|NO.Z.00091|——————————|^^ 升级 ^^|——|KuberNetes&二进制升级.V02|——|kubernetes_master|k8s-master01组件|
一、kubernetes升级说明
### --- kubernetes升级说明
~~~ 在升级的时候建议最后去升级controller-manager
~~~ 在升级的过程中最好把controller-manager停掉，
~~~ 在升级的过程中可能会造成pod的漂移，造成业务pod漂移。
~~~ 在升级的过程中节点挂掉，pod漂移，业务宕机
~~~ 若是controller-manager的版本过高，kubelet版本过低；就会造成循环的去创建pod或者有些api版本不符合报错
~~~ 在升级过程中，一般都是把controller-manager停掉；等升级完所有的节点，包括升级完kubelet再启动controller-manager，再升级
~~~ # 第二种方案：把这个节点下线在升级，与controller-manager在哪里都是没有影响的### --- kubernetes升级方案
~~~ 方案一：本地直接升级，停止controller-manager再去升级
~~~ 方案二：停止所升级节点，再升级kubernetes相关组件；controller-manager是不可以停止掉的
~~~ 方案三：设置成不可调度的方式，再去升级### --- 此环境升级采用方案：
~~~ 此环境升级kubernetes方案：把节点下线；容器会漂移到其它节点，然后再升级这个节点相关组件
~~~ 优点一：升级节点pod提前漂移了，系统业务不会受到影响
~~~ 缺点一：若是不下线不漂移；直接升级kubelet，在升级完之后节点上的容器会重启；升级完之后哈希值会变，哈希值变化之后，容器就会重启。先下线，或者设置成不可调度的方式，再滚动发布。pod会滚动重新发布
~~~ 缺点：升级过程中，pod来回漂移会造成pod多次重启；若是在本地直接升级会造成服务中断；若是多副本，是没有任何问题的。二、下载kubernetes.v1.19.5版本包
### --- 下载kubernetes-server.v1.19.5
~~~ # 下载kubernetes-server.1.19.5版本的资源包
[root@k8s-master01 master]# wget -c https://dl.k8s.io/v1.19.5/kubernetes-server-linux-amd64.tar.gz~~~ # 查看下载的版本包
[root@k8s-master01 master]# ll
-rw-r--r-- 1 root root 309852159 Dec 10 2020 kubernetes-server-linux-amd64.tar.gz~~~ # 解压kubernetes版本包
[root@k8s-master01 master]# tar -zxvf kubernetes-server-linux-amd64.tar.gz ~~~ # 查看当前版本包下kubernetes版本
~~~ 注：Client可以看到它的版本号为:1.19.5
~~~ 注：Server版本号为1.17.0；还是历史版本号；升级完成之后会发生更改
[root@k8s-master01 bin]# pwd
/root/upgrade/master/kubernetes/server/bin
[root@k8s-master01 bin]# ./kubectl version
Client Version: version.Info{Major:"1", Minor:"19", GitVersion:"v1.19.5", GitCommit:"e338cf2c6d297aa603b50ad3a301f761b4173aa6", GitTreeState:"clean", BuildDate:"2020-12-09T11:18:51Z", GoVersion:"go1.15.2", Compiler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{Major:"1", Minor:"20", GitVersion:"v1.17.0", GitCommit:"8a62859e515889f07e3e3be6a1080413f17cf2c3", GitTreeState:"clean", BuildDate:"2021-04-15T03:19:55Z", GoVersion:"go1.15.10", Compiler:"gc", Platform:"linux/amd64"}一、kubernetes升级k8s-master01节点组件：kubernetes升级kube-apiserver
### --- kubernetes升级kube-apiserver
~~~ # 查看kube-apiserver历史版本
[root@k8s-master01 ~]# /usr/local/bin/kube-apiserver --version
Kubernetes v1.17.0~~~ # 备份kube-apiserver版本
[root@k8s-master01 ~]# which kube-apiserver
/usr/local/bin/kube-apiserver
[root@k8s-master01 ~]# mkdir /usr/local/bin/api.bac
[root@k8s-master01 ~]# cp /usr/local/bin/kube-apiserver /usr/local/bin/api.bac/~~~ # 停止kube-apiserver服务
[root@k8s-master01 ~]# systemctl stop kube-apiserver~~~ # 升级kube-apiserver服务
[root@k8s-master01 ~]# cp upgrade/master/kubernetes/server/bin/kube-apiserver /usr/local/bin/
cp: overwrite ‘/usr/local/bin/kube-apiserver’? y~~~ # 查看kube-apiserver版本
[root@k8s-master01 ~]# /usr/local/bin/kube-apiserver --version
Kubernetes v1.19.5~~~ # 启动kube-apiserver服务
[root@k8s-master01 ~]# systemctl daemon-reload
[root@k8s-master01 ~]# systemctl restart kube-apiserver二、kubernetes升级kube-controller-manager
### --- kubernetes升级kube-controller-manager
~~~ # 查看kube-controller-manager历史版本
