ICode9
精准搜索请尝试： 精确搜索
首页
编程语言>
数据库
系统相关
互联网
其他分享
Python
Java
JavaScript
android
PHP
首页 > 其他分享> 文章详细

三、升级kubeasz部署的kubernetes

2021-09-25 22:02:56  阅读：261  来源： 互联网

标签：kube kubernetes kubeasz 172.168 升级 v1.21 Ready k8s root


1、升级前实验环境

OS: 
root@harbor:~# cat /etc/issue
Ubuntu 20.04.2 LTS \n \l

root@harbor:~# uname -r
5.4.0-81-generic

IP分配：
172.168.33.201 harbor.ywx.net  k8s-deploy
172.168.33.202 haproxy01
172.168.33.203 haproxy02
172.168.33.204 ecd01
172.168.33.205 ecd02
172.168.33.206 ecd03
172.168.33.207 k8s-master01
172.168.33.208 k8s-master02
172.168.33.209 k8s-master03
172.168.33.210 k8s-node01
172.168.33.211 k8s-node02
172.168.33.212 k8s-node03

VIP：
172.168.33.50  api-server的VIP
172.168.33.51
172.168.33.52
172.168.33.53

Kubernetes: v1.12.0

root@k8s-master01:/usr/local/src# kubectl get nodes
NAME             STATUS                     ROLES    AGE     VERSION
172.168.33.207   Ready,SchedulingDisabled   master   7d23h   v1.21.0
172.168.33.208   Ready,SchedulingDisabled   master   7d23h   v1.21.0
172.168.33.209   Ready,SchedulingDisabled   master   7d23h   v1.21.0
172.168.33.210   Ready                      node     7d23h   v1.21.0
172.168.33.211   Ready                      node     7d23h   v1.21.0
172.168.33.212   Ready                      node     7d23h   v1.21.0

　　

kubeasz部署v1.12.0见：

https://www.cnblogs.com/yaokaka/p/15308917.html

2、升级kubernetes到v1.12.5

1)下载kubernetes v1.12.5版本的二进制软件

https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.21.md#downloads-for-v1215

下载：
Source Code: kubernetes.tar.gz	

Client Binaries: kubernetes-client-linux-amd64.tar.gz

Server Binaries: kubernetes-server-linux-amd64.tar.gz

Node Binaries: kubernetes-node-linux-amd64.tar.gz

#注意：不要下载错误的cpu架构
root@k8s-master01:/usr/local/src# pwd
/usr/local/src
root@k8s-master01:/usr/local/src# ls
kubernetes-client-linux-amd64.tar.gz  kubernetes-node-linux-amd64.tar.gz  kubernetes-server-linux-amd64.tar.gz  kubernetes.tar.gz

root@k8s-master01:/usr/local/src# tar -xf kubernetes-client-linux-amd64.tar.gz 
root@k8s-master01:/usr/local/src# tar -xf kubernetes-node-linux-amd64.tar.gz 
root@k8s-master01:/usr/local/src# tar -xf kubernetes-server-linux-amd64.tar.gz 
root@k8s-master01:/usr/local/src# tar -xf kubernetes.tar.gz 

root@k8s-master01:/usr/local/src# ll kubernetes
total 35284
drwxr-xr-x 10 root root      176 Sep 16 05:27 ./
drwxr-xr-x  3 root root      179 Sep 25 20:44 ../
drwxr-xr-x  3 root root       49 Sep 16 05:27 LICENSES/
-rw-r--r--  1 root root     3387 Sep 16 05:27 README.md
drwxr-xr-x  2 root root        6 Sep 16 05:22 addons/
drwxr-xr-x  3 root root       31 Sep 16 05:27 client/
drwxr-xr-x  9 root root      311 Sep 16 05:27 cluster/
drwxr-xr-x  2 root root       38 Sep 16 05:27 docs/
drwxr-xr-x  3 root root       17 Sep 16 05:27 hack/
-rw-r--r--  1 root root 36121220 Sep 16 05:22 kubernetes-src.tar.gz
drwxr-xr-x  3 root root       17 Sep 16 05:21 node/
drwxr-xr-x  3 root root       66 Sep 16 05:27 server/
-rw-r--r--  1 root root        8 Sep 16 05:27 version

