<p>Kubernetes version 1.35 introduced several new features and bug fixes, and AWS is excited to announce that you can now use <a href="https://aws.amazon.com/eks/">Amazon Elastic Kubernetes Service</a> (EKS) and <a href="https://aws.amazon.com/eks/eks-distro/">Amazon EKS Distro</a> to run Kubernetes version 1.35. Starting today, you can create new EKS clusters using version 1.35 and upgrade existing clusters to version 1.35 using the EKS console, the eksctl command line interface, or through an infrastructure-as-code tool.<br> <br> Kubernetes version 1.35 introduces several key improvements, including In-Place Pod Resource Updates allowing CPU and memory adjustments without restarting Pods, and PreferSameNode Traffic Distribution prioritizing local endpoints before routing to remote nodes for reduced latency. The release brings Node Topology Labels via Downward API enabling Pods to access region and zone information without API server queries, alongside Image Volumes delivering data artifacts like AI models using OCI container images. To learn more about the changes in Kubernetes version 1.35, see our <a href="https://docs.aws.amazon.com/eks/latest/userguide/kubernetes-versions-standard.html">documentation</a> and the Kubernetes project <a href="https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.34.md">release notes.</a><br> <br> EKS now supports Kubernetes version 1.35 in all the AWS Regions where EKS is available, including the AWS GovCloud (US) Regions.<br> <br> You can learn more about the Kubernetes versions available on EKS and instructions to update your cluster to version 1.35 by visiting EKS documentation. You can use <a href="https://docs.aws.amazon.com/eks/latest/userguide/cluster-insights.html">EKS cluster insights</a> to check if there are any issues that can impact your Kubernetes cluster upgrades. EKS Distro builds of Kubernetes version 1.35 are available through ECR Public Gallery and GitHub. Learn more about the EKS version lifecycle policies in the <a href="https://docs.aws.amazon.com/eks/latest/userguide/kubernetes-versions.html">documentation</a>.</p>

What's Changed • fix(transform): update parent default logic for schemas with required fields by @jakobmoellerdev in https://github.com/kubernetes-sigs/kro/pull/983 Full Changelog: https://github.com/kubernetes-sigs/kro/compare/v0.8.0...v0.8.1

