v1.16.1 calico3.9
This commit is contained in:
huanqing.shao
507
install/history-k8s/install-kubernetes-1.15.3.md
Normal file
507
install/history-k8s/install-kubernetes-1.15.3.md
Normal file
@ -0,0 +1,507 @@
|
||||
---
|
||||
vssueId: 16
|
||||
description: Kubernete安装文档_使用kubeadm安装高可用的Kubernetes_v1.15.3集群_可用于生产环境
|
||||
meta:
|
||||
- name: keywords
|
||||
content: Kubernetes集群,Kubernetes高可用,Kubernetes生产环境
|
||||
# storyBook:
|
||||
# title: '使用 kubeadm 安装 kubernetes v1.15.3（高可用）'
|
||||
# initial: FullPage
|
||||
# pages:
|
||||
# - name: overview
|
||||
# title: 配置要求
|
||||
# - name: step1
|
||||
# title: 检查环境
|
||||
# - name: step2
|
||||
# title: 安装 docker/kubelet
|
||||
# - name: step3
|
||||
# title: 初始化 apiserver 集群
|
||||
# - name: step4
|
||||
# title: 初始化 worker 节点
|
||||
# - name: step5
|
||||
# title: 安装 Ingress Controller
|
||||
# - name: step6
|
||||
# title: 总结
|
||||
---
|
||||
|
||||
# 安装Kubernetes高可用
|
||||
|
||||
<AdSenseTitle/>
|
||||
|
||||
<!-- <StoryBook>
|
||||
|
||||
<div slot="overview"> -->
|
||||
|
||||
::: tip
|
||||
推荐初学者按照 [安装Kubernetes 单Master节点](install-k8s.html) 文档进行 Kubernetes 集群搭建
|
||||
:::
|
||||
|
||||
## 介绍
|
||||
|
||||
kubernetes 安装有多种选择，本文档描述的集群安装具备如下特点：
|
||||
|
||||
* Kubernetes 1.15.3
|
||||
* calico 3.8.2
|
||||
* nginx-ingress 1.5.3
|
||||
* Docker 18.09.7
|
||||
* 三个 master 组成主节点集群，通过内网 loader balancer 实现负载均衡；至少需要三个 master 节点才可组成高可用集群，否则会出现 ***脑裂*** 现象
|
||||
* 多个 worker 组成工作节点集群，通过外网 loader balancer 实现负载均衡
|
||||
|
||||
安装后的拓扑图如下：<a :href="$withBase('/kuboard.rp')" download="www.kuboard.cn.rp">下载拓扑图源文件</a> <font color="#999">使用Axure RP 9.0可打开该文件</font>
|
||||
|
||||
|
||||
|
||||
* **在线答疑**
|
||||
|
||||
<Qq></Qq> 也可以扫描二维码加群
|
||||
<p>
|
||||
<img src="/images/kuboard_qq.png" alt="Kubernetes教程：QQ群在线答疑"/>
|
||||
</p>
|
||||
|
||||
<!-- <div>
|
||||
<div style="margin-top: 10px;">未打赏用户可进 QQ 群聊，<span style="color: red;">打赏用户可进微信群聊</span>。</div>
|
||||
<div style="margin-top: 10px;">
|
||||
<span>扫第一个二维码完成打赏，扫第二个进微信群聊。</span> <span style="color: #CCC">QQ 群聊二维码在左侧导航栏下方。</span>
|
||||
<p style="margin-top: 10px;">
|
||||
<img src="/images/dz.png" style="width: 150px; margin-right: 150px;"></img>
|
||||
<img src="/images/dz2.jpeg" style="width: 150px;"></img>
|
||||
</p>
|
||||
</div>
|
||||
</div> -->
|
||||
|
||||
<!-- </div>
|
||||
|
||||
<div slot="step1"> -->
|
||||
|
||||
## 检查 centos / hostname
|
||||
|
||||
``` sh
|
||||
# 在 master 节点和 worker 节点都要执行
|
||||
cat /etc/redhat-release
|
||||
|
||||
# 此处 hostname 的输出将会是该机器在 Kubernetes 集群中的节点名字
|
||||
hostname
|
||||
```
|
||||
|
||||
**操作系统兼容性**
|
||||
|
||||
| CentOS 版本 | 本文档是否兼容 | 备注 |
|
||||
| ----------- | --------------------------------------- | ----------------------------------- |
|
||||
| 7.7 | <span style="font-size: 24px;">😄</span> | 已验证 |
|
||||
| 7.6 | <span style="font-size: 24px;">😄</span> | 已验证 |
|
||||
| 7.5 | <span style="font-size: 24px;">😄</span> | 已验证 |
|
||||
| 7.4 | <span style="font-size: 24px;">🤔</span> | 待验证 |
|
||||
| 7.3 | <span style="font-size: 24px;">🤔</span> | 待验证 |
|
||||
| 7.2 | <span style="font-size: 24px;">😞</span> | 已证实会出现 kubelet 无法启动的问题 |
|
||||
|
||||
::: tip 修改 hostname
|
||||
如果您需要修改 hostname，可执行如下指令：
|
||||
``` sh
|
||||
# 修改 hostname
|
||||
hostnamectl set-hostname your-new-host-name
|
||||
# 查看修改结果
|
||||
hostnamectl status
|
||||
```
|
||||
:::
|
||||
|
||||
<!-- </div>
|
||||
|
||||
<div slot="step2"> -->
|
||||
|
||||
## 安装 docker / kubelet
|
||||
|
||||
使用 root 身份在所有节点执行如下代码，以安装软件：
|
||||
- docker
|
||||
- nfs-utils
|
||||
- kubectl / kubeadm / kubelet
|
||||
|
||||
<el-tabs type="border-card">
|
||||
<el-tab-pane label="快速安装">
|
||||
|
||||
``` sh
|
||||
# 在 master 节点和 worker 节点都要执行
|
||||
|
||||
