Kubernetes(k8s)1.18.3高可用集群部署（多master多node）（三）

第三部分：部署k8s集群（多master多node）

创建kubeadm配置文件  （在主master上）

cat > ./kubeadm-config.yaml <<EOF

apiVersion: kubeadm.k8s.io/v1beta2

kind: ClusterConfiguration

kubernetesVersion: v1.18.3

apiServer:

  certSANs:

  - "127.0.0.1"

  - "k8s-vip"

  - "k8s-m1"

  - "k8s-m2"

  - "k8s-m3"

  - "192.168.40.9"

  - "192.168.40.81"

  - "192.168.40.82"

  - "192.168.40.83"

imageRepository: registry.cn-hangzhou.aliyuncs.com/google_containers

controlPlaneEndpoint: "k8s-vip:16443"

networking:

  serviceSubnet: "10.96.0.0/16"

  podSubnet: "10.244.0.0/16"

EOF

初始化master 

（在主master上）

# kubeadm init --config=kubeadm-config.yaml --upload-certs

按照说明进行设置：

[[email protected] ~]# mkdir -p $HOME/.kube

[[email protected] ~]# cp -i /etc/kubernetes/admin.conf $HOME/.kube/config

[[email protected] ~]# chown $(id -u):$(id -g) $HOME/.kube/config

部署flannel网络

[[email protected] ~]# mkdir -p ~/k8s_flannel && cd ~/k8s_flannel

[[email protected] k8s_flannel]# wget https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml

修改kube-flannel.yml配置文件

# vim kube-flannel.yml

集群容器网段要与集群的pod-network一致

flanneld启动参数加上–-iface=<iface-name>  本例中的网卡为 eth0（单网卡可选  多个网卡必须指定网络接口否则nds解析错误）

启动flannel：

# kubectl apply -f  kube-flannel.yml       启动flannel网络插件

检查pod与node

master other 加入集群

 

先复制密钥及相关文件：

在vip的master1上把相关密钥与文件scp到其他所有master上

 

# ssh [email protected] mkdir -p /etc/kubernetes/pki/etcd

# scp /etc/kubernetes/admin.conf [email protected]:/etc/kubernetes/admin.conf

# scp /etc/kubernetes/pki/{ca.*,sa.*,front-proxy-ca.*} [email protected]:/etc/kubernetes/pki

# scp /etc/kubernetes/pki/etcd/ca.* [email protected]:/etc/kubernetes/pki/etcd

 

# ssh [email protected] mkdir -p /etc/kubernetes/pki/etcd

# scp /etc/kubernetes/admin.conf [email protected]:/etc/kubernetes/admin.conf

# scp /etc/kubernetes/pki/{ca.*,sa.*,front-proxy-ca.*} [email protected]:/etc/kubernetes/pki

# scp /etc/kubernetes/pki/etcd/ca.* [email protected]:/etc/kubernetes/pki/etcd

加入集群

在k8s-m2 k8s-m3 分别输入join命令：

kubeadm join k8s-vip:16443 --token cf1xoc.etmcuyk8d8ms138j \

    --discovery-token-ca-cert-hash sha256:1b177dfd0be798642fb643f5a7c23aa92e8f1f11fd5619c7dcf8e815c69ce2c1 \

    --control-plane --certificate-key 680e691bf851059d00ba25e96a4d2f68825965c56e15d2020b305a17fc6cc1b0