What's Changed • Replace panic in CRD Graph Builder with proper error handling by @AnshulPatil2005 in https://github.com/kubernetes-sigs/kro/pull/901 • docs: correct apiVersion for Application example by @birapjr in https://github.com/kubernetes-sigs/kro/pull/899 • docs: cutoff 0.7.1 by @jakobmoellerdev in https://github.com/kubernetes-sigs/kro/pull/896 • cleanup: use variadic append for enum values by @PhantomInTheWire in https://github.com/kubernetes-sigs/kro/pull/910 • fix: Fixed Quickstart instance.yaml sample on document by @ricky9408 in https://github.com/kubernetes-sigs/kro/pull/913 • Prevent random.* from being classified as a resource by @benzaidfoued in https://github.com/kubernetes-sigs/kro/pull/919 • Improve CRD cleanup skip log clarity by @skools-here in https://github.com/kubernetes-sigs/kro/pull/923 • Docs: manifests download URL changed to match actual URL by @hatofmonkeys in https://github.com/kubernetes-sigs/kro/pull/925 • chore(deps): bump qs from 6.13.0 to 6.14.1 in /website in the npm_and_yarn group across 1 directory by @dependabot[bot] in https://github.com/kubernetes-sigs/kro/pull/929 • Fix url in kubectl commands for upgrade and delete using kubectl manifests by @Fsero in https://github.com/kubernetes-sigs/kro/pull/938 • fix: Support []object and map[string]object types in RGD schema by @kunalvirwal in https://github.com/kubernetes-sigs/kro/pull/939 • fix: Convert CEL type to Go type recursively by @cirias in https://github.com/kubernetes-sigs/kro/pull/940 • chore: cache compiled CEL programs by @bschaatsbergen in https://github.com/kubernetes-sigs/kro/pull/943 • chore: bump pinned version by @bschaatsbergen in https://github.com/kubernetes-sigs/kro/pull/945 • Update AWS cluster management example to use EKS capabilities by @iamahgoub in https://github.com/kubernetes-sigs/kro/pull/946 • feat: add bytes conversion to go native types by @shabbskagalwala in https://github.com/kubernetes-sigs/kro/pull/951 • chore: bump kro in kro command by @tjamet in https://github.com/kubernetes-sigs/kro/pull/952 • test: improve unit test coverage for pkg/graph/variable by @shivansh-gohem in https://github.com/kubernetes-sigs/kro/pull/949 • chore: bump controller-runtime to v0.23.0 and k8s deps to v0.35.0 by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/956 • feat: add controller warmup support for improved leader failover by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/955 • fix(dag): preserve vertex order when dependencies are satisfied by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/958 • feat: add DurationType and TimestampType conversion to go native types by @shabbskagalwala in https://github.com/kubernetes-sigs/kro/pull/960 • chore(deps): bump qs from 6.13.0 to 6.14.1 in /website in the npm_and_yarn group across 1 directory by @dependabot[bot] in https://github.com/kubernetes-sigs/kro/pull/964 • chore: dependency bumps (go1.25.6, golangci-lint, helm, ko, chainsaw, and more) by @jakobmoellerdev in https://github.com/kubernetes-sigs/kro/pull/963 • chore(deps): bump lodash from 4.17.21 to 4.17.23 in /website in the npm_and_yarn group across 1 directory by @dependabot[bot] in https://github.com/kubernetes-sigs/kro/pull/965 • add support for labels and annotations in the generated CRDs by @cnvergence in https://github.com/kubernetes-sigs/kro/pull/916 • Cluster management example: ensure access to workload cluster is granted before Argo CD secret creation by @iamahgoub in https://github.com/kubernetes-sigs/kro/pull/966 • feat: add Collections support + runtime/controller rewrite by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/936 • fix(schema): use date-time format for timestamps by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/973 • Regenerate CRDs to reflect new schema.metadata field by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/977 • fix(schema): preserve nested array/object paths in status schema by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/972 • docs(cel): explain multiline expressions and YAML chomping by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/974 • docs: expand collections gotchas and cross-references by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/971 • fix(website): version CRD with docs snapshots by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/978 • fix(graph): reject cluster-scoped resources with namespace set by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/976 • KREP-002: add declarative resource collections support by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/679 • refactor(instance): align node state tracking by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/970 • feat(crd): Prevent breaking changes in RGD schemas by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/352 • feat: support recursive references in custom types (Supersedes #714) by @shivansh-gohem in https://github.com/kubernetes-sigs/kro/pull/950 • fix(release): capture GIT_VERSION once to prevent -dirty suffix in LDFAGS by @a-hilaly in https://github.com/kubernetes-sigs/kro/pull/982 • feat: add early validation for apiVersion and kind to fail fast by @antcybersec in https://github.com/kubernetes-sigs/kro/pull/980 New Contributors • @AnshulPatil2005 made their first contribution in https://github.com/kubernetes-sigs/kro/pull/901 • @birapjr made their first contribution in https://github.com/kubernetes-sigs/kro/pull/899 • @PhantomInTheWire made their first contribution in https://github.com/kubernetes-sigs/kro/pull/910 • @ricky9408 made their first contribution in https://github.com/kubernetes-sigs/kro/pull/913 • @benzaidfoued made their first contribution in https://github.com/kubernetes-sigs/kro/pull/919 • @skools-here made their first contribution in https://github.com/kubernetes-sigs/kro/pull/923 • @hatofmonkeys made their first contribution in https://github.com/kubernetes-sigs/kro/pull/925 • @Fsero made their first contribution in https://github.com/kubernetes-sigs/kro/pull/938 • @kunalvirwal made their first contribution in https://github.com/kubernetes-sigs/kro/pull/939 • @cirias made their first contribution in https://github.com/kubernetes-sigs/kro/pull/940 • @shabbskagalwala made their first contribution in https://github.com/kubernetes-sigs/kro/pull/951 • @shivansh-gohem made their first contribution in https://github.com/kubernetes-sigs/kro/pull/949 • @cnvergence made their first contribution in https://github.com/kubernetes-sigs/kro/pull/916 Full Changelog: https://github.com/kubernetes-sigs/kro/compare/v0.7.1...v0.8.0

What's Changed • feat: allow disabling mount of SOCI directory by @fletcherw in https://github.com/awslabs/amazon-eks-ami/pull/2588 • Add retry to net.LookupHost() in nodeadm by @nkvetsinski in https://github.com/awslabs/amazon-eks-ami/pull/2585 Full Changelog: https://github.com/awslabs/amazon-eks-ami/compare/v20260114...v20260120

AWS EBS CSI Driver CHANGELOG See CHANGELOG for full list of changes

Note There are no changes to the AMI template in this release.

