以对话的形式管理你的Kubernetes集群

BotKube

BotKube 是一个用于监控和调试 Kubernetes 集群的消息传递工具。 BotKube 可以与多个消息传递平台(如 Slack、Mattermost 或 Microsoft Teams)集成,以帮助您监控 Kubernetes 集群、调试关键部署,并通过对 Kubernetes 资源运行检查来提供标准实践建议。
以对话的形式管理你的Kubernetes集群

安装BotKube

1、本文使用slack作为消息接收平台,需要自行注册slack,拿到Access token

xoxb-2169032185141-2256603089394-qeLWxe0sUtwfdlwnk6VWbw11

2、将 BotKube 用户添加到 Slack 频道
将 BotKube 应用程序安装到您的 Slack 工作区后,您会看到一个名为“BotKube”的新机器人用户添加到您的工作区中。 将该机器人添加到您想要接收通知的 Slack 频道。
(您可以通过在频道中邀请@BotKube 来添加它)

3、使用helm把BotKube安装到kubernrtes集群

helm repo add infracloudio https://infracloudio.github.io/charts
helm repo update

请自行替换<>内的信息:
helm install --version v0.12.1 botkube --namespace botkube   --set communications.slack.enabled=true   --set communications.slack.channel=<SLACK_CHANNEL_NAME>   --set communications.slack.token=<SLACK_API_TOKEN_FOR_THE_BOT>   --set config.settings.clustername=<CLUSTER_NAME>   --set config.settings.kubectl.enabled=<ALLOW_KUBECTL>   --set image.repository=infracloudio/botkube   --set image.tag=v0.12.1   infracloudio/botkube


- SLACK_CHANNEL_NAME 是添加@BotKube 的频道名称
- SLACK_API_TOKEN_FOR_THE_BOT 是将 BotKube 应用程序安装到 Slack 工作区后收到的令牌
- CLUSTER_NAME 是在传入消息中设置的集群名称
- ALLOW_KUBECTL 设置为 true 以允许 BotKube 在集群上执行 kubectl 命令

4、查看安装信息

# helm list -n botkube
NAME   	NAMESPACE	REVISION	UPDATED                                	STATUS  	CHART          	APP VERSION
botkube	botkube  	1       	2021-07-09 10:41:35.813245746 +0800 CST	deployed	botkube-v0.12.1	v0.12.1
# kubectl get all -n botkube
NAME                           READY   STATUS    RESTARTS   AGE
pod/botkube-747ff4dc5d-795hz   1/1     Running   0          2d22h

NAME                      READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/botkube   1/1     1            1           2d22h

NAME                                 DESIRED   CURRENT   READY   AGE
replicaset.apps/botkube-747ff4dc5d   1         1         1       2d22h

5、如果你需要修改配置,可以更新botkube-configmap,默认设置如下

recommendations: true
resources:
- events:
  - create
  - delete
  - error
  name: v1/pods
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: v1/services
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - update
  - delete
  - error
  name: apps/v1/deployments
  namespaces:
    ignore:
    - null
    include:
    - all
  updateSetting:
    fields:
    - spec.template.spec.containers[*].image
    - status.availableReplicas
    includeDiff: true
- events:
  - create
  - update
  - delete
  - error
  name: apps/v1/statefulsets
  namespaces:
    ignore:
    - null
    include:
    - all
  updateSetting:
    fields:
    - spec.template.spec.containers[*].image
    - status.readyReplicas
    includeDiff: true
- events:
  - create
  - delete
  - error
  name: networking.k8s.io/v1beta1/ingresses
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: v1/nodes
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: v1/namespaces
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: v1/persistentvolumes
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: v1/persistentvolumeclaims
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: v1/configmaps
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - update
  - delete
  - error
  name: apps/v1/daemonsets
  namespaces:
    ignore:
    - null
    include:
    - all
  updateSetting:
    fields:
    - spec.template.spec.containers[*].image
    - status.numberReady
    includeDiff: true
- events:
  - create
  - update
  - delete
  - error
  name: batch/v1/jobs
  namespaces:
    ignore:
    - null
    include:
    - all
  updateSetting:
    fields:
    - spec.template.spec.containers[*].image
    - status.conditions[*].type
    includeDiff: true
- events:
  - create
  - delete
  - error
  name: rbac.authorization.k8s.io/v1/roles
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: rbac.authorization.k8s.io/v1/rolebindings
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: rbac.authorization.k8s.io/v1/clusterrolebindings
  namespaces:
    ignore:
    - null
    include:
    - all
- events:
  - create
  - delete
  - error
  name: rbac.authorization.k8s.io/v1/clusterroles
  namespaces:
    ignore:
    - null
    include:
    - all
settings:
  clustername: k8s-2
  configwatcher: true
  kubectl:
    commands:
      resources:
      - deployments
      - pods
      - namespaces
      - daemonsets
      - statefulsets
      - storageclasses
      - nodes
      verbs:
      - api-resources
      - api-versions
      - cluster-info
      - describe
      - diff
      - explain
      - get
      - logs
      - top
      - auth
      - describe
    defaultNamespace:
    - default
    - pro
    - qa
    enabled: true
    restrictAccess: false
  upgradeNotifier: true
ssl:
  enabled: false

使用Slack与BotKube对话

1、先ping下BotKube,看看输出
以对话的形式管理你的Kubernetes集群

2、查看可使用的命令
以对话的形式管理你的Kubernetes集群

3、以对话的形式执行kubectl命令,例如获取pod,更多命令可以自己测试...
以对话的形式管理你的Kubernetes集群

4、验证slack收集kubernetes事件信息,例如创建一个新的pod,这里以nginx为例
我这里直接用lens连接集群,创建一个名为nginx的deployment,
以对话的形式管理你的Kubernetes集群
直接创建即可,然后来到slack可以看到如下信息,即botkube将这个create事件消息推送到了你的slack频道。
以对话的形式管理你的Kubernetes集群
现在删除deployment,再看slack频道,收到delete的消息
以对话的形式管理你的Kubernetes集群
更多好玩的功能,请自行体会吧。。。

以对话的形式管理你的Kubernetes集群

上一篇:.Net Core GRPC报错


下一篇:HTML 链接