-
Pre-Upgrade Planning: ✔ Backup Everything, especially all YAML files. ✔ Review the release notes for changes and breaking updates. ✔ First, upgrade the staging environment and run it for at least 15 days in non-prod before upgrading production. ✔ Ensure proper observability with logging and monitoring tools.
-
Upgrading the Control Plane: ✔ Upgrade through AWS Console or Terraform (Terraform is preferred). ✔ Upgrade EKS Add-ons as per your requirements.
-
Upgrading Node Groups: ✔ Upgrade node groups one by one to prevent disruptions. ✔ Before upgrading, reduce the number of nodes in the target node group to speed up the process. ✔ Increase node count in another Auto Scaling Group to maintain capacity. ✔ Recheck Cluster Autoscaler logs and ensure it’s functioning correctly. ✔ Monitor application logs and observability tools to detect any issues. ✔ Check logs for DNS and other supporting applications to ensure stability.
This structured approach (in Short) ensures a smooth EKS upgrade with minimal downtime while keeping applications highly available.