Kubeadm 搭建
- 一、搭建准备
- 1.1 环境准备
- 1.2 所有节点安装docker
- 1.3 所有主机安装 cri-dockerd
- 1.4 所有节点安装kubeadm,kubelet和kubectl
- 1.5 部署K8S集群
- 1.6 设定kubectl
- 1.7 部署 Dashboard
一、搭建准备
master(2C/4G,cpu核心数要求大于2) 192.168.80.10 docker、kubeadm、kubelet、kubectl、flannel
node01(2C/2G) 192.168.80.11 docker、kubeadm、kubelet、kubectl、flannel
node02(2C/2G) 192.168.80.12 docker、kubeadm、kubelet、kubectl、flannel
Harbor节点(hub.kgc.com) 192.168.80.13 docker、docker-compose、harbor-offline-v1.2.2
1.1 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
sed -i 's/enforcing/disabled/' /etc/selinux/config
iptables -F && iptables -t nat -F && iptables -t mangle -F && iptables -X
swapoff -a
sed -ri 's/.*swap.*/#&/' /etc/fstab
#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//修改主机名
hostnamectl set-hostname master01
hostnamectl set-hostname node01
hostnamectl set-hostname node02//所有节点修改hosts文件
vim /etc/hosts
192.168.80.10 master01
192.168.80.11 node01
192.168.80.12 node02//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//生效参数
sysctl --system
1.2 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.iomkdir /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"
Cgroup Driver: systemd
1.3 所有主机安装 cri-dockerd
-
Kubernetes自v1.24移除了对docker-shim的支持,而Docker Engine默认又不支持CRI规范,因而二者将无法直接完成整合。 为此,Mirantis和Docker联合创建了cri-dockerd项目,用于为Docker Engine提供一个能够支持到CRI规范的垫片,从而能够让Kubernetes基于CRI控制Docker 。
-
项目地址
https://github.com/Mirantis/cri-dockerd
cd /opt/
wget https://github.com/Mirantis/cri-dockerd/releases/download/v0.2.6/cri-dockerd-0.2.6-3.el7.x86_64.rpmyum localinstall -y cri-dockerd-0.2.6-3.el7.x86_64.rpmvim /lib/systemd/system/cri-docker.service
#修改ExecStart行如下
ExecStart=/usr/bin/cri-dockerd --network-plugin=cni --pod-infra-container-image=registry.aliyuncs.com/google_containers/pause:3.8systemctl daemon-reload
systemctl enable --now cri-docker
1.4 所有节点安装kubeadm,kubelet和kubectl
//定义kubernetes源
cat > /etc/yum.repos.d/kubernetes.repo << EOF
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=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
EOFyum install -y kubelet-1.25.4 kubeadm-1.25.4 kubectl-1.25.4kubeadm version//开机自启kubelet
systemctl enable kubelet.service
#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
1.5 部署K8S集群
//在 master01 节点上设置集群初始化配置文件
kubeadm config print init-defaults > /opt/kubeadm-config.yamlcd /opt/
vim kubeadm-config.yaml
apiVersion: kubeadm.k8s.io/v1beta3
bootstrapTokens:
- groups:- system:bootstrappers:kubeadm:default-node-tokentoken: abcdef.0123456789abcdefttl: 24h0m0susages:- signing- authentication
kind: InitConfiguration
localAPIEndpoint:advertiseAddress: 192.168.80.10bindPort: 6443
nodeRegistration:criSocket: unix:///var/run/cri-dockerd.sockimagePullPolicy: IfNotPresentname: master01taints:- effect: NoSchedulekey: node-role.kubernetes.io/master
---
apiServer:timeoutForControlPlane: 4m0s
apiVersion: kubeadm.k8s.io/v1beta3
certificatesDir: /etc/kubernetes/pki
clusterName: kubernetes
controllerManager: {}
dns: {}
etcd:local:dataDir: /var/lib/etcd
imageRepository: registry.aliyuncs.com/google_containers
kind: ClusterConfiguration
kubernetesVersion: v1.25.4
networking:dnsDomain: cluster.localpodSubnet: 10.244.0.0/16serviceSubnet: 10.96.0.0/16
scheduler: {}
---
apiVersion: kubeproxy.config.k8s.io/v1alpha1
kind: KubeProxyConfiguration
mode: ipvs//查看初始化需要的镜像
kubeadm config images list --kubernetes-version 1.25.4//所有节点拉取镜像
#拷贝yaml配置文件给其他主机,通过配置文件进行拉取镜像
for i in node01 node02; do scp /opt/kubeadm-config.yaml $i:/opt/; donekubeadm config images pull --config /opt/kubeadm-config.yaml//master01 节点初始化kubeadm
方法一:
kubeadm init --config=kubeadm-config.yaml --upload-certs | tee kubeadm-init.log
#--upload-certs 参数可以在后续执行加入节点时自动分发证书文件
#tee kubeadm-init.log 用以输出日志//查看 kubeadm-init 日志
less kubeadm-init.log//kubernetes配置文件目录
ls /etc/kubernetes///存放ca等证书和密码的目录
ls /etc/kubernetes/pki 方法二:
kubeadm init \
--apiserver-advertise-address=192.168.80.10 \
--image-repository registry.aliyuncs.com/google_containers \
