node01:/usr/lib/systemd/system && \
scp /data/kubernetes/ssl/ca.pem root@k8s-node01:/data/kubernetes/ssl
scp -r /data/kubernetes root@k8s-node02:/data/ && \
scp -r /usr/lib/systemd/system/{kubelet,kube-proxy}.service root@k8s-node02:/usr/lib/systemd/system && \
scp /data/kubernetes/ssl/ca.pem root@k8s-node02:/data/kubernetes/ssl

scp -r /data/kubernetes root@k8s-node03:/data/ && \
scp -r /usr/lib/systemd/system/{kubelet,kube-proxy}.service root@k8s-node03:/usr/lib/systemd/system && \
scp /data/kubernetes/ssl/ca.pem root@k8s-node03:/data/kubernetes/ssl

scp -r /data/kubernetes root@k8s-node04:/data/ && \
scp -r /usr/lib/systemd/system/{kubelet,kube-proxy}.service root@k8s-node04:/usr/lib/systemd/system && \
scp /data/kubernetes/ssl/ca.pem root@k8s-node04:/data/kubernetes/ssl

scp -r /data/kubernetes root@k8s-node05:/data/ && \
scp -r /usr/lib/systemd/system/{kubelet,kube-proxy}.service root@k8s-node05:/usr/lib/systemd/system && \
scp /data/kubernetes/ssl/ca.pem root@k8s-node05:/data/kubernetes/ssl

删除kubelet证书和kubeconfig文件

因为这几个文件是证书申请审批后自动生成的，每个Node不同，所以必须删除。

在node1-5节点操作

rm -f /data/kubernetes/cfg/kubelet.kubeconfig && \
rm -f /data/kubernetes/ssl/kubelet*

修改配置文件中的主机名

在node1-5节点操作

node01
sed -i 's#k8s-master01#k8s-node01#' /data/kubernetes/cfg/kubelet.conf
sed -i 's#k8s-master01#k8s-node01#' /data/kubernetes/cfg/kube-proxy-config.yml

hostname-override=k8s-node01
hostnameOverride: k8s-node01
node02
sed -i 's#k8s-master01#k8s-node02#' /data/kubernetes/cfg/kubelet.conf
sed -i 's#k8s-master01#k8s-node02#' /data/kubernetes/cfg/kube-proxy-config.yml
node03
sed -i 's#k8s-master01#k8s-node03#' /data/kubernetes/cfg/kubelet.conf
sed -i 's#k8s-master01#k8s-node03#' /data/kubernetes/cfg/kube-proxy-config.yml
node04
sed -i 's#k8s-master01#k8s-node04#' /data/kubernetes/cfg/kubelet.conf
sed -i 's#k8s-master01#k8s-node04#' /data/kubernetes/cfg/kube-proxy-config.yml
node05
sed -i 's#k8s-master01#k8s-node05#' /data/kubernetes/cfg/kubelet.conf
sed -i 's#k8s-master01#k8s-node05#' /data/kubernetes/cfg/kube-proxy-config.yml

启动并设置开机启动

node01-05上操作

systemctl daemon-reload && \
systemctl start kubelet kube-proxy && \
systemctl enable kubelet kube-proxy && \
systemctl status kubelet kube-proxy

在Master上批准所有Node01-05的 kubelet证书申请

在master01上操作

# 查看证书请求
[root@k8s-master01 deploy]# kubectl get csr
NAME                                                   AGE     SIGNERNAME                                    REQUESTOR           CONDITION
node-csr-AbCwM2KlOtGYE2INPtBxQOH207FBmxdDmr-vtVnmCEk   8m22s   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Pending
node-csr-CYGaTHLaDRVcqnJBfnQ2Qd6l0M7kJC4wkJLnD1HYYR4   8m25s   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Pending
node-csr-Dc0tb1P25edvnyCujTbv4wQWiQFhGULVuafdiQooA_g   8m24s   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Pending
。。。

# 同意授权请求
kubectl certificate approve node-csr-所有请求码

查看节点信息,稍等一会等待ready，到此 1主5从已经部署好，下面可以扩容master节点了：

[root@k8s-master01 deploy]# kubectl  get no
NAME           STATUS   ROLES    AGE   VERSION
k8s-master01   Ready    <none>   63m   v1.20.15
k8s-node01     Ready    <none>   34m   v1.20.15
k8s-node02     Ready    <none>   30s   v1.20.15
k8s-node03     Ready    <none>   55s   v1.20.15
k8s-node04     Ready    <none>   54s   v1.20.15
k8s-node05     Ready    <none>   84s   v1.20.15