　　

2）升级master节点

#在node节点操作，剔除k8s-master节点
#第一步从所有的node节点的负载均衡中剔除k8s-master01节点
#以k8s-node01为例：k8s-node02和k8s-node03操作一样
root@k8s-node01:~# vim /etc/kube-lb/conf/kube-lb.conf
user root;
worker_processes 1;

error_log  /etc/kube-lb/logs/error.log warn;

events {
    worker_connections  3000;
}

stream {
    upstream backend {
        #server 172.168.33.207:6443    max_fails=2 fail_timeout=3s; #剔除k8s-master01节点
        server 172.168.33.208:6443    max_fails=2 fail_timeout=3s;
        server 172.168.33.209:6443    max_fails=2 fail_timeout=3s;
    }

    server {
        listen 127.0.0.1:6443;
        proxy_connect_timeout 1s;
        proxy_pass backend;
    }
}

#第二步所有node节点重启kube-lb
root@k8s-node01:~# systemctl restart kube-lb.service

#以上2步需要在所有node节点操作

#在master节点上操作，升级master节点
#以k8s-master01为例
#第三步停止master节点上kube-apiserver kube-controller-manager kube-proxy kube-scheduler kubelet服务
root@k8s-master01:~# systemctl  stop kube-apiserver kube-controller-manager kube-proxy kube-scheduler kubelet

#第四步拷贝v1.12.5的kube-apiserver kube-controller-manager kube-proxy kube-scheduler kubelet二进制文件覆盖老版本的二进制文件
root@k8s-master01:~# cd /usr/local/src/kubernetes/server/bin/
root@k8s-master01:/usr/local/src/kubernetes/server/bin# cp kube-apiserver kube-controller-manager kube-scheduler kube-proxy kubelet kubectl /usr/local/bin/

#第五步重启master节点上kube-apiserver kube-controller-manager kube-proxy kube-scheduler kubelet服务
root@k8s-master01:~# systemctl  start kube-apiserver kube-controller-manager kube-proxy kube-scheduler kubelet

#在所有node节点操作
#第六步将k8s-master01重新添加进所有node节点的kube-lb文件并重启kube-lb
#以k8s-node01为例
root@k8s-node01:~# vim /etc/kube-lb/conf/kube-lb.conf
user root;
worker_processes 1;

error_log  /etc/kube-lb/logs/error.log warn;

events {
    worker_connections  3000;
}

stream {
    upstream backend {
        server 172.168.33.207:6443    max_fails=2 fail_timeout=3s;
        server 172.168.33.208:6443    max_fails=2 fail_timeout=3s;
        server 172.168.33.209:6443    max_fails=2 fail_timeout=3s;
    }

    server {
        listen 127.0.0.1:6443;
        proxy_connect_timeout 1s;
        proxy_pass backend;
    }
}

root@k8s-node01:~# systemctl restart kube-lb.service 

#第七步验证k8s-master01的版本信息
root@k8s-master01:~# kubectl get nodes
NAME             STATUS                     ROLES    AGE     VERSION
172.168.33.207   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.208   Ready,SchedulingDisabled   master   7d23h   v1.21.0
172.168.33.209   Ready,SchedulingDisabled   master   7d23h   v1.21.0
172.168.33.210   Ready                      node     7d23h   v1.21.0
172.168.33.211   Ready                      node     7d23h   v1.21.0
172.168.33.212   Ready                      node     7d23h   v1.21.0

#第八步升级k8s-master02和k8s-master03
#k8s-master02和k8s-master03升级步骤与k8s-master01一致
#验证master节点
root@k8s-master01:~# kubectl get nodes
NAME             STATUS                     ROLES    AGE     VERSION
172.168.33.207   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.208   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.209   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.210   Ready                      node     7d23h   v1.21.0
172.168.33.211   Ready                      node     7d23h   v1.21.0
172.168.33.212   Ready                      node     7d23h   v1.21.0

　　

