README.md

aws-deployment-pipeline terraform module

Overview

The aws-deployment-pipeline terraform module, which creates a set of resources that allow to automate the building process of containerized applications.

In the main.tf file there are definitions of the resources responsible for triggering the entire process, the most important of which are:

• github_repository_webhook.github_proxy_api_gateway_webhook - GitHub repository wehbook enabling the capture of PUSH actions,
• api_gateway_webhook_proxy module (build using cloudposse/api-gateway/aws source module) - AWS Api Gateway Proxy used in GitHub repository wehbook configuration,
• webhook_proxy_lambda module (build using terraform-aws-modules/lambda/aws" source module) - AWS Lambda, which triggers the deployment pipeline process using lambda-webhook-proxy/index.js code,
• deployment_pipeline module (build using local ./deployment_pipeline module) - description of the most important file main.tf below.

The deployment_pipeline/main.tf file contains definitions of deployment pipelines built primarily using AWS CodePipeline and AWS Codebuild, as well as resources responsible for notifying the status of individual pipelines:

• aws_codebuild_project.deployment_test_code - AWS CodeBuild project for the application code testing stage,
• aws_codebuild_project.deployment_docker_image_build - AWS CodeBuild project for the application docker image build stage,
• aws_codebuild_project.deployment_gitops_push - AWS CodeBuild project for updating image tags in kustomization files,
• aws_codepipeline.deployment_pipeline - AWS CodePipeline project, which consists of 4 phases: Source (downloading source code from indicated GitHub respository), Test, BuildImage and PushToGitOps,
• aws_s3_bucket.deployment_pipeline_artifacts - AWS S3 Bucket for storing aws_codepipeline.deployment_pipeline artifacts,
• aws_ecr_repository.ecr_repository - AWS ECR registry for storing docker images,
• deployment_pipeline_notifications module build using https://github.com/kjagiello/terraform-aws-codepipeline-slack-notifications source module) - module used to send notifications about aws_codepipeline.deployment_pipeline status,
• deployment_pipeline_notifications_failed module build using https://github.com/kjagiello/terraform-aws-codepipeline-slack-notifications source module) - module used to send notifications about aws_codepipeline.deployment_pipeline in failed or canceled status.

Usage

Required vars:

• name,
• env,
• vpc_id,
• vpc_private_subnets,
• default_security_group_id,
• github_repository_organization,
• github_repository_name,
• github_repository_branch,
• github_access_token,
• devops_slack_webhook,
• devops_slack_channel_name,
• devops_slack_webhook_failed,
• devops_slack_channel_name_failed,
• target_gitops_repository,
• github_access_token_secret_name,
• github_user,
• pipelines.pipeline_name.name,
• pipelines.pipeline_name.short_name,
• pipelines.pipeline_name.secret_arns[],
• pipelines.pipeline_name.file_path_pattern_trigger,
• pipelines.pipeline_name.buildspec,
• pipelines.pipeline_name.buildspec_test_code,
• pipelines.pipeline_name.environment_variables.

Optional vars, defaults values in brackets:

• cloudwatch_log_retention_in_days (60),
• pipelines.pipeline_name.enable_test_stage (false).

module "deployment_pipeline_cosmos_graphql_repository" {
  source = "github.com/Airnauts/terraform-aws-deployment-pipeline.git?ref=v0.0.72"

  name = "cosmos-graphql"
  env  = "dev"

  vpc_id                    = module.vpc.vpc_id
  vpc_private_subnets       = module.vpc.private_subnets
  default_security_group_id = module.vpc.default_security_group_id

  github_repository_organization = "Airnauts"
  github_repository_name         = "cosmos-graphql"
  github_repository_branch       = "dev"
  github_access_token            = local.github_access_token

  devops_slack_webhook             = local.devops_slack_webhook
  devops_slack_channel_name        = "#cosmos-devops"
  devops_slack_webhook_failed      = local.dev_backend_slack_webhook
  devops_slack_channel_name_failed = "#cosmos-dev-backend"

  target_gitops_repository        = "Airnauts"
  github_access_token_secret_name = "GITHUB_ACCESS_TOKEN"
  github_user                     = "airnauts-cosmosbot"

  cloudwatch_log_retention_in_days = 30

  pipelines = {
    analytics = {
      name                      = "app-name"
      shorter_name              = "app"
      secret_arns               = []
      file_path_pattern_trigger = ".*"
      buildspec                 = "buildspec.yml"
      buildspec_test_code       = "buildspec-test-code.yml"
      enable_test_stage         = true
      environment_variables     = [
        {
          name : "PATH",
          type : "PLAINTEXT",
          value : "."
        }
      ]
    }
}

How to use deployment pipeline module in your project?

The aws-deployment-pipeline terraform module was used in the Cosmos project for creating a CI/CD pipeline for docker images.