平台插件设置

KubeGems 内置了 核心组件 和 Kubernetes 组件 两种类型的插件,它们是 KubeGems 结合云原生生态的重要组成部分

核心组件 内的插件与 KubeGems 核心服务进行了深度绑定，并影响页面功能加载

info

部分核心组件的启（禁）用会影响 kubegems-api 服务的重启。当插件启（禁）用时，插件相关的访问地址会作为全局变量，触发kubegems-api 重新更新，例如 monitoring 插件

Kubernetes 组件 是 KubeGems 与第三方或云原生社区集成的插件，他们并不直接与 KubeGems 核心服务交互。

KubeGems 即将上线新的插件类型 `Operator控制器`，用于统一管理 kubernetes 集群的控制

前置条件

KubeGems 成功安装且运行正常
KubeGems 至少有一个 Kubernetes 集群
需要具备 admin 权限的账号操作

启用插件

进入到【集群管理】的插件列表页面，在界面中，KubeGems 提供了插件的各项配置，您可以根据实际情况进行配置。完成后点击 “确定” 即可完成插件的部署

部分插件存在一些依赖，在启用时请注意插件左边的介绍说明

更新插件

KubeGems 插件默认启用了从在线 Charts 仓库中更新索引，当 Plugin 仓库有更新时，点击【检查插件版本】即可出发更新

提示

当插件在第一次部署成功后，该插件源就会从本地切换到在线源，如果您的集群不能访问公网，就无法更新插件！

卸载插件

在集群的插件页面中，直接点击【卸载】按钮即可完成插件卸载。

扩展

查看插件状态

KubeGems 的父插件均位于kubegems-installer命名空间下，执行以下命令可查看各父插件状态

$ kubectl get plugin -n kubegems-installer                  

NAME                       KIND       STATUS      NAMESPACE             VERSION          APPVERSION       UPGRADETIMESTAMP   AGE
cert-manager               template   Installed   cert-manager          1.8.0                             3d23h              15d
eventer                    template   Installed   kubegems-eventer      1.4.12                            3d23h              15d
gateway                    template   Failed      kubegems-gateway      1.0.0                             15d                15d
global                     template   Installed   kubegems-installer    1.0.0                             3d23h              15d
istio                      template   Installed   istio-system          1.11.7                            3d23h              15d
kubegems                   helm       Installed   kubegems              1.23.0-alpha.2   1.23.0-alpha.2   3d23h              15d
kubegems-edge              helm       Installed   kubegems-edge         1.23.0-alpha.2   1.23.0-alpha.2   3d23h              15d
kubegems-installer         helm       Installed   kubegems-installer    1.23.0-alpha.2   1.23.0-alpha.2   3d23h              15d
kubegems-local             helm       Installed   kubegems-local        1.23.0-alpha.2   1.23.0-alpha.2   3d23h              15d
kubegems-models            template   Installed   kubegems              1.23.0-alpha.2                    3d23h              15d
logging                    template   Installed   kubegems-logging      3.17.6                            3d23h              15d
metrics-server             template   Installed   kube-system           3.8.2                             3d23h              15d
monitoring                 template   Installed   kubegems-monitoring   43.2.1                            3d23h              15d
node-problem-detector      template   Installed   kube-system           2.2.2                             3d23h              15d
opentelemetry              template   Installed   observability         1.0.0                             3d23h              15d
prometheus-cloud-dial      template   Installed   kubegems-monitoring   0.22.0                            3d23h              14d
prometheus-node-exporter   template   Installed   kubegems-monitoring   3.3.0                             3d23h              15d
seldon                     template   Installed   seldon-system         1.0.0                             3d23h              15d
tracing                    template   Installed   observability         2.34.0                            3d23h              15d

各命名空间下的plugin对象为子插件,他们受父插件管理，并将状态消息上报给父插件，可以如下命令查看它们状态

$ kubectl get plugin -n kubegems-monitoring

NAME                           KIND   STATUS      NAMESPACE             VERSION   APPVERSION   UPGRADETIMESTAMP   AGE
kube-prometheus-stack          helm   Installed   kubegems-monitoring   43.2.1    0.61.1       15d                15d
prometheus-blackbox-exporter   helm   Installed   kubegems-monitoring   7.1.3     0.22.0       14d                14d
prometheus-node-exporter       helm   Installed   kubegems-monitoring   3.3.0     1.3.1        15d                15d

