一、kubeadm 部署 K8S 集群架构

主机名

IP地址

安装组件

master(2C/4G,cpu核心数要求大于2)

192.168.2.66

docker、kubeadm、kubelet、kubectl、flannel

node01(2C/2G)

192.168.2.200

docker、kubeadm、kubelet、kubectl、flannel

node02(2C/2G)

192.168.2.77

docker、kubeadm、kubelet、kubectl、flannel

Harbor节点(hub.kgc.com)

192.168.2.22

docker、docker-compose、harbor-offline-v1.2.2

部署的总体步骤如下:

1、在所有节点上安装Docker和kubeadm

2、部署Kubernetes Master

3、部署容器网络插件

4、部署 Kubernetes Node,将节点加入Kubernetes集群中

5、部署 Dashboard Web 页面,可视化查看Kubernetes资源

6、部署 Harbor 私有仓库,存放镜像资源

1、环境准备

//所有节点,关闭防火墙规则,关闭selinux,关闭swap交换

systemctl stop firewalld
systemctl disable firewalld
setenforce 0
iptables -F && iptables -t nat -F && iptables -t mangle -F && iptables -X
swapoff -a  #交换分区必须要关闭
sed -ri 's/.*swap.*/#&/' /etc/fstab #永久关闭swap分区,&符号在sed命令中代表上次匹配的结果
#加载 ip_vs 模块
for i in $(ls /usr/lib/modules/$(uname -r)/kernel/net/netfilter/ipvs|grep -o "^[^.]*");do echo $i; /sbin/modinfo -F filename $i >/dev/null 2>&1 && /sbin/modprobe $i;done
复制代码

k8s harbor拉取 k8s harbor部署_上传

//修改主机名

hostnamectl set-hostname master
hostnamectl set-hostname node01
hostnamectl set-hostname node02
复制代码

k8s harbor拉取 k8s harbor部署_云原生_02

k8s harbor拉取 k8s harbor部署_docker_03

k8s harbor拉取 k8s harbor部署_云原生_04

//所有节点修改hosts文件

vim /etc/hosts
192.168.2.66 master
192.168.2.200 node01
192.168.2.77 node02
复制代码

k8s harbor拉取 k8s harbor部署_云原生_05

//调整内核参数

cat > /etc/sysctl.d/kubernetes.conf << EOF
#开启网桥模式,可将网桥的流量传递给iptables链
net.bridge.bridge-nf-call-ip6tables=1
net.bridge.bridge-nf-call-iptables=1
#关闭ipv6协议
net.ipv6.conf.all.disable_ipv6=1
net.ipv4.ip_forward=1
EOF
复制代码

k8s harbor拉取 k8s harbor部署_docker_06

//生效参数

sysctl --system
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_07

2、所有节点安装docker

yum install -y yum-utils device-mapper-persistent-data lvm2
yum-config-manager --add-repo https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
yum install -y docker-ce docker-ce-cli containerd.io
 
mkdir /etc/docker
cat > /etc/docker/daemon.json <<EOF
{
"registry-mirrors": ["https://6ijb8ubo.mirror.aliyuncs.com"],
"exec-opts": ["native.cgroupdriver=systemd"],
"log-driver": "json-file",
"log-opts": {
"max-size": "100m"
}
}
EOF
#使用Systemd管理的Cgroup来进行资源控制与管理,因为相对Cgroupfs而言,Systemd限制CPU、内存等资源更加简单和成熟稳定。
#日志使用json-file格式类型存储,大小为100M,保存在/var/log/containers目录下,方便ELK等日志系统收集和管理日志。
 
systemctl daemon-reload
systemctl restart docker.service
systemctl enable docker.service
 
