In case you're still using AWS ECS and lack a neat CLI tool to manage your resources, meet this great project, which brings you a K9s-like experience.

e1s is a terminal app that allows you to browse and manage AWS ECS (Elastic Container Service) resources. Its interface is inspired by K9s, a well-known Kubernetes TUI. The first public version of e1s was released last June, and today offers tons of features. Here are some of them:
- EC2 ECS and Fargate launch types’ support.
- Describing clusters, services, tasks and task definitions, containers, and service autoscaling.
- Editing services, registering new task definitions, stopping tasks.
- Launching interactive exec in containers; starting port forwarding; transferring files.
- Displaying CloudWatch logs and utilisation metrics (CPU, memory).
- Numerous key bindings; theme and colour customisations.
- Available for Linux, macOS and Windows. The installation methods are: pre-built binaries, Homebrew, Docker image, AWS CloudShell, and go install.

▶️ GitHub repo: https://github.com/keidarcy/e1s

#tools #CLI #AWS

As this chart suggests (source), major cloud providers are now distinctly faster when it comes to making the latest Kubernetes releases available for their users.

We are definitely leaving behind the times when it took 100+ days to be able to run the newest Kubernetes in the preferred managed service. Now, it takes just about a month to get it for early adopters using AKS and GKE. EKS is close to making it generally available in a month.

#news #reports #AWS #GCP #Azure

Here are two important Kubernetes/EKS-related announcements during AWS re:Invent 2024 happening these days (December 2-6):

1. EKS Auto Mode. This new feature “automates compute, storage, and networking management for Kubernetes clusters [any new or existing EKS cluster].” Its main idea is to offload cluster operations to AWS and benefit from improved performance and security of apps as well as optimised compute costs. It’s available today for EKS clusters with Kubernetes 1.29+. It is built on top of Karpenter and is called “possibly the most meaningful since EKS first launched 6 years ago” by some AWSers. Find more details in this announcement.

2. EKS Hybrid Nodes that allow you “to attach your on-premises and edge infrastructure as nodes to EKS clusters in the cloud.” By doing so, you offload managing Kubernetes control planes to EKS and get a centralised hybrid infrastructure, which can also benefit from various AWS services, including monitoring, logging, and identity management. More technical details are available in this blog.

#news #AWS #events

Amazon EKS introduced a new catalogue of community add-ons, simplifying the use and management of well-known Open Source components in Kubernetes clusters.

Currently, it features metrics-server, kube-state-metrics, cert-manager, prometheus-node-exporter, and external-dns. All of them were packaged and validated for EKS, and hosted in the EKS-owned private ECR. You can work with add-ons via EKS Console, API, CLI, eksctl, and CloudFormation.

Find more details in this announcement and relevant documentation.

#news #AWS

Amazon EKS announced it now supports Kubernetes clusters with up to 100,000 nodes. The authors name massive AI/ML workloads as a possible use case for such setups, which can fit 1.6m AWS Trainium chips or 800k NVIDIA GPUs in a single K8s cluster.

It’s been a significant effort in AWS involving a comprehensive set of improvements aimed at achieving such a scale. Here’s what the engineers did:

- Re-architecting etcd (its new design is illustrated in this post). It involved switching from a Raft-based consensus backend to an internal component built at AWS, moving BoltDB from network-attached Amazon EBS volumes to in-memory storage (tmpfs), and choosing an optimal partitioning scheme.
- Tuning API servers by elaborating specific configurations, enabling strongly-consistent reads from cache and streaming list responses, and using CBOR (Concise Binary Object Representation) encoding for custom resources.
- Optimizing cluster controllers' performance, enhancing Karpenter, scaling the cluster network, and introducing SOCI (Seekable OCI) fast pull for container image pulls.

Find more details, including the resulting benchmark charts, in this blog post.

#news #AWS

CNOE AWS reference implementation

CNOE (Cloud Native Operational Excellence) is an Open Source organisation that focuses on building Kubernetes-based Internal Developer Platforms (IDPs) for enterprises. IDP Builder for spinning up a complete IDP featuring Docker as the only dependency is the most well-known project by CNOE.

Last week, the organisation published a GitHub repo with its AWS reference IDP implementation. It aims to create a production-ready IDP on EKS based on Crossplane, Backstage, Argo CD, Argo Workflows, Keycloak, and other Open Source tools.

UPD: In this YouTube video, Miguel Fontanilla, Platform Engineering Lead at Sennder, demonstrates the CNOE AWS IDP.

#news #tools #AWS

AWS Load Balancer Controller now supports Gateway API

Previously, AWS Load Balancer Controller relied on Application Load Balancer and Network Load Balancer as Ingress and Service in Kubernetes environments. Now, you can also use the Gateway API.

P.S. According to this GitHub issue, Azure Kubernetes Service is also expected to introduce Gateway API support for App Routing in March. Google Cloud has been recommending using its Gateway API implementation in the GKE Gateway controller to expose apps in Kubernetes for a while.

#news #networking #aws