Variables and Outputs

In the previous chapter, we hardcoded values like ami-0c55b159cbfafe1f0. In a professional MERN-stack workflow, this is a bad practice. What if you want to deploy the same app to a different region or use a larger server for production?

Variables allow you to parameterize your code, while Outputs allow you to extract information from your infrastructure after it is built.

1. Input Variables (The Inputs)

Think of variables as the "Arguments" of a function. Instead of hardcoding values, you define a variable and pass the value when you run Terraform.

Definition Syntax

We typically store variables in a separate file named variables.tf.

variables.tf

variable "instance_type" {
  description = "The size of the EC2 instance"
  type        = string
  default     = "t2.micro" # Optional: provides a fallback
}

variable "server_port" {
  description = "The port the server will use for HTTP requests"
  type        = number
  default     = 80
}

Using Variables in Code

To use a variable, use the var.<NAME> syntax in your main.tf:

main.tf

resource "aws_instance" "app_server" {
  ami           = "ami-0c55b159cbfafe1f0"
  instance_type = var.instance_type # Using the variable here

  tags = {
    Name = "CodeHarborHub-Server"
  }
}

2. Output Values (The Results)

Outputs are like the "Return" value of a function. After Terraform finishes building your cloud, you might need to know the Public IP of your server or the URL of your database.

Definition Syntax

We typically store these in outputs.tf.

outputs.tf

output "server_public_ip" {
  description = "The public IP address of the web server"
  value       = aws_instance.app_server.public_ip
}

When you run terraform apply, Terraform will print these values to your terminal at the very end.

Ways to Assign Variable Values

Terraform provides multiple ways to set your variables. At CodeHarborHub, we use these in order of priority:

Method	Best For...	Example
Variable Files (.tfvars)	Environment-specific settings.	terraform apply -var-file="prod.tfvars"
Command Line Flags	Quick testing/one-off changes.	terraform apply -var="instance_type=t2.large"
Environment Variables	CI/CD pipelines (GitHub Actions).	export TF_VAR_instance_type=t2.small
Default Values	Sensible defaults for beginners.	Defined inside variables.tf.

Variable Types & Validation

Terraform is a "Strongly Typed" language. This prevents errors before they ever reach the cloud.

Pro Tip: Validation

You can even add rules to ensure developers don't use expensive servers by accident!

variables.tf

variable "instance_type" {
  type = string
  validation {
    condition     = contains(["t2.micro", "t3.micro"], var.instance_type)
    error_message = "At CodeHarborHub, we only allow free-tier instances (t2/t3.micro)."
  }
}

Practical Workflow: The .tfvars file

To keep your code clean, create a file named terraform.tfvars:

terraform.tfvars

# Values assigned here override the defaults in variables.tf
instance_type = "t2.micro"
server_port   = 8080

When you run terraform apply, Terraform automatically loads terraform.tfvars and applies those values.

Learning Challenge

1. Take your main.tf from the last lesson.
2. Create a variables.tf and move the bucket_name into a variable.
3. Create an outputs.tf to display the ARN (Amazon Resource Name) of the bucket.
4. Run terraform apply and watch the output appear in your terminal!