EKS upgrade service · Quoted after review
An upgrade is a sequence, not a button.
Get EKS upgrade help grounded in the cluster you actually run. We start with readiness, work through the dependencies and agree a change plan with explicit acceptance and recovery criteria.
Check the whole path
Control plane & APIs
Identify the supported minor-version sequence and removed APIs. Inspect available upgrade insights and actual workload definitions; a quiet test environment can miss infrequent callers.
Add-ons & compute
Check VPC CNI, CoreDNS, kube-proxy and storage drivers against the target version. Review managed or self-managed nodes, AMIs and Karpenter compatibility before replacing compute.
Workloads & disruption
Test admission webhooks, controllers, ingress, storage and application behavior. Review PDBs, replica counts, spare capacity and how traffic behaves during node drains.
A deliberate change process
- Review first. Establish the inventory, blockers and support timeline.
- Prepare and test. Fix incompatible manifests, agree add-on order and rehearse against representative workloads.
- Approve the window. Your team signs off on the exact changes, access, owners and recovery options.
- Execute the agreed scope. Upgrade one control-plane minor version at a time, then reconcile compatible add-ons and compute using the agreed sequence.
- Validate and hand over. Check workload health, application paths, alerts and version state against the acceptance criteria.
AWS announced EKS version rollback on July 1, 2026: eligible clusters can revert to the previous minor version within 7 days of an upgrade, at no extra charge, after readiness checks for API compatibility, version skew, add-on compatibility and cluster health. EKS Auto Mode rolls back worker nodes first. Rollback does not restore application data or automatically undo every add-on or manifest change. The bounded window is not a substitute for preflight testing; check eligibility in the EKS user guide. See the AWS announcement and EKS user guide.
What your team owns
You keep authority over your AWS account and the production change.
Provide application owners, representative tests and a decision-maker for the change window. Confirm backup and restore evidence, dependencies outside Kubernetes and the business tolerance for interruption. Approve write access only for the implementation scope; the initial review remains read-only.
Use the upgrade checklist →Check your support-cost exposure · Review access boundaries
Questions before you start
How much does an upgrade cost?
Upgrade work is quoted after the $1,500 readiness review. The quote depends on version distance, add-ons, node provisioning, workload constraints and testing needs. There is no fixed price before those dependencies are understood.
Can you roll back an EKS control-plane upgrade?
AWS announced EKS version rollback on July 1, 2026: eligible clusters can revert to the previous minor version within 7 days of an upgrade, at no extra charge, after readiness checks for API compatibility, version skew, add-on compatibility and cluster health. EKS Auto Mode rolls back worker nodes first. Rollback does not restore application data or automatically undo every add-on or manifest change. The bounded window is not a substitute for preflight testing; check eligibility in the EKS user guide.
Can we skip Kubernetes minor versions?
EKS control-plane upgrades proceed one minor version at a time. Node and add-on compatibility must be checked for each step. An upgrade plan accounts for the whole path rather than only the destination.
Is zero downtime guaranteed?
No. Availability depends on workload behavior, replicas, capacity, disruption constraints and external dependencies. We agree checks and acceptance criteria with your team instead of promising an outcome we cannot verify.
A clear next step
Know what needs attention.
Then decide what to change.
A one-week, read-only EKS Production Readiness Review. $1,500. A prioritized plan your team can use.