Terraform from scratch

Terraform from scratch

This is not a crash course. This is not a tutorial. This document is written as a **continuous learning handbook** intended to be read over many days. Expect depth. Expect repetition. Expect real-world thinking.

Terraform is a foundational skill for modern infrastructure engineers. If you understand Terraform deeply, cloud providers become interchangeable, systems become predictable, and scaling stops being scary.

---

1. The Real Problem Terraform Solves

To understand Terraform, you must first understand the **pain it eliminates**. Before Infrastructure as Code, infrastructure lived in three places:

• Human memory
• Manual cloud consoles
• Outdated documentation

This created environments where:

• Production ? Staging
• Hotfixes were undocumented
• No one knew who changed what
• Rebuilding was impossible under pressure

Terraform solves this by enforcing **source-controlled infrastructure truth**.

---

2. Infrastructure as Code - Beyond the Buzzword

Infrastructure as Code (IaC) is often misunderstood as "automation". Automation is only a side effect.

The real benefits are:

• Reproducibility
• Auditability
• Predictability
• Disaster recovery

With IaC, infrastructure becomes:

• Versioned (Git)
• Reviewable (Pull Requests)
• Testable (Plans)

Terraform is a **declarative IaC engine**, meaning you describe the final state, not the steps to reach it.

---

3. Declarative vs Imperative - Why Terraform Wins

Imperative tools say:

"Create a server, then configure it, then attach storage."

Declarative tools say:

"I want one server with storage attached."

Terraform builds a dependency graph and figures out execution order automatically.

---

4. Terraform Architecture - How It Actually Works

Terraform has four major internal components:

• Core engine
• Providers
• State management
• Execution planner

When you run Terraform:

• Configuration is parsed
• Providers are initialized
• State is loaded
• A dependency graph is created
• A plan is generated

---

5. Installing Terraform and Verifying Environment

After installation, always verify:

terraform version

In professional environments, Terraform versions are pinned to avoid breaking changes between releases.

Many companies use:

.tfenv-version

to enforce consistent Terraform versions across teams.

---

6. Understanding Terraform Configuration Files

Terraform configuration files use the HashiCorp Configuration Language (HCL). It is designed to be readable, structured, and machine-friendly.

Key characteristics of HCL:

• Blocks define intent
• Arguments define configuration
• Expressions reference data

Example structure:

resource "aws_instance" "web" { ami = "ami-123456" instance_type = "t3.micro" }

This block defines a **desired object**, not a command.

---

7. Providers - Terraform's API Bridge

Providers allow Terraform to communicate with external systems such as AWS, Azure, GCP, Kubernetes, GitHub, and hundreds more.

Example AWS provider:

provider "aws" { region = "us-east-1" }

Providers manage:

• Authentication
• API requests
• Resource lifecycle

---

8. Resources - Real Infrastructure Objects

Resources are the heart of Terraform. Each resource represents a real object in the target system.

Examples of resources:

• Virtual machines
• Networks
• Databases
• Load balancers
• DNS records

Example:

resource "aws_s3_bucket" "logs" { bucket = "company-prod-logs" acl = "private" }

Terraform tracks this resource using state.

---

9. Terraform Workflow - Professional Usage

The standard workflow:

terraform init terraform plan terraform apply

In enterprise environments:

• Plan is reviewed
• Apply is gated
• Changes are logged

---

10. Terraform State - The Single Most Important Concept

Terraform state maps configuration to real infrastructure. It answers the question:

"What exists right now?"

Without state, Terraform cannot:

• Detect drift
• Plan changes
• Safely update resources

State files must be:

• Protected
• Backed up
• Locked