docker info | grep "Cgroup Driver"
复制代码

k8s harbor拉取 k8s harbor部署_docker_08

k8s harbor拉取 k8s harbor部署_云原生_09

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_10

3、所有节点安装kubeadm,kubelet和kubectl

//定义kubernetes源,安装kubeadm,kubelet和kubectl

cat > /etc/yum.repos.d/kubernetes.repo << EOF
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=0
repo_gpgcheck=0
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
EOF
 
yum install -y kubelet-1.15.1 kubeadm-1.15.1 kubectl-1.15.1
复制代码

k8s harbor拉取 k8s harbor部署_上传_11

//开机自启kubelet

systemctl enable kubelet.service  
复制代码

#K8S通过kubeadm安装出来以后都是以Pod方式存在,即底层是以容器方式运行,所以kubelet必须设置开机自启

k8s harbor拉取 k8s harbor部署_上传_12

3、部署K8S集群

//查看初始化需要的镜像

kubeadm config images list
复制代码

k8s harbor拉取 k8s harbor部署_kubernetes_13

  
//在 master 节点上传 kubeadm-basic.images.tar.gz 压缩包至 /opt 目录

cd /opt
tar zxvf kubeadm-basic.images.tar.gz
 
for i in $(ls /opt/kubeadm-basic.images/*.tar); do docker load -i $i; done
复制代码

k8s harbor拉取 k8s harbor部署_上传_14

k8s harbor拉取 k8s harbor部署_上传_15

//复制镜像和脚本到 node 节点,并在 node 节点上执行脚本 bash /opt/load-images.sh

scp -r kubeadm-basic.images root@node01:/opt
scp -r kubeadm-basic.images root@node02:/opt
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_16

k8s harbor拉取 k8s harbor部署_docker_17

 
//初始化kubeadm

方法一:

kubeadm config print init-defaults > /opt/kubeadm-config.yaml
 
cd /opt/
vim kubeadm-config.yaml
......
11 localAPIEndpoint:
12 advertiseAddress: 192.168.229.90 #指定master节点的IP地址
13 bindPort: 6443
......
34 kubernetesVersion: v1.15.1   #指定kubernetes版本号
35 networking:
36 dnsDomain: cluster.local
37 podSubnet: "10.244.0.0/16"   #指定pod网段,10.244.0.0/16用于匹配flannel默认网段
38 serviceSubnet: 10.96.0.0/16  #指定service网段
39 scheduler: {}
--- #末尾再添加以下内容
apiVersion: kubeproxy.config.k8s.io/v1alpha1
kind: KubeProxyConfiguration
mode: ipvs  #把默认的service调度方式改为ipvs模式
 
 
kubeadm init --config=kubeadm-config.yaml --experimental-upload-certs | tee kubeadm-init.log
#--experimental-upload-certs 参数可以在后续执行加入节点时自动分发证书文件,k8sV1.16版本开始替换为 --upload-certs
#tee kubeadm-init.log 用以输出日志
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_18

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_19

k8s harbor拉取 k8s harbor部署_docker_20

k8s harbor拉取 k8s harbor部署_docker_21

//查看 kubeadm-init 日志

less kubeadm-init.log
复制代码

k8s harbor拉取 k8s harbor部署_云原生_22

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_23

//kubernetes配置文件目录

ls /etc/kubernetes/
复制代码

k8s harbor拉取 k8s harbor部署_docker_24

//存放ca等证书和密码的目录

ls /etc/kubernetes/pki
复制代码

k8s harbor拉取 k8s harbor部署_kubernetes_25

方法二:

kubeadm init \
--apiserver-advertise-address=0.0.0.0 \
--image-repository registry.aliyuncs.com/google_containers \
--kubernetes-version=v1.15.1 \
--service-cidr=10.1.0.0/16 \
--pod-network-cidr=10.244.0.0/16  
复制代码

初始化集群需使用kubeadm init命令,可以指定具体参数初始化,也可以指定配置文件初始化。
可选参数:

