个人笔记，仅供参考，他人按此步骤不一定能顺利搭建
1、配置阿里云docker源，安装组件，安装docker-ce(所有节点)
yum install -y yum-utils device-mapper-persistent-data lvm2
yum-config-manager --add-repo http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
yum install docker-ce-19.03.14
2、配置阿里云kubernetes源，安装kubeadm,kubectl,kubelet(kubeadm,kubelet在work节点，kubectl在master节点)
[root@master manifests]# cat /etc/yum.repos.d/kubernetes.repo
[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
yum install -y kubelet
yum install -y kubectl
yum install -y kubeadm
3、安装完成后，我们还需要对docker进行配置，因为用yum源的方式安装的kubelet生成的配置文件将参数--cgroup-driver改成了systemd，而 docker 的cgroup-driver是cgroupfs，这二者必须一致才行，我们可以通过docker info命令查看：
[root@master manifests]# docker info |grep Cgroup
Cgroup Driver: systemd
修改
[root@master manifests]# cat /etc/docker/daemon.json
{
"exec-opts": ["native.cgroupdriver=systemd"],
"log-driver": "json-file",
"log-opts": {
"max-size": "100m"
},
"storage-driver": "overlay2",
"storage-opts": [
"overlay2.override_kernel_check=true"
]
}
4、关闭swap
swapoff -a
5、在master节点提前下载好k8s组件镜像，再使用docker tag重命名，因为K8s默认是从grc.io下载镜像的，而我本地无法访问
查看需要的镜像
[root@master ~]# kubeadm config images list
k8s.gcr.io/kube-apiserver:v1.20.2
k8s.gcr.io/kube-controller-manager:v1.20.2
k8s.gcr.io/kube-scheduler:v1.20.2
k8s.gcr.io/kube-proxy:v1.20.2
k8s.gcr.io/pause:3.2
k8s.gcr.io/etcd:3.4.13-0
k8s.gcr.io/coredns:1.7.0
从dockerhub可以下载的地址下载镜像
docker pull aiotceo/kube-apiserver:v1.20.2
docker pull aiotceo/kube-controller-manager:v1.20.2
docker pull aiotceo/kube-proxy:v1.20.2
docker pull aiotceo/kube-scheduler:v1.20.2
docker pull aiotceo/pause:3.2
docker pull aiotceo/coredns:1.7.0
docker pull aiotceo/etcd:3.4.13-alpine
docker pull aiotceo/etcd:3.4.13-ubuntu
重命名
1109 docker tag docker.io/aiotceo/kube-apiserver:v1.20.2 k8s.gcr.io/kube-apiserver:v1.20.2
1111 docker tag aiotceo/kube-proxy:v1.20.2 k8s.gcr.io/kube-proxy:v1.20.2
1112 docker tag aiotceo/kube-scheduler:v1.20.2 k8s.gcr.io/kube-scheduler:v1.20.2
1113 docker tag docker.io/aiotceo/pause:3.2 k8s.gcr.io/pause:3.2
1114 docker tag docker.io/aiotceo/coredns:1.7.0 k8s.gcr.io/coredns:1.7.0
1115 docker tag docker.io/aiotceo/etcd:3.4.13-ubuntu k8s.gcr.io/etcd:3.4.13-0
6、集群初始化
kubeadm init --kubernetes-version=v1.20.2 --pod-network-cidr=10.244.0.0/16 --apiserver-advertise-address=10.1.1.206
初始化完成后你会看到下面这样的提示：
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 10.1.1.206:6443 --token hr8pxx.x7hnskwkz7dp20tq
--discovery-token-ca-cert-hash sha256:a894653ab32c92d89a4a43f6486bbe7cfbbeee1e601b5b3a99ffdcd68367737b
意思说你可以使用kubeadm join加入本集群，但是需要安装网络插件
此时使用kubectl get cs查看集群状态，会有以下提示：
[root@master manifests]# kubectl get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME STATUS MESSAGE ERROR
scheduler Unhealthy Get "http://127.0.0.1:10251/healthz": dial tcp 127.0.0.1:10251: connect: connection refused
controller-manager Unhealthy Get "http://127.0.0.1:10252/healthz": dial tcp 127.0.0.1:10252: connect: connection refused
etcd-0 Healthy {"health":"true"}
这是因为集群初始化默认从官方下载的kube-scheduler和kube-controller-manaer配置文件里面的port端口为0，将--port=0注释掉，然后重启Kubelet
...