curl -sSL https://kuboard.cn/install-script/v1.15.3/install_kubelet.sh | sh
|
||||
|
||||
```
|
||||
|
||||
</el-tab-pane>
|
||||
<el-tab-pane label="手动安装">
|
||||
|
||||
手动执行以下代码，效果与快速安装完全相同。
|
||||
|
||||
<<< @/.vuepress/public/install-script/v1.15.3/install_kubelet.sh
|
||||
|
||||
::: warning
|
||||
如果此时执行 `service status kubelet` 命令，将得到 kubelet 启动失败的错误提示，请忽略此错误，因为必须完成后续步骤中 kubeadm init 的操作，kubelet 才能正常启动
|
||||
:::
|
||||
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
|
||||
<!-- </div>
|
||||
|
||||
<div slot="step3"> -->
|
||||
|
||||
## 初始化API Server
|
||||
|
||||
### 创建 ApiServer 的 Load Balancer（私网）
|
||||
|
||||
监听端口：6443 / TCP
|
||||
|
||||
后端资源组：包含 demo-master-a-1, demo-master-b-1, demo-master-b-2
|
||||
|
||||
后端端口：6443
|
||||
|
||||
开启 按源地址保持会话
|
||||
|
||||
假设完成创建以后，Load Balancer的 ip 地址为 x.x.x.x
|
||||
|
||||
### 初始化第一个master节点
|
||||
|
||||
|
||||
::: tip
|
||||
* 以 root 身份在 demo-master-a-1 机器上执行
|
||||
* 初始化 master 节点时，如果因为中间某些步骤的配置出错，想要重新初始化 master 节点，请先执行 `kubeadm reset` 操作
|
||||
:::
|
||||
|
||||
::: danger 关于初始化时用到的环境变量
|
||||
* **APISERVER_NAME** 不能是 master 的 hostname
|
||||
* **APISERVER_NAME** 必须全为小写字母、数字、小数点，不能包含减号
|
||||
* **POD_SUBNET** 所使用的网段不能与 ***master节点/worker节点*** 所在的网段重叠。该字段的取值为一个 <a href="/glossary/cidr.html" target="_blank">CIDR</a> 值，如果您对 CIDR 这个概念还不熟悉，请不要修改这个字段的取值 10.100.0.1/16
|
||||
:::
|
||||
|
||||
<el-tabs type="border-card">
|
||||
<el-tab-pane label="快速初始化">
|
||||
|
||||
在第一个 master 节点 demo-master-a-1 上执行
|
||||
|
||||
``` sh
|
||||
# 只在 master 节点执行
|
||||
# 替换 apiserver.demo 为 您想要的 dnsName
|
||||
export APISERVER_NAME=apiserver.demo
|
||||
# Kubernetes 容器组所在的网段，该网段安装完成后，由 kubernetes 创建，事先并不存在于您的物理网络中
|
||||
export POD_SUBNET=10.100.0.1/16
|
||||
echo "127.0.0.1 ${APISERVER_NAME}" >> /etc/hosts
|
||||
curl -sSL https://kuboard.cn/install-script/v1.15.3/init_master.sh | sh
|
||||
```
|
||||
|
||||
</el-tab-pane>
|
||||
<el-tab-pane label="手工初始化">
|
||||
|
||||
``` sh
|
||||
# 只在 master 节点执行
|
||||
# 替换 apiserver.demo 为 您想要的 dnsName
|
||||
export APISERVER_NAME=apiserver.demo
|
||||
# Kubernetes 容器组所在的网段，该网段安装完成后，由 kubernetes 创建，事先并不存在于您的物理网络中
|
||||
export POD_SUBNET=10.100.0.1/16
|
||||
echo "127.0.0.1 ${APISERVER_NAME}" >> /etc/hosts
|
||||
```
|
||||
|
||||
<<< @/.vuepress/public/install-script/v1.15.3/init_master.sh
|
||||
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
|
||||
***执行结果***
|
||||
|
||||
执行结果中：
|
||||
* 第15、16、17行，用于初始化第二、三个 master 节点
|
||||
* 第25、26行，用于初始化 worker 节点
|
||||
|
||||
``` sh {15,16,17,25,26}
|
||||
Your Kubernetes control-plane has initialized successfully!
|
||||
|
||||
To start using your cluster, you need to run the following as a regular user:
|
||||
|
||||
mkdir -p $HOME/.kube
|
||||
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
|
||||
sudo chown $(id -u):$(id -g) $HOME/.kube/config
|
||||
|
||||
You should now deploy a pod network to the cluster.
|
||||
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
|
||||
https://kubernetes.io/docs/concepts/cluster-administration/addons/
|
||||
|
||||
You can now join any number of the control-plane node running the following command on each as root:
|
||||
|
||||
kubeadm join apiserver.k8s:6443 --token 4z3r2v.2p43g28ons3b475v \
|
||||
--discovery-token-ca-cert-hash sha256:959569cbaaf0cf3fad744f8bd8b798ea9e11eb1e568c15825355879cf4cdc5d6 \
|
||||
--control-plane --certificate-key 41a741533a038a936759aff43b5680f0e8c41375614a873ea49fde8944614dd6
|
||||
|
||||
Please note that the certificate-key gives access to cluster sensitive data, keep it secret!
|
||||
As a safeguard, uploaded-certs will be deleted in two hours; If necessary, you can use
|
||||
"kubeadm init phase upload-certs --upload-certs" to reload certs afterward.
