# TaskFlow — Cloud Deployment (sk1)

> A task management web application deployed on Microsoft Azure.
> Live at: **https://taskflow-14802.azurewebsites.net**

---

## 1. What the Application Does

TaskFlow is a simple web-based task manager for individuals or small teams.

Users can:
- **Create** tasks with a title and optional description
- **Advance** tasks through three stages: `To Do → In Progress → Done`
- **Delete** tasks they no longer need

All data is stored in a managed PostgreSQL database and persists across restarts and redeployments.

---

## 2. Public Cloud and Infrastructure Description

### Cloud Provider
**Microsoft Azure** — Azure for Students subscription.

### Azure Services Used

| Service | Purpose |
|---|---|
| **Azure Container Registry (ACR)** — Basic | Stores the Docker image for the Flask app |
| **Azure App Service Plan** — Linux B1 | Managed hosting platform for the container |
| **Azure Web App** (container) | Runs the Flask app with built-in HTTPS and auto-restart |
| **Azure Database for PostgreSQL Flexible Server** — Burstable B