Plan and apply separation in pipelines in Terraform - Mini Project: Build & Apply

Plan and apply separation in pipelines

📖 Scenario: You are working in a team managing cloud infrastructure using Terraform. To keep your deployment process safe and organized, you want to separate the planning and applying steps into two different pipeline stages. This helps catch errors before making changes live.

🎯 Goal: Build a Terraform pipeline configuration that clearly separates the terraform plan step from the terraform apply step using two distinct pipeline jobs.

📋 What You'll Learn

Create a Terraform pipeline with two jobs: plan and apply

The plan job runs terraform init and terraform plan

The apply job runs terraform apply only after plan succeeds

Use a variable to control the Terraform workspace name

💡 Why This Matters

🌍 Real World

Separating plan and apply steps in Terraform pipelines is a common practice in real-world cloud infrastructure management to prevent accidental changes and improve safety.

💼 Career

Cloud engineers and DevOps professionals use this pattern daily to maintain reliable and controlled infrastructure deployments.

1

Define Terraform workspace variable

Create a Terraform variable called workspace with default value dev in a file named variables.tf. This variable will hold the workspace name.

Terraform

# Your code here

Need a hint?

Use the variable block with a default value.

2

Create plan job configuration

Add a pipeline job named plan that runs terraform init and terraform plan using the workspace variable var.workspace. Use a shell script block for commands.

Terraform

variable "workspace" {
  description = "The Terraform workspace name"
  type        = string
  default     = "dev"
}

# Your code here

Need a hint?

Define a job block with steps running terraform init and terraform plan using the variable.

3

Create apply job with dependency

Add a pipeline job named apply that runs terraform apply -auto-approve using the workspace variable var.workspace. Make sure this job depends on the plan job so it runs only after planning succeeds.

Terraform

variable "workspace" {
  description = "The Terraform workspace name"
  type        = string
  default     = "dev"
}

job "plan" {
  steps = [
    "terraform init",
    "terraform plan -var='workspace=${var.workspace}'"
  ]
}

# Your code here

Need a hint?

Use depends_on to link the apply job to plan.

4

Complete pipeline with workspace usage

Ensure the pipeline uses the workspace variable in both jobs and that the apply job runs only after the plan job completes successfully. The final pipeline configuration should include the variable and both jobs with correct commands and dependencies.

Terraform

variable "workspace" {
  description = "The Terraform workspace name"
  type        = string
  default     = "dev"
}

job "plan" {
  steps = [
    "terraform init",
    "terraform plan -var='workspace=${var.workspace}'"
  ]
}

job "apply" {
  depends_on = ["plan"]
  steps = [
    "terraform apply -auto-approve -var='workspace=${var.workspace}'"
  ]
}

# Your code here

Need a hint?

Review that both jobs use the workspace variable and apply depends on plan.