|
||||
|
||||
Then you can join any number of worker nodes by running the following on each as root:
|
||||
|
||||
kubeadm join apiserver.k8s:6443 --token 4z3r2v.2p43g28ons3b475v \
|
||||
--discovery-token-ca-cert-hash sha256:959569cbaaf0cf3fad744f8bd8b798ea9e11eb1e568c15825355879cf4cdc5d6
|
||||
|
||||
```
|
||||
|
||||
**检查 master 初始化结果**
|
||||
|
||||
``` sh
|
||||
# 只在第一个 master 节点执行
|
||||
|
||||
# 执行如下命令，等待 3-10 分钟，直到所有的容器组处于 Running 状态
|
||||
watch kubectl get pod -n kube-system -o wide
|
||||
|
||||
# 查看 master 节点初始化结果
|
||||
kubectl get nodes
|
||||
```
|
||||
|
||||
### 初始化第二、三个master节点
|
||||
|
||||
**获得 master 节点的 join 命令**
|
||||
|
||||
|
||||
<el-tabs type="border-card">
|
||||
<el-tab-pane label="和第一个Master节点一起初始化">
|
||||
|
||||
初始化第一个 master 节点时的输出内容中，第15、16、17行就是用来初始化第二、三个 master 节点的命令，如下所示：<font color="red">此时请不要执行该命令</font>
|
||||
|
||||
``` sh
|
||||
kubeadm join apiserver.k8s:6443 --token 4z3r2v.2p43g28ons3b475v \
|
||||
--discovery-token-ca-cert-hash sha256:959569cbaaf0cf3fad744f8bd8b798ea9e11eb1e568c15825355879cf4cdc5d6 \
|
||||
--control-plane --certificate-key 41a741533a038a936759aff43b5680f0e8c41375614a873ea49fde8944614dd6
|
||||
```
|
||||
|
||||
</el-tab-pane>
|
||||
<el-tab-pane label="第一个Master节点初始化2个小时后再初始化">
|
||||
|
||||
**获得 certificate key**
|
||||
|
||||
在 demo-master-a-1 上执行
|
||||
|
||||
```sh
|
||||
# 只在 第一个 master 节点 demo-master-a-1 上执行
|
||||
kubeadm init phase upload-certs --upload-certs
|
||||
```
|
||||
|
||||
输出结果如下：
|
||||
|
||||
``` sh {6}
|
||||
[root@demo-master-a-1 ~]# kubeadm init phase upload-certs --upload-certs
|
||||
W0902 09:05:28.355623 1046 version.go:98] could not fetch a Kubernetes version from the internet: unable to get URL "https://dl.k8s.io/release/stable-1.txt": Get https://dl.k8s.io/release/stable-1.txt: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)
|
||||
W0902 09:05:28.355718 1046 version.go:99] falling back to the local client version: v1.15.3
|
||||
[upload-certs] Storing the certificates in Secret "kubeadm-certs" in the "kube-system" Namespace
|
||||
[upload-certs] Using certificate key:
|
||||
70eb87e62f052d2d5de759969d5b42f372d0ad798f98df38f7fe73efdf63a13c
|
||||
```
|
||||
|
||||
**获得 join 命令**
|
||||
|
||||
在 demo-master-a-1 上执行
|
||||
``` sh
|
||||
# 只在 第一个 master 节点 demo-master-a-1 上执行
|
||||
kubeadm token create --print-join-command
|
||||
```
|
||||
|
||||
输出结果如下：
|
||||
``` sh {2}
|
||||
[root@demo-master-a-1 ~]# kubeadm token create --print-join-command
|
||||
kubeadm join apiserver.demo:6443 --token bl80xo.hfewon9l5jlpmjft --discovery-token-ca-cert-hash sha256:b4d2bed371fe4603b83e7504051dcfcdebcbdcacd8be27884223c4ccc13059a4
|
||||
```
|
||||
|
||||
则，第二、三个 master 节点的 join 命令如下：
|
||||
|
||||
* 命令行中，蓝色部分来自于前面获得的 join 命令，红色部分来自于前面获得的 certificate key
|
||||
|
||||
<div style="background-color: #ddd; padding: 20px; line-height: 20px;">
|
||||
<font color="blue">kubeadm join apiserver.demo:6443 --token ejwx62.vqwog6il5p83uk7y \<br/>
|
||||
--discovery-token-ca-cert-hash sha256:6f7a8e40a810323672de5eee6f4d19aa2dbdb38411845a1bf5dd63485c43d303 </font>\<br/>
|
||||
--control-plane --certificate-key <font color="red">70eb87e62f052d2d5de759969d5b42f372d0ad798f98df38f7fe73efdf63a13c</font>
|
||||
</div>
|
||||
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
|
||||
**初始化第二、三个 master 节点**
|
||||
|
||||
在 demo-master-b-1 和 demo-master-b-2 机器上执行
|
||||
|
||||
``` sh
|
||||
# 只在第二、三个 master 节点 demo-master-b-1 和 demo-master-b-2 执行
|
||||
# 替换 x.x.x.x 为 ApiServer LoadBalancer 的 IP 地址
|
||||
export APISERVER_IP=x.x.x.x
|
||||
# 替换 apiserver.demo 为 前面已经使用的 dnsName
|
||||
export APISERVER_NAME=apiserver.demo
|
||||
echo "${APISERVER_IP} ${APISERVER_NAME}" >> /etc/hosts
|
||||
# 使用前面步骤中获得的第二、三个 master 节点的 join 命令
|
||||
kubeadm join apiserver.demo:6443 --token ejwx62.vqwog6il5p83uk7y \
|
||||