[root@k8s-master01 ~]# /usr/local/bin/kube-controller-manager --version
Kubernetes v1.17.0~~~ # 备份kube-controller-manager版本
[root@k8s-master01 ~]# which kube-controller-manager
/usr/local/bin/kube-controller-manager
[root@k8s-master01 ~]# mkdir /usr/local/bin/controller.bak
[root@k8s-master01 ~]# cp /usr/local/bin/kube-controller-manager /usr/local/bin/controller.bak/~~~ # 停止kube-controller-manager服务
[root@k8s-master01 ~]# systemctl stop kube-controller-manager~~~ # 升级kube-controller-manager服务
[root@k8s-master01 ~]# cp upgrade/master/kubernetes/server/bin/kube-controller-manager /usr/local/bin/
cp: overwrite ‘/usr/local/bin/kube-controller-manager’? y~~~ # 查看kube-controller-manager版本
[root@k8s-master01 ~]# /usr/local/bin/kube-controller-manager --version
Kubernetes v1.19.5~~~ # 启动kube-controller-manager服务
[root@k8s-master01 ~]# systemctl daemon-reload
[root@k8s-master01 ~]# systemctl restart kube-controller-manager三、kubernetes升级kube-scheduler
### --- kubernetes升级kube-scheduler
~~~ # 查看kube-scheduler历史版本
[root@k8s-master01 ~]# /usr/local/bin/kube-scheduler --version
Kubernetes v1.17.0~~~ # 备份kube-scheduler版本
[root@k8s-master01 ~]# which kube-scheduler
/usr/local/bin/kube-scheduler
[root@k8s-master01 ~]# mkdir /usr/local/bin/scheduler.bak
[root@k8s-master01 ~]# cp /usr/local/bin/kube-scheduler /usr/local/bin/scheduler.bak/~~~ # 停止kube-scheduler服务
[root@k8s-master01 ~]# systemctl stop kube-scheduler~~~ # 升级kube-scheduler服务
[root@k8s-master01 ~]# cp upgrade/master/kubernetes/server/bin/kube-scheduler /usr/local/bin/
cp: overwrite ‘/usr/local/bin/kube-scheduler’? y~~~ # 查看kube-scheduler版本
[root@k8s-master01 ~]# /usr/local/bin/kube-scheduler --version
Kubernetes v1.19.5~~~ # 启动kube-scheduler服务
[root@k8s-master01 ~]# systemctl daemon-reload
[root@k8s-master01 ~]# systemctl restart kube-scheduler四、kubernetes升级kube-proxy
### --- kubernetes升级kube-proxy
~~~ # 查看kube-proxy历史版本
[root@k8s-master01 ~]# /usr/local/bin/kube-proxy --version
Kubernetes v1.17.0~~~ # 备份kube-proxy版本
[root@k8s-master01 ~]# which kube-proxy
/usr/local/bin/kube-proxy
[root@k8s-master01 ~]# mkdir /usr/local/bin/proxy.bak
[root@k8s-master01 ~]# cp /usr/local/bin/kube-proxy /usr/local/bin/proxy.bak/~~~ # 停止kube-proxy服务
[root@k8s-master01 ~]# systemctl stop kube-proxy~~~ # 升级kube-proxy服务
[root@k8s-master01 ~]# cp upgrade/master/kubernetes/server/bin/kube-proxy /usr/local/bin/
cp: overwrite ‘/usr/local/bin/kube-proxy’? y~~~ # 查看kube-proxy版本
[root@k8s-master01 ~]# /usr/local/bin/kube-proxy --version
Kubernetes v1.19.5~~~ # 启动kube-proxy服务
[root@k8s-master01 ~]# systemctl daemon-reload
[root@k8s-master01 ~]# systemctl restart kube-proxy五、查看集群状态
### --- 查看集群状态
[root@k8s-master01 ~]# kubectl cluster-info
Kubernetes master is running at https://192.168.1.20:8443
CoreDNS is running at https://192.168.1.20:8443/api/v1/namespaces/kube-system/services/kube-dns:dns/proxyWalter Savage Landor:strove with none,for none was worth my strife.Nature I loved and, next to Nature, Art:I warm'd both hands before the fire of life.It sinks, and I am ready to depart
——W.S.Landor
浙公网安备 33010602011771号