A Helm chart for AWS EBS CSI Driver

📚 Quick Links v3.0.0 (requires Kubernetes 1.22+) Image: public.ecr.aws/eks/aws-load-balancer-controller:v3.0.0 Documentation Thanks to all our contributors!💜💜💜 🎉 Gateway API is Now GA! We are excited to announce that Gateway API support is now Generally Available (GA) in AWS Load Balancer Controller v3.0.0! This milestone marks the production-ready status of Gateway API features for managing AWS Application Load Balancers and Network Load Balancers through the Kubernetes Gateway API. We encourage you to try it out and welcome any feedback via GitHub Issues. For more gateway api details, please refer to our live doc. ⚠️ Action Required Webhook Certificate Issue • Issue: #4359 attempted to fix a race condition in webhook certificate renewal but introduced a critical bug. The aws-load-balancer-tls Secret persists but the Certificate that owned and renewed it was removed, causing eventual certificate expiration and webhook TLS failures (#4541). This prevents the controller from updating target group bindings and can lead to outages. This bug impacts users deploying with Helm and utilizing the enableCertManager=true flag. • Action for users staying on older versions: Set keepTLSSecret=false in your helm chart to mitigate the issue • Action for users upgrading to v3.0.0: No action required - the fix is included in this release 🔧 Enhancements and Fixes • Helm Chart Version Alignment: Helm chart version now aligns with LBC version. Previously, LBC v2.x used Helm chart v1.x (e.g., LBC v2.17 = Helm v1.17). Starting with v3.0.0, both versions match. • Gateway Deletion: Removed route count check when deleting gateways, allowing deletion of gateways with attached routes (#4549) • Subnet Ordering: Fixed subnet order preservation when using aws-load-balancer-subnets annotation - now maintains requested order instead of non-deterministic ordering (#4504) • AZ Mismatch Fix: Fixed orphaned targets issue caused by AvailabilityZone mismatch in refreshUnhealthyTargets - targets are now properly deregistered regardless of cached AZ (#4544) • NLB Target Group Limit: Fixed target group association limit error for weighted configs by including base service UID in target group name generation (#4540) • Listener Error Propagation: Fixed target group tuple error messages not being propagated to end users (#4545) • Webhook Certificate: Reverted race condition fix in webhook certificate renewal that caused issues (#4542) 📋 Full Changelog • Revert "fix: Race condition in webhook certificate renewal with cert-… by @zac-nixon in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4542 • Fix NLB target group association limit issue for weighted configs by @shraddhabang in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4540 • Fix AZ mismatch in refreshUnhealthyTargets causing orphaned targets by @MinhNguyen-at in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4544 • Update model_build_listener.go by @zac-nixon in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4545 • Fix: preserve requested order for subnets when using aws-load-balancer-subnets annotation by @nelsen129 in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4504 • Remove KeepTLS parameter in helm chart by @zac-nixon in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4548 • [gateway api] remove route count check for deleting gateway by @zac-nixon in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4549 • [feat gateway-api]update gw api doc by @shuqz in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4550 • cut v3.0.0 release by @shuqz in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4551 New Contributors • @MinhNguyen-at made their first contribution in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4544 • @nelsen129 made their first contribution in https://github.com/kubernetes-sigs/aws-load-balancer-controller/pull/4504 Full Changelog: https://github.com/kubernetes-sigs/aws-load-balancer-controller/compare/v2.17.1...v3.0.0

This blog post dives deeper into optimizing Amazon EKS Auto Mode clusters using AWS&nbsp;Graviton and Amazon EC2&nbsp;Spot instances. AWS customers adopt AWS Graviton instances to achieve up to 40% higher price-performance ratio and up to 60% less energy to meet their sustainability goals. Additionally, AWS customers use Amazon EC2 Spot instances for eligible workloads to save up to 90% on Amazon Elastic Compute Cloud (Amazon EC2) On-Demand costs.

Commits • f29079c: (Backport v1.8.x) chore: update aws api data for new g7e instance family (#8869) (Joshua Guo) #8869

In this blog post, we show how to create and manage a fleet of Amazon Elastic Kubernetes Service (Amazon EKS) clusters using Kube Resource Orchestrator (kro), AWS Controllers for Kubernetes (ACK), and Argo CD. These tools allow you to implement a GitOps-based cluster management solution to increase productivity and improve consistency and standardization by using the Kubernetes API for end-to-end operations.

In this post, we walk you through building an end-to-end recovery system, showing you how to configure traffic shifting that protects your applications from Availability Zone impairments while maintaining the high availability that your business demands.

