当前位置：文江博客话题详情

Prometheus kubernetes-helm kustomize

用kustomize部署kube-prometheus-stack not

发布于 2025-01-27 01:51:35 字数 1581 浏览 3 评论 0 原文

我正在尝试部署kube-prometheus-stack 。但是我从部署中得到：

unable to recognize ".": no matches for kind "Alertmanager" in version "monitoring.coreos.com/v1"
unable to recognize ".": no matches for kind "Prometheus" in version "monitoring.coreos.com/v1"
unable to recognize ".": no matches for kind "PrometheusRule" in version "monitoring.coreos.com/v1"
...(and so on)

完整输出下：

我试图在牧场主桌面上部署本地，但是在docker桌面上，我得到了相同的。

我做了什么：该堆栈的文件也没有kustomize文件，因此我进行了清单

helm template prometheus-community/kube-prometheus-stack > prometheus.yaml

：没有工作后，我试图将直接放在我的kustomize

 helmCharts:
 - name: kube-prometheus-stack 
   repo: https://prometheus-community.github.io/helm-charts
   version: 35.0.3
   releaseName: prometheus

中

kubectl kustomize . --enable-helm | kubectl apply -f -

。

如果使用helm，则

helm install prometheus prometheus-community/kube-prometheus-stack

可以使用以下操作。

有趣的是，当我卸载它时：

helm uninstall prometheus

然后通过kustomize

kubectl apply -k .

的工作再次部署它，但这不是我需要的解决方案。那我做错了什么？

原文

I am trying to deploy kube-prometheus-stack
https://github.com/prometheus-community/helm-charts/tree/main/charts/kube-prometheus-stack through kustomize. But I am getting from the deployment:

unable to recognize ".": no matches for kind "Alertmanager" in version "monitoring.coreos.com/v1"
unable to recognize ".": no matches for kind "Prometheus" in version "monitoring.coreos.com/v1"
unable to recognize ".": no matches for kind "PrometheusRule" in version "monitoring.coreos.com/v1"
...(and so on)

Full output under: https://app.warp.dev/block/JJwOYMJZng9CyBdVlBaIIF

I tried to deploy local on rancher desktop but on docker desktop I get the same.

What I did:
There are nor kustomize file for this stack so I take the manifest with:

helm template prometheus-community/kube-prometheus-stack > prometheus.yaml

after this didn't work I tried to take directly in my kustomize.yaml the helm with:

 helmCharts:
 - name: kube-prometheus-stack 
   repo: https://prometheus-community.github.io/helm-charts
   version: 35.0.3
   releaseName: prometheus

and started kustomize with:

kubectl kustomize . --enable-helm | kubectl apply -f -

both have the same problem.

If use helm with:

helm install prometheus prometheus-community/kube-prometheus-stack

its working.

What interesting is that when I uninstall it:

helm uninstall prometheus

and then deploy it again through kustomize

kubectl apply -k .

Its working, but it is not the solution which I need. So what am I doing wrong?

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

浮云落日 2025-02-03 01:51:35

在尝试部署“使用”这些CRD的对象之前，您需要给Kubernetes一些时间来创建CRD。

下次只做一次申请，等待10秒，然后再次申请。甚至几次，直到错误消失。

回复收藏 0 原文

花间憩 2025-02-03 01:51:35

我在Helmfile也有类似的问题。当CRD与资源同时部署（在同一图表中甚至是Helmfile依赖性）时，Helmfile验证将失败：

Error: unable to build kubernetes objects from release manifest: [unable to recognize "": no matches for kind "Alertmanager" in version "monitoring.coreos.com/v1", unable to recognize "": no matches for kind "Prometheus" in version "monitoring.coreos.com/v1", unable to recognize "": no matches for kind "ServiceMonitor" in version "monitoring.coreos.com/v1"]

但是，Kubernetes代理完全能够同时处理CRD/资源创建。使用Helmfile的解决方案是禁用第一次部署的验证：

helmfile.yaml

releases:
- name: prometheus
  disableValidation: true # Missing CRDs on first deploy
  installed: true
  values:
    - values.yaml

这是我到目前为止唯一发现的解决方法，
我对Kustomize不太熟悉，但是我想也必须有一种跳过验证的方法吗？

希望这对

I had a similar problem with Helmfile. When the CRDs are deployed at the same time (in the same chart or even as Helmfile dependency) as the resources, Helmfile validation will fail :

Error: unable to build kubernetes objects from release manifest: [unable to recognize "": no matches for kind "Alertmanager" in version "monitoring.coreos.com/v1", unable to recognize "": no matches for kind "Prometheus" in version "monitoring.coreos.com/v1", unable to recognize "": no matches for kind "ServiceMonitor" in version "monitoring.coreos.com/v1"]

However, Kubernetes agent is perfectly capable of handling crd/resource creation at the same time. The solution with Helmfile was to disable validation on first deploy :

helmfile.yaml

releases:
- name: prometheus
  disableValidation: true # Missing CRDs on first deploy
  installed: true
  values:
    - values.yaml

It's the only workaround I have found so far,
I'm not very familiar with Kustomize, but I suppose there must be a way to skip validation as well ?

Hope this helps anyway

回复收藏 0 原文

南城旧梦 2025-02-03 01:51:35

我通过设置 helmcharts.includecrds：true 在 kustomization.yaml 中解决了这个问题。

apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization
namespace: monitoring

helmCharts:
  - name: kube-prometheus-stack
    repo: https://prometheus-community.github.io/helm-charts
    version: 45.28.1
    includeCRDs: true
    releaseName: kube-prometheus-stack
    namespace: monitoring
    valuesFile: values.yaml

I solved this by setting helmCharts.includeCRDs: true in kustomization.yaml.

apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization
namespace: monitoring

helmCharts:
  - name: kube-prometheus-stack
    repo: https://prometheus-community.github.io/helm-charts
    version: 45.28.1
    includeCRDs: true
    releaseName: kube-prometheus-stack
    namespace: monitoring
    valuesFile: values.yaml

回复收藏 0 原文

~没有更多了~