Automated Terraform Deployments Using Bitbucket Pipelines

README.md

Automated Terraform Deployments

Requirements

• AWS Account
• AWS API User API Key/Secret
• Bitbucket Account

Setup

1. Create or update an existing Bitbucket repo
2. Open https://bitbucket.org/USERNAME/REPO/addon/pipelines/deployments
3. Set your bitbucket-pipelines.yml with the content of the attached file
4. Commit file
5. Open your repo settings > Repository Variables
6. Create AWS_ACCESS_KEY_ID with your AWS Api Key (mark as secret)
7. Create AWS_SECRET_ACCESS_KEY with your AWS Secret Api Key (mark as secret)
8. Create AWS_DEFAULT_REGION with an AWS Region (optinal if not set in your AWS provider file)
9. Create and commit a file called aws.tf with the content of the attached file
   • For more information about the Terraform AWS provider check here
10. Create an S3 bucket (should be private, enable versioning)
11. Create a dynamo DB table
12. Create and commit a file called terraform.tf with the content of the attached file
    • Replace BUCKET-NAME with your bucket name
    • Replace DYNAMODB-NAME with your dynamo DB table name
    • For more information about the AWS Terraform backend check here
13. You are now ready to start deploying

Deploying

To use this for deployments:

• Create a file with Terraform AWS resources.
• Commit to a development branch
• The pipeline will test to make sure the code is valid and show a plan of what will be created/updated/removed
• If the pipeline passes, merge the changes to the master branch
• The pipeline will again test and plan the steps before going ahead and deploying the new code

aws.tf

provider "aws" {
}

bitbucket-pipelines.yml

image: hashicorp/terraform:full
pipelines:
    default:
        - step:
            script:
                - terraform init
                - terraform validate
                - terraform plan
    branches:
        master:
            - step:
                script:
                    - terraform init
                    - terraform validate
                    - terraform plan
                    - terraform apply -input=false -auto-approve

terraform.tf

terraform {
  required_version = ">= 0.9.1"

  backend "s3" {
    # This is an s3bucket you will need to create in your aws 
    # space
    bucket = "BUCKET-NAME"
     
    # The key should be unique to each stack, because we want to
    # have multiple enviornments alongside each other we set
    # this dynamically in the bitbucket-pipelines.yml with the
    # --backend
    key = "example-01"

    region = "us-west-2"

    # This is a DynamoDB table with the Primary Key set to LockID
    dynamodb_table = "DYNAMODB-NAME"

    #Enable server side encryption on your terraform state
    encrypt = true
  }
}