Migration of GitHub actions-runner-controller to gha-runner-scale-set

As of today we are supporting the new (officially by GitHub supported) deployment method of GitHub Actions runners, the gha-runner-scale-set-controller. This new controller is a more efficient and scalable way to deploy self-hosted GitHub runners (controlled by a new gha-runner-scale-set) on Kubernetes. Next to improved stability and ongoing development, this new controller adds autoscaling of the runner pool based on the number of pending jobs, resulting in a more scalable and cost-effective solution.

More …

Upgraded Teleport to version 15.0.1

We’ve upgraded all Teleport clusters from version 14.0.1 to 15.0.1. Teleport is a tool we mostly use internally to provide secure and auditted access to (EC2) instances, Kubernetes clusters and several dashboards. The nodes will gradually be upgraded to the new version when new instances are launched.

More …

Karpenter now deployed by default and migrated system pool

In September last year we announced the addition of Karpenter as experimental feature. Since then we have been making improvements to our implementation and gradually used some pilots internally and with some customers. We are happy to announce that Karpenter is now deployed by default (via AWS Fargate) on all our EKS clusters and we’ve migrated the system NodePool to use it instead of the standard Cluster Autoscaler.

More …

Upgraded Teleport to version 14.0.1

We’ve upgraded all Teleport clusters from version 13.3.8 to 14.0.1. Teleport is a tool we mostly use internally to provide secure and auditted access to (EC2) instances, Kubernetes clusters and several dashboards. The nodes will gradually be upgraded to the new version when new instances are launched.

More …

Removed Calico as NetworkPolicies engine - this functionality has merged with the AWS VPC CNI

Historically we’ve been using Calico as controller to provide NetworkPolicies support. This was offered as an optional feature only, considering the resource (and thus possible cost) impact of running this component. As announced in our K8s 1.27 upgrade post, the latest version of the AWS VPC CNI, responsible for providing cluster networking, now has native support for NetworkPolicies built-in.

More …

Support for GPU node pools on AWS (EKS)

We’re adding support for GPU node pools in EKS. GPU nodes are great for compute-intensive workloads such as graphics and visualization workloads, or machine-learning processes. AWS uses the NVIDIA device plugin to make the GPU capacity of a node available to Kubernetes workloads.

More …