DEPLOY HELM CHARTS LIKE JENKINS + AWX AND EKS USING TERRAFORM 👾

Tired of creating clusters manually just to try something out for a few minutes? Then run this. Need some apps on that cluster really fast? Use Terraforms Helm Provider on top of their kubernetes provider to quickly deploy EKS clusters with your apps on top of them. In this example, I'm deploying Jenkins + AWX on EKS. It takes about 15 minutes from start to finish and minimal work.

Pre-requisites

• Terraform installed
• AWS cli installed on a host to connect to the cluster
• AWS credentials configured
• kubectl installed on a host to deploy to the cluster

Deployment Instructions

• Install Terraform
• Clone this repository
• Edit the jenkins-values.yaml and the awx-values.yaml to match your values
• Edit or remove the set_sensitive arguments in helm_release.tf if you would like to. These values will override the values in the *-values.yaml files.
• Run a terraform init to grab providers and modules
• Run aws_configure and establish your credentials for aws
• Run a terraform_apply and wait 10 - 15 minutes. Note: If it fails for HTTP timeout while waiting to apply the Helm chart, retry terraform_apply
• Run aws eks --region us-east-1 update-kubeconfig --name dev-cluster to add the cluster context to your kubeconfig
• Run kubectl get pods and kubectl get svc to ensure the applications deployed as expected
• Run kubectl patch svc <SERVICE_NAME> -p '{"spec": {"type": "LoadBalancer"}}' to change the Service to type LoadBalancer
• Run kubectl get svc again to grab the AWS created DNS address
• Go to your browser and navigate to http://<dns-address>:<port> Note: This may take 3 - 5 minutes to resolve while waiting for the apps and the ELB to fully initialize.
• Log in with the credentials you set in the *-values.yaml or the helm_release.tf
• Profit

Connecting

• Run aws eks --region us-east-1 update-kubeconfig --name dev-cluster to add the context to your kubeconfig

Troubleshooting

Pods stuck in Pending

• Possibility of resources not efficient. The instances in the worker group could be too small to assign IP addresses to all the pods

Workers not joining the cluster

• Ensure the workers are getting public IP addresses