Help improve this page
To contribute to this user guide, choose the Edit this page on GitHub link that is located in the right pane of every page.
This page documents updates to Amazon EKS Auto Mode. You can periodically check this page for announcements about features, bug fixes, known issues, and deprecated functionality.
April 18, 2025
Feature: Support for resolving .local domains (typically reserved for Multicast DNS) via unicast DNS.
April 11, 2025
Feature: Added certificateBundles and ephemeralStorage.kmsKeyID to NodeClass. For more information, see Node Class Specification.
Feature: Improved image pull speed, particularly for instance types with local instance storage that can take advantage of the faster image decompression.
Bug Fix: Resolved a race condition which caused FailedCreatePodSandBox , Error while dialing: dial tcp 127.0.0.1:50051: connect: connection refused to sometimes occur for Pods scheduling to a Node immediately at startup.
April 4, 2025
Feature: Increase registryPullQPS from 5 to 25 and registryBurst from 10 to 50 to reduce client enforced image pull throttling (Failed to pull image xyz: pull QPS exceeded)
March 31, 2025
Bug Fix: Fixes an issue where if a Core DNS Pod is running on an Auto Mode node, DNS queries from Pods on the node would hit that Core DNS Pod instead of the node local DNS server. DNS queries from Pods on an Auto Mode node will always go to the node local DNS.
March 21, 2025
Bug Fix: Auto Mode nodes now resolve kube-dns.kube-system.svc.cluster.local correctly when there isn’t a kube-dns service installed in the cluster. Addresses GitHub issue #2546
March 14, 2025
Feature: IPv4 egress enabled in IPv6 clusters. IPv4 traffic egressing from IPv6 Auto Mode clusters will now be automatically translated to the v4 address of the node primary ENI.
