Managing Cloud Infrastructure as Code with Terraform

Creating servers and network resources by clicking through a web panel carries serious risks for repeatability and auditability. Terraform is an open-source IaC tool that lets you define your infrastructure as code. This guide covers HCL syntax, state management, module structure, and production best practices.

What Is Infrastructure as Code?

Infrastructure as Code (IaC) is the approach of defining servers, networks, storage, and other infrastructure components through code files instead of manual operations. This way, infrastructure changes are tracked in version control, go through code review, and are applied automatically.

HCL Syntax and Basic Structure

Terraform uses HashiCorp Configuration Language (HCL). Each resource is defined with a resource block. Providers determine which cloud platform you work with.

terraform {
  required_providers {
    hcloud = {
      source  = "hetznercloud/hcloud"
      version = "~> 1.45"
    }
  }
}

provider "hcloud" {
  token = var.hcloud_token
}

resource "hcloud_server" "web" {
  name        = "web-server-01"
  server_type = "cx21"
  image       = "ubuntu-22.04"
  location    = "fsn1"
  ssh_keys    = [hcloud_ssh_key.deploy.id]
}

resource "hcloud_ssh_key" "deploy" {
  name       = "deploy-key"
  public_key = file("~/.ssh/id_rsa.pub")
}

State Management

Terraform stores the state of created resources in a terraform.tfstate file. This file holds the mapping between real infrastructure and code. For team collaboration, storing the state file in a remote backend (S3, GCS, Terraform Cloud) is essential.

terraform {
  backend "s3" {
    bucket         = "my-terraform-state"
    key            = "production/terraform.tfstate"
    region         = "eu-central-1"
    encrypt        = true
    dynamodb_table = "terraform-locks"
  }
}

Terraform Modules

Modules let you package repeating infrastructure patterns. You can define a VPS + firewall + DNS record combination as a module and reuse it across different environments.

variable "server_name" {
  type = string
}

variable "server_type" {
  type    = string
  default = "cx21"
}

resource "hcloud_server" "this" {
  name        = var.server_name
  server_type = var.server_type
  image       = "ubuntu-22.04"
  location    = "fsn1"
}

output "server_ip" {
  value = hcloud_server.this.ipv4_address
}

Plan, Apply and Destroy Workflow

Terraform's core workflow consists of three commands: plan previews changes, apply executes them, and destroy removes resources. In production, always review the plan output first.

# Download providers and initialize
$ terraform init

# Preview changes (nothing is applied)
$ terraform plan -out=tfplan

# Apply the plan file
$ terraform apply tfplan

# Delete all resources (use with caution)
$ terraform destroy

For infrastructure automation, check our GitHub Actions CI/CD guide. For configuration management, see our Ansible Playbook guide. For disaster recovery planning, explore our IaC Disaster Recovery guide. The Terraform documentation and Terraform Registry are valuable additional resources.

Frequently Asked Questions

Conclusion

By managing your infrastructure as code with Terraform, you gain repeatability, auditability, and speed. Store state files in a remote backend, package repeating patterns with modules, and integrate the plan-apply workflow into your CI/CD pipeline. Start with a single server and gradually convert your entire infrastructure to code.