k8s集群搭建（centos）

时间 2019-11-19

原文原文链接

0.准备： centos ip为 192.168.26.129（master） centos ip为 192.168.26.131（node1）两台机器都操做node

/etc/hosts添加
192.168.26.129 master
192.168.26.129 etcd
192.168.26.131 node1

1.中止centos的firewalld，关闭linuxlinux

systemctl disable firewalld
systemctl stop firewalld

编辑/etc/selinux/config，将SELINUX=enforcing改成SELINUX=disabled （为的是重启后也不启动selinux）本次也不启动，先临时关闭docker

setenforce 0

2.安装 etcd,kubernetesvim

yum update
yum install -y etcd kubernetes

3.修改docker配置文件 vim /etc/sysconfig/dockercentos

OPTIONS='--selinux-enabled=false --insecure-registry gcr.io'

master节点api

4.修改etcd配置文件 vim /etc/etcd/etcd.confide

ETCD_LISTEN_CLIENT_URLS="http://0.0.0.0:2379"
ETCD_ADVERTISE_CLIENT_URLS="http://etcd:2379"

5.修改 apiserver 的配置文件 vim /etc/kubernetes/apiserver，绑定etcdspa

KUBE_API_ADDRESS="--insecure-bind-address=0.0.0.0" ## 必须
KUBE_ETCD_SERVERS="--etcd-servers=http://etcd:2379" ## 必须
KUBE_ADMISSION_CONTROL="--admission-control=NamespaceLifecycle,NamespaceExists,LimitRanger,ResourceQuota"

依次启动服务code

systemctl start etcd
systemctl start docker(不是必须)
systemctl start kube-apiserver
systemctl start kube-controller-manager
systemctl start kube-scheduler

###### 若是master节点也是工做节点的话就启动kubelet，不然不须要
systemctl start kubelet
systemctl start kube-proxy

192.168.26.131的node节点配置 vim /etc/kubernetes/kubeletserver

KUBELET_ADDRESS="--address=0.0.0.0"
KUBELET_HOSTNAME="--hostname-override=node1"
KUBELET_API_SERVER="--api-servers=http://master:8080"

节点启动

systemctl start docker
systemctl start kubelet
systemctl start kube-proxy

此时在 master 节点上执行 kubectl get node

看到node1，ready表示成功