Hack-a-Mae 3.0 Environment Setup

Introduction

This repository contains code and instructions for teams to setup their AWS environment for the Hack-a-Mae 3.0 event.

Teams should have received an email with instructions on how to login to their AWS burner account. Note that access to the AWS burner account is not permitted from the Fannie Mae network, so you should use your personal computer for access.

Note that AWS burner accounts are only available for 3 days, so each team will receive 2 accounts. The first account is active April 12-15, and the second account is active April 15-18. Please plan accordingly to move your work to the second account on April 15 and use that account for presentations on April 17.

Setup Instructions

1. Click on the link provided in the instructions, review/accept the terms and conditions, and join the event. Links on the left side panel lead to AWS console and credentials if you choose to connect locally.

2. Open the AWS console and change region to us-east-1.

3. Create roles

• Open the IAM service and select Roles
• Create admin role for EC2 access
  • Click Create Role
  • Select EC2 use case and click Next
  • Add Administrator permission and click Next
  • Enter MyEC2AdminRole for the role name and click Create Role
• Create admin role for CloudFormation access
  • Click Create Role
  • Select CloudFormation use case and click Next
  • Add Administrator permission and click Next
  • Enter MyCloudFormationAdminRole for the role name and click Create Role

4. Update params.json to set values specific to your AWS burner account (available from VPC service in AWS console):

• VPC id
• Subnet ids for 2 different subnets of your choosing
• Default security group id
• Team number (1-20)

5. Create a new CI/CD pipeline

• Open the CodePipeline service and click Create Pipeline
• Name the pipeline whatever you want and click Next
• Select GitHub (Version 2) source provider
• Click Connect to GitHub and follow the prompts to connect to your account
• Select the repository and branch assigned to your team
• Select No filter trigger type and click Next
• Click Skip build stage
• Select AWS CloudFormation for the deploy provider
  • Select Create or update a stack for Action Mode
  • Name the stack whatever you want
  • Enter source artifact filename cft-starter.yml
  • Enable configuration file and enter source artifact filename params.json
  • Select the MyCloudFormationServiceRole created in step 3
  • Click Next
• Review the settings and click Create Pipeline
• Pipeline should trigger automatically. Monitor the status in CodePipeline and CloudFormation until stack is created successfully.

Environment Details

The cft-starter.yml template creates 2 Application Load Balancers, each with 2 EC2 instances, one in each subnet that you selected. ALB1 is configured with an AutoScaling group, while ALB2 simply has 2 standalone EC2 instances.

Each EC2 instance is setup with the following items:

• Amazon Linux 2 AMI for x86_64 created on Feb 23, 2024
• EBS root filesystem that comes with the AMI
• EBS data filesystem mounted as /export/appl/data
• EFS data2 filesystem mounted as /export/appl/data2 (on standalone instances only)
• 1 or more security groups
• 1 or more private IP addresses
• 1 or more tags
• Additional software
  • Apache web server with PHP
  • Java 17 Corretto
  • jq
  • SpringBoot demo application deployed to /export/appl/data and run as a service on startup

There is also a web page created to summarize the configuration, read data from several text files, and report the status of the SpringBoot application. To access that page, go to http://DNS name from your ALB/index.php

Challenge Details

The challenge for Hack-a-Mae 3.0 is to:

• Identify the latest available Amazon Linux 2 AMI
• Identify EC2 instances in us-east-1 that are running an older version of the AMI
• Take a snapshot of the status before AMI update
• Upgrade each EC2 instance to the latest AMI, while preserving data and configuration, including the following:
  • EBS root filesystem, with any files and packages created outside of the AMI
  • EBS data filesystem
  • EFS data2 filesystem, if present
  • Security group(s)
  • Private IP address(es)
  • Tag(s)
• Take a snapshot of the status after AMI update
• Validate the status page for each ALB to confirm the environment is working properly. The only values that should change are Instance Id and AMI Id/description.
• Build the solution into a automated pipeline capable of running within the Fannie Mae environment.