[root@k8s-master01 ~]# kubectl  get pod -A -o wide
NAMESPACE     NAME                                       READY   STATUS    RESTARTS   AGE     IP              NODE           NOMINATED NODE   READINESS GATES
kube-system   calico-kube-controllers-54b8449c7d-kfcqk   1/1     Running   1          54m     172.16.32.130   k8s-master01   <none>           <none>
kube-system   calico-node-j7qnq                          1/1     Running   1          5m48s   10.168.1.112    k8s-node02     <none>           <none>
kube-system   calico-node-nnjlr                          1/1     Running   1          39m     10.168.1.111    k8s-node01     <none>           <none>
kube-system   calico-node-rvzdd                          1/1     Running   1          6m12s   10.168.1.113    k8s-node03     <none>           <none>
kube-system   calico-node-wvdbz                          1/1     Running   1          6m11s   10.168.1.114    k8s-node04     <none>           <none>
kube-system   calico-node-xmr8j                          1/1     Running   1          6m41s   10.168.1.115    k8s-node05     <none>           <none>
kube-system   calico-node-xqsdb                          1/1     Running   1          54m     10.168.1.101    k8s-master01   <none>           <none>

测试

清除calico一次性启动的exited的docker容器

docker system prune

重启master01、node1-5主机，再次kubectl get nodes查看功能是否正常

3.2 部署CoreDNS

  • master01

CoreDNS用于集群内部Service名称解析：

coredns.yaml

需要修改下clusterIP字段，把IP修改你自己service网段所在的第二个IP。

注意image使用coredns/coredns:1.2.2版本，测试1.6 、1.8不可兼容k8s1.20。

spec:
  selector:
    k8s-app: kube-dns
  clusterIP: 10.244.0.2
  ports:
  - name: dns

部署：

cd /opt/deploy/package
kubectl apply -f coredns.yaml

查看：

kubectl get po,svc,deploy -A -o wide

[root@k8s-master01 package]# kubectl  get   pods  -n  kube-system
NAME                                       READY   STATUS    RESTARTS   AGE
calico-kube-controllers-54b8449c7d-kfcqk   1/1     Running   1          2d
calico-node-j7qnq                          1/1     Running   1          2d
calico-node-nnjlr                          1/1     Running   1          2d
calico-node-rvzdd                          1/1     Running   1          2d
calico-node-wvdbz                          1/1     Running   1          2d
calico-node-xmr8j                          1/1     Running   1          2d
calico-node-xqsdb                          1/1     Running   1          2d
coredns-6d8f96d957-m7hlr                   1/1     Running   0          106s

[root@k8s-master01 package]# kubectl  get   pods  -n  kube-system
NAME                                       READY   STATUS    RESTARTS   AGE
calico-kube-controllers-54b8449c7d-kfcqk   1/1     Running   1          2d
calico-node-j7qnq                          1/1     Running   1          2d
calico-node-nnjlr                          1/1     Running   1          2d
calico-node-rvzdd                          1/1     Running   1          2d
calico-node-wvdbz                          1/1     Running   1          2d
calico-node-xmr8j                          1/1     Running   1          2d
calico-node-xqsdb                          1/1     Running   1          2d
coredns-6d8f96d957-m7hlr                   1/1     Running   0          106s

至此，一个完整的master可用集群（1主5从）就完成了，下面需要配置api-server的高可用，即多master节点环境。

4 部署流程三:扩容多Master节点（高可用）

4.1 部署master02

部署Master02 节点

master02 IP: 10.168.1.102

Master02 与已部署的Master01所有操作一致。所以我们只需将Master1所有K8s文件拷贝过来，再修改下服务器IP和主机名启动即可。

安装docker

在master02节点操作

在基础环境配置中已安装，这一步忽略。

导入离线镜像包

在master01上
scp -r  /opt/deploy/images/ root@k8s-master02:/opt/deploy && \
scp -r  /opt/deploy/images/ root@k8s-master03:/opt/deploy

在所有master02-03节点对应的目录下导入操作
cd /opt/deploy/images/ &&  for i in `ls ./*`;do docker load <$i;done

创建etcd证书目录

在Master02创建etcd证书目录

mkdir -p /data/etcd/ssl

拷贝master01配置文件到master02

拷贝Master01上所有K8s文件和etcd证书到Master02

在master01节点操作：

scp -r /data/kubernetes k8s-master02:/data && \
scp -r /data/etcd/ssl k8s-master02:/data/etcd && \
scp /usr/lib/systemd/system/kube* k8s-master02:/usr/lib/systemd/system && \
scp /usr/bin/kubectl k8s-master02:/usr/bin && \
scp -r ~/.kube k8s-master02:~

