Merkle Tree File Verification Backend

A backend server for demonstrating the functionality of a Merkle tree for a file storage and integrity verification system. Built using Go and PostgreSQL. The server exposes endpoints for uploading, downloading and generating Merkle proofs to ensure file integrity.

This backend is written to be used in conjuction with the Merkle Tree File Verification Client.

Prerequisites

• Docker
• Docker Compose

Environment Variables

• DB_URL
  • This specifies the URL of the DB.
  • Defaults to postgresql://admin:admin@localhost:5432 (points to the default URL of the DB from this service).
  • If you want to connect to a deployed DB, set DB_URL to the URL of your DB (must be a PostgreSQL DB):

    export DB_URL=postgres://db_user:db_password@db_postgres.db.com/db_name

Getting Started

You can run the backend locally with go, make or Docker.

Note The backend runs on port 8080 and the DB on port 5432 so make sure those ports are free if running locally.

Running with Make

make run

Running with Docker Compose

Step 1: Build the Docker Images (if needed)

docker-compose -f docker-compose.yml build

Step 2: Run the server and DB

docker-compose -f docker-compose.yml up

Backend Routes

The backend server exposes the following routes:

• POST /files/upload-batch/{id}: Upload files to the database.
  • queryParams add ?batch-complete=true to signal the last batch
• POST /files/delete-all: Delete all files from the database.
• GET /files/download/{id}: Download a file from the database.
• GET /files/get-proof/{id}: Get a Merkle proof for a file from the database.
• POST /files/corrupt-file/{id}: Simulate file corruption in the database by modifying a file and not the hash.

Database Table Structure

The backend server utilizes a PostgreSQL database to store file metadata and content. Below is the structure of the files table used by the server:

Column	Type	Description
id	integer	Sequence-generated integer used as the primary key for files.
batch_id	uuid	Universally unique identifier (UUID) used to identify which batch a file belongs to.
name	text	The name of the file.
file	bytea	The content of the file stored as binary.
original_hash	bytea	The SHA256 hash of the file stored as binary.

Purpose of Each Column

• id: Primary key used for unique identification of files.
• batch_id: Files are uploaded in batches and this batch_id is used to group the files together in a single batch.
• name: Original filename of the uploaded file.
• file: Binary content (file data) of the uploaded file.
• original_hash: This plays a crucial role in generating Merkle proofs for file integrity verification. By using the hash of each file's content, the server or client can detect any modifications or tampering attempts on the file data. This field allows the server to simulate malicious activity (e.g., file corruption) by modifying the file content while keeping a reference to original unmodified file hash.