目录
1.安装准备
2.安装过程
3.参考文档
一、安装准备
docker加速器
由于国内网络问题，强烈建议使用docker hub加速器，
阿里云请参考：https://yq.aliyun.com/articles/29941
daocloud请参考：https://www.daocloud.io/mirror#accelerator-doc
我在此记录为
https://abc.mirror.aliyuncs.com
操作系统
centos 7.2.1511上yum源配置(base epel)
新增k8s源
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
ntp同步配置(chrony.conf)
可选：dns配置，如忽略采用/etc/hosts方式
二、安装过程
主机名相关
| 主机名 | ip | 角色 |
|---|---|---|
| kcluster1 | 10.9.5.91 | master |
| kcluster2 | 10.9.5.90 | node |
| kcluster3 | 10.9.5.19 | node |
如没有用dns，需更新所有服务器的/etc/hosts
安装docker（所有服务器）
sudo yum install docker -y
============================================================================================================================================================================================================================================
Package Arch Version Repository Size
============================================================================================================================================================================================================================================
Installing:
docker x86_64 2:1.10.3-59.el7.centos extras 12 M
修改docker启动文件，增加加速器地址
sudo vim /usr/lib/systemd/system/docker.service
ExecStart=/usr/bin/docker-current daemon --registry-mirror=https://abc.mirror.aliyuncs.com \
--exec-opt native.cgroupdriver=systemd \
$OPTIONS \
$DOCKER_STORAGE_OPTIONS \
$DOCKER_NETWORK_OPTIONS \
$ADD_REGISTRY \
$BLOCK_REGISTRY \
$INSECURE_REGISTRY
启动docker
sudo systemctl daemon-reload && sudo systemctl start docker
sudo systemctl enable docker.service
安装k8s软件
sudo yum install socat kubelet kubeadm kubectl kubernetes-cni -y
sudo systemctl enable kubelet.service && sudo systemctl start kubelet.service
配置master(root用户下)
设置环境变量
export KUBE_REPO_PREFIX=registry.cn-hangzhou.aliyuncs.com/google-containers
KUBE_HYPERKUBE_IMAGE=registry.cn-hangzhou.aliyuncs.com/google-containers/hyperkube-amd64:v1.5.1
KUBE_DISCOVERY_IMAGE=registry.cn-hangzhou.aliyuncs.com/google-containers/kube-discovery-amd64:1.0
KUBE_ETCD_IMAGE=registry.cn-hangzhou.aliyuncs.com/google-containers/etcd-amd64:3.0.4
执行配置
kubeadm init --pod-network-cidr="192.168.0.0/16"
执行需要一段时间
最后出现类似如下
Your Kubernetes master has initialized successfully!
You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
http://kubernetes.io/docs/admin/addons/
You can now join any number of machines by running the following on each node:
kubeadm join --token=xxxxxxxxxxxxxxx 10.9.5.91
配置node(root用户下)
kubeadm join --token=xxxxxxxxxxxxxx 10.9.5.91
下面是执行命令前后的变化
[root@cloud4ourself-kcluster1 ~]# kubectl get nodes
NAME STATUS AGE
cloud4ourself-kcluster1.novalocal Ready,master 16m
[root@cloud4ourself-kcluster1 ~]# kubectl get nodes
NAME STATUS AGE
cloud4ourself-kcluster1.novalocal Ready,master 16m
cloud4ourself-kcluster2.novalocal Ready 3s
配置网络（master节点root用户下）
wget https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml
修改网段(也可不修改)和image
由
"Network": "10.244.0.0/16",
改为
"Network": "192.168.0.0/16",
两个image
