0%

详解K8s安装及陷阱

kubernetes-install

操作系统初始化

  • 关闭防火墙(all)
1
2
3
4
5
6
# 临时关闭防火墙
systemctl stop firewalld
# 永久关闭防火墙
systemctl disable firewalld
# 验证
systemctl status firewalld
  • 关闭selinux(all)
1
2
3
4
# 临时关闭
setenforce 0
# 永久
sed -i 's/enforcing/disabled/' /etc/selinux/config
  • 关闭swap(all)
1
2
3
4
# 临时
swapoff -a
# 永久
sed -ri 's/.*swap.*/#&/' /etc/fstab
  • 设置主机名称(all)
1
2
3
4
# 设置名称(k8s-m-1)忽略大写字母
hostnamectl set-hostname k8s-m-1
# 验证
hostname
  • Master添加Hostname(master)
1
2
3
4
5
6
7
8
9
10
11
12
# 设置
cat >> /etc/hosts << EOF
masterIp master
node1Ip node1
node2Ip node2
EOF
# eg
cat >> /etc/hosts << EOF
192.168.50.212 k8s-m-1
192.168.50.87 k8s-n-1
192.168.50.85 k8s-n-2
EOF
  • 将桥接的IPV4 流量传递到iptables的链(all)
1
2
3
4
5
6
cat > /etc/sysctl.d/k8s.conf << EOF
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-ip6tables = 1
EOF
# 生效
sysctl --system
  • 时间同步(All)
1
2
3
yum install -y ntpdate 
ntpdate time.windows.com
# 三台机子输出如下则成功(相差几秒或几分为正常现象)

安装Docker

官方文档-安装

  • Docker安装sh Script:(All)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
# You can use scripts for one click installation,You may need to type enter at the end
# remove docker
sudo yum remove docker \
docker-client \
docker-client-latest \
docker-common \
docker-latest \
docker-latest-logrotate \
docker-logrotate \
docker-engine
# Set up repository
sudo yum install -y yum-utils
# Use Aliyun Docker
sudo yum-config-manager --add-repo http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
# install docker from yum
yum install -y docker-ce docker-ce-cli containerd.io
# restart docker
systemctl restart docker
# cat version
docker --version

  • 配置加速(all)
j
1
2
3
4
5
6
7
8
9
10
11
12
sudo mkdir -p /etc/docker
sudo tee /etc/docker/daemon.json <<-'EOF'
{
"registry-mirrors": ["https://etdea28s.mirror.aliyuncs.com"]
}
EOF

# reload
sudo systemctl daemon-reload
sudo systemctl restart docker

# 检查阿里云加速

kubernetes安装

  • 配置kubernetes源(all)
1
2
3
4
5
6
7
8
9
cat <<EOF > /etc/yum.repos.d/kubernetes.repo
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64/
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
EOF

由于官网kubernetes源在国外有墙,直接使用官方源会导致安装失败。所以我们配置国内的阿里源

  • 安装 kubectl kubelet kubeadm(all)
1
2
3
4
# install kubectl kubelet kubeadm
yum install -y kubectl kubelet kubeadm
# set boot on opening computer
systemctl enable kubelet
  • 初始化k8s部署(Master)
1
2
3
4
5
6
7
8
9
10
11
12
13
kubeadm init \
--apiserver-advertise-address=youselfIp of Master \
--image-repository registry.aliyuncs.com/google_containers \
# 不冲突即可
--service-cidr=10.10.0.0/16 \
--pod-network-cidr=10.122.0.0/16

# eg
kubeadm init \
--apiserver-advertise-address=192.168.50.212 \
--image-repository registry.aliyuncs.com/google_containers \
--service-cidr=10.10.0.0/16 \
--pod-network-cidr=10.122.0.0/16

常见错误:running with swap on is not supported. Please disable swap

[preflight] If you know what you are doing, you can make a check non-fatal with `—ignore-preflight-

errors=…`

原因:系统自动进行分区

解决:

1
2
3
4
# 临时
swapoff -a
# 永久
sed -ri 's/.*swap.*/#&/' /etc/fstab
  • following as a regular user(Master)
1
2
3
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config
  • join master node(node)
1
kubeadm join 172.16.164.136:6443 --token 9oilao.bpbxcm5zkk0jjcgm --discovery-token-ca-cert-hash sha256:609794bd03915be382bdb130c4c180e89cdc863d35cf99be79cf4ddcbfacee24

加入成功,如下图

此时我们在Master节点上使用命令kubectl get nodes查看节点信息:如下图所示

此时的kubectl get nodes的status都是NotNotReady:

查看kubernetes运行状态:

kubectl get pods -n kube-system

如图:

果然,两个Pending犹豫未决

此时我们部署CNI网络,配置如下

1
2
3
# 根据官方文档提示配置CNI网络
kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml
# 报错:The connection to the server raw.githubusercontent.com was refused - did you specify the right host or port? 原因:外网不可访问 -> 在https://www.ipaddress.com/查询raw.githubusercontent.com的真实IP。
1
2
3
sudo vi /etc/hosts
199.232.28.133 raw.githubusercontent.com
# 如下

1
2
3
4
5
# 开启IPVS,修改ConfigMap的kube-system/kube-proxy中的模式为ipvs

kubectl edit cm kube-proxy -n kube-system
# 将空的data -> ipvs -> mode中替换如下
mode: "ipvs"

在此运行kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml成功,如图

此时运行kubectl get nodes效果图如下->成功。(肯能并不一定会立马成功,上面确定没问题,请稍等片刻即可)

测试kubernetes

1
2
3
4
5
6
7
# 创建nginx镜像 Create a deployment with the specified name
# kubectl create deployment NAME --image=image -- [COMMAND] [args...] [options]
kubectl create deployment nginx --image=nginx
# 对外暴露端口
kubectl expose deployment nginx --port=80 --type=NodePort
# 查看pod服务
kubectl get pod,svc

成功