Terraform AWS Tutorial: Your First Resource

Hands-on Terraform from zero to applied. The fastest path to understanding what state, plan, apply mean.

Step 1: Install Terraform

brew install terraform (macOS); verify with terraform version.

Have AWS credentials configured (aws CLI works).

Step 2: Write configuration

• resource "aws_s3_bucket" "learn" {
    bucket = "tf-learn-$(uuid)"
  }

• Save as main.tf in a fresh directory.

Step 3: Plan and apply

The init-plan-apply triplet is the entire Terraform workflow. Every change goes through these three commands; learn them once and the rest is parameters.

• init. terraform init downloads the AWS provider and sets up the backend.
• plan. terraform plan shows what will change; read it before applying.
• apply. terraform apply creates the bucket; type 'yes' at the prompt to confirm.
• State written. terraform.tfstate appears locally; this is the source of truth Terraform compares against next time.

Step 4: Destroy

Destroy is the inverse of apply. Same plan format, same confirmation, only the diff is negative. Run it when you are done with the lab.

• destroy. terraform destroy tears down everything in state; confirm at the prompt.
• Inspect state. terraform state list shows what Terraform manages; useful before destroy to confirm scope.
• Targeted destroy. terraform destroy -target=aws_s3_bucket.learn removes one resource without touching the rest.
• Cleanup. rm -rf .terraform terraform.tfstate* when finished; do not commit state files to git.

Antipatterns

• Skipping plan. Apply blindly is dangerous.
• Committing tfstate to git. Use remote state.
• Hard-coded names. Use random_id or variables.

What to do this week

Three moves. (1) Run the tutorial end-to-end on your own laptop / sandbox. (2) Apply the pattern to one production workload. (3) Document the variations you needed; share with the team.