Kubernetes部署
MEOW
安装
curl -L git.io/meowproxy | bash
# 建议安装目录 /usr/bin/
复制自己的配置文件
# in local computer
xclip -sel clipboard < ~/.meow/rc
# in CentOS server
mv ~/.meow/rc ~/.meow/rc.bak
xclip -o -sel clipboard > ~/.meow/rc
运行
MEOW
Hostname
执行下面脚本。不要执行多次 如果出错则手动修改
sed -i s/localhost\.localdomain/NAME/g /etc/hosts /etc/hostname
hostname="10-10-103-144.node"
rm -rf /etc/hostname
echo $hostname > /etc/hostname
sed -i "s/\slocalhost\s/ localhost $hostname /g" /etc/hosts
sysctl kernel.hostname=$hostname
Screenfetch
curl -L https://raw.githubusercontent.com/KittyKatt/screenFetch/master/screenfetch-dev -o /usr/bin/screenfetch
AWK简述
awk -v var1=xxx BEGIN{[something]} [ifstatments or regex]{something match the condition}{[always run]}END{[something]}
# 支持if else,for,i++
# 支持数组（支持for in遍历，但是顺序会打乱）
# 最好的批量初始化的方式：
arr[""]=0 # 批量设为0
str="a b c"; split(str,arr[," "]);# ["a","b","c"]
加Kubernetes源
cat <<EOF > /etc/yum.repos.d/kubernetes.repo
[kubernetes]
name=Kubernetes
baseurl=http://yum.kubernetes.io/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=0
repo_gpgcheck=0
EOF
安装Kubernetes
yum install docker kubelet kubeadm kubectl kubernetes-cni
启动进程&关闭防火墙
systemctl start docker
systemctl start kubelet
systemctl enable docker
systemctl enable kubelet
systemctl stop firewalld
systemctl disable firewalld
Tear Down(当kubeadm init失败时执行)
kubeadm reset && systemctl start kubelet
Pull Docker Images
此方式十分艰难。后面可以通过代理直接安装。见Tech Log@1-12
#! /bin/bash
declare -a images=(
'dnsmasq-metrics-amd64:1.0'
'etcd-amd64:3.0.14-kubeadm'
'exechealthz-amd64:v1.2.0'
'kube-apiserver-amd64:v1.5.1'
'kube-controller-manager-amd64:v1.5.1'
'kube-discovery-amd64:1.0'
'kube-dnsmasq-amd64:1.4'
'kube-proxy-amd64:v1.5.1'
'kube-scheduler-amd64:v1.5.1'
'kubedns-amd64:1.9'
'kubernetes-dashboard-amd64:v1.5.0'
'pause-amd64:3.0'
# 'weave-kube:1.8.1'
# 'weave-npc:1.8.1'
)
for imageName in ${images[@]}; do
docker pull stkevintan/$imageName
docker tag stkevintan/$imageName gcr.io/google_containers/$imageName
docker rmi stkevintan/$imageName
done
xclip -o -sel clipboard > pull_docker
chmod +x pull_docker
./pull_docker
初始化（Master）
# 首先需要使用@17-1-3的脚本设置代理。
# Del environment
# Set curl
# Set kube env
# Set docker
# 在/etc/default/docker的DOCKER_OPTS中加入:--insecure-registry gcr.io
root@10-10-103-144 ~ $ kubeadm init
[kubeadm] WARNING: kubeadm is in alpha, please do not use it for production clusters.
[preflight] Running pre-flight checks
[init] Using Kubernetes version: v1.5.2
[tokens] Generated token: "c30875.954dc1dda4d0184b"
[certificates] Generated Certificate Authority key and certificate.
[certificates] Generated API Server key and certificate
[certificates] Generated Service Account signing keys
[certificates] Created keys and certificates in "/etc/kubernetes/pki"
[kubeconfig] Wrote KubeConfig file to disk: "/etc/kubernetes/kubelet.conf"
[kubeconfig] Wrote KubeConfig file to disk: "/etc/kubernetes/admin.conf"
[apiclient] Created API client, waiting for the control plane to become ready
[apiclient] All control plane components are healthy after 16.391731 seconds
[apiclient] Waiting for at least one node to register and become ready
[apiclient] First node is ready after 0.509209 seconds
[apiclient] Creating a test deployment
[apiclient] Test deployment succeeded
[token-discovery] Created the kube-discovery deployment, waiting for it to become ready
[token-discovery] kube-discovery is ready after 3.505676 seconds
[addons] Created essential addon: kube-proxy
[addons] Created essential addon: kube-dns
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=c30875.954dc1dda4d0184b 10.10.103.144
初始化（Node）
# 需要在子节点上安装kubelet kubectl kubernetes-cni kubeadm
# 需要使用@17-1-3的脚本设置代理。
# Del environment
# Set curl
# Set kube env
# Set docker
# 在/etc/default/docker的DOCKER_OPTS中加入:--insecure-registry gcr.io
systemctl enable kubelet
systemctl start kubelet
kubeadm join --token=ace10c.73fcdec44f7a4045 10.10.103.146
可以在master节点上看到该子节点：
root@master ~ $ kubectl get node
NAME STATUS AGE
master Ready,master 5d
node1 Ready 1m
Weave Network
# download config file
curl -L https://git.io/weave-kube -o weave-kube.yaml
# pull docker image
docker pull stkevintan/weave-kube:1.8.1
docker tag stkevintan/weave-kube:1.8.1 weaveworks/weave-kube:1.8.1
docker rmi stkevintan/weave-kube:1.8.1
# create
kubectl create -f weave-kube.yaml
# 或者
kubectl create -f https://git.io/weave-kube
可以看到结果：
root@master ~ $ kubectl get pod --namespace=kube-system
NAME READY STATUS RESTARTS AGE
dummy-2088944543-bv5gj 1/1 Running 0 5d
etcd-master 1/1 Running 0 5d
kube-apiserver-master 1/1 Running 27 5d
kube-controller-manager-master 1/1 Running 0 5d
kube-discovery-1769846148-xbm2h 1/1 Running 0 5d
kube-dns-2924299975-phn03 0/4 ContainerCreating 0 5d
kube-proxy-081gt 1/1 Running 0 5d
kube-proxy-qc40l 1/1 Running 0 1h
kube-scheduler-master 1/1 Running 1 5d
weave-net-4wvkm 0/2 ContainerCreating 0 30s
weave-net-l7dh0 0/2 ContainerCreating 0 30s
Dashboard
curl -L https://rawgit.com/kubernetes/dashboard/master/src/deploy/kubernetes-dashboard.yaml -o kubernetes-dashboard.yaml
sed -i 's/v1.5.1/v1.5.0/g' kubernetes-dashboard.yaml # 注意此时gcr上面并没有v1.5.1
sed -i 's/Alaways/IfNotPresent/g' kubernetes-dashboard.yaml
kubectl create -f kubernetes-dashboard.yaml
kubectl create -f https://rawgit.com/kubernetes/dashboard/master/src/deploy/kubernetes-dashboard.yaml
Delete Pod
kubectl delete -f xxx.yaml