3）升级node节点

#以k8s-node01为例，k8s-node02和k8s-node03通k8s-node01一样
#第一步:驱逐该k8s-node01节点上的所有pod
#在master节点上操作
root@k8s-node01:~# kubectl drain 172.168.33.210
......
There are pending nodes to be drained:
 172.168.33.210
cannot delete DaemonSet-managed Pods (use --ignore-daemonsets to ignore): kube-system/calico-node-ccv26
cannot delete Pods with local storage (use --delete-emptydir-data to override): kubernetes-dashboard/kubernetes-dashboard-79b875f7f8-phwtv
#需要添加--ignore-daemonsets和--delete-emptydir-data来清除所有pod

root@k8s-master01:~# kubectl drain 172.168.33.210 --force --ignore-daemonsets --delete-emptydir-data
#清楚DaemonSet-managed Pods和local storage

root@k8s-master01:~# kubectl get nodes
NAME             STATUS                     ROLES    AGE     VERSION
172.168.33.207   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.208   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.209   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.210   Ready,SchedulingDisabled   node     7d23h   v1.21.0 #k8s-node01不在调度pod
172.168.33.211   Ready                      node     7d23h   v1.21.0
172.168.33.212   Ready                      node     7d23h   v1.21.0



#第二步:停止K8s-node01上的kubelet和kube-proxy服务
root@k8s-node01:~# systemctl stop kubelet kube-proxy

#第三步:把v1.12.5版本的kubelet和kube-proxy从k8s-master01上拷贝过来
root@k8s-master01:/usr/local/src/kubernetes/server/bin# pwd
/usr/local/src/kubernetes/server/bin
root@k8s-master01:/usr/local/src/kubernetes/server/bin# scp kubelet kube-proxy 172.168.33.210:/usr/local/bin/

#第四步:在k8s-node01上重启kubelet和kube-proxy服务
root@k8s-node01:~# systemctl start kubelet kube-proxy

#第五步:取消k8s-node01上不调度pod的设置
root@k8s-master01:~# kubectl uncordon 172.168.33.210
node/172.168.33.210 uncordoned
root@k8s-master01:~# kubectl get nodes
NAME             STATUS                     ROLES    AGE     VERSION
172.168.33.207   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.208   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.209   Ready,SchedulingDisabled   master   7d23h   v1.21.5
172.168.33.210   Ready                      node     7d23h   v1.21.5
172.168.33.211   Ready                      node     7d23h   v1.21.0
172.168.33.212   Ready                      node     7d23h   v1.21.0

#第六步:升级k8s-node02和k8s-node03，操作与k8s-node01一样
#第七步:验证
root@k8s-master01:~# kubectl get nodes
NAME             STATUS                     ROLES    AGE   VERSION
172.168.33.207   Ready,SchedulingDisabled   master   8d    v1.21.5
172.168.33.208   Ready,SchedulingDisabled   master   8d    v1.21.5
172.168.33.209   Ready,SchedulingDisabled   master   8d    v1.21.5
172.168.33.210   Ready                      node     8d    v1.21.5
172.168.33.211   Ready                      node     8d    v1.21.5
172.168.33.212   Ready                      node     8d    v1.21.5

　　

标签：kube,kubernetes,kubeasz,172.168,升级,v1.21,Ready,k8s,root
来源： https://www.cnblogs.com/yaokaka/p/15335719.html

本站声明： 1. iCode9 技术分享网（下文简称本站）提供的所有内容，仅供技术学习、探讨和分享；
2. 关于本站的所有留言、评论、转载及引用，纯属内容发起人的个人观点，与本站观点和立场无关；
3. 关于本站的所有言论和文字，纯属内容发起人的个人观点，与本站观点和立场无关；
4. 本站文章均是网友提供，不完全保证技术分享内容的完整性、准确性、时效性、风险性和版权归属；如您发现该文章侵犯了您的权益，可联系我们第一时间进行删除；
5. 本站为非盈利性的个人网站，所有内容不会用来进行牟利，也不会利用任何形式的广告来间接获益，纯粹是为了广大技术爱好者提供技术内容和技术思想的分享性交流网站。

关于我们 | 联系我们 | 留言反馈

专注分享技术，共同学习，共同进步。侵权联系[81616952@qq.com]

Copyright (C)ICode9.com, All Rights Reserved.

ICode9版权所有