由
image: quay.io/coreos/flannel:v0.7.0-amd64
改为
image: docker.io/4admin2root/flannel:v0.7.0-amd64
执行kubectl apply -f kube-flannel.yml
此处也可以改为其他网络模式，请参考https://kubernetes.io/docs/admin/addons/
检查pod状态
kubectl --namespace=kube-system get po -o wide
直到所有都为 Running
配置dashboard(master节点root用户下)
kubectl apply -f http://k8s.oss-cn-shanghai.aliyuncs.com/kube/kubernetes-dashboard1.5.0.yaml
kubectl get svc --namespace=kube-system
取得访问地址
如http://kcluster1:31810/#/workload?namespace=_all
验证
[root@cloud4ourself-kcluster1 ~]# kubectl create -f https://raw.githubusercontent.com/4admin2root/daocloud/master/my-calc.yaml
replicationcontroller "my-calc-rc" created
service "my-calc-service" created
[root@cloud4ourself-kcluster1 ~]# kubectl create -f https://raw.githubusercontent.com/4admin2root/daocloud/master/my-frontend.yaml
replicationcontroller "my-frontend-rc" created
service "my-frontend-service" created
[root@cloud4ourself-kcluster1 ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
kube-flannel-ds-7f69s 2/2 Running 6 40m
kube-flannel-ds-c7d00 2/2 Running 6 40m
kube-flannel-ds-p3lww 2/2 Running 0 40m
my-calc-rc-6fzh6 0/1 ContainerCreating 0 1m
my-calc-rc-pvhx0 0/1 ContainerCreating 0 1m
my-frontend-rc-jbt2p 0/1 ContainerCreating 0 58s
my-frontend-rc-m2svn 0/1 ContainerCreating 0 58s
[root@cloud4ourself-kcluster1 ~]# kubectl get pod
NAME READY STATUS RESTARTS AGE
kube-flannel-ds-7f69s 2/2 Running 6 44m
kube-flannel-ds-c7d00 2/2 Running 6 44m
kube-flannel-ds-p3lww 2/2 Running 0 44m
my-calc-rc-6fzh6 1/1 Running 0 4m
my-calc-rc-pvhx0 1/1 Running 0 4m
my-frontend-rc-jbt2p 1/1 Running 0 4m
my-frontend-rc-m2svn 1/1 Running 0 4m
[root@cloud4ourself-kcluster1 ~]# kubectl get pod -o wide
NAME READY STATUS RESTARTS AGE IP NODE
kube-flannel-ds-7f69s 2/2 Running 6 45m 10.9.5.90 cloud4ourself-kcluster2.novalocal
kube-flannel-ds-c7d00 2/2 Running 6 45m 10.9.5.19 cloud4ourself-kcluster3.novalocal
kube-flannel-ds-p3lww 2/2 Running 0 45m 10.9.5.91 cloud4ourself-kcluster1.novalocal
my-calc-rc-6fzh6 1/1 Running 0 6m 192.168.2.2 cloud4ourself-kcluster3.novalocal
my-calc-rc-pvhx0 1/1 Running 0 6m 192.168.1.3 cloud4ourself-kcluster2.novalocal
my-frontend-rc-jbt2p 1/1 Running 0 5m 192.168.1.4 cloud4ourself-kcluster2.novalocal
my-frontend-rc-m2svn 1/1 Running 0 5m 192.168.2.3 cloud4ourself-kcluster3.novalocal
[root@cloud4ourself-kcluster1 ~]# kubectl describe svc my-frontend-service
Name: my-frontend-service
Namespace: default
Labels:
Selector: app=my-frontend
Type: NodePort
IP: 10.98.166.14
Port: 5000/TCP
NodePort: 30080/TCP
Endpoints: 192.168.1.4:5000,192.168.2.3:5000
Session Affinity: None
No events.
[root@cloud4ourself-kcluster1 ~]# kubectl describe svc my-calc-service
Name: my-calc-service
Namespace: default
Labels:
Selector: app=my-calc
Type: ClusterIP
IP: 10.105.114.170
Port: 5000/TCP
Endpoints: 192.168.1.3:5000,192.168.2.2:5000
Session Affinity: None
No events.
三、参考文档
https://kubernetes.io/docs/admin/addons/
https://yq.aliyun.com/articles/66474?commentId=6660