--apiserver-advertise-address:apiserver通告给其他组件的IP地址,一般应该为Master节点的用于集群内部通信的IP地址,0.0.0.0表示节点上所有可用地址
 --apiserver-bind-port:apiserver的监听端口,默认是6443
 --cert-dir:通讯的ssl证书文件,默认/etc/kubernetes/pki
 --control-plane-endpoint:控制台平面的共享终端,可以是负载均衡的ip地址或者dns域名,高可用集群时需要添加
 --image-repository:拉取镜像的镜像仓库,默认是k8s.gcr.io
 --kubernetes-version:指定kubernetes版本
 --pod-network-cidr:pod资源的网段,需与pod网络插件的值设置一致。通常,Flannel网络插件的默认为10.244.0.0/16,Calico插件的默认值为192.168.0.0/16;
 --service-cidr:service资源的网段
 --service-dns-domain:service全域名的后缀,默认是cluster.local

方法二初始化后需要修改 kube-proxy 的 configmap,开启 ipvs

kubectl edit cm kube-proxy -n=kube-system
修改mode: ipvs
 
  
 
提示:
......
Your Kubernetes control-plane has initialized successfully!
 
To start using your cluster, you need to run the following as a regular user:
 
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
 
You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
https://kubernetes.io/docs/concepts/cluster-administration/addons/
 
Then you can join any number of worker nodes by running the following on each as root:
 
kubeadm join 192.168.229.90:6443 --token rc0kfs.a1sfe3gl4dvopck5 \
--discovery-token-ca-cert-hash sha256:864fe553c812df2af262b406b707db68b0fd450dc08b34efb73dd5a4771d37a2
复制代码

//设定kubectl
kubectl需经由API server认证及授权后方能执行相应的管理操作,kubeadm 部署的集群为其生成了一个具有管理员权限的认证配置文件 /etc/kubernetes/admin.conf,它可由 kubectl 通过默认的 “$HOME/.kube/config” 的路径进行加载。

mkdir -p $HOME/.kube
cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
chown $(id -u):$(id -g) $HOME/.kube/config
复制代码

k8s harbor拉取 k8s harbor部署_kubernetes_26

k8s harbor拉取 k8s harbor部署_上传_27

//在 node 节点上执行 kubeadm join 命令加入群集

kubeadm join 192.168.2.66:6443 --token abcdef.0123456789abcdef \
    --discovery-token-ca-cert-hash sha256:aeeaa0e74c0854944d152aaf29f7b2f10863fc05a8dc1d73a7687b04bfde9e75
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_28

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_29

//所有节点部署网络插件flannel
方法一:
//所有节点上传flannel镜像 flannel.tar 到 /opt 目录,master节点上传 kube-flannel.yml 文件

cd /opt
docker load -i flannel.tar
复制代码

k8s harbor拉取 k8s harbor部署_kubernetes_30

k8s harbor拉取 k8s harbor部署_云原生_31

k8s harbor拉取 k8s harbor部署_上传_32

//在 master 节点创建 flannel 资源

kubectl apply -f kube-flannel.yml
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_33

如果状态还是NotReady,就替换新的cni成旧的cni,因为新的cni有污点,还有flannel.tar,重新下载镜像,kube-flannel.yml替换好了就行了,所有节点都需要更新


方法二:

kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml 
复制代码

//在master节点查看节点状态(需要等几分钟)

kubectl get pods -n kube-system
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_34

//测试 pod 资源创建

kubectl create deployment nginx --image=nginx
 
kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx-554b9c67f9-zr2xs 1/1 Running 0 14m 10.244.1.2 node01 <none> <none>
复制代码

k8s harbor拉取 k8s harbor部署_docker_35

//暴露端口提供服务

kubectl expose deployment nginx --port=80 --type=NodePort
kubectl get svc
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_36