删除证书文件

master02操作

删除kubelet证书和kubeconfig文件，bootstrap自动注册生成的。 

rm -f /data/kubernetes/cfg/kubelet.kubeconfig && \
rm -f /data/kubernetes/ssl/kubelet*

修改配置文件IP和主机名

master02操作

修改apiserver、kubelet和kube-proxy配置文件为本地IP

vim /data/kubernetes/cfg/kube-apiserver.conf 
...
--bind-address=10.168.1.102 \
--advertise-address=10.168.1.102 \
...

vim /data/kubernetes/cfg/kubelet.conf
--hostname-override=k8s-master02

vim /opt/kubernetes/cfg/kube-proxy-config.yml
hostnameOverride: k8s-master02

启动并设置开机启动

master02操作

启动kube-apiserver、kube-controller-manage、kube-scheduler、kubelet、kube-proxy

systemctl daemon-reload && \
systemctl start kube-apiserver kube-controller-manager kube-scheduler kubelet kube-proxy && \
systemctl enable kube-apiserver kube-controller-manager kube-scheduler kubelet kube-proxy && \
systemctl status kube-apiserver kube-controller-manager kube-scheduler kubelet kube-proxy

查看集群状态

master02操作，查看本地api-server是否可用

# 修改连接master为本机IP
vim ~/.kube/config
...
server: https://10.168.1.102:6443

[root@k8s-master02 ssl]# kubectl get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME                 STATUS    MESSAGE             ERROR
scheduler            Healthy   ok                  
controller-manager   Healthy   ok                  
etcd-2               Healthy   {"health":"true"}   
etcd-0               Healthy   {"health":"true"}   
etcd-1               Healthy   {"health":"true"}

批准kubelet证书申请

master02操作

# 查看证书请求，这里填写你自己生成的。
[root@k8s-master02 ~]# kubectl get csr
NAME                                                   AGE   SIGNERNAME                                    REQUESTOR           CONDITION
node-csr-S3RaMdAqgH805EJpWII0zgFC3AWpzBxliok3MYOHFpU   68s   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Pending

# 授权请求在master02上
 kubectl certificate approve node-csr-S3RaMdAqgH805EJpWII0zgFC3AWpzBxliok3MYOHFpU

# 查看Node，如果状态为NotReady，稍微等待一会儿就好了。30s,如果not ready可以查看下calico是否running
[root@k8s-master02 images]# kubectl get  no
NAME           STATUS   ROLES    AGE    VERSION
k8s-master01   Ready    <none>   2d1h   v1.20.15
k8s-master02   Ready    <none>   10m    v1.20.15
k8s-node01     Ready    <none>   2d1h   v1.20.15
k8s-node02     Ready    <none>   2d     v1.20.15
k8s-node03     Ready    <none>   2d     v1.20.15
k8s-node04     Ready    <none>   2d     v1.20.15
k8s-node05     Ready    <none>   2d     v1.20.15

kubectl 命令自动补全

master02操作

yum install -y bash-completion && \
source /usr/share/bash-completion/bash_completion && \
source <(kubectl completion bash) && \
kubectl completion bash > ~/.kube/completion.bash.inc && \
source '/root/.kube/completion.bash.inc' && \
source $HOME/.bash_profile && \
echo "source <(kubectl completion bash)" >> ~/.bashrc && \
source ~/.bashrc

4.2 部署master03

部署Master03 节点

master03 IP: 10.168.1.103

Master03 与已部署的Master01所有操作一致。所以我们只需将Master1所有K8s文件拷贝过来，再修改下服务器IP和主机名启动即可。

安装docker

在master03节点操作

在基础环境配置中已安装，这一步忽略。

创建etcd证书目录

在Master03创建etcd证书目录

mkdir -p /opt/etcd/ssl

拷贝master01配置文件到master02

拷贝Master01上所有K8s文件和etcd证书到Master02

在master01节点操作：

scp -r /data/kubernetes k8s-master03:/data && \
scp -r /data/etcd/ssl k8s-master03:/data/etcd && \
scp /usr/lib/systemd/system/kube* k8s-master03:/usr/lib/systemd/system && \
scp /usr/bin/kubectl k8s-master03:/usr/bin && \
scp -r ~/.kube k8s-master03:~

删除证书文件

master03操作

删除kubelet证书和kubeconfig文件 

rm -f /data/kubernetes/cfg/kubelet.kubeconfig && \
rm -f /data/kubernetes/ssl/kubelet*

修改配置文件IP和主机名

master03操作

修改apiserver、kubelet和kube-proxy配置文件为本地IP

vim /data/kubernetes/cfg/kube-apiserver.conf 
...
--bind-address=10.168.1.103 \
--advertise-address=10.168.1.103 \
...