What's Changed • fix(neuron): install aws-neuronx-dkms-2.21 at boot on inf1 by @mselim00 in https://github.com/awslabs/amazon-eks-ami/pull/2579 • chore(docs): backfill CHANGELOG.md for release v20251217 by @mselim00 in https://github.com/awslabs/amazon-eks-ami/pull/2583 • Revert "fix(al2023): Only enable DNS for the primary interface to avoid duplicate entries (#2578)" by @shvbsle in https://github.com/awslabs/amazon-eks-ami/pull/2584 Full Changelog: https://github.com/awslabs/amazon-eks-ami/compare/v20260107...v20260114

Chores • bump upstream karpenter to v1.8.2 (#8847) #8847 (Jigisha Patil)

Tests • Bump upgrade commit pin (#8768) #8769 (Ryan Mistretta) Chores • Bump Upstream Karpenter to include patch changes (#8834) #8834 (Amanuel Engeda) • update static instance data (#8833) #8833 (Jason Deal) • Bump Upstream Karpenter to v1.8.1 (#8839) #8839 (Amanuel Engeda)

This blog post examines how Salesforce, operating one of the world's largest Kubernetes deployments, successfully migrated from Cluster Autoscaler to Karpenter across their fleet of 1,000 plus Amazon Elastic Kubernetes Service (Amazon EKS) clusters.

What's Changed • feat(nodeadm): add nodeadm config dump command by @vflaux in https://github.com/awslabs/amazon-eks-ami/pull/2107 • fix(nodeadm): clarify missing cluster ID error for outposts by @HusainZafar in https://github.com/awslabs/amazon-eks-ami/pull/2573 • fix(al2023): Only enable DNS for the primary interface to avoid duplicate entries by @shvbsle in https://github.com/awslabs/amazon-eks-ami/pull/2578 Full Changelog: https://github.com/awslabs/amazon-eks-ami/compare/v20251217...v20260107

v2.17.1 (requires Kubernetes 1.22+) Documentation Image: public.ecr.aws/eks/aws-load-balancer-controller:v2.17.1 Thanks to all our contributors! 😊 🚀 What's New QUIC Protocol Support: Added QUIC protocol support for Gateway API and Service API JWT Validation: Support for JWT validation in Gateway API Default Load Balancer Scheme: Added support for specifying —default-load-balancer-scheme flag in Helm chart 🔧 Enhancements and Fixes Bug Fixes • Helm Chart: Duplicated CRD in helm kustomization Documentation Updates • Service Actions: Fixed service.beta.kubernetes.io/actions example in documentation • Conformance Report: Generated v2.17.0 conformance test report Changelog since v2.17.0 • Merge Main to release-2.17 (#4533, @wweiwei-li) • [GW API] Add QUIC support (#4530, @zac-nixon) • docs: fix service.beta.kubernetes.io/actions example (#4529, @davidxia) • Add E2E tests for QUIC support in Service API. (#4527, @zac-nixon) • feat(chart): add support for specifying —default-load-balancer-scheme flag (#4141, @ysam12345) • fix: duplicated CRD in helm kustomization (#4518, @wweiwei-li) • generate v2.17.0 conformance test report (#4521, @shuqz) • [feat gw-api]support jwt validation (#4516, @shuqz) • fix helm chart version (#4515, @wweiwei-li)

This post explores using the Amazon CloudWatch monitoring, including new Amazon EKS metrics and the CloudWatch Observability Operator, to gain deeper visibility into cluster operations, detect issues, understand bottlenecks, and maintain healthy EKS clusters.

Amazon Web Services (AWS) announced the availability of DNS-based and Admin network policies for Amazon Elastic Kubernetes Service (EKS) Auto mode and Admin network policies for both EKS Auto mode and EKS on Amazon Elastic Compute Cloud (EC2), providing enhanced capabilities to secure network traffic both within your clusters and to external endpoints. In this post, we explore practical use cases that demonstrate how these policies solve real-world challenges and remove the need to rely on third-party software across different deployment scenarios, from securing access to external services to hybrid cloud integration and multi-tenant environments.

This post details how platform engineering teams can build an assurance pipeline for Amazon EKS deployments, incorporating validation frameworks that verify configurations, test infrastructure as code (IaC), assess application resilience, and establish compliance with organizational standards.

This post looks at various options for container image caching, model training, and inferencing workloads. This post also discusses various storage options such as Amazon Simple Storage Service (Amazon S3), FSx for lustre, S3 Express One Zone, and Amazon S3 Connector for PyTorch.

In this post, we focus on observing and scaling ML operations (MLOps) infrastructure on Kubernetes. MLOps platforms running on Amazon EKS provide powerful built-in capabilities for logging, monitoring, and alerting that are essential for maintaining healthy ML systems at scale.

In this post we demonstrate how using GitLab Runners on EKS Auto Mode, combined with Amazon Elastic Compute Cloud (Amazon EC2) Spot Instances, can deliver enterprise-scale CI/CD capabilities while achieving up to 90% cost reduction when compared to traditional deployment models. This approach not only optimizes operational expenses, but also provides resilient, scalable pipeline execution.