--discovery-token-ca-cert-hash sha256:6f7a8e40a810323672de5eee6f4d19aa2dbdb38411845a1bf5dd63485c43d303 \
|
||||
--control-plane --certificate-key 70eb87e62f052d2d5de759969d5b42f372d0ad798f98df38f7fe73efdf63a13c
|
||||
```
|
||||
|
||||
**检查 master 初始化结果**
|
||||
|
||||
``` sh
|
||||
# 只在第一个 master 节点 demo-master-a-1 执行
|
||||
# 查看 master 节点初始化结果
|
||||
kubectl get nodes
|
||||
```
|
||||
|
||||
<!-- </div>
|
||||
|
||||
<div slot="step4"> -->
|
||||
|
||||
## 初始化 worker节点
|
||||
|
||||
### 获得 join命令参数
|
||||
|
||||
<el-tabs type="border-card">
|
||||
<el-tab-pane label="和第一个Master节点一起初始化">
|
||||
|
||||
初始化第一个 master 节点时的输出内容中，第25、26行就是用来初始化 worker 节点的命令，如下所示：<font color="red">此时请不要执行该命令</font>
|
||||
|
||||
``` sh
|
||||
kubeadm join apiserver.k8s:6443 --token 4z3r2v.2p43g28ons3b475v \
|
||||
--discovery-token-ca-cert-hash sha256:959569cbaaf0cf3fad744f8bd8b798ea9e11eb1e568c15825355879cf4cdc5d6
|
||||
```
|
||||
|
||||
</el-tab-pane>
|
||||
<el-tab-pane label="第一个Master节点初始化2个小时后再初始化">
|
||||
|
||||
**在第一个 master 节点 demo-master-a-1 节点执行**
|
||||
|
||||
```bash
|
||||
# 只在第一个 master 节点 demo-master-a-1 上执行
|
||||
kubeadm token create --print-join-command
|
||||
```
|
||||
|
||||
可获取kubeadm join 命令及参数，如下所示
|
||||
|
||||
```bash
|
||||
kubeadm join apiserver.demo:6443 --token mpfjma.4vjjg8flqihor4vt --discovery-token-ca-cert-hash sha256:6f7a8e40a810323672de5eee6f4d19aa2dbdb38411845a1bf5dd63485c43d303
|
||||
```
|
||||
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
|
||||
### 初始化worker
|
||||
|
||||
**针对所有的 worker 节点执行**
|
||||
|
||||
```sh
|
||||
# 只在 worker 节点执行
|
||||
# 替换 ${APISERVER_IP} 为 ApiServer LoadBalancer 的 IP 地址
|
||||
# 替换 ${APISERVER_NAME} 为 前面已经使用的 dnsName
|
||||
echo "${APISERVER_IP} ${APISERVER_NAME}" >> /etc/hosts
|
||||
|
||||
# 替换为前面 kubeadm token create --print-join-command 的输出结果
|
||||
kubeadm join apiserver.demo:6443 --token mpfjma.4vjjg8flqihor4vt --discovery-token-ca-cert-hash sha256:6f7a8e40a810323672de5eee6f4d19aa2dbdb38411845a1bf5dd63485c43d303
|
||||
```
|
||||
|
||||
### 检查 worker 初始化结果
|
||||
|
||||
在第一个master节点 demo-master-a-1 上执行
|
||||
|
||||
```sh
|
||||
# 只在第一个 master 节点 demo-master-a-1 上执行
|
||||
kubectl get nodes
|
||||
```
|
||||
|
||||
|
||||
|
||||
## 移除 worker 节点
|
||||
|
||||
::: warning
|
||||
正常情况下，您无需移除 worker 节点
|
||||
:::
|
||||
|
||||
在准备移除的 worker 节点上执行
|
||||
|
||||
```sh
|
||||
kubeadm reset
|
||||
```
|
||||
|
||||
在第一个 master 节点 demo-master-a-1 上执行
|
||||
|
||||
```sh
|
||||
kubectl delete node demo-worker-x-x
|
||||
```
|
||||
|
||||
> * 将 demo-worker-x-x 替换为要移除的 worker 节点的名字
|
||||
> * worker 节点的名字可以通过在第一个 master 节点 demo-master-a-1 上执行 kubectl get nodes 命令获得
|
||||
|
||||
<!-- </div>
|
||||
|
||||
<div slot="step5"> -->
|
||||
|
||||
## 安装 Ingress Controller
|
||||
|
||||
> Ingress官方文档：https://kubernetes.io/docs/concepts/services-networking/ingress/
|
||||
>
|
||||
> Ingress Controllers官网介绍：https://kubernetes.io/docs/concepts/services-networking/ingress-controllers/
|
||||
>
|
||||
> 本文中使用如下部署方式：https://kubernetes.github.io/ingress-nginx/deploy/baremetal/#using-a-self-provisioned-edge
|
||||
>
|
||||
> kubernetes支持多种Ingress Controllers (traefic / Kong / Istio / Nginx 等)，本文推荐使用 https://github.com/nginxinc/kubernetes-ingress
|
||||
|
||||
|
||||
<el-tabs type="border-card">
|
||||
<el-tab-pane label="快速安装">
|
||||
|
||||
**在 master 节点上执行**
|
||||
|
||||
``` sh
|
||||
# 只在第一个 master 节点 demo-master-a-1 上执行
|
||||
kubectl apply -f https://kuboard.cn/install-script/v1.15.3/nginx-ingress.yaml
|
||||
```
|
||||
|
||||
</el-tab-pane>
|
||||
<el-tab-pane label="YAML文件">
|
||||
|
||||
<<< @/.vuepress/public/install-script/v1.15.3/nginx-ingress.yaml
|
||||
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
|
||||
::: warning
|
||||
如果您打算将 Kubernetes 用于生产环境，请参考此文档 [Installing Ingress Controller](https://github.com/nginxinc/kubernetes-ingress/blob/v1.5.3/docs/installation.md)，完善 Ingress 的配置
|
||||
:::
|