vim /data/kubernetes/cfg/kubelet.conf
--hostname-override=k8s-master03

vim /data/kubernetes/cfg/kube-proxy-config.yml
hostnameOverride: k8s-master03

启动并设置开机启动

master03操作

启动kube-apiserver、kube-controller-manage、kube-scheduler、kubelet、kube-proxy

systemctl daemon-reload && \
systemctl start kube-apiserver kube-controller-manager kube-scheduler kubelet kube-proxy && \
systemctl enable kube-apiserver kube-controller-manager kube-scheduler kubelet kube-proxy && \
systemctl status kube-apiserver kube-controller-manager kube-scheduler kubelet kube-proxy

查看集群状态

master03操作

# 修改连接master为本机IP
vim ~/.kube/config
...
server: https://10.168.103:6443

[root@k8s-master03 ~]# kubectl get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME                 STATUS    MESSAGE             ERROR
scheduler            Healthy   ok                  
controller-manager   Healthy   ok                  
etcd-0               Healthy   {"health":"true"}   
etcd-2               Healthy   {"health":"true"}   
etcd-1               Healthy   {"health":"true"}

批准kubelet证书申请

master01操作

# 查看证书请求
[root@k8s-master03 images]# kubectl get csr
NAME                                                   AGE   SIGNERNAME                                    REQUESTOR           CONDITION
node-csr-S3RaMdAqgH805EJpWII0zgFC3AWpzBxliok3MYOHFpU   17m   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Approved,Issued
node-csr-nOFJ0egY262TxgZ92mgFgUsHFBC2s60lzLjh4cMiZp0   26s   kubernetes.io/kube-apiserver-client-kubelet   kubelet-bootstrap   Pending

# 授权请求
kubectl certificate approve node-csr-nOFJ0egY262TxgZ92mgFgUsHFBC2s60lzLjh4cMiZp0

# 查看Node，如果状态为NotReady，稍微等待一会儿就好了
[root@k8s-master03 images]#  kubectl  get   node
NAME           STATUS   ROLES    AGE    VERSION
k8s-master01   Ready    <none>   2d1h   v1.20.15
k8s-master02   Ready    <none>   16m    v1.20.15
k8s-master03   Ready    <none>   39s    v1.20.15
k8s-node01     Ready    <none>   2d1h   v1.20.15
k8s-node02     Ready    <none>   2d     v1.20.15
k8s-node03     Ready    <none>   2d     v1.20.15
k8s-node04     Ready    <none>   2d     v1.20.15
k8s-node05     Ready    <none>   2d     v1.20.15

kubectl 命令自动补全

master03操作

yum install -y bash-completion && \
source /usr/share/bash-completion/bash_completion && \
source <(kubectl completion bash) && \
kubectl completion bash > ~/.kube/completion.bash.inc && \
source '/root/.kube/completion.bash.inc' && \
source $HOME/.bash_profile && \
echo "source <(kubectl completion bash)" >> ~/.bashrc && \
source ~/.bashrc

测试

依次重启node与master节点后执行 kubectl get node

至此，多master + 多node k8s集群部署完毕（3主5 从，全work node节点）

5 部署流程四:API-Server高可用

  • 目前是每一台master使用的api-server网关是自己，配置信息从每一个节点写入到etcd集群中，现在要使用lvs将所有的api请求定向到统一api地址

5.1 部署Nginx+Keepalived高可用负载均衡器

在3台master节点安装软件包

注1：为了节省机器，这里与K8s Master节点机器复用。也可以独立于k8s集群之外部署，只要nginx与apiserver能通信就行。

注2：如果你是在公有云上，一般都不支持keepalived，那么你可以直接用它们的负载均衡器产品，直接负载均衡多台Master kube-apiserver，架构与上面一样。

在3台Master节点操作：

yum -y install nginx-all-modules.noarch && yum install nginx keepalived -y

Nginx配置文件（3台配置一样）

cat > /etc/nginx/nginx.conf << "EOF"
user nginx;
worker_processes auto;
error_log /var/log/nginx/error.log;
pid /run/nginx.pid;

include /usr/share/nginx/modules/*.conf;

events {
    worker_connections 1024;
}

# 四层负载均衡，为3台Master apiserver组件提供负载均衡
stream {

    log_format  main  '$remote_addr $upstream_addr - [$time_local] $status $upstream_bytes_sent';

    access_log  /var/log/nginx/k8s-access.log  main;

    upstream k8s-apiserver {
       server 10.168.1.101:6443;    # Master1 APISERVER IP:PORT
       server 10.168.1.102:6443;    # Master2 APISERVER IP:PORT
       server 10.168.1.103:6443;      # Master3 APISERVER IP:PORT
    }

