AWS eks medium security documentation change
Summary
Updated data plane options to include EKS Auto Mode, added details about automatic AMI updates for CVE fixes and security patches, and corrected terminology from 'EKS Auto Mode' to 'Cluster Autoscaler'
Security assessment
Explicitly mentions automatic deployment of CVE fixes and security patches through Auto Mode AMI updates, which directly addresses security vulnerabilities
Diff
diff --git a/eks/latest/best-practices/reliability.md b/eks/latest/best-practices/reliability.md index 93af27d64..1428585f8 100644 --- a//eks/latest/best-practices/reliability.md +++ b//eks/latest/best-practices/reliability.md @@ -36 +36,5 @@ In EKS, AWS is responsible for the reliability of the Kubernetes control plane. -The responsibility for the data plane’s reliability is shared between you, the customer, and AWS. EKS offers three worker node options for deploying the Kubernetes data plane. Fargate, which is the most managed option, handles provisioning and scaling of the data plane. The second option, managed nodes groups, handles provisioning, and updates of the data plane. And finally, self-managed nodes is the least managed option for the data plane. The more AWS-managed data plane you use, the less responsibility you have. +The responsibility for the data plane’s reliability is shared between you, the customer, and AWS. EKS offers four worker node options for deploying the Kubernetes data plane. + +[EKS Auto Mode](https://docs.aws.amazon.com/eks/latest/userguide/automode.html), which is the most managed option, handles provisioning, scaling and updates of the data plane along with providing managed Compute, Networking, and Storage capabilities. Auto Mode AMIs are released frequently and clusters are updated to the latest AMI automatically to deploy CVE fixes and security patches. You have the ability to control when this occurs by configuring [disruption controls](https://docs.aws.amazon.com/eks/latest/userguide/create-node-pool.html#_disruption) on your Auto Mode NodePools. + +Fargate handles provisioning and scaling of the data plane by running one Pod per Node. The third option, managed nodes groups, handles provisioning, and updates of the data plane. And finally, self-managed nodes is the least managed option for the data plane. The more AWS-managed data plane you use, the less responsibility you have. @@ -66 +70 @@ To use the Amazon Web Services Documentation, Javascript must be enabled. Please -EKS Auto Mode +Cluster Autoscaler