AWS Expense Tracker - Infrastructure Setup

Project Overview

Building a personal expense tracker using AWS services to learn cloud infrastructure fundamentals.

Architecture: EC2 + S3 (with future Lambda integration)

Learning Focus: VPC networking, EC2 deployment, S3 integration, IAM permissions

Infrastructure Components

1. Custom VPC Network

• VPC CIDR: 10.0.0.0/16
• Public Subnet: 10.0.1.0/24 (eu-north-1a)
• Private Subnet: 10.0.2.0/24 (eu-north-1a) - for future use
• Internet Gateway: Attached for public internet access
• Route Table: Custom routing for public subnet (0.0.0.0/0 → IGW)

2. Security Groups

• Web Server Security Group:
  • SSH (22) - Administrative access
  • HTTP (80) - Web traffic
  • HTTPS (443) - Secure web traffic
  • Port 5000 - Flask development server

3. S3 Storage

• Bucket: Globally unique name for receipt storage
• Versioning: Enabled for data history and avoid override
• CORS: Configured for web uploads
• Purpose: Store uploaded receipt images

4. EC2 Instance

• Type: t3.micro (free tier eligible)
• AMI: Amazon Linux 2023 (ami-0b83c7f5e2823d1f4)
• Placement: Public subnet with Elastic IP
• Key Pair: Created for SSH access
• IAM: Using same IAM as the AWS Cli

Commands Summary

See infrastructure folder. Mostly used bash commands with the AWS cli and later combined together to create a bash script call 'setup-commands.sh'.

Key Learning Outcomes

VPC Networking

• ✅ Created custom VPC with proper CIDR planning (used subnet mask 16 which provides plenty of IP addresses for subnetting)
• ✅ Configured public/private subnet architecture (although only public subnet is used)
• ✅ Set up Internet Gateway and routing tables
• ✅ Understood how traffic flows in AWS networks

Security

• ✅ Implemented least-privilege security groups (only added policies involving S3 and EC2)
• ✅ Configured proper SSH key-based authentication

Service Integration

• ✅ Connected EC2 instance to custom VPC
• ✅ Integrated S3 storage with CORS configuration
• ✅ Set up cross-service communication between Flask App running on EC2 and S3.

Next Steps

1. Potential Future Enhancements
   • Lambda function for receipt OCR processing
   • Database integration (RDS) with User Access

Troubleshooting Notes

ARM64 Architecture

• Required specific AWS CLI download for aarch64 (developed on a raspberry pi)
• Used ARM-compatible AMI selection

Cost

• All resources within AWS Free Tier limits
• t3.micro instance (750 hours/month free)
• S3 free tier (5GB storage, 20K GET, 2K PUT requests)
• No NAT Gateway costs (using public subnet only)

Architecture Diagram

Internet
    |
Internet Gateway
    |
Public Subnet (10.0.1.0/24)
    |
EC2 Instance (Web Server)
    |
S3 Bucket (Receipt Storage)

---

Project Status: ✅ Completed Total Setup Time: ~30 minutes AWS Services Used: VPC, EC2, S3, IAM