    server {
       listen 16443; # 由于nginx与master节点复用，这个监听端口不能是6443，否则会冲突
       proxy_pass k8s-apiserver;
    }
}

http {
    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';

    access_log  /var/log/nginx/access.log  main;

    sendfile            on;
    tcp_nopush          on;
    tcp_nodelay         on;
    keepalive_timeout   65;
    types_hash_max_size 2048;

    include             /etc/nginx/mime.types;
    default_type        application/octet-stream;
}
EOF

keepalived配置文件（Nginx Master01）

master01上操作：

cat > /etc/keepalived/keepalived.conf << EOF
global_defs { 
    notification_email { 
    acassen@firewall.loc 
    failover@firewall.loc 
    sysadmin@firewall.loc 
} 
   notification_email_from Alexandre.Cassen@firewall.loc  
   smtp_server 127.0.0.1 
   smtp_connect_timeout 30 
   router_id NGINX_MASTER
} 

vrrp_script check_nginx {
    script "/etc/keepalived/check_nginx.sh"
}

vrrp_instance VI_1 { 
    state MASTER 
    interface ens192 # 修改为实际网卡名
    virtual_router_id 49 # VRRP 路由 ID实例，每个实例是唯一的 
    priority 100    # 优先级，备服务器设置 90 
    advert_int 1    # 指定VRRP 心跳包通告间隔时间，默认1秒 
    authentication { 
        auth_type PASS      
        auth_pass 1111 
    }  
    # 虚拟IP
    virtual_ipaddress { 
        10.168.1.110/24
    } 
    track_script {
        check_nginx
    } 
}
EOF

参数说明：

vrrp_script：指定检查nginx工作状态脚本（根据nginx状态判断是否故障转移）

virtual_ipaddress：虚拟IP（VIP）

准备上述配置文件中检查nginx运行状态的脚本：

master01上操作：

cat > /etc/keepalived/check_nginx.sh  << "EOF"
#!/bin/bash
count=$(ss -antp |grep 16443 |egrep -cv "grep|$$")

if [ "$count" -eq 0 ];then
    exit 1
else
    exit 0
fi
EOF
chmod +x /etc/keepalived/check_nginx.sh

注：keepalived根据脚本返回状态码（0为工作正常，非0不正常）判断是否故障转移

keepalived配置文件（Nginx 从节点）

在master02上操作：

cat > /etc/keepalived/keepalived.conf << EOF
global_defs { 
   notification_email { 
     acassen@firewall.loc 
     failover@firewall.loc 
     sysadmin@firewall.loc 
   } 
   notification_email_from Alexandre.Cassen@firewall.loc  
   smtp_server 127.0.0.1 
   smtp_connect_timeout 30 
   router_id NGINX_BACKUP
} 

vrrp_script check_nginx {
    script "/etc/keepalived/check_nginx.sh"
}

vrrp_instance VI_1 { 
    state BACKUP 
    interface ens192
    virtual_router_id 49 # VRRP 路由 ID实例，每个实例是唯一的 
    priority 90
    advert_int 1
    authentication { 
        auth_type PASS      
        auth_pass 1111 
    }  
    virtual_ipaddress { 
        10.168.1.110/24
    } 
    track_script {
        check_nginx
    } 
}
EOF

准备上述配置文件中检查nginx运行状态的脚本：

cat > /etc/keepalived/check_nginx.sh  << "EOF"
#!/bin/bash
count=$(ss -antp |grep 16443 |egrep -cv "grep|$$")

if [ "$count" -eq 0 ];then
    exit 1
else
    exit 0
fi
EOF
chmod +x /etc/keepalived/check_nginx.sh

在master03上操作：

cat > /etc/keepalived/keepalived.conf << EOF
global_defs { 
   notification_email { 
   acassen@firewall.loc 
   failover@firewall.loc 
   sysadmin@firewall.loc 
} 
   notification_email_from Alexandre.Cassen@firewall.loc  
   smtp_server 127.0.0.1 
   smtp_connect_timeout 30 
   router_id NGINX_BACKUP
} 

vrrp_script check_nginx {
    script "/etc/keepalived/check_nginx.sh"
}

vrrp_instance VI_1 { 
    state BACKUP 
    interface ens192
    virtual_router_id 49 # VRRP 路由 ID实例，每个实例是唯一的 
    priority 80
    advert_int 1
    authentication { 
        auth_type PASS      
        auth_pass 1111 
    }  
    virtual_ipaddress { 
        10.168.1.110/24
    } 
    track_script {
        check_nginx
    } 
}
EOF

