Automate EKS Upgrades with AWS DevOps Agent and Kiro

In today's fast-paced cloud environment, managing upgrades for your EKS clusters can be a daunting task. Planned Lifecycle Events (PLEs) signal when a managed service version is nearing the end of its support, and failing to act can lead to service disruptions. Automating these upgrades not only saves time but also enhances reliability and performance across your applications.

The automation pipeline kicks off when AWS Health publishes an AWS_EKS_PLANNED_LIFECYCLE_EVENT to the default Amazon EventBridge bus. An EventBridge rule named eks-health-planned-lifecycle matches this event and invokes the AWS Lambda function devops-agent-health-event. This function extracts critical information and directs the AWS DevOps Agent to utilize the eks-upgrade-planning skill. The agent then discovers the cluster topology, validates the version increment, checks addon compatibility, and determines the upgrade sequence. The output is a structured AWS CDK Change Spec. Following this, a second EventBridge rule triggers the Trigger Upgrade Lambda function, which retrieves the investigation's journal records and dispatches the eks-upgrade.yml GitHub Actions workflow via the GitHub API. This workflow validates the investigation output and installs Kiro CLI in headless mode to apply the necessary changes.

In production, it's crucial to understand the dependencies and the sequence of events in this automation. Ensure that your EventBridge rules are correctly configured to catch the relevant AWS Health events. Pay attention to the output from the AWS DevOps Agent; it contains vital information that must be followed precisely when applying changes. The command for Kiro CLI is straightforward but requires careful adherence to the instructions in the CDK Change Spec to avoid errors. Remember, this automation is designed to enhance your upgrade process, but it relies heavily on the correct setup of your AWS infrastructure and services.