//测试访问

curl http://node01:32557  #使用Node1或者node2的IP进行访问测试
复制代码

k8s harbor拉取 k8s harbor部署_docker_37

//扩展3个副本

kubectl scale deployment nginx --replicas=3
kubectl get pods -o wide
复制代码

k8s harbor拉取 k8s harbor部署_docker_38

二、dashboard 部署

1、 安装dashboard

//所有节点安装dashboard
方法一:
//所有节点上传dashboard镜像 dashboard.tar 到 /opt 目录,master节点上传kubernetes-dashboard.yaml文件

cd /opt/
docker load -i dashboard.tar
 
kubectl apply -f kubernetes-dashboard.yaml
复制代码

k8s harbor拉取 k8s harbor部署_kubernetes_39

k8s harbor拉取 k8s harbor部署_上传_40

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_41

方法二:

kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.0.0/aio/deploy/recommended.yaml
复制代码

//查看所有容器运行状态

kubectl get pods -A
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_42

2、使用火狐或者360浏览器访问

https://node02:30001/
https://192.168.2.200:30001/    #使用Node1或者node2访问
复制代码

k8s harbor拉取 k8s harbor部署_docker_43

//创建service account并绑定默认cluster-admin管理员集群角色

kubectl create serviceaccount dashboard-admin -n kube-system
 
kubectl create clusterrolebinding dashboard-admin --clusterrole=cluster-admin --serviceaccount=kube-system:dashboard-admin
复制代码

k8s harbor拉取 k8s harbor部署_云原生_44

//获取令牌密钥

kubectl describe secrets -n kube-system $(kubectl -n kube-system get secret | awk '/dashboard-admin/{print $1}')
复制代码

k8s harbor拉取 k8s harbor部署_云原生_45

//复制token令牌直接登录网站

k8s harbor拉取 k8s harbor部署_云原生_46

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_47

三 、安装Harbor私有仓库

//修改主机名

hostnamectl set-hostname hub.xkq.com
复制代码

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_48

//所有节点加上主机名映射

echo '192.168.2.22 hub.xkq.com' >> /etc/hosts
复制代码

k8s harbor拉取 k8s harbor部署_docker_49

//安装 docker

yum install -y yum-utils device-mapper-persistent-data lvm2
yum-config-manager --add-repo https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
yum install -y docker-ce docker-ce-cli containerd.io
 
mkdir /etc/docker
cat > /etc/docker/daemon.json <<EOF### 下面命令也需要在master和node节点重新执行,因为之前没有指定harbor仓库地址
{
"registry-mirrors": ["https://6ijb8ubo.mirror.aliyuncs.com"],
"exec-opts": ["native.cgroupdriver=systemd"],
"log-driver": "json-file",
"log-opts": {
"max-size": "100m"
},
"insecure-registries": ["https://hub.xkq.com"]   
}
EOF
 
systemctl start docker
systemctl enable docker
复制代码

k8s harbor拉取 k8s harbor部署_云原生_50

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_51

//所有 node 节点都修改 docker 配置文件,加上私有仓库配置

cat > /etc/docker/daemon.json <<EOF
{
"registry-mirrors": ["https://6ijb8ubo.mirror.aliyuncs.com"],
"exec-opts": ["native.cgroupdriver=systemd"],
"log-driver": "json-file",
"log-opts": {
"max-size": "100m"
},
"insecure-registries": ["https://hub.xkq.com"]
}
EOF
 
systemctl daemon-reload
systemctl restart docker
复制代码

k8s harbor拉取 k8s harbor部署_云原生_52

//安装 Harbor
//上传 harbor-offline-installer-v1.2.2.tgz 和 docker-compose 文件到 /opt 目录

cd /opt
cp docker-compose /usr/local/bin/
chmod +x /usr/local/bin/docker-compose
 
tar zxvf harbor-offline-installer-v1.2.2.tgz
cd harbor/
vim harbor.cfg
5 hostname = hub.xkq.com
9 ui_url_protocol = https
24 ssl_cert = /data/cert/server.crt
25 ssl_cert_key = /data/cert/server.key
59 harbor_admin_password = Harbor12345
复制代码

