⏱️5 min read · 932 words
Terraform is the industry standard for Infrastructure as Code (IaC). In 2026, every DevOps team uses Terraform to provision and manage cloud infrastructure declaratively. This guide goes from first resource to production-grade modules and state management.
📋 Table of Contents
Why Terraform?
- Provider-agnostic — works with AWS, GCP, Azure, Kubernetes, and 3,000+ providers
- Declarative — describe what you want, not how to create it
- State management — tracks real infrastructure vs your code
- Plan before apply — preview changes before making them
- Modules — reusable, shareable infrastructure components
Installation and Setup
# Install Terraform (macOS)
brew install terraform
# Install via tfenv (version manager)
brew install tfenv
tfenv install 1.9.0
tfenv use 1.9.0
# Verify
terraform --version
# Install providers (runs automatically on init)
terraform init
Core Concepts
# main.tf — basic structure
# Provider — which cloud to use
terraform {
required_version = ">= 1.9"
required_providers {
aws = {
source = "hashicorp/aws"
version = "~> 5.0"
}
}
}
provider "aws" {
region = var.aws_region
}
# Variables
variable "aws_region" {
type = string
description = "AWS region"
default = "us-east-1"
}
variable "instance_type" {
type = string
default = "t3.micro"
}
# Resource — actual infrastructure
resource "aws_instance" "web" {
ami = "ami-0c55b159cbfafe1f0"
instance_type = var.instance_type
tags = {
Name = "web-server"
Environment = var.environment
}
}
# Output
output "instance_ip" {
value = aws_instance.web.public_ip
description = "Public IP of web server"
}
Terraform Workflow
# 1. Initialize — download providers
terraform init
# 2. Format — auto-format code
terraform fmt
# 3. Validate — check syntax
terraform validate
# 4. Plan — preview changes
terraform plan
terraform plan -out=tfplan # save plan to file
# 5. Apply — create/update infrastructure
terraform apply
terraform apply tfplan # apply saved plan
# 6. Destroy — remove all resources
terraform destroy
# Targeted operations
terraform plan -target=aws_instance.web
terraform apply -target=module.networking
Real Example: AWS Web Stack
# VPC
resource "aws_vpc" "main" {
cidr_block = "10.0.0.0/16"
enable_dns_hostnames = true
enable_dns_support = true
tags = { Name = "${var.project}-vpc" }
}
# Public subnet
resource "aws_subnet" "public" {
count = 2
vpc_id = aws_vpc.main.id
cidr_block = "10.0.${count.index}.0/24"
availability_zone = data.aws_availability_zones.available.names[count.index]
map_public_ip_on_launch = true
tags = { Name = "${var.project}-public-${count.index}" }
}
# Security group
resource "aws_security_group" "web" {
name = "${var.project}-web-sg"
vpc_id = aws_vpc.main.id
ingress {
from_port = 80
to_port = 80
protocol = "tcp"
cidr_blocks = ["0.0.0.0/0"]
}
ingress {
from_port = 443
to_port = 443
protocol = "tcp"
cidr_blocks = ["0.0.0.0/0"]
}
egress {
from_port = 0
to_port = 0
protocol = "-1"
cidr_blocks = ["0.0.0.0/0"]
}
}
# Application Load Balancer
resource "aws_lb" "web" {
name = "${var.project}-alb"
internal = false
load_balancer_type = "application"
security_groups = [aws_security_group.web.id]
subnets = aws_subnet.public[*].id
}
Modules — Reusable Infrastructure
# modules/rds/main.tf
variable "db_name" { type = string }
variable "db_user" { type = string }
variable "db_password" {
type = string
sensitive = true
}
variable "subnet_ids" { type = list(string) }
variable "sg_ids" { type = list(string) }
resource "aws_db_instance" "main" {
identifier = var.db_name
engine = "postgres"
engine_version = "16.2"
instance_class = "db.t3.medium"
allocated_storage = 20
db_name = var.db_name
username = var.db_user
password = var.db_password
db_subnet_group_name = aws_db_subnet_group.main.name
vpc_security_group_ids = var.sg_ids
backup_retention_period = 7
skip_final_snapshot = false
deletion_protection = true
storage_encrypted = true
tags = { Name = var.db_name }
}
output "endpoint" { value = aws_db_instance.main.endpoint }
# Use module in root
module "database" {
source = "./modules/rds"
db_name = "myapp"
db_user = var.db_user
db_password = var.db_password
subnet_ids = module.networking.private_subnet_ids
sg_ids = [aws_security_group.db.id]
}
Remote State — Team Collaboration
# Store state in S3 (never commit terraform.tfstate to git!)
terraform {
backend "s3" {
bucket = "mycompany-terraform-state"
key = "production/terraform.tfstate"
region = "us-east-1"
encrypt = true
dynamodb_table = "terraform-state-lock" # prevents concurrent applies
}
}
Workspaces — Multiple Environments
# Create workspaces for dev/staging/prod
terraform workspace new dev
terraform workspace new staging
terraform workspace new prod
terraform workspace list
# * dev
# staging
# prod
terraform workspace select prod
terraform plan
# Reference workspace in code
locals {
instance_type = terraform.workspace == "prod" ? "t3.large" : "t3.micro"
}
Data Sources
# Look up existing resources
data "aws_vpc" "existing" {
filter {
name = "tag:Name"
values = ["production-vpc"]
}
}
data "aws_ami" "ubuntu" {
most_recent = true
owners = ["099720109477"] # Canonical
filter {
name = "name"
values = ["ubuntu/images/hvm-ssd/ubuntu-*-22.04-amd64-server-*"]
}
}
resource "aws_instance" "web" {
ami = data.aws_ami.ubuntu.id
subnet_id = data.aws_vpc.existing.id
}
Best Practices
- Never commit
terraform.tfstate— use remote state (S3, Terraform Cloud) - Use
terraform planbefore every apply — review changes - Lock provider versions —
version = "~> 5.0"prevents surprises - Use modules — DRY infrastructure for dev/staging/prod
- Tag all resources —
Environment,Project,Owner - Store secrets in AWS Secrets Manager — not in Terraform variables
- Use
-targetsparingly — can leave state inconsistent
Terraform is now essential for any cloud infrastructure work. Start with simple EC2 instances, build up to VPC, RDS, and ALB, then extract reusable modules. Remote state and workspaces enable team collaboration and multi-environment management.
📚 You might also like
🔗 Share this article




✍️ Leave a Comment