Carter
Carter
**What would you like to be added**: Support for cgroup v2, specifically for `systemd`. **Why is this needed**: Kubernetes [recommends](https://kubernetes.io/docs/setup/production-environment/container-runtimes/#cgroup-v2) that, when using `systemd` as the cgroup driver, the "unified...
**Issue #, if available:** Closes #999 **Description of changes:** Counts failures of `aws eks wait cluster-active` towards the retry count of `describe-cluster`. (If the cluster isn't active, we can't do...
*(relayed from an internal ticket)* **What happened**: [`aws eks wait cluster-active`](https://github.com/awslabs/amazon-eks-ami/blob/master/files/bootstrap.sh#L369-L371) may get rate-limited (`TooManyRequestsException`) and cause the bootstrap script to terminate, instead of falling back to the retry logic...
With the latest Amazon Linux 2 kernels, customers running EC2 P4d, P4de and P5 instances may be unable to use the GPUDirect RDMA feature, which allows for faster communication between...
**Description of changes:** This adds validation for the external API types using the same mechanism that the API server uses for CRD's. Validation directives like `required` are only expressed in...
When listing releases here: https://github.com/awslabs/amazon-eks-ami/releases GitHub will truncate the notes at (apparently) 9500 characters. That results in the older k8s versions not being rendered. We'll need to compact our notes,...
**Description of changes:** Configures a node shutdown period of 45 seconds, with the last 15 seconds reserved for critical pods, for Kubernetes versions 1.30 and above. By submitting this pull...
**What would you like to be added**: Add a flag to enable tracing for `kubelet` and `containerd`. **Why is this needed**: To aid troubleshooting.
**Issue #, if available:** Fixes #1734 **Description of changes:** Decompresses user data if it's found to be GZIP. Currently only decompresses the full user data content, does not handle compressed...
**Description of changes:** It's nice for the CI workflow to scope down its jobs when a change only applies to one OS, but it's kind of a pain to declare...