k8s harbor拉取 k8s harbor部署_云原生_53

k8s harbor拉取 k8s harbor部署_kubernetes_54

//生成证书

mkdir -p /data/cert
cd /data/cert  
复制代码

#生成私钥

openssl genrsa -des3 -out server.key 2048
复制代码

输入两遍密码:123456

k8s harbor拉取 k8s harbor部署_上传_55

#生成证书签名请求文件

openssl req -new -key server.key -out server.csr
复制代码

输入私钥密码:123456
输入国家名:CN
输入省名:BJ
输入市名:BJ
输入组织名:LV
输入机构名:LV
输入域名:hub.ly.com

k8s harbor拉取 k8s harbor部署_kubernetes_56

k8s harbor拉取 k8s harbor部署_docker_57

#备份私钥

cp server.key server.key.org
复制代码

k8s harbor拉取 k8s harbor部署_docker_58

#清除私钥密码

openssl rsa -in server.key.org -out server.key  
复制代码

输入私钥密码:123456

k8s harbor拉取 k8s harbor部署_上传_59

#签名证书

openssl x509 -req -days 1000 -in server.csr -signkey server.key -out server.crt
 
chmod +x /data/cert/*
 
cd /opt/harbor/
./install.sh
复制代码

k8s harbor拉取 k8s harbor部署_云原生_60

浏览器访问:hub.xkq.com 用户名:admin
密码:Harbor12345

k8s harbor拉取 k8s harbor部署_docker_61

k8s harbor拉取 k8s harbor部署_kubernetes_62

//在一个node节点上登录harbor

docker login -u admin -p Harbor12345 https://hub.xkq.com
复制代码

k8s harbor拉取 k8s harbor部署_docker_63

//上传镜像

docker tag nginx:latest hub.xkq.com/library/nginx:v1
docker push hub.xkq.com/library/nginx:v1
复制代码

k8s harbor拉取 k8s harbor部署_kubernetes_64

k8s harbor拉取 k8s harbor部署_云原生_65

//在master节点上删除之前创建的nginx资源

kubectl delete deployment nginx
 
kubectl run nginx-deployment --image=hub.xkq.com/library/nginx:v1 --port=80 --replicas=3
 
kubectl expose deployment nginx-deployment --port=30000 --target-port=80
kubectl get svc,pods
 
yum install ipvsadm -y
ipvsadm -Ln
 
curl 10.96.222.161:30000

kubectl edit svc nginx-deployment
25 type: NodePort   #把调度策略改成NodePort
 
kubectl get svc
复制代码

k8s harbor拉取 k8s harbor部署_docker_66

k8s harbor拉取 k8s harbor部署_上传_67

k8s harbor拉取 k8s harbor部署_docker_68

k8s harbor拉取 k8s harbor部署_docker_69

k8s harbor拉取 k8s harbor部署_云原生_70

k8s harbor拉取 k8s harbor部署_上传_71

浏览器访问:

192.168.2.66:32122
192.168.2.200:32122
192.168.2.77:32122
复制代码

k8s harbor拉取 k8s harbor部署_docker_72

k8s harbor拉取 k8s harbor部署_k8s harbor拉取_73

k8s harbor拉取 k8s harbor部署_docker_74

四、 内核参数优化方案

cat > /etc/sysctl.d/kubernetes.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 #禁止使用 swap 空间,只有当系统内存不足(OOM)时才允许使用它
vm.overcommit_memory=1  #不检查物理内存是否够用
vm.panic_on_oom=0   #开启 OOM
fs.inotify.max_user_instances=8192
fs.inotify.max_user_watches=1048576
fs.file-max=52706963    #指定最大文件句柄数
fs.nr_open=52706963 #仅4.4以上版本支持
net.ipv6.conf.all.disable_ipv6=1
net.netfilter.nf_conntrack_max=2310720
EOF