查看插件配置

可以使用 edit 或者 get 查看插件配置，例如：

$ kubectl get plugin prometheus-blackbox-exporter -o yaml -n kubegems-monitoring

apiVersion: plugins.kubegems.io/v1beta1
kind: Plugin
metadata:
  creationTimestamp: "2023-01-05T10:50:54Z"
  finalizers:
  - plugins.kubegems.io/finalizer
  generation: 1
  name: prometheus-blackbox-exporter
  namespace: kubegems-monitoring
  resourceVersion: "736290"
  uid: d91689c2-1c48-4a64-b1cd-b6a5046fb6dc
spec:
  kind: helm
  url: https://charts.kubegems.io/kubegems
  values:
    config:
      modules:
        http_2xx:
          http:
            follow_redirects: true
            preferred_ip_protocol: ip4
            valid_http_versions:
            - HTTP/1.1
            - HTTP/2.0
          prober: http
          timeout: 5s
        http_post_2xx:
          http:
            method: POST
            preferred_ip_protocol: ip4
            valid_http_versions:
            - HTTP/1.1
            - HTTP/2
          prober: http
          timeout: 5s
        ping:
          icmp:
            preferred_ip_protocol: ip4
          prober: icmp
          timeout: 5s
        tcp_connect:
          prober: tcp
          timeout: 5s
    hostNetwork: false
    image:
      repository: registry.cn-beijing.aliyuncs.com/kubegems/blackbox-exporter
    namespaceOverride: kubegems-monitoring
    pspEnabled: false
    replicas: 1
    serviceAccount:
      create: false
  version: 7.1.3
status:
  appVersion: 0.22.0
  creationTimestamp: "2023-01-05T10:51:11Z"
  message: |
    See https://github.com/prometheus/blackbox_exporter/ for how to configure Prometheus and the Blackbox Exporter.

    1. Get the application URL by running these commands:
      export POD_NAME=$(kubectl get pods --namespace kubegems-monitoring -l "app.kubernetes.io/name=prometheus-blackbox-exporter,app.kubernetes.io/instance=prometheus-blackbox-exporter" -o jsonpath="{.items[0].metadata.name}")
      export CONTAINER_PORT=$(kubectl get pod --namespace kubegems-monitoring $POD_NAME -o jsonpath="{.spec.containers[0].ports[0].containerPort}")
      echo "Visit http://127.0.0.1:8080 to use your application
      kubectl --namespace kubegems-monitoring port-forward $POD_NAME 8080:$CONTAINER_PORT
  namespace: kubegems-monitoring
  phase: Installed
  resources:
  - apiVersion: v1
    kind: ConfigMap
    name: prometheus-blackbox-exporter
    namespace: kubegems-monitoring
  - apiVersion: v1
    kind: Service
    name: prometheus-blackbox-exporter
    namespace: kubegems-monitoring
  - apiVersion: apps/v1
    kind: Deployment
    name: prometheus-blackbox-exporter
    namespace: kubegems-monitoring
  upgradeTimestamp: "2023-01-05T10:51:11Z"
  values:
    config:
      modules:
        http_2xx:
          http:
            follow_redirects: true
            preferred_ip_protocol: ip4
            valid_http_versions:
            - HTTP/1.1
            - HTTP/2.0
          prober: http
          timeout: 5s
        http_post_2xx:
          http:
            method: POST
            preferred_ip_protocol: ip4
            valid_http_versions:
            - HTTP/1.1
            - HTTP/2
          prober: http
          timeout: 5s
        ping:
          icmp:
            preferred_ip_protocol: ip4
          prober: icmp
          timeout: 5s
        tcp_connect:
          prober: tcp
          timeout: 5s
    hostNetwork: false
    image:
      repository: registry.cn-beijing.aliyuncs.com/kubegems/blackbox-exporter
    namespaceOverride: kubegems-monitoring
    pspEnabled: false
    replicas: 1
    serviceAccount:
      create: false
  version: 7.1.3#

定位插件状态为`Failed`的报错

kubectl get  plugin <插件名称> -n kubegems-installer -ojsonpath='{.status.message}'

以 loki 插件状态异常举例

# 找到状态未 faild 的插件

$ kubectl get plugin -n kubegems-logging

NAME                       KIND   STATUS      NAMESPACE          VERSION   APPVERSION   UPGRADETIMESTAMP   AGE
logging-operator           helm   Installed   kubegems-logging   3.17.6    3.17.6       31d                144d
logging-operator-logging   helm   Installed   kubegems-logging   3.17.6    3.17.6       22d                144d
loki                       helm   Failed      kubegems-logging   2.11.1    v2.5.0       136d               144d
loki-redis                 helm   Installed   kubegems-logging   16.9.11   6.2.7        144d               144d

# 找到报错消息
kubectl get plugin  -n kubegems-logging loki -ojsonpath='{.status.message}'

cannot patch "loki" with kind StatefulSet: StatefulSet.apps "loki" is invalid: spec: 
Forbidden: updates to statefulset spec for fields other than 'replicas', 'template', and 'updateStrategy' are forbidden

可以看到上述报错是 plugin 在执行更新时，修改的statefulset类型资源不合法导致请求被拒绝

平台插件设置​

info

KubeGems 即将上线新的插件类型 Operator控制器，用于统一管理 kubernetes 集群的控制

前置条件​

启用插件​

部分插件存在一些依赖，在启用时请注意插件左边的介绍说明

更新插件​

提示

卸载插件​

扩展​

查看插件状态​

查看插件配置​

定位插件状态为Failed的报错​

平台插件设置

KubeGems 即将上线新的插件类型 `Operator控制器`，用于统一管理 kubernetes 集群的控制

前置条件

启用插件

更新插件

卸载插件

扩展

查看插件状态

查看插件配置

定位插件状态为`Failed`的报错