准备上述配置文件中检查nginx运行状态的脚本：

cat > /etc/keepalived/check_nginx.sh  << "EOF"
#!/bin/bash
count=$(ss -antp |grep 16443 |egrep -cv "grep|$$")

if [ "$count" -eq 0 ];then
    exit 1
else
    exit 0
fi
EOF
chmod +x /etc/keepalived/check_nginx.sh

启动所有节点的keepalived与nginx并设置开机启动

systemctl daemon-reload && \
systemctl start nginx keepalived && \
systemctl enable nginx keepalived && \
systemctl status nginx keepalived

查看keepalived工作状态

在master01节点执行以下命令可以看到网卡多了一个虚拟IP

[root@k8s-master01 ~]# ip a |grep   secondary
    inet 10.42.221.40/24 scope global secondary ens192

Nginx+Keepalived高可用测试

关闭主节点Nginx，测试VIP是否漂移到备节点服务器。

在Nginx Master执行systemctl stop nginx;

在Nginx Backup，ip addr命令查看已成功绑定VIP。

访问负载均衡器测试

找K8s集群中任意一个节点，使用curl查看K8s版本测试，使用VIP访问:

[root@k8s-master01 ~]#  curl -k https://10.168.1.110:16443/version
{
  "major": "1",
  "minor": "20",
  "gitVersion": "v1.20.15",
  "gitCommit": "8f1e5bf0b9729a899b8df86249b56e2c74aebc55",
  "gitTreeState": "clean",
  "buildDate": "2022-01-19T17:23:01Z",
  "goVersion": "go1.15.15",
  "compiler": "gc",
  "platform": "linux/amd64"

可以正确获取到K8s版本信息，说明负载均衡器搭建正常。该请求数据流程：curl -> vip(nginx) -> apiserver

通过查看Nginx日志也可以看到转发apiserver IP,由于VIP漂移到了master02上，所以要看02的nginx请求日志

[root@k8s-master01 ~]#  cat /var/log/nginx/k8s-access.log
10.168.1.101 10.168.1.101:6443 - [19/Jun/2023:00:11:18 +0800] 200 422

还原VIP

开启master01的nginx,此时VIP恢复到了master01主机上。

5.2 修改所有Worker Node连接LB VIP

  • 之前再3台master上部署了VIP负载后端3个api-server，现在要将所有work nodes的api-server地址修改成LB VIP

试想下，虽然我们增加了Master02/03 Node和负载均衡器，但是我们是从单Master架构扩容的，也就是说目前所有的Worker Node组件连接都还是Master01 Node，如果不改为连接VIP走负载均衡器，那么Master还是单点故障。

因此接下来就是要改所有Worker Node（kubectl get node命令查看到的节点）组件配置文件，由原来的10.168.1.101修改为10.168.110（VIP）。

在所有Worker Node(由于master也是work node，因此也要修改ip)执行：

#在所有kubelet kube-proxy （work node角色节点上执行）使work组件与vip api通信
sed -i 's#10.168.1.101:6443#10.168.1.110:16443#' /data/kubernetes/cfg/*
systemctl restart kubelet kube-proxy
#如果需要还原
sed -i 's#10.168.110:16443#10.168.1.101:6443#' /opt/kubernetes/cfg/*

再次重启所有节点，检查节点状态：

[root@k8s-master01 cfg]# kubectl get nodes
NAME           STATUS   ROLES    AGE    VERSION
k8s-master01   Ready    <none>   23h    v1.20.15
k8s-master02   Ready    <none>   141m   v1.20.15
k8s-master03   Ready    <none>   47m    v1.20.15
k8s-node01     Ready    <none>   22h    v1.20.15
k8s-node02     Ready    <none>   22h    v1.20.15

6 生命周期之work node扩容节点脚本（暂未更新）

下面是新增node节点的shell脚本，写脚本的时候用的单master环境，如果是多master，在脚本里把其它master节点的主机名和ip补全即可。

以新增节点node03为例

#!/bin/bash

# 集群节点主机名
master01=k8s-master01
master02=k8s-master02
master03=k8s-master03
node01=k8s-node01
node02=k8s-node02
node03=k8s-node03
lb=k8s-master-lb

# 集群节点IP
master01Ip=10.168.1.101
master02Ip=10.168.102
master03Ip=10.168.103
node01IP=10.168.1.111
node02IP=10.168.1.112
node03IP=10.42.221.137
lbIP=10.168.110

# 计时器
timer() {
    echo 5
    sleep 1
    echo 4
    sleep 1
    echo 3
    sleep 1
    echo 2
    sleep 1
    echo 1
    sleep 1
}

# 执行命令状态码检测
checkStatusCode() {
    if [ $? -eq 0 ]; then
        echo "返回状态码: $?,   Ok !"
    else
        echo "返回状态码: $?,   Fail !"
    fi
}

