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
- 腾讯云
- centos7.3 64位
安装
?
|
1
2
3
4
5
6
|
yum-config-manager --add-repo https://docs.docker.com/v1.13/engine/installation/linux/repo_files/centos/docker.repo
yum makecache fast
yum -y installdocker-engine-1.13.1
yum installepel-release -y
yum remove -y docker-engine*
yum install-y kubernetes etcd docker flannel
|
修改配置文件
注意下面的10.135.163.237换成自己服务器ip
?
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
|
sed-i "s/localhost:2379/10.135.163.237:2379/g"/etc/etcd/etcd.conf
sed-i "s/localhost:2380/10.135.163.237:2380/g"/etc/etcd/etcd.conf
sed-i "s/10.135.163.237:2379/10.135.163.237:2379,http:\/\/127.0.0.1:2379/g"/etc/etcd/etcd.conf
sed-i "s/127.0.0.1:2379/10.135.163.237:2379/g"/etc/kubernetes/apiserver
sed-i "s/--insecure-bind-address=127.0.0.1/--insecure-bind-address=0.0.0.0/g"/etc/kubernetes/apiserver
sed-i "s/--admission-control=NamespaceLifecycle,NamespaceExists,LimitRanger,SecurityContextDeny,ServiceAccount,ResourceQuota/--admission_control=NamespaceLifecycle,NamespaceExists,LimitRanger,ResourceQuota/g"/etc/kubernetes/apiserver
sed-i "s/--hostname-override=127.0.0.1/--hostname-override=10.135.163.237/g"/etc/kubernetes/kubelet
sed-i "s/127.0.0.1:8080/10.135.163.237:8080/g"/etc/kubernetes/kubelet
sed-i "s/--address=127.0.0.1/--address=0.0.0.0/g"/etc/kubernetes/kubelet
sed-i "s/127.0.0.1:8080/10.135.163.237:8080/g"/etc/kubernetes/config
sed-i "s/127.0.0.1:2379/10.135.163.237:2379/g"/etc/sysconfig/flanneld
|
修改host
?
|
1
2
|
vi/etc/hosts
10.135.163.237 k8s_master
|
添加网络
?
|
1
2
3
4
5
|
systemctl enableetcd.service
systemctl start etcd.service
etcdctl mk //atomic.io/network/config'{"Network":"172.17.0.0/16"}'
|
启动服务
?
|
1
2
|
service docker start
forSERVICES inetcd kube-apiserver kube-controller-manager kube-scheduler kube-proxy kubelet docker flanneld ; dosystemctl restart $SERVICES; systemctl enable$SERVICES; systemctl status $SERVICES; done;
|
第一个demo
编写文件a.yaml
?
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
|
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: my-app
spec:
replicas: 2
template:
metadata:
labels:
app: my-app
spec:
containers:
- name: my-app
image: registry.alauda.cn/yubang/paas_base_test
ports:
- containerPort: 80
command: ["/bin/bash", "/var/start.sh"]
resources:
limits:
cpu: 0.5
memory: 64Mi
|
编写文件b.yaml
?
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
|
apiVersion: v1
kind: Service
metadata:
name: my-app-svc
labels:
app: my-app
spec:
ports:
- port: 80
targetPort: 80
nodePort: 30964
type: NodePort
selector:
app: my-app
|
创建服务
?
|
1
2
|
kubectl create -f a.yaml --validate
kubectl create -f b.yaml --validate
|
删除服务
?
|
1
2
|
kubectl delete -f a.yaml
kubectl delete -f b.yaml
|
增加子节点
安装软件
?
|
1
2
3
4
5
6
|
yum-config-manager --add-repo https://docs.docker.com/v1.13/engine/installation/linux/repo_files/centos/docker.repo
yum makecache fast
yum -y installdocker-engine-1.13.1
yum installepel-release -y
yum remove -y docker-engine*
yum install-y kubernetes docker flannel
|
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
2
3
4
5
6
7
8
9
10
11
12
13
|
sed-i "s/--hostname-override=127.0.0.1/--hostname-override=139.199.0.29/g"/etc/kubernetes/kubelet
sed-i "s/127.0.0.1:8080/10.135.163.237:8080/g"/etc/kubernetes/kubelet
sed-i "s/--address=127.0.0.1/--address=0.0.0.0/g"/etc/kubernetes/kubelet
sed-i "s/127.0.0.1:8080/10.135.163.237:8080/g"/etc/kubernetes/config
sed-i "s/127.0.0.1:2379/10.135.163.237:2379/g"/etc/sysconfig/flanneld
sed-i "s/--admission-control=NamespaceLifecycle,NamespaceExists,LimitRanger,SecurityContextDeny,ServiceAccount,ResourceQuota/--admission_control=NamespaceLifecycle,NamespaceExists,LimitRanger,ResourceQuota/g"/etc/kubernetes/apiserver
|
启动服务
?
|
1
2
3
4
5
6
7
|
service docker start
forSERVICES inkube-proxy kubelet docker flanneld; do
systemctl restart $SERVICES
systemctl enable$SERVICES
systemctl status $SERVICES
done;
|
在主服务器查看节点
?
|
1
|
kubectl get node
|
重启服务（重新加入集群）
?
|
1
|
systemctl restart kube-apiserver.service
|
删除节点
?
|
1
|
kubectl delete node 节点ip
|
总结