spec:
containers:
- command:
- kube-scheduler
- --authentication-kubeconfig=/etc/kubernetes/scheduler.conf
- --authorization-kubeconfig=/etc/kubernetes/scheduler.conf
- --bind-address=127.0.0.1
- --kubeconfig=/etc/kubernetes/scheduler.conf
- --leader-elect=true
- --port=0
image: k8s.gcr.io/kube-scheduler:v1.20.2
...
systemctl restart kubelet
此时再查看cs，状态都为ready
[root@master manifests]# 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"}
7、安装网络插件
下载flannel配置文件,同样需要提前在worker节点下载kube-proxy镜像，并重命名
wget https://github.com/coreos/flannel/blob/master/Documentation/kube-flannel.yml
注意，需要在浏览器里面直接打开然后复制内容,再创建kube-flannel.yml，否则用kubectl apply -f 这个链接会报错
kubectl apply -f kube-flannel.yml
查看
kubectl -n kube-system get pod
[root@master ~]# kubectl -n kube-system get pod
NAME READY STATUS RESTARTS AGE
etcd-master 1/1 Running 0 71m
kube-apiserver-master 1/1 Running 0 71m
kube-controller-manager-master 1/1 Running 0 71m
kube-flannel-ds-5xsbn 1/1 Running 0 34m
kube-flannel-ds-9dxlm 1/1 Running 0 34m
kube-flannel-ds-xt564 1/1 Running 0 35m
8、工作节点加入集群
分别在工作节点使用上面初始化的生成的token和hash值加入集群
kubeadm join 10.1.1.206:6443 --token hr8pxx.x7hnskwkz7dp20tq
--discovery-token-ca-cert-hash sha256:a894653ab32c92d89a4a43f6486bbe7cfbbeee1e601b5b3a99ffdcd68367737b
vim /etc/profile
在底部增加新的环境变量 export KUBECONFIG=/etc/kubernetes/admin.conf
source /etc/profile
把主节点文件拷贝到node节点
scp -r /etc/kubernetes/admin.conf root@200.200.4.151:/etc/kubernetes/admin.conf
给从节点打标签
kubectl label node k8s-node-151 node-role.kubernetes.io/node=
[root@master ~]# kubectl get csr
NAME AGE SIGNERNAME REQUESTOR CONDITION
csr-c4tqr 78m kubernetes.io/kube-apiserver-client-kubelet system:bootstrap:hr8pxx Approved,Issued
csr-pdsnh 78m kubernetes.io/kube-apiserver-client-kubelet system:bootstrap:hr8pxx Approved,Issued
[root@master ~]# kubectl get nodes
NAME STATUS ROLES AGE VERSION
master Ready control-plane,master 86m v1.20.2
node01 Ready <none> 78m v1.20.2
node02 Ready <none> 78m v1.20.2
9、创建coredns
提前将镜像下载好，并重命名
docker pull aiotceo/coredns:1.7.0
docker tag docker.io/aiotceo/coredns:1.7.0 k8s.gcr.io/coredns:1.7.0
[root@master ~]# kubectl -n kube-system get pod
NAME READY STATUS RESTARTS AGE
coredns-74ff55c5b-ckd4v 1/1 Running 0 90m
coredns-74ff55c5b-pgw2h 1/1 Running 0 90m
PS：安装过程中如果有出错，使用以下命令重置，并且重新初始化，worker节点也一样
kubeadm reset
systemctl daemon-reload
systemctl restart kubelet
iptables -F && iptables -t nat -F && iptables -t mangle -F && iptables -X