||||
|
||||
### 在 IaaS 层完成如下配置（**公网Load Balancer**）
|
||||
|
||||
创建负载均衡 Load Balancer：
|
||||
|
||||
* 监听器 1：80 / TCP， SOURCE_ADDRESS 会话保持
|
||||
* 服务器资源池 1： demo-worker-x-x 的所有节点的 80端口
|
||||
* 监听器 2：443 / TCP， SOURCE_ADDRESS 会话保持
|
||||
* 服务器资源池 2： demo-worker-x-x 的所有节点的443端口
|
||||
|
||||
假设刚创建的负载均衡 Load Balancer 的 IP 地址为： z.z.z.z
|
||||
|
||||
|
||||
|
||||
### 配置域名解析
|
||||
|
||||
将域名 *.demo.yourdomain.com 解析到地址负载均衡服务器 的 IP 地址 z.z.z.z
|
||||
|
||||
|
||||
|
||||
### 验证配置
|
||||
|
||||
在浏览器访问 a.demo.yourdomain.com，将得到 404 NotFound 错误页面
|
||||
|
||||
|
||||
<!-- </div>
|
||||
|
||||
<div slot="step6"> -->
|
||||
|
||||
## 下一步
|
||||
:tada: :tada: :tada:
|
||||
|
||||
您已经完成了 Kubernetes 集群的安装，下一步请：
|
||||
|
||||
[安装 Kuboard](/install/install-dashboard.html)
|
||||
|
||||
安装 Kuboard 之前先
|
||||
<a target="_blank" :href="`http://demo.kuboard.cn/#/dashboard?k8sToken=${$site.themeConfig.kuboardToken}`">
|
||||
在线体验 Kuboard
|
||||
</a>
|
||||
|
||||
<!-- </div>
|
||||
|
||||
</StoryBook> -->
|
||||
@ -31,78 +31,6 @@ meta:
|
||||
|
||||
<!-- <AdSenseTitle/> -->
|
||||
|
||||
<script>
|
||||
|
||||
const ENV_COUNT = 5
|
||||
|
||||
export default {
|
||||
data () {
|
||||
let validateEnv = (rule, value, callback) => {
|
||||
if (value.length < ENV_COUNT) {
|
||||
callback(new Error('请确认您的环境符合上述条件'));
|
||||
} else {
|
||||
callback();
|
||||
}
|
||||
};
|
||||
return {
|
||||
form: {
|
||||
checked: []
|
||||
},
|
||||
rules: {
|
||||
checked: [{validator: validateEnv, trigger: 'change'}]
|
||||
}
|
||||
}
|
||||
},
|
||||
computed: {
|
||||
envOk () {
|
||||
if (this.form.checked.length === ENV_COUNT) {
|
||||
return true
|
||||
}
|
||||
return false
|
||||
}
|
||||
},
|
||||
mounted () {
|
||||
},
|
||||
watch: {
|
||||
'form.checked' () {
|
||||
if (this.form.checked.length === ENV_COUNT) {
|
||||
|
||||
}
|
||||
},
|
||||
envOk () {
|
||||
if (this.envOk) {
|
||||
this.$sendGaEvent('install-k8s', 'envOk', '安装k8s-已确认环境符合条件')
|
||||
}
|
||||
}
|
||||
},
|
||||
methods: {
|
||||
downloadDiagram () {
|
||||
console.log('尝试发送 ga event')
|
||||
if (window.ga) {
|
||||
window.ga('send', {
|
||||
hitType: 'event',
|
||||
eventCategory: '安装K8S',
|
||||
eventAction: 'Download',
|
||||
eventLabel: '下载拓扑图源文件'
|
||||
});
|
||||
console.log('发送成功 ga event')
|
||||
} else {
|
||||
console.log('开发环境，不发送 ga event')
|
||||
}
|
||||
},
|
||||
canSlideNext (currentName) {
|
||||
if (currentName === 'step1' && this.form.checked.length < ENV_COUNT) {
|
||||
this.$refs.envForm.validate(valid => {
|
||||
|
||||
})
|
||||
return { flag: false, message: '请翻到本页最下方，并确认您的环境符合要求的条件' }
|
||||
}
|
||||
return { flag: true, message: 'can slide next' }
|
||||
}
|
||||
}
|
||||
}
|
||||
</script>
|
||||
|
||||
<!-- <StoryBook>
|
||||
<div slot="introduction"> -->
|
||||
|
||||
@ -122,7 +50,7 @@ export default {
|
||||
|
||||
* **持续更新和完善**
|
||||
* 始终有最新的 Kubernetes 稳定版安装文档，当前版本 v1.16.1
|
||||
* 当前已更新了 <font color="red"> 60 次 </font>， [查看更新历史](https://github.com/eip-work/kuboard-press/commits/master/install/install-k8s.md)
|
||||
* 当前已更新了 <font color="red"> 61 次 </font>， [查看更新历史](https://github.com/eip-work/kuboard-press/commits/master/install/install-k8s.md)
|
||||
|
||||
* **在线答疑**
|
||||
|
||||
@ -165,7 +93,7 @@ export default {
|
||||
**安装后的软件版本为**
|
||||
|
||||
* Kubernetes v1.16.1
|
||||
* calico 3.8.2
|
||||
* calico 3.9
|
||||
* nginx-ingress 1.5.5
|
||||
* Docker 18.09.7
|
||||
|
||||
@ -184,7 +112,7 @@ export default {
|
||||
</grid-item>
|
||||
</grid>
|
||||
|
||||
安装后的拓扑图如下：<span v-on:click="downloadDiagram"><a :href="$withBase('/kuboard.rp')" download="www.kuboard.cn.rp">下载拓扑图源文件</a></span> <font color="#999">使用Axure RP 9.0可打开该文件</font>
|
||||
安装后的拓扑图如下：<span v-on:click="$sendGaEvent('下载拓扑图-kubernetes', '下载拓扑图-kubernetes', 'Download-install-kubernetes.html')"><a :href="$withBase('/kuboard.rp')" download="www.kuboard.cn.rp">下载拓扑图源文件</a></span> <font color="#999">使用Axure RP 9.0可打开该文件</font>
