Deploying Apps to AWS with Terraform - S3 Backend

6. Configuring Terraform State with AWS S3 Backend

Introduction

This lesson is focused on learning how to persist your Terraform state in an AWS S3 backend. Terraform backends are crucial as they determine how the state is stored, which can be either locally on your disk or remotely in a solution like AWS S3.

Understanding Terraform Backends

Terraform backends play a vital role in managing the state of your infrastructure deployment. By default, Terraform stores state files locally, but for more robust solutions, especially in collaborative environments, remote storage options such as AWS S3 are preferred.

Configuring the AWS S3 Backend

To configure Terraform to use AWS S3 as a backend, you must modify the Terraform block in your project code. This includes specifying the backend type (in this case, S3), the region, the S3 bucket name, and other necessary configurations. It's important to note that currently, Terraform does not allow interpolation of variables in the backend configuration, though this may change in the future.

Prerequisites

Before you begin writing your Terraform code to utilize an S3 backend, you need to have an S3 bucket ready. Ensure that you have the necessary IAM permissions to create, delete, and modify buckets as required for your Terraform project.

Step-by-Step Guide

1. Create an S3 Bucket: First, create an S3 bucket where Terraform will store its state files.

2. Write Terraform Configuration: In your Terraform project, create a backend.tf file and define the backend configuration with details like the S3 bucket name and AWS region.

backend.tf

terraform {
  required_version = ">= 1.2.0"
  backend "s3" {
    region  = "eu-west-2"
    profile = "default"
    key     = "terraformstatefile"
    bucket  = "terraform-bucket1995"
  }
}

3. Initialize Terraform: Run terraform init to initialize the backend configuration. This step is crucial as it sets up Terraform to use the specified S3 bucket for state storage.

4. Format Terraform Code: Optionally, run terraform fmt to format your Terraform code, ensuring consistency and readability.

5. Init the Terraform Code: Run, terraform init