Skip to content

태그: EKS

총 13개의 글이 있습니다.

AWS Load Balancer Controller

AWS Load Balancer Controller is a controller to help manage Elastic Load Balancers for a Kubernetes cluster. It satisfies Kubernetes Ingress resources by provisioning Application Load Balancers. It satisfies Kubernetes Service resources by provisioning Network Load Balancers. This project was formerly known as “AWS ALB Ingress Controller”, we rebranded it to be “AWS Load Balancer Controller”. De

Add IAM to RBAC

Add new IAM user or role to the Kubernetes RBAC, using kubectl or eksctl Before you choose the kubectl or eksctl tool to edit the aws-auth ConfigMap, make sure that you complete step 1. Then, follow steps 2-4 to edit with kubectl. To edit with eksctl, proceed to step 5. After you identify the cluster creator or admin, configure AWS CLI to use the cluster creator IAM. See Configuration basics for

The AWS-provided VPC CNI is the default networking add-on that runs on Kubernetes worker nodes for EKS clusters. VPC CNI add-on is installed by default when you provision EKS clusters. The VPC CNI provides configuration options for pre-allocation of ENIs and IP addresses for fast Pod startup times.\ VPC CNI components Amazon VPC CNI has two components: CNI Binary: which will setup Pod netwo

If you want to persist your own data to EBS in EKS environment, you can use EBS CSI driver addon provided by AWS. Beforehand we need to enable the IAM OIDC provider and create the IAM role for the EBS CSI driver. The easiest way to do both is to use eksctl (other ways like using plain aws cli or the AWS GUI are described in the docs). 1. Enable IAM OIDC provider A prerequisite for the EBS CSI driv

When you create a Kubernetes ingress, an AWS Application Load Balancer (ALB) is provisioned that load balances application traffic. ALBs can be used with Pods that are deployed to nodes or to AWS Fargate. You can deploy an ALB to public or private subnets. Before you can load balance application traffic to an application, you must meet the following requirements. Prerequisites Have an existing c

EKS Control Plane

EKS is a managed Kubernetes service that mekes it easy for you to run Kubernetes on AWS without needing to install, operate, and main your own Kubernetes control plane or worker nodes. It runs ipstream Kubernetes and is certified Kubernetes conformant. EKS automatically manages the availability and scalability of the Kubernetes control plane nodes, and it automatically replaces unhealthy control p

EKS Network BestPractice

It is critical to understand Kubernetes networking to operate you cluster and applications efficiently. Pod networkin, also called the cluster networking, is the center of Kubernetes networkin. Kubernetes supports Container Network Interface(CNI) plugins for cluster networking. Amazon EKS officially supports VPC CNI plugin to implement Kubernetes Pod networking. The VPC CNI provides native integra

EKS 기본 환경 Setting 방법을 알아보자. 쿠버네티스는 kubectl에 대한 인증정보를 kubeconfig라는 yaml파일에 저장한다. EKS 클러스터는 어떻게 접근을 하는지, 어떻게 인증을 받아오는지 알아보자. Kubeconfig EKS 클러스터에 접근 가능한 kubeconfig파일을 생성하는 것은 간단하다. aws command 한 줄이면 인증 정보를 kubeconfig 파일에 저장할 수 있다. Terminal windowaws eks --region &x3C;region-code> update-kubeconfig --name &x3C;cluster_name> update-kubeconfig --name "> kubeconfig 파일 내용을 살펴보자. apiVersion: v1clusters:cl

EKS 인증과정

kubectl에서 EKS Cluster의 K8s API Server에 접근하거나, Worker Node에서 동작하는 kubelet에서 EKS Cluster의 K8s API Server 접근시에는 AWS IAM Authenticator가 이용된다. 왼쪽은 kubelet의 kubeconfig이고, 오른쪽은 kubectl의 kubeconfig이다. 두 kubeconfig 모두 user 부분을 확인해보면 aws eks get-token 명령어를 수행하는 것을 확인할 수 있다. aws eks get-token 명령어는 해당 명령어를 수행하는 Identity가 누구인지 알려주는 AWS STS의 GetCallerIdentity API의 Presigned URL을 생성하고, 생성한 URL을 Encoding하여 Tok

IP addresse prefix

Each Amazon EC2 instance supports a maximum number of elastic network interfaces and a maximum number of IP addresses that can be assigned to each network interface. Each node requires one IP address for each network interface. All other available IP addresses can be assigned to Pods. Each Pod requires its own IP address. As a result, you might have nodes that have available compute and memory res

AWS Load Balancer Controller supports Network Load Balancer (NLB) with IP tergets for pods runing on Amazon EC2 instances and AWS Fargate through Kubernetes service of type LoadBalancer with proper annotation. In this mode, the AWS NLB targets traffic directly to the Kubernetes pods behind the service, aliminationg the need for an extra network hop through the worker nodes in the Kubernetes cluste

A Kubernetes node is a physical or virtual machine participating in a Kubernetes cluster, which can e used to run pods. When a node shuts down or crashed, it enters the NotReasy state, meaning it cannot be used to run pods. All stateful pods running on the node then becom unavailable. Common reasons for a Kubernetes node not ready error include lack of resources on the node, a problem with the kub

Spot 인스턴스에서 서버 가용성 개선하기

대덕SW마이스터고 동아리에서 개발한 6개의 교내 서비스(약 250명의 교사와 학생이 사용 중)를 대상으로 서비스 인프라를 통합하는 프로젝트를 진행했습니다. 이 프로젝트에서 Karpenter를 코드를 커스텀하여, Spot 인스턴스 환경의 Replica 1인 Deployment의 가용성을 96% -99.95%로 개선한 과정을 정리한 글입니다. 문제 상황 해당 인프라에선 각 동아리가 새 프로젝트를 개발할 때마다 배포할 수 있어야 했기 때문에, 자주 변동하는 컨테이너, 컴퓨팅 리소스를 쉽게 관리하기 위해 Kubernetes(EKS)를 사용하였다. 초기에 가용성을 크게 고려하지 않고 각 Deployment의 Replica 수를 1로 유지했음에도, 수십개 컨테이너를 실행하기엔 서버비 예산에 압박이 있었다. 따라서 모든