|
||||
|
||||
<p style="max-width: 720px;">
|
||||
<img src="/images/topology/k8s.png" style="max-width: 100%;" alt="Kubernetes安装：Kubernetes安装拓扑图">
|
||||
@ -261,43 +189,7 @@ echo "127.0.0.1 $(hostname)" >> /etc/hosts
|
||||
|
||||
## 安装 docker / kubelet
|
||||
|
||||
<!-- <transition-group name="el-zoom-in-top"> -->
|
||||
<div v-show="!envOk" key="not">
|
||||
|
||||
<grid :rwd="{compact: 'stack'}">
|
||||
<grid-item size="2/3" :rwd="{tablet: '1/1', compact: '1/1'}" style="padding: 1rem 0 1rem 1rem;">
|
||||
|
||||
<div>
|
||||
|
||||
::: danger 必须选中下面的四个勾选框才能继续
|
||||
* 选中后显示 **安装 docker/kubelet 的文档**
|
||||
|
||||
<div style="display: inline-block; width: 480px; max-width: calc(100vw - 100px); overflow: hidden; line-height: 40px; background-color: rgba(255,229,100,0.3); padding: 20px 0 0 20px; margin-bottom: 20px; border: 1px solid #d7dae2;">
|
||||
<el-form :model="form" ref="envForm" :rules="rules" style="text-align: left;">
|
||||
<el-form-item prop="checked" class="env-form-item">
|
||||
<el-checkbox-group v-model="form.checked">
|
||||
<li style="height: 40px;"> <el-checkbox style="width: 300px; max-width: calc(100vw - 100px); text-align: left;" label="centos">我的任意节点 centos 版本在兼容列表中</el-checkbox> </li>
|
||||
<li style="height: 40px;"> <el-checkbox style="width: 300px; max-width: calc(100vw - 100px); text-align: left;" label="hostname">我的任意节点 hostname 不是 localhost，且不包含下划线</el-checkbox> </li>
|
||||
<li style="height: 40px;"> <el-checkbox style="width: 300px; max-width: calc(100vw - 100px); text-align: left;" label="cpu">我的任意节点 CPU 内核数量大于等于 2</el-checkbox> </li>
|
||||
<li style="height: 40px;"> <el-checkbox style="width: 300px; max-width: calc(100vw - 100px); text-align: left;" label="docker">我的任意节点不会直接使用 docker run 或 docker-compose 运行容器</el-checkbox> </li>
|
||||
<li style="height: 40px;"> <el-checkbox style="width: 300px; max-width: calc(100vw - 100px); text-align: left;" label="networkcard">我的任意节点只有一块网卡（可以在完成K8S安装后再添加网卡）</el-checkbox> </li>
|
||||
</el-checkbox-group>
|
||||
</el-form-item>
|
||||
</el-form>
|
||||
</div>
|
||||
:::
|
||||
|
||||
</div>
|
||||
|
||||
</grid-item>
|
||||
<grid-item size="1/3" :rwd="{tablet: '1/1', compact: '0/1'}" style="padding: 2rem 1rem 1rem 1rem;">
|
||||
<AdSenseSquare/>
|
||||
</grid-item>
|
||||
</grid>
|
||||
|
||||
</div>
|
||||
<el-collapse-transition>
|
||||
<div v-show="envOk" key="ok">
|
||||
<InstallEnvCheck type="k8s">
|
||||
|
||||
使用 root 身份在所有节点执行如下代码，以安装软件：
|
||||
- docker
|
||||
@ -329,8 +221,7 @@ curl -sSL https://kuboard.cn/install-script/v1.16.1/install_kubelet.sh | sh
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
|
||||
</div>
|
||||
</el-collapse-transition>
|
||||
</InstallEnvCheck>
|
||||
<!-- </div>
|
||||
|
||||
<div slot="step3"> -->
|
||||
|
||||
@ -1,11 +1,11 @@
|
||||
---
|
||||
vssueId: 16
|
||||
description: Kubernete安装文档_使用kubeadm安装高可用的Kubernetes_v1.15.3集群_可用于生产环境
|
||||
description: Kubernete安装文档_使用kubeadm安装高可用的Kubernetes_v1.16.1集群_可用于生产环境
|
||||
meta:
|
||||
- name: keywords
|
||||
content: Kubernetes集群,Kubernetes高可用,Kubernetes生产环境
|
||||
# storyBook:
|
||||
# title: '使用 kubeadm 安装 kubernetes v1.15.3（高可用）'
|
||||
# title: '使用 kubeadm 安装 kubernetes v1.16.1（高可用）'
|
||||
# initial: FullPage
|
||||
# pages:
|
||||
# - name: overview
|
||||
@ -26,7 +26,7 @@ meta:
|
||||
|
||||
# 安装Kubernetes高可用
|
||||
|
||||
<AdSenseTitle/>
|
||||
<!-- <AdSenseTitle/> -->
|
||||
|
||||
<!-- <StoryBook>
|
||||
|
||||
@ -40,14 +40,22 @@ meta:
|
||||
|
||||
kubernetes 安装有多种选择，本文档描述的集群安装具备如下特点：
|
||||
|
||||
* Kubernetes 1.15.3
|
||||
* calico 3.8.2
|
||||
* Kubernetes 1.16.1
|
||||
* calico 3.9
|
||||
* nginx-ingress 1.5.3
|
||||
* Docker 18.09.7
|
||||
* 三个 master 组成主节点集群，通过内网 loader balancer 实现负载均衡；至少需要三个 master 节点才可组成高可用集群，否则会出现 ***脑裂*** 现象
|
||||
* 多个 worker 组成工作节点集群，通过外网 loader balancer 实现负载均衡