# 修改主机名
changeHostName() {
    hostnamectl set-hostname $node03
}

# 把集群其他节点hosts信息加入到本节点hosts
addHosts() {
    hostsPath=/etc/hosts
cat >> $hostsPath << EOF
$master01Ip $master01
$master02Ip $master02
$master03Ip $master03
$node01IP $node01
$node02IP $node02
$node03IP $node03
$lbIP $lb
EOF
}

# 把本节点ssh公钥拷贝到集群其他节点
cpSshPub() {
    ssh-keygen
    for i in $master01 $node01 $node02; do
        ssh-copy-id -i ~/.ssh/id_rsa.pub $i  
    done
    checkStatusCode
}

# 把本节点hosts信息拷贝到集群其他节点
sshRemote() {
    ssh $master01 "echo $node03IP $node03 >> $hostsPath"
    checkStatusCode
    ssh $node01 "echo $node03IP $node03 >> $hostsPath"
    checkStatusCode
    ssh $node02 "echo $node03IP $node03 >> $hostsPath"
    checkStatusCode
}

# 修改系统配置
changeSystemCfg() {
    selinuxConfigPath=/etc/selinux/config
    swapConfigPath=/etc/fstab
    # 关闭防火墙
    systemctl stop firewalld && systemctl disable firewalld
    # 临时关闭selinux
    setenforce 0
    # 永久关闭selinux
    sed -i 's/enforcing/disabled/' $selinuxConfigPath
    # 临时关闭swap
    swapoff -a
    # 永久关闭swap
    sed -ri 's/.*swap.*/#&/' $swapConfigPath
}

# ntpdate时间同步
installNtpSvc() {
    # 安装ntp服务
    wlnmpRpm=http://mirrors.wlnmp.com/centos/wlnmp-release-centos.noarch.rpm
    shanghaiTime=/usr/share/zoneinfo/Asia/Shanghai
    localTime=/etc/localtime
    timezone=/etc/timezone
    rpm -ivh $wlnmpRpm && yum install ntpdate -y
    checkStatusCode

    # 同步时间
    ln -sf $shanghaiTime $localTime && \
    echo 'Asia/Shanghai' > $timezone && \
    ntpdate time2.aliyun.com
    checkStatusCode

    # 加入到定时任务
    crontabConfig=/var/spool/cron/root
cat >> $crontabConfig << EOF
*/5 * * * * ntpdate time2.aliyun.com &> /dev/null
30 04 */3 * * yum update -y && yum clean all && yum makecache &> /dev/null
EOF
    # 重启crontab
    systemctl restart crond.service
    crontab -l
}

# 将桥接的IPv4流量传递到iptables的链
k8sIpv4() {
    kubenetesIpv4Config=/etc/sysctl.d/k8s.conf
cat > $kubenetesIpv4Config << EOF 
net.bridge.bridge-nf-call-ip6tables = 1 
net.bridge.bridge-nf-call-iptables = 1 
EOF

    # 重载
    sysctl --system
}

# 开启ipv4转发
ipv4Forward() {
    sysctlConf=/etc/sysctl.conf
cat >> $sysctlConf << EOF
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
net.bridge.bridge-nf-call-arptables = 1
net.ipv4.ip_forward = 1
EOF

    sysctl -p
}

# 升级指定内核版本：4.19.5
updateKernel() {
    kernelPath=./kernel
    kernelDownloadAddr=http://mirrors.coreix.net/elrepo-archive-archive/kernel/el7/x86_64/RPMS
    kernelRpm1=kernel-ml-4.19.5-1.el7.elrepo.x86_64.rpm
    kernelRpm2=kernel-ml-devel-4.19.5-1.el7.elrepo.x86_64.rpm
    kernelRpm3=kernel-ml-headers-4.19.5-1.el7.elrepo.x86_64.rpm
    if [ ! -d $kernelPath ]; then
        echo "内核rpm包不存在, 开始手动下载:"
        mkdir $kernelPath && cd $kernelPath
        wget $kernelDownloadAddr/$kernelRpm1 && \
        wget $kernelDownloadAddr/$kernelRpm2 && \
        wget $kernelDownloadAddr/$kernelRpm3
        checkStatusCode
    fi
    cd $kernelPath
    rpm -ivh kernel-ml-4.19.5-1.el7.elrepo.x86_64.rpm && \
    rpm -ivh kernel-ml-devel-4.19.5-1.el7.elrepo.x86_64.rpm && \
    rpm -ivh kernel-ml-headers-4.19.5-1.el7.elrepo.x86_64.rpm
    checkStatusCode
    # 修改 GRUB 配置
    grubPath=/etc/default/grub
    sed -i 's/GRUB_DEFAULT=saved/GRUB_DEFAULT=0/g' $grubPath
    # 重建内核配置并重启
    grub2-mkconfig -o /boot/grub2/grub.cfg
    checkStatusCode
    echo "系统初始化配置结束, 系统将在5秒后重启, 重启后请继续执行脚本:"
    timer && reboot
}

