Raft3D

Raft3D is a distributed 3D printer management system with data persistence implemented via the Raft Consensus Algorithm rather than traditional centralized databases.

Features

• Distributed 3D printer management with Raft consensus
• REST API for managing printers, filaments, and print jobs
• Fault tolerance with leader election and data replication
• Support for snapshotting and log compaction

Requirements

• Go 1.21 or higher
• Internet connection to download dependencies

Installation

1. Clone the repository:

git clone https://github.com/yourusername/raft3d.git
cd raft3d

2. Build the application:

go build -o raft3d ./cmd/server

Running the Cluster

To demonstrate Raft3D, you need to run at least 3 nodes. Here's how to start a 3-node cluster:

Node 1 (Bootstrap node)

mkdir -p data/node1
./raft3d -id node1 -raft-addr localhost:7000 -raft-dir data/node1 -http-addr localhost:8000 -bootstrap -peers localhost:7000,localhost:7001,localhost:7002

Node 2

mkdir -p data/node2
./raft3d -id node2 -raft-addr localhost:7001 -raft-dir data/node2 -http-addr localhost:8001 -peers localhost:7000,localhost:7001,localhost:7002

Node 3

mkdir -p data/node3
./raft3d -id node3 -raft-addr localhost:7002 -raft-dir data/node3 -http-addr localhost:8002 -peers localhost:7000,localhost:7001,localhost:7002

Testing the API

You can use curl or a tool like Postman to test the API endpoints. Here are some examples:

Create a Printer

curl -X POST http://localhost:8000/api/v1/printers -H "Content-Type: application/json" -d '{"company": "Creality", "model": "Ender 3"}'

List Printers

curl -X GET http://localhost:8000/api/v1/printers

Create a Filament

curl -X POST http://localhost:8000/api/v1/filaments -H "Content-Type: application/json" -d '{"type": "PLA", "color": "red", "total_weight_in_grams": 1000, "remaining_weight_in_grams": 1000}'

List Filaments

curl -X GET http://localhost:8000/api/v1/filaments

Create a Print Job

curl -X POST http://localhost:8000/api/v1/print_jobs -H "Content-Type: application/json" -d '{"printer_id": "PRINTER_ID", "filament_id": "FILAMENT_ID", "filepath": "prints/model.gcode", "print_weight_in_grams": 100}'

Replace PRINTER_ID and FILAMENT_ID with the actual IDs returned from the create operations.

List Print Jobs

curl -X GET http://localhost:8000/api/v1/print_jobs

Update Print Job Status

curl -X POST "http://localhost:8000/api/v1/print_jobs/JOB_ID/status?status=Running"

Replace JOB_ID with the actual ID of the print job.

Testing Raft Functionality

To test the Raft consensus functionality, you can:

1. Send a request to create a printer to the leader node
2. Verify that the printer exists by querying any node
3. Kill the leader node (Ctrl+C on the leader's terminal)
4. Observe that a new leader is elected
5. Create a new printer by sending a request to the new leader
6. Restart the old leader node
7. Verify that the old leader node has received all the data by querying it

Checking Node Status

You can check the status of any node by sending a GET request to the /status endpoint:

curl -X GET http://localhost:8000/status

This will return information about the node, including whether it's the leader and the current state of the Raft consensus.

Architecture

Raft3D is built using the HashiCorp Raft library and follows the Raft consensus algorithm. The application consists of:

1. Raft Node: Responsible for maintaining the distributed state and consensus
2. Finite State Machine (FSM): Handles the application state (printers, filaments, print jobs)
3. HTTP API: Provides RESTful endpoints for interacting with the system

License

This project is licensed under the MIT License.