|
||||
|
||||
安装后的拓扑图如下：<a :href="$withBase('/kuboard.rp')" download="www.kuboard.cn.rp">下载拓扑图源文件</a> <font color="#999">使用Axure RP 9.0可打开该文件</font>
|
||||
[领取腾讯云最高2860元代金券](https://cloud.tencent.com/act/cps/redirect?redirect=1040&cps_key=2ee6baa049659f4713ddc55a51314372&from=console)
|
||||
|
||||
[腾讯云限时1折秒杀](https://cloud.tencent.com/act/cps/redirect?redirect=1044&cps_key=2ee6baa049659f4713ddc55a51314372&from=console)
|
||||
|
||||
[领取阿里云最高2000元红包](https://promotion.aliyun.com/ntms/yunparter/invite.html?userCode=obezo3pg)
|
||||
|
||||
[阿里云服务器限时2折](https://www.aliyun.com/acts/limit-buy?userCode=obezo3pg)
|
||||
|
||||
安装后的拓扑图如下：<span v-on:click="$sendGaEvent('下载拓扑图-kubernetes', '下载拓扑图-kubernetes', 'Download-install-kubernetes.html')"> <a :href="$withBase('/kuboard.rp')" download="www.kuboard.cn.rp" >下载拓扑图源文件</a> </span> <font color="#999">使用Axure RP 9.0可打开该文件</font>
|
||||
|
||||
|
||||
|
||||
@ -57,17 +65,6 @@ kubernetes 安装有多种选择，本文档描述的集群安装具备如下特
|
||||
<p>
|
||||
<img src="/images/kuboard_qq.png" alt="Kubernetes教程：QQ群在线答疑"/>
|
||||
</p>
|
||||
|
||||
<!-- <div>
|
||||
<div style="margin-top: 10px;">未打赏用户可进 QQ 群聊，<span style="color: red;">打赏用户可进微信群聊</span>。</div>
|
||||
<div style="margin-top: 10px;">
|
||||
<span>扫第一个二维码完成打赏，扫第二个进微信群聊。</span> <span style="color: #CCC">QQ 群聊二维码在左侧导航栏下方。</span>
|
||||
<p style="margin-top: 10px;">
|
||||
<img src="/images/dz.png" style="width: 150px; margin-right: 150px;"></img>
|
||||
<img src="/images/dz2.jpeg" style="width: 150px;"></img>
|
||||
</p>
|
||||
</div>
|
||||
</div> -->
|
||||
|
||||
<!-- </div>
|
||||
|
||||
@ -101,6 +98,8 @@ hostname
|
||||
hostnamectl set-hostname your-new-host-name
|
||||
# 查看修改结果
|
||||
hostnamectl status
|
||||
# 设置 hostname 解析
|
||||
echo "127.0.0.1 $(hostname)" >> /etc/hosts
|
||||
```
|
||||
:::
|
||||
|
||||
@ -110,6 +109,8 @@ hostnamectl status
|
||||
|
||||
## 安装 docker / kubelet
|
||||
|
||||
<InstallEnvCheck type="kubernetes">
|
||||
|
||||
使用 root 身份在所有节点执行如下代码，以安装软件：
|
||||
- docker
|
||||
- nfs-utils
|
||||
@ -121,7 +122,7 @@ hostnamectl status
|
||||
``` sh
|
||||
# 在 master 节点和 worker 节点都要执行
|
||||
|
||||
curl -sSL https://kuboard.cn/install-script/v1.15.3/install-kubelet.sh | sh
|
||||
curl -sSL https://kuboard.cn/install-script/v1.16.1/install_kubelet.sh | sh
|
||||
|
||||
```
|
||||
|
||||
@ -130,7 +131,7 @@ curl -sSL https://kuboard.cn/install-script/v1.15.3/install-kubelet.sh | sh
|
||||
|
||||
手动执行以下代码，效果与快速安装完全相同。
|
||||
|
||||
<<< @/.vuepress/public/install-script/v1.15.3/install-kubelet.sh
|
||||
<<< @/.vuepress/public/install-script/v1.16.1/install_kubelet.sh
|
||||
|
||||
::: warning
|
||||
如果此时执行 `service status kubelet` 命令，将得到 kubelet 启动失败的错误提示，请忽略此错误，因为必须完成后续步骤中 kubeadm init 的操作，kubelet 才能正常启动
|
||||
@ -139,6 +140,8 @@ curl -sSL https://kuboard.cn/install-script/v1.15.3/install-kubelet.sh | sh
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
|
||||
</InstallEnvCheck>
|
||||
|
||||
<!-- </div>
|
||||
|
||||
<div slot="step3"> -->
|
||||
@ -157,8 +160,9 @@ curl -sSL https://kuboard.cn/install-script/v1.15.3/install-kubelet.sh | sh
|
||||
|
||||
假设完成创建以后，Load Balancer的 ip 地址为 x.x.x.x
|
||||
|
||||
### 初始化第一个master节点
|
||||
> 根据每个人实际的情况不同，实现 LoadBalancer 的方式不一样，本文不详细阐述如何搭建 LoadBalancer，请读者自行解决
|
||||
|
||||
### 初始化第一个master节点
|
||||
|
||||
::: tip
|
||||
* 以 root 身份在 demo-master-a-1 机器上执行
|
||||
@ -177,20 +181,20 @@ curl -sSL https://kuboard.cn/install-script/v1.15.3/install-kubelet.sh | sh
|
||||
在第一个 master 节点 demo-master-a-1 上执行
|
||||
|
||||
``` sh
|
||||
# 只在 master 节点执行
|
||||
# 只在第一个 master 节点执行
|
||||
# 替换 apiserver.demo 为 您想要的 dnsName
|
||||
export APISERVER_NAME=apiserver.demo
|
||||
# Kubernetes 容器组所在的网段，该网段安装完成后，由 kubernetes 创建，事先并不存在于您的物理网络中
|
||||
export POD_SUBNET=10.100.0.1/16
|
||||
echo "127.0.0.1 ${APISERVER_NAME}" >> /etc/hosts
|
||||
curl -sSL https://kuboard.cn/install-script/v1.15.3/init-master.sh | sh