updateYum() {
    echo "内核版本:"
    uname -a
    sleep 2
    echo "开始更新yum:"
    yum update -y && yum clean all && yum makecache
    checkStatusCode
    initdPath=/etc/rc.d/init.d
cat >> $initdPath/updateYum.sh << EOF
#!/bin/bash
#chkconfig:2345 80 90
#decription:auto update yum repo
yum update -y && yum clean all && yum makecache
EOF

    chmod +x $initdPath/updateYum.sh
    cd $initdPath
    chkconfig --add updateYum.sh
    chkconfig updateYum.sh on
    chkconfig
    cd -
}

# 安装docker
installDockerEngine() {
    docker ps
    if [ $? -ne 0 ]; then
        echo "Docker不存在, 开始安装..."
        # 一键安装
        curl -fsSL https://get.docker.com/ | sh
        systemctl start docker && systemctl enable docker && systemctl status docker
        # 配置Docker国内镜像加速
cat > /etc/docker/daemon.json << EOF
{"registry-mirrors": ["https://u8n2zdxj.mirror.aliyuncs.com"]}
EOF
        # 重载
        systemctl daemon-reload && systemctl restart docker && systemctl status docker
    else
        echo "Docker已存在"
        echo "版本:"
        docker version
    fi
}

# 初始化配置
initSystem() {
    changeHostName
    addHosts
    cpSshPub
    sshRemote
    changeSystemCfg
    installNtpSvc
    k8sIpv4
    ipv4Forward
    updateKernel
}

# 部署node节点
addNodeWork() {
    # 拷贝已部署好的Node相关文件到新节点
    dataDir=/opt
    sslDir=$dataDir/kubernetes/ssl
    cfgDir=$dataDir/kubernetes/cfg
    logsDir=$dataDir/kubernetes/logs
    kubeSystemSvc=/usr/lib/systemd/system

    scp -r root@$master01:$dataDir/kubernetes $dataDir && \
    scp -r root@$master01:$kubeSystemSvc/{kubelet,kube-proxy}.service $kubeSystemSvc && \
    scp root@$master01:$sslDir/ca.pem $sslDir
    rm $logsDir/* -rf

    # 删除旧的kubelet证书和kubeconfig文件
    rm -f $cfgDir/kubelet.kubeconfig && \
    rm -f $sslDir/kubelet*

    # 修改配置文件中的主机名
    sed -i 's/k8s-master01/k8s-node03/g' $cfgDir/kubelet.conf
    sed -i 's/k8s-master01/k8s-node03/g' $cfgDir/kube-proxy-config.yml

    # 启动并设置开机自启
    systemctl daemon-reload && \
    systemctl start kubelet kube-proxy && \
    systemctl enable kubelet kube-proxy && \
    systemctl status kubelet kube-proxy

    # 把新增的node节点加入到集群中
    ssh master01 "kubectl certificate approve $(kubectl get csr | awk '{if (NR > 1 && NR < 3){print $1}}'); kubectl certificate approve $(kubectl get csr | awk '{if (NR > 2){print $1}}')"
}

echo "输入编号执行对应安装程序:"
echo "1、系统初始化 (内核升级后系统会自动重启)"
echo "2、检查内核是否升级成功并更新yum"
echo "3、安装Docker"
echo "4、新增node节点加入到k8s集群"
echo "请输入数字:"
read num

case $num in
    1)  echo "系统初始化开始..."
        initSystem
    ;;
    2)  echo "检查内核版本, 更新yum..."
        updateYum
    ;;
    3)  echo "开始安装Docker..."
        installDockerEngine
    ;;
    4) echo "开始配置node节点, 请稍等..."
        addNodeWork
    ;;
    *)  echo "请输入对应数字"
        exit
    ;;
esac

至此一个可拥有完整生命周期的k8s集群部署完成了。

转载请注明-MrZ-个人博客
THE END
分享
二维码
海报
K8s1.20 二进制部署&高可用扩容
目录1 1 环境说明1.1 1.1 主机规划1.2 1.2 IP规划2 2 部署流程一 : 3 节点master集群部署2.1 2.1 master规划2.2 2.2 修改hosts2.3 2.3 OS优化2.4 2.4 免密登录2.5 2.5 部署ETCD集群2.5.1 2.5.1 安装cfssl2.5.2 ……
<<上一篇
下一篇>>