AWS EKS
AWS EKS¶
Description¶
This implementation allows you to create entire AWS infrastructure from scratch and provides Kubernetes cluster (EKS) to deploy db1000n project.
Prerequisites¶
Configure AWS profile¶
The following example shows sample values:
$ aws configure
AWS Access Key ID [None]: AKIAIOSFODNN7EXAMPLE
AWS Secret Access Key [None]: wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY
Default region name [None]: us-west-2
Default output format [None]: json
Deployment¶
Deploy infrastructure¶
cd db1000n/terraform/aws/eks-cluster/
terraform init
terraform plan
terraform apply
NOTE: You can create multilpe *.tfvars configuration files with various variables, regions and AWS accounts
using terraform workspace command:
cd db1000n/terraform/aws/eks-cluster/
terraform init
terraform workspace new $your_workspace
terraform plan -var-file $your_file.tfvars
terraform apply -var-file $your_file.tfvars
Update kubeconfig¶
aws --profile $your_aws_profile eks update-kubeconfig --name $your_eks_cluster_name
Connect to EKS cluster¶
$ kubectl get nodes
NAME STATUS ROLES AGE VERSION
ip-xxx-xxx-x-xx.us-east-1.compute.internal Ready <none> 107m v1.21.5-eks-9017834
ip-xxx-xxx-x-xx.us-east-1.compute.internal Ready <none> 107m v1.21.5-eks-9017834
ip-xxx-xxx-x-xx.us-east-1.compute.internal Ready <none> 107m v1.21.5-eks-9017834
Install application¶
$ cd db1000n/kubernetes/helm-charts/
$ helm upgrade --install \
--create-namespace \
--namespace=db1000n \
-f values.yaml db1000n .
Check installation¶
$ kubectl -n db1000n get pods
NAME READY STATUS RESTARTS AGE
db1000n-54d8744b54-8hffr 1/1 Running 0 2m10s
db1000n-54d8744b54-8vml4 1/1 Running 0 2m10s
db1000n-54d8744b54-9stzv 1/1 Running 0 2m10s
Deletion¶
Delete application¶
helm uninstall db1000n -n db1000n
Delete infrastructure¶
terraform destroy