2026-01-27
Upgraded cluster add-ons
#add-on #kubernetes #update #upgrade #component #eks
The following updates have been rolled out to all non-production clusters, and will be deployed to production in the coming week. As usual there are also improvements across various other add-ons, ensuring enhanced performance and security:
- amazon-eks-ami v20260114
- aws-ebs-csi-driver v1.54.0-eksbuild.1
- Node-local volume support enabling a single cluster-wide PV/PVC to mount node-specific EBS volumes with ReadWriteMany access
- Fix for rare volume leak edge case in CreateVolume error handling
- aws-efs-csi-driver v2.3.0
- aws-load-balancer-controller v2.17.1
- aws-mountpoint-s3-csi-driver v2.2.1-eksbuild.1
- aws-vpc-cni v1.21.1-eksbuild.1
- Support for Cluster Network Policy in AWS Network Policy Agent, allowing cluster administrators to define network policies with Layer 3 and 4 rules at the cluster scope
- cert-manager v1.19.2
- eks-node-monitoring-agent v1.4.3-eksbuild.2
- external-dns v0.20.0
- New flags to allow migration of OwnerID
- Custom annotation prefix support for split horizon DNS
- fluent-bit v4.2.2
- flux v2.7.5
- Image Automation APIs reach General Availability (GA)
- New ExternalArtifact and ArtifactGenerator APIs for advanced source composition and decomposition patterns
- grafana-loki v3.6.4
- grafana-tempo v2.9.0
- TraceQL metrics sampling with new query hints for faster approximate results
- Cached querier responses now excluded from SLO metrics for more accurate measurements
- ingress-nginx v1.14.1
- karpenter v1.8.5
- Fix for hash collisions while resolving subnets, security groups and AMIs from NodeClass selectors
- Updated AWS API data for new g7e instance family
- keda v2.18.3
- kube-prometheus-stack 81.2.0
- Grafana v12.3.1
- Redesigned logs visualization with improved performance and new display options
- Prometheus v3.9.1
- Native Histograms is no longer experimental and now a stable feature
- Prometheus Operator v0.88.0
- Native histograms scraping support via new
scrapeNativeHistogramsfield in multiple CRDs
- Native histograms scraping support via new
- Alertmanager v0.30.1
- Added incident.io and Mattermost receiver support in Alertmanager config
- Grafana v12.3.1
- kube-proxy v1.33.7-eksbuild.2
- metrics-server v0.8.0-eksbuild.6
- oauth2-proxy v7.13.0
- secrets-store-csi-driver v1.5.5
- snapshot-controller v8.4.0-eksbuild.3
- traefik v3.6.7
- velero v1.17.1