|
||||
curl -sSL https://kuboard.cn/install-script/v1.16.1/init_master.sh | sh
|
||||
```
|
||||
|
||||
</el-tab-pane>
|
||||
<el-tab-pane label="手工初始化">
|
||||
|
||||
``` sh
|
||||
# 只在 master 节点执行
|
||||
# 只在第一个 master 节点执行
|
||||
# 替换 apiserver.demo 为 您想要的 dnsName
|
||||
export APISERVER_NAME=apiserver.demo
|
||||
# Kubernetes 容器组所在的网段，该网段安装完成后，由 kubernetes 创建，事先并不存在于您的物理网络中
|
||||
@ -198,7 +202,7 @@ export POD_SUBNET=10.100.0.1/16
|
||||
echo "127.0.0.1 ${APISERVER_NAME}" >> /etc/hosts
|
||||
```
|
||||
|
||||
<<< @/.vuepress/public/install-script/v1.15.3/init-master.sh
|
||||
<<< @/.vuepress/public/install-script/v1.16.1/init_master.sh
|
||||
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
@ -251,10 +255,18 @@ watch kubectl get pod -n kube-system -o wide
|
||||
kubectl get nodes
|
||||
```
|
||||
|
||||
::: danger
|
||||
请等到所有容器组（大约9个）全部处于 Running 状态，才进行下一步
|
||||
:::
|
||||
|
||||
### 初始化第二、三个master节点
|
||||
|
||||
**获得 master 节点的 join 命令**
|
||||
|
||||
> 可以和第一个Master节点一起初始化第二、三个Master节点，也可以从单Master节点调整过来，只需要
|
||||
> * 增加Master的 LoadBalancer
|
||||
> * 将所有节点的 /etc/hosts 文件中 apiserver.demo 解析为 LoadBalancer 的地址
|
||||
> * 添加第二、三个Master节点
|
||||
|
||||
<el-tabs type="border-card">
|
||||
<el-tab-pane label="和第一个Master节点一起初始化">
|
||||
@ -284,7 +296,7 @@ kubeadm init phase upload-certs --upload-certs
|
||||
``` sh {6}
|
||||
[root@demo-master-a-1 ~]# kubeadm init phase upload-certs --upload-certs
|
||||
W0902 09:05:28.355623 1046 version.go:98] could not fetch a Kubernetes version from the internet: unable to get URL "https://dl.k8s.io/release/stable-1.txt": Get https://dl.k8s.io/release/stable-1.txt: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)
|
||||
W0902 09:05:28.355718 1046 version.go:99] falling back to the local client version: v1.15.3
|
||||
W0902 09:05:28.355718 1046 version.go:99] falling back to the local client version: v1.16.1
|
||||
[upload-certs] Storing the certificates in Secret "kubeadm-certs" in the "kube-system" Namespace
|
||||
[upload-certs] Using certificate key:
|
||||
70eb87e62f052d2d5de759969d5b42f372d0ad798f98df38f7fe73efdf63a13c
|
||||
@ -350,6 +362,7 @@ kubectl get nodes
|
||||
|
||||
### 获得 join命令参数
|
||||
|
||||
|
||||
<el-tabs type="border-card">
|
||||
<el-tab-pane label="和第一个Master节点一起初始化">
|
||||
|
||||
@ -385,9 +398,11 @@ kubeadm join apiserver.demo:6443 --token mpfjma.4vjjg8flqihor4vt --discovery
|
||||
|
||||
```sh
|
||||
# 只在 worker 节点执行
|
||||
# 替换 ${APISERVER_IP} 为 ApiServer LoadBalancer 的 IP 地址
|
||||
# 替换 ${APISERVER_NAME} 为 前面已经使用的 dnsName
|
||||
echo "${APISERVER_IP} ${APISERVER_NAME}" >> /etc/hosts
|
||||
# 替换 x.x.x.x 为 ApiServer LoadBalancer 的 IP 地址
|
||||
export MASTER_IP=x.x.x.x
|
||||
# 替换 apiserver.demo 为初始化 master 节点时所使用的 APISERVER_NAME
|
||||
export APISERVER_NAME=apiserver.demo
|
||||
echo "${MASTER_IP} ${APISERVER_NAME}" >> /etc/hosts
|
||||
|
||||
# 替换为前面 kubeadm token create --print-join-command 的输出结果
|
||||
kubeadm join apiserver.demo:6443 --token mpfjma.4vjjg8flqihor4vt --discovery-token-ca-cert-hash sha256:6f7a8e40a810323672de5eee6f4d19aa2dbdb38411845a1bf5dd63485c43d303
|
||||
@ -403,7 +418,6 @@ kubectl get nodes
|
||||
```
|
||||
|
||||
|
||||
|
||||
## 移除 worker 节点
|
||||
|
||||
::: warning
|
||||
@ -447,13 +461,13 @@ kubectl delete node demo-worker-x-x
|
||||
|
||||
``` sh
|
||||
# 只在第一个 master 节点 demo-master-a-1 上执行
|
||||
kubectl apply -f https://kuboard.cn/install-script/v1.15.3/nginx-ingress.yaml
|
||||
kubectl apply -f https://kuboard.cn/install-script/v1.16.1/nginx-ingress.yaml
|
||||
```
|
||||
|
||||
</el-tab-pane>
|
||||
<el-tab-pane label="YAML文件">
|
||||
|
||||
<<< @/.vuepress/public/install-script/v1.15.3/nginx-ingress.yaml
|
||||
<<< @/.vuepress/public/install-script/v1.16.1/nginx-ingress.yaml
|
||||
|
||||
</el-tab-pane>
|
||||
</el-tabs>
|
||||
|
||||
Reference in New Issue
Block a user