-
Notifications
You must be signed in to change notification settings - Fork 205
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[nri-kubernetes] Update version of kube-state-metrics dependency #1076
Comments
Jira CommentId: 201402 Just FYI: Latest version is: [https://artifacthub.io/packages/helm/newrelic/nri-bundle] 5.0.15 |
version 5.0.15 includes kube-state-metrics: version 4.23.0. |
Jira CommentId: 204167 Per [https://github.com/kubernetes/kube-state-metrics/issues/1711,] it looks like this was fixed last November in KSM v2.7.0, which is included in [v4.24.0 |https://github.com/prometheus-community/helm-charts/releases/tag/kube-state-metrics-4.24.0]and later of the KSM helm chart. |
Jira CommentId: 204177 [~bthaker] I'm not sure how this stacks against other planned agent work, but before updating the kube-state-metrics version we deploy as part of the nri-bundle, we'll need to do some testing to make sure the new version works with our integration. |
Jira CommentId: 204858 [~csong] [~jjaramillo] Do we have a manual or automated test for the KSM compatibility check with nri-bundle for every Kubernetes version/release that we support? Any suggestions for the next steps for the ask in this Jira issue? |
Is there any update on this? Version 5.0.24 also includes kube-state-metrics:4.23.0 which forbids using nri-bundle on kubernetes 1.26 |
hi @marinegreff, we are planning how to proceed with this upgrade because this upgrade requires deprecating support for older k8s versions. We'll keep you posted |
Hi @davidgit , thank you for the update. I understand the problem, but I just wanted to mention that older k8s versions are being deprecated quickly for cloud-managed clusters (1.23 is already deprecated, 1.24 is already deprecated for AKS and soon on GKE and EKS). We currently face the deprecation of k8s 1.25 by the end of this year, hence my question.
I'll keep an eye open |
Thanks @marinegreff! We are aware of it and that's one of the main reasons to deprecate its support too |
Are there any timescales for this? |
Hi @catt231, Our plan is to address this during the next couple of weeks. |
<!-- Thank you for contributing to New Relic's Helm charts. Before you submit this PR we'd like to make sure you are aware of our technical requirements: * https://github.com/newrelic-experimental/helm-charts/blob/master/CONTRIBUTING.md#technical-requirements For a quick overview across what we will look at reviewing your PR, please read our review guidelines: * https://github.com/newrelic-experimental/helm-charts/blob/master/REVIEW_GUIDELINES.md Following our best practices right from the start will accelerate the review process and help get your PR merged quicker. When updates to your PR are requested, please add new commits and do not squash the history. This will make it easier to identify new changes. The PR will be squashed anyways when it is merged. Thanks. For fast feedback, please @-mention maintainers that are listed in the Chart.yaml file. Please make sure you test your changes before you push them. Once pushed, a Github Action will run across your changes and do some initial checks and linting. These checks run very quickly. Please check the results. We would like these checks to pass before we even continue reviewing your changes. --> #### What this PR does / why we need it: Updates KSM version installed with the bundle, so we can fully support the latest Kubernetes versions (>v1.23) #### Which issue this PR fixes - fixes #1076 #### Special notes for your reviewer: #### Checklist - [x] Title of the PR starts with chart name (e.g. `[mychartname]`)
Bug description
Kubernetes: 1.26.2-gke.1000
nri-bundle: 5.0.11 which includes kube-state-metrics: version 4.23.0. This cluster version has deprecated API autoscaling.apiserver.k8s.io/v2beta2 looks like kube-state-metrics version 4.23.0 does not know about it, and generate lot of errors:
E0504 14:57:41.765197 1 reflector.go:138] pkg/mod/k8s.io/client-go@v0.24.4/tools/cache/reflector.go:167: Failed to watch *v2beta2.HorizontalPodAutoscaler: failed to list *v2beta2.HorizontalPodAutoscaler: the server could not find the requested resource 86 W0504 14:58:12.635910 1 reflector.go:324] pkg/mod/k8s.io/client-go@v0.24.4/tools/cache/reflector.go:167: failed to list *v2beta2.HorizontalPodAutoscaler: the server could not find the requested resource 85 E0504 14:58:12.635943 1 reflector.go:138] pkg/mod/k8s.io/client-go@v0.24.4/tools/cache/reflector.go:167: Failed to watch *v2beta2.HorizontalPodAutoscaler: failed to list *v2beta2.HorizontalPodAutoscaler: the server could not find the requested resource 84 W0504 14:59:04.231191 1 reflector.go:324] pkg/mod/k8s.io/client-go@v0.24.4/tools/cache/reflector.go:167: failed to list *v2beta2.HorizontalPodAutoscaler: the server could not find the requested resource
Version of Helm and Kubernetes
Helm: v3.9.3
Kubernetes: 1.26.2-gke.1000 (GCP GKE)
Which chart?
nri-bundle
version: 5.0.11
What happened?
We updated claster to version 1.26.2-gke.1000 and nri-bundle to version 5.0.11, after that i can see lot of errors in pod newrelic-kube-state-metrics
W0504 14:40:45.760497 1 reflector.go:324] pkg/mod/k8s.io/client-go@v0.24.4/tools/cache/reflector.go:167: failed to list *v2beta2.HorizontalPodAutoscaler: the server could not find the requested resource
What you expected to happen?
Update to latest version newrelic-kube-state-metrics in nri-bundle
How to reproduce it?
GKE: 1.26.2 + nri-bundle: 5.0.11
The text was updated successfully, but these errors were encountered: