ukhomeoffice / dq-tf-datapipeline

dq-tf-datapipeline

This Terraform module has one private subnet and deploys an EC2 instance representing a web server and an SQL database instance. Allowing inbound TCP traffic on port 135, inbound RDP TCP traffic on port 3389 and SQL TCP traffic on 5432.

Connectivity

In/Out	Type	Protocol	FromPort	To Port	Description
INBOUND	RDP	TCP	3389	3389	Datapipeline Web
INBOUND	RDP	TCP	3389	3389	Datapipeline PostgreSQL
INBOUND	PostgreSQL Custom port	TCP	1433	1433	Datapipeline PostgreSQL

Content overview

This repo controls the deployment of an application module.

It consists of the following core elements:

main.tf

This file has the basic components for EC2 instances

• Private subnet and route table association
• Two EC2 instances using the connectivity tester module
• Security group for the SQL and Web server

outputs.tf

Various data outputs for other modules/consumers.

variables.tf

Input data for resources within this repo.

tests/dp_test.py

Code and resource tester with mock data. It can be expanded by adding further definitions to the unit.

User guide

Prepare your local environment

This project currently depends on:

• drone v0.5+dev
• terraform v0.11.1+
• terragrunt v0.13.21+
• python v3.6.3+

Please ensure that you have the correct versions installed (it is not currently tested against the latest version of Drone)

How to run/deploy

To run tests using the tf testsuite:

drone exec --repo.trusted

To launch:

terragrunt plan
terragrunt apply

FAQs

The remote state isn't updating, what do I do?

If the CI process appears to be stuck with a stale tf state then run the following command to force a refresh:

terragrunt refresh

If the CI process is still failing after a refresh look for errors about items no longer available in AWS - say something that was deleted manually via the AWS console or CLI. To explicitly delete the stale resource from TF state use the following command below. Note:terragrunt state rm will not delete the resource from AWS it will unlink it from state only.

terragrunt state rm aws_resource_name