Skip to content

Latest commit

 

History

History
77 lines (62 loc) · 1.99 KB

README.md

File metadata and controls

77 lines (62 loc) · 1.99 KB

Terraform deployment for the ELK Stack in AWS EKS

This is the documentation of the Terraform scripts that deploy the ELK stack in an existing AWS EKS cluster.

Resources

The following AWS resources are deployed, but not limited to:

  • Kubernetes
    • Namespace
    • Service Account
    • Secrets
  • Helm Charts
    • ElasticSearch
    • Kibana
    • Logstash
    • Filebeat
  • AWS EKS
    • Node Group
    • EBS Driver addon
  • AWS IAM Roles
    • ElasticSearch access to S3 role
    • EKS Node Group role
    • EBS CSI role
  • Amazon S3 bucket

Getting Started

Requirements

Variables

Create a variables file, such as variables.tfvars, and fill the following variables:

infra_env = "test"

aws_access_key_id     = ""
aws_secret_access_key = ""
aws_region            = "us-east-1"

k8s_namespace     = "monitoring"
eks_cluster_name  = ""
bucket_name       = ""

elasticsearch_username = "elastic"
elasticsearch_password = ""

private_subnets_name = ""

eks_instance_types = [ "t3.medium" ]

The variables that are already filled are mere examples and may be changed accordingly if needed.

Deployment

Running Terraform

First, it is necessary to install the Terraform dependencies by running the following command:

terraform init

Then, plan and apply:

terraform plan -out=tfplan -var-file=variables.tfvars
terraform apply tfplan

The commands shown above can also be run to make changes to the architecture, as it will check what resources need to be created/modified/destroyed.

Destroying

In order to destroy the whole infrastructure, first make sure to:

  • Delete all the contents of the S3 bucket created by Terraform

Then run the following command:

terraform destroy -var-file=variables.tfvars