The following updates have been rolled out to all non-production clusters. Notable updates include a major release for Prometheus, bringing a new UI. As usual there’s also improvements across various other add-ons, ensuring enhanced performance and security. Finally we’d also like to remind you again for Actions to take regarding the Grafana AngularJS deprecation!
- amazon-eks-ami v20250116
- aws-ebs-csi-driver v1.38.1-eksbuild.1
- aws-efs-csi-driver v2.1.4
- aws-load-balancer-controller v2.11.0
- Support Load balancer Capacity Unit (LCU) Reservation for ALB and NLBs. For more info checkout what’s new post.
- aws-node-termination-handler v1.23.1
- aws-vpc-cni v1.19.2-eksbuild.1
- cert-manager v1.16.3
- fluent-bit v3.2.4
- gha-runner-scale-set-controller 0.10.1
- This release includes major improvements to the runner provisioning duration. In short, you should see less latency between queueing a workflow run and having a runner available to execute the job.
- grafana-agent v0.43.4
- ingress-nginx v1.12.0
- k8s-dns-node-cache v1.24.0
- karpenter v1.1.1
- keda v2.16.1
- kube-prometheus-stack 68.3.0
- Alertmanager v0.28.0: Add a new Jira integration
- Grafana v11.4.0: Query Cloudwatch Logs Insights with PPL and SQL
- Prometheus v3.1.0:
- Brand new UI and UTF-8 support enabled by default
- Several PromQL changes
- Prometheus Operator v0.79.2
- loki v3.3.2
- oauth2-proxy v7.8.1
- prometheus-pushgateway v1.11.0
- secrets-store-csi-driver v1.4.7
- velero v1.15.2
- vertical-pod-autoscaler v1.2.2
Actions to take
An important change has come to Grafana where the use of AngularJS is being removed in favor of React. Support for AngularJS will be completely removed in Grafana 12, the next major release. As a result you will see warning messages in Grafana dashboards, helping you to identify which panels are using AngularJS plugins.
Many of our customers deploy their own custom dashboards, so we recommend you to check these for AngularJS plugin warnings and update them accordingly. For more information on how to do this, you can refer to the Grafana pages on this topic:
- Removal of AngularJS support in Grafana: what you need to know (blog)
- List of lugins using AngularJS and alternatives to use
Often, you can just click the Try migration button in the warning and Grafana will auto-migrate your dashboard. After verifying the results, you can save your updated dashboard:
- If using persistence, just click save and you’re done
- If not using persistence, instead relying on dynamic ConfigMap dashboard loading, export the dashboard’s json and backport it to your ConfigMap.
The most commonly used core plugins to change are Graph (old), Singlestat, Table (old), and Worldmap. While these can be auto-migrated, you can also update them to the newer versions, for example:
graph->timeseriessinglestat->stat
Some old custom plugins also have found there way as core plugins. For example you can replace grafana-piechart-panel with the build-in Pie Chart.
[!NOTE] While we have updated our own custom dashboards to use newer plugin versions, unfortunately there are a handfull of leftover dashboards, included in the upstream
kube-prometheus-stackthat we deploy and have not yet been updated. You can safely ignore the warnings for these dashboards for now. Affected dashboards are:
- Alertmanager / Overview
- Grafana / Overview
- Node Exporter / MacOS
- Node Exporter / Nodes
- Node Exporter / USE Method / Cluster
- Node Exporter / USE Method / Node
- Prometheus / Overview