Problem Statement

1. Main Functionality

The application’s main functionality is to enable users to:

• Evaluate important aspects of their life using the Wheel of Life
• Receive AI-generated suggestions and personalized feedback for improvement
• Track their evaluations over time and visualize trends
• Write reflective notes and input category ratings

AI will analyze both current and past data to provide actionable, mostly positive feedback and tailored suggestions.

---

2. Wheel of Life Categories

1. Career
2. Finances
3. Growth
4. Purpose
5. Health
   • Mental
   • Physical
6. Relationships
   • Family
   • Friends
   • Romance
7. Other
   • Spirituality
   • Personal Development
   • Entertainment
8. Custom Categories
   • Users can add (e.g., Social Engagement) or remove categories as they wish.

---

3. Intended Users

• Students
• Professionals
• Anyone aiming to improve and balance their life

---

4. GenAI Integration

• Analysis: AI processes weekly and historical ratings plus reflective notes.
• Feedback: Provides mostly positive, data-driven feedback.
• Suggestions: Offers small, actionable steps tailored to the user’s current state and trends.
• Privacy: Acts as a private coach, lowering the barrier to sharing personal reflections.

---

5. Example User Scenario

1. Login & Review:

   • User signs in via OAuth.
   • Views previous weeks’ Wheel of Life reports.

2. Weekly Report:

   • Rates satisfaction in each category.
   • Adds a short reflective note (e.g., "Feeling a bit lonely...")

3. AI-Generated Feedback:

   • Reflects the user’s notes and ratings.
   • Gives supportive commentary and action items.

4. Ongoing Reflection:

   • User interacts with AI assistant and reviews progress over time.

---

Initial Product Backlog

• User Authentication
• Life Category Management
• Weekly Reflection Logging
• Personalized AI Feedback
• AI Assistant Interaction
• Progress Visualization

---

Tech Stack

Frontend: Vite + React
Backend: Spring Boot (Java)
AI Service: Node.js + OpenAI API
Auth: Spring Security + JWT
Database: PostgreSQL
Containerization: Docker & Docker Compose
Monitoring: Prometheus + Grafana
Reverse Proxy: Traefik (AWS deploy)

---

Monitoring & Observability

Uses Prometheus (metrics) and Grafana (dashboards and alerting).

• Metrics: uptime, requests, errors, report counts.
• Dashboards: service-specific, CPU, trends.
• Alerts: Grafana rules for critical issues.

---

CI/CD Workflows

• GitHub Actions-based
• Automated testing for each service
• Matrix build system builds 8 Docker images: 4 for EC2, 4 for Kubernetes
• Build and push images to GHCR
• Manual EC2 deployment
• Automatic Helm deployment to Rancher on main or deployment branches

Note: EC2 and Rancher images differ by build args and tags. This distinction is currently handled via build arguments. For future development, a runtime configuration strategy (e.g., environment-based config injection or service discovery) will be adopted.

---

Build Details

Services:

• authentication
• wheel-of-life
• client
• genai

Matrix Target:

• ec2
• rancher

Tags:

• latest-ec2, latest-rancher, branch names, PR refs

---

AWS EC2 Deployment

• SCPs compose-aws.yml to EC2
• Generates .env file using GitHub secrets
• Runs Docker Compose

Domains (Traefik via nip.io):

• client.<IP>.nip.io
• authentication.<IP>.nip.io
• wheel-of-life.<IP>.nip.io
• genai.<IP>.nip.io

---

Rancher Helm Deployment

• Runs on push to main or deployment
• Loads kubeconfig
• Installs Helm and deploys using Helm chart
• Uses latest-rancher image tag

---

Secrets & Environment Variables

Name	Type	Description
EC2_PUBLIC_IP	Variable	EC2 IP for nip.io domains
AWS_EC2_USER	Variable	SSH user
AWS_EC2_PRIVATE_KEY	Secret	SSH key
POSTGRES_USER	Variable	Database user
POSTGRES_PASSWORD	Secret	Database password
GRAFANA_USER	Variable	Grafana login
GRAFANA_PASSWORD	Secret	Grafana password
JWT_SECRET	Secret	JWT signing key
OPENAI_API_KEY	Secret	GenAI service
KUBE_CONFIG	Secret	kubeconfig for Rancher
GITHUB_TOKEN	Secret	GHCR access

---

Workflow Diagram

graph TD
    A[Push to GitHub] --> B[Run Tests]
    B --> C[Build Docker Images]
    C --> D[Push to GitHub Container Registry]
    D --> E[Manual Deployment to EC2]
    D --> F[Automatic Helm Deploy to Rancher]

Loading

---

---

Diagrams

Use Case Diagram

Component Diagram

Analysis Object Model

---

API Documentation (OpenAPI/Swagger)

You can explore and test the API endpoints for each service using OpenAPI documentation:

• Authentication Service: http://localhost:8080/swagger-ui.html
• Wheel-of-Life Service: http://localhost:8081/swagger-ui.html
• GenAI Service: http://localhost:5001/apidocs

---

Getting Started (Local)

1. Clone the repository and create a .env file using .env.example as a reference.
2. Start the development environment:

bash

docker-compose up --build

3. Access the services:

• Frontend: http://localhost:5173
• Wheel of Life Service: http://localhost:8080
• Authentication Service: http://localhost:8081
• GenAI Service: http://localhost:5001

---

Deployment (AWS with Traefik)

For production deployment, use:

docker-compose -f compose-aws.yml up -d

Make sure the following environment variables are configured:

• CLIENT_HOST
• AUTH_HOST
• GENAI_HOST
• GRAFANA_HOST
• JWT_SECRET
• OPENAI_API_KEY

---

Environment Variables

Reference .env.example for required variables. Key ones include:

• POSTGRES_USER, POSTGRES_PASSWORD, POSTGRES_DB
• JWT_SECRET
• VITE_SERVER_URL, VITE_AUTH_URL, VITE_GENAI_URL
• OPENAI_API_KEY, OPENAI_MODEL
• GRAFANA_USER, GRAFANA_PASSWORD

---

Project Structure

├── client/ # Frontend (Vite + React) ├── server/ │ ├── wheel-of-life/ # Self-reflection microservice (Spring Boot) │ └── authentication/ # Auth microservice (Spring Boot) ├── genai/ # AI assistant backend ├── grafana/ # Monitoring dashboards ├── prometheus/ # Prometheus config for scraping metrics ├── docker-compose.yml # Local deployment ├── compose-aws.yml # Production deployment (Traefik + HTTPS) └── .env # Environment configuration

---

License

Licensed under the MIT License.

---

Authors

Developed by Team OpsOnTheRocks – TU Munich

---

Contact

For questions, contributions, or suggestions — feel free to reach out or open an issue.