在centos 7中安装配置k8s集群的步骤详解

配置背景介绍

kubernetes是google开源的容器集群管理系统，提供应用部署、维护、扩展机制等功能，
利用kubernetes能方便管理跨集群运行容器化的应用，简称：k8s（k与s之间有8个字母）

为什么要用kubernetes这么复杂的docker集群管理工具呢？一开始接触了docker内置的swarm，
这个工具非常简单快捷的完成docker集群功能。但是在使用docker1.13内置的swarm做集群的时候
遇到vip负载均衡没有正确映射端口到外网，或者出现地址被占用的情况，这对高可用性的需求是不利的，
然而又没找到一个解决方案，只能转投k8s。

实验环境

centos7.3 64位

安装

1. yum-config-manager --add-repo https://docs.docker.com/v1.13/engine/inst
   
2. allation/linux/repo_files/centos/docker.repo
   
3. yum makecache fast
   
4. yum -y install docker-engine-1.13.1
   
5. yum install epel-release -y
   
6. yum remove -y docker-engine*
   
7. yum install -y kubernetes etcd docker flannel
   

复制代码

修改配置文件

注意下面的10.135.163.237换成自己服务器ip

1. sed -i "s/localhost:2379/10.135.163.237:2379/g" /etc/etcd/etcd.conf
   
2. sed -i "s/localhost:2380/10.135.163.237:2380/g" /etc/etcd/etcd.conf
   
3. sed -i "s/10.135.163.237:2379/10.135.163.237:2379,http:\/\/127.0.0.1:2379/g" /etc/etcd/etcd.conf
   
4. sed -i "s/127.0.0.1:2379/10.135.163.237:2379/g" /etc/kubernetes/apiserver
   
5. sed -i "s/--insecure-bind-address=127.0.0.1/--insecure-bind-address=0.0.0.0/g" /etc/kubernetes/apiserver
   
6. sed -i "s/--admission-control=NamespaceLifecycle,NamespaceExists,LimitRanger,SecurityContextDeny,Servi
   
7. ceAccount,ResourceQuota/--admission_control=NamespaceLifecycle,NamespaceExists,LimitRanger,ResourceQ
   
8. uota/g" /etc/kubernetes/apiserver
   
9. sed -i "s/--hostname-override=127.0.0.1/--hostname-override=10.135.163.237/g" /etc/kubernetes/kubelet
   
10. sed -i "s/127.0.0.1:8080/10.135.163.237:8080/g" /etc/kubernetes/kubelet
    
11. sed -i "s/--address=127.0.0.1/--address=0.0.0.0/g" /etc/kubernetes/kubelet
    
12. sed -i "s/127.0.0.1:8080/10.135.163.237:8080/g" /etc/kubernetes/config
    
13. sed -i "s/127.0.0.1:2379/10.135.163.237:2379/g" /etc/sysconfig/flanneld
    

复制代码

修改host

1. vi /etc/hosts
   
2. 10.135.163.237 k8s_master

复制代码

添加网络

1. systemctl enable etcd.service
   
2. systemctl start etcd.service
   
3. etcdctl mk //atomic.io/network/config '{"Network":"172.17.0.0/16"}'

复制代码

启动服务

1. service docker start
   
2. for SERVICES in etcd kube-apiserver kube-controller-manager kube-sc
   
3. heduler kube-proxy kubelet docker flanneld ; do systemctl restart $SERVICES;
   
4. systemctl enable $SERVICES; systemctl status $SERVICES; done;
   

复制代码

第一个demo

编写文件a.yaml

1. apiVersion: extensions/v1beta1
   
2. kind: Deployment
   
3. metadata:
   
4. name: my-app
   
5. spec:
   
6. replicas: 2
   
7. template:
   
8. metadata:
   
9. labels:
   
10. app: my-app
    
11. spec:
    
12. containers:
    
13. - name: my-app
    
14. image: registry.alauda.cn/yubang/paas_base_test
    
15. ports:
    
16. - containerPort: 80
    
17. command: ["/bin/bash", "/var/start.sh"]
    
18. resources:
    
19.   limits:
    
20.   cpu: 0.5
    
21.   memory: 64Mi

复制代码

编写文件b.yaml

1. codecodeapiVersion: v1
   
2. kind: Service
   
3. metadata:
   
4. name: my-app-svc
   
5. labels:
   
6. app: my-app
   
7. spec:
   
8. ports:
   
9. - port: 80
   
10. targetPort: 80
    
11. nodePort: 30964
    
12. type: NodePort
    
13. selector:
    
14. app: my-app
    
15. /code/code

复制代码

创建服务

1. kubectl create -f a.yaml --validate
   
2. kubectl create -f b.yaml --validate

复制代码

删除服务

1. kubectl delete -f a.yaml
   
2. kubectl delete -f b.yaml

复制代码

增加子节点

安装软件

1. yum-config-manager --add-repo https://docs.docker.com/v1.13/en
   
2. gine/installation/linux/repo_files/centos/docker.repo
   
3. yum makecache fast
   
4. yum -y install docker-engine-1.13.1
   
5. yum install epel-release -y
   
6. yum remove -y docker-engine*
   
7. yum install -y kubernetes docker flannel

复制代码

修改配置文件（10.135.163.237为主节点ip，139.199.0.29为当前节点ip）

1. sed -i "s/--hostname-override=127.0.0.1/--hostname-override=139.199.0.29/g" /etc/kubernetes/kubelet
   
2. sed -i "s/127.0.0.1:8080/10.135.163.237:8080/g" /etc/kubernetes/kubelet
   
3. sed -i "s/--address=127.0.0.1/--address=0.0.0.0/g" /etc/kubernetes/kubelet
   
4. sed -i "s/127.0.0.1:8080/10.135.163.237:8080/g" /etc/kubernetes/config
   
5. sed -i "s/127.0.0.1:2379/10.135.163.237:2379/g" /etc/sysconfig/flanneld
   
6. sed -i "s/--admission-control=NamespaceLifecycle,NamespaceExists,LimitRanger,Security
   
7. ContextDeny,ServiceAccount,ResourceQuota/--admission_control=NamespaceLifecycle,Na
   
8. mespaceExists,LimitRanger,ResourceQuota/g" /etc/kubernetes/apiserver
   

复制代码

启动服务

1. service docker start
   
2. 
   
3. for SERVICES in kube-proxy kubelet docker flanneld; do
   
4.   systemctl restart $SERVICES
   
5.   systemctl enable $SERVICES
   
6.   systemctl status $SERVICES
   
7. done;
   

复制代码

在主服务器查看节点

1. kubectl get node

复制代码

重启服务（重新加入集群）

1. systemctl restart kube-apiserver.service
   

复制代码

删除节点

1. kubectl delete node 节点ip
   

复制代码