DevOps Challenge

A DevOps coding challenge

The goal of this exercise is to demonstrate how I think about and approach problems through code. The solution is be an example of my work and demonstrates community best practices for the tools and languages involved.

Requirements

Create a containerized Ansible project that provisions a Linux VM running in Amazon Web Services. Once provisioned, the Ansible project should execute a custom Ansible module created for this exercise that returns a list of installed systemd units and their current state.

The returned data structure should be a list of dictionaries each containing two keys representing the name of the unit file and the current state. For example:

[

{

    "unit": "docker.socket",
    
    "state": "enabled"
    
},

{

    "unit": "docker.service",
    
    "state": "enabled"
    
}

]

Other Requirements

• The solution should be delivered as a link to a public GitHub repository containing the project (you are here).
• The project should contain a working Dockerfile that builds the solution.
• The project should have three playbooks:
  • A playbook to provision the VM and any other resources needed
  • A playbook to connect to the VM and return the solution
  • A playbook to delete the VM and any other AWS resources created by the solution.
• The created Linux VM should reside in the default public VPC with a default, dynamically assigned public IP address.
• The playbook that connects to the VM and returns the solution should be runnable multiple times and produce the result without having to provision or deprovision the instance each time.
• The playbook that connects to the VM and returns the solution should continue to work without modification if the public IP address of the Linux VM changes.
• All playbooks should be idempotent and not return false changed responses.
• Forcefully terminating the running Linux VM should automatically create a duplicate instance to replace it
• The custom Ansible module you create for this test must only use functionality present in the Python standard library on the running Linux VM.
• When designing the Docker image, follow best practices to reduce the image size to be as small as is reasonably possible.
• The container should take AWS credentials as its only input.
• The project should require no existing AWS resources other than an AWS account with valid credentials and permissions.
• Using roles and other Ansible best practices for project layout is highly encouraged.
• Throughout the project, pay particular attention to modularizing repetitive code with an eye toward producing the most readable and elegant solution possible.

To run through this exercise, follow along with the commands in demo.sh. Enjoy!