--kubernetes-version v1.25.4 \
--service-cidr=10.96.0.0/16 \
--pod-network-cidr=10.244.0.0/16 \
--cri-socket /var/run/cri-dockerd.sock \
--upload-certs
- 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:多主节点必选项,用于指定控制平面的固定访问地址。注意:kubeadm不支持将没有–control-plane-endpoint参数的单个控制平面集群转换为高可用性集群
–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。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
–token-ttl:默认token的有效期为24小时,0表示永不过期
–cri-socket #v1.24版之后指定连接cri的socket文件路径,注意;不同的CRI连接文件不同
#如果是cRI是containerd,则使用–cri-socket unix:///run/containerd/containerd.sock
#如果是cRI是docker,则使用–cri-socket unix:///var/run/cri-dockerd.sock
#如果是CRI是CRI-o,则使用–cri-socket unix:///var/run/crio/crio.sock
#注意:CRI-o与containerd的容器管理机制不一样,所以镜像文件不能通用。
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/.kubesudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/configsudo chown $(id -u):$(id -g) $HOME/.kube/configYou should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:https://kubernetes.io/docs/concepts/cluster-administration/addons/Then you can join any number of worker nodes by running the following on each as root:kubeadm join 192.168.80.10:6443 --token rc0kfs.a1sfe3gl4dvopck5 \--discovery-token-ca-cert-hash sha256:864fe553c812df2af262b406b707db68b0fd450dc08b34efb73dd5a4771d37a2
1.6 设定kubectl
- kubectl需经由API server认证及授权后方能执行相应的管理操作,kubeadm 部署的集群为其生成了一个具有管理员权限的认证配置文件 /etc/kubernetes/admin.conf,它可由 kubectl 通过默认的 “$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//所有节点部署网络插件flannel
#所有节点上传 flannel 镜像 flannel.tar、flannel-cni-plugin.tar 和网络插件 cni-plugins-linux-amd64-v1.1.1.tgz 到 /opt 目录,master 节点上传 kube-flannel.yml 文件
mv /opt/cni /opt/cni_bak
mkdir -p /opt/cni/bin
tar zxvf cni-plugins-linux-amd64-v1.1.1.tgz -C /opt/cni/bin方法一:
cd /opt
docker load < flannel.tar
docker load < flannel-cni-plugin.tarkubectl apply -f kube-flannel.yml 方法二:
kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml//在 node 节点上执行 kubeadm join 命令加入群集
kubeadm join 192.168.80.10:6443 --token rc0kfs.a1sfe3gl4dvopck5 \--discovery-token-ca-cert-hash sha256:864fe553c812df2af262b406b707db68b0fd450dc08b34efb73dd5a4771d37a2 \--cri-socket /var/run/cri-dockerd.sock #需要额外指定 cri-dockerd//在master节点查看节点状态
kubectl get nodes
NAME STATUS ROLES AGE VERSION
master01 Ready control-plane 88m v1.25.4
node01 Ready <none> 110s v1.25.4
node02 Ready <none> 105s v1.25.4kubectl get pods -A
NAMESPACE NAME READY STATUS RESTARTS AGE
kube-flannel kube-flannel-ds-lc7lg 1/1 Running 0 83m
kube-flannel kube-flannel-ds-phlnb 1/1 Running 0 88s
kube-flannel kube-flannel-ds-wlvvk 1/1 Running 0 93s
kube-system coredns-c676cc86f-5x7b5 1/1 Running 0 88m
kube-system coredns-c676cc86f-8wxg7 1/1 Running 0 88m
kube-system etcd-master01 1/1 Running 0 88m
kube-system kube-apiserver-master01 1/1 Running 0 88m
kube-system kube-controller-manager-master01 1/1 Running 0 88m
kube-system kube-proxy-rjs6g 1/1 Running 0 88s
kube-system kube-proxy-vp2b5 1/1 Running 0 88m
kube-system kube-proxy-xnllf 1/1 Running 0 93s
kube-system kube-scheduler-master01 1/1 Running 0 88m//测试 pod 资源创建
kubectl create deployment nginx --image=nginxkubectl 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>//暴露端口提供服务
kubectl expose deployment nginx --port=80 --type=NodePortkubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 25h
nginx NodePort 10.96.15.132 <none> 80:32698/TCP 4s//测试访问
curl http://node01:32698//扩展3个副本
kubectl scale deployment nginx --replicas=3kubectl get pods -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx-554b9c67f9-9kh4s 1/1 Running 0 66s 10.244.1.3 node01 <none> <none>
nginx-554b9c67f9-rv77q 1/1 Running 0 66s 10.244.2.2 node02 <none> <none>
nginx-554b9c67f9-zr2xs 1/1 Running 0 17m 10.244.1.2 node01 <none> <none>
1.7 部署 Dashboard
kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.6.1/aio/deploy/recommended.yaml//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
小结:初始化失败,进行的操作
kubeadm reset -f
ipvsadm --clear
rm -rf ~/.kube
再次进行初始化