多k8s集群管理

多集群的切換是K8s運維中比不可少的問題，常見的基於多個集群進行切換的方法有三種：

• 切換config文件
• 通過context進行集群切換
• 用kubecm進行集群切換

切換config文件

我們先看看放在.kube下默認的config集群：

$ kubectl config view
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: DATA+OMITTED
    server: 
  name: kubernetes
contexts:
- context:
    cluster: kubernetes
    user: "2xxxxxxxxx7"
  name: 2xxxxxxxxxxxxxxxx8
current-context: 2xxxxxxxxxxxx8
kind: Config
preferences: {}
users:
- name: "2xxxxxxxxxxx27"
  user:
    client-certificate-data: 
    client-key-data:

通過export更改默認集群配置：

$ export  KUBECONFIG=$HOME/.kube/rancher-config

查看：

$ kubectl config view
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: DATA+OMITTED
    server: 
  name: stage
contexts:
- context:
    cluster: stage
    user: stage
  name: stage
current-context: stage
kind: Config
preferences: {}
users:
- name: stage
  user:
    token:

這種方法就是特別麻煩，每次切換都要通過export設置，非常麻煩，同時也不方便。

通過context進行集群切換

把要切換的所有config文件添加到KUBECONFIG環境變量中

$ export  KUBECONFIG=$HOME/.kube/config:$HOME/.kube/rancher-config

或者可以將命令直接寫到.bashrc文件中，這樣就不需要每次手動設置。

通過kubectl config view可以看到這兩個文件已經合並到一起了：

$ kubectl config view
apiVersion: v1
clusters:
- cluster:
    certificate-authority-data: DATA+OMITTED
    server: 
  name: kubernetes
contexts:
- context:
    cluster: kubernetes
    user: "2xxxxxxxxx7"
  name: 2xxxxxxxxxxxxxxxx8
current-context: 2xxxxxxxxxxxx8
kind: Config
preferences: {}
users:
- name: "2xxxxxxxxxxx27"
  user:
    client-certificate-data: 
    client-key-data:

合並一起后就可以通過use-context切換上下文來切換集群：

# 切換stage集群
$ kubectl config use-context stage
Switched to context "stage".

切換到第二個集群：

$ kubectl config use-context 2xxxxxxxxxxx8
Switched to context "21xxxxxxxxxxxxx9648".

但這種方法不方便的地方就在如果context很多的時候，不方便查看。

用kubecm進行集群切換

安裝篇

基礎環境

mac

k8s 1.16.9

# wget https://github.com.cnpmjs.org/sunny0826/kubecm/releases/download/v0.10.3/kubecm_0.10.3_Darwin_x86_64.tar.gz
# tar xf kubecm_0.10.3_Darwin_x86_64.tar.gz
# mv kubecm /usr/local/bin
# kubecm --help

KubeConfig Manager
_          _
| | ___   _| |__   ___  ___ _ __ ___
| |/ / | | | '_ \ / _ \/ __| '_ \ _ \
|   <| |_| | |_) |  __/ (__| | | | | |
|_|\_\\__,_|_.__/ \___|\___|_| |_| |_|

Find more information at: https://github.com/sunny0826/kubecm

Usage:
kubecm [command]

Available Commands:
add         Merge configuration file with $HOME/.kube/config
alias       Generate alias for all contexts
completion  Generates bash/zsh completion scripts
delete      Delete the specified context from the kubeconfig
help        Help about any command
ls          List kubeconfig
merge       Merge the kubeconfig files in the specified directory
namespace   Switch or change namespace interactively
rename      Rename the contexts of kubeconfig
switch      Switch Kube Context interactively
version     Print version info

Flags:
    --config string   path of kubeconfig (default "/var/root/.kube/config")
-h, --help            help for kubecm

Use "kubecm [command] --help" for more information about a command.

補全

bash

# ource <(kubecm completion bash)
# echo "source <(kubecm completion bash)" >> ~/.bashrc
# source  ~/.bashrc

zsh

# source <(kubecm completion zsh)
# echo "source <(kubecm completion zsh)" >> ~/.zshrc
# source  ~/.zshrc

基本使用

查看

# kubecm ls
+------------+-----------------------------------+---------------+-----------------------+--------------------------------+--------------+
|   CURRENT  |                NAME               |    CLUSTER    |          USER         |             SERVER             |   Namespace  |
+============+===================================+===============+=======================+================================+==============+
|      *     |   283387751775489594-c02e82b6ada  |   kubernetes  |   283387751775489594  |   https://3.3.3.3:6443  |    default   |
|            |       f044bd94d251fea1279648      |               |                       |                                |              |
+------------+-----------------------------------+---------------+-----------------------+--------------------------------+--------------+

Cluster check succeeded!
Kubernetes master is running at https://3.3.3.3:6443
Contains components: [scheduler controller-manager etcd-5 etcd-0 etcd-4 etcd-1 etcd-3 etcd-2]

添加

# ll
-rw-r--r--    1 root  wheel   5.9K  8 26 15:39 config.shengchan
-rw-r--r--    1 root  wheel   5.9K 11 18 15:59 config.yufa

# kubecm add -f config.yufa -c -n stage
# kubecm add -f config.shengchan -c -n production
#  kubecm ls
+------------+---------------+-----------------------+--------------------+--------------------------------+--------------+
|   CURRENT  |      NAME     |        CLUSTER        |        USER        |             SERVER             |   Namespace  |
+============+===============+=======================+====================+================================+==============+
|      *     |   production  |   cluster-kg7h48f8tb  |   user-kg7h48f8tb  |   https://3.3.3.3:6443  |    default   |
+------------+---------------+-----------------------+--------------------+--------------------------------+--------------+
|            |     stage     |   cluster-kfc6fbfcdt  |   user-kfc6fbfcdt  |   https://4.4.4.4:6443  |    default   |
+------------+---------------+-----------------------+--------------------+--------------------------------+--------------+

刪除

# kubecm delete
😸 Select:production
👻 True
Context Delete:「production」
「/var/root/.kube/config」 write successful!
+------------+---------------+-----------------------+--------------------+--------------------------------+--------------+
|   CURRENT  |      NAME     |        CLUSTER        |        USER        |             SERVER             |   Namespace  |
+============+===============+=======================+====================+================================+==============+
|     *      |     stage     |   cluster-kfc6fbfcdt  |   user-kfc6fbfcdt  |   https://4.4.4.4:6443  |    default   |
+------------+---------------+-----------------------+--------------------+--------------------------------+--------------+

切換

# kubecm switch
😸 Select:production
「/var/root/.kube/config」 write successful!
+------------+---------------+-----------------------+--------------------+--------------------------------+--------------+
|   CURRENT  |      NAME     |        CLUSTER        |        USER        |             SERVER             |   Namespace  |
+============+===============+=======================+====================+================================+==============+
|      *     |   production  |   cluster-kg7h48f8tb  |   user-kg7h48f8tb  |   https://3.3.3.3:6443  |    default   |
+------------+---------------+-----------------------+--------------------+--------------------------------+--------------+
|            |     stage     |   cluster-kfc6fbfcdt  |   user-kfc6fbfcdt  |   https://4.4.4.4:6443  |    default   |
+------------+---------------+-----------------------+--------------------+--------------------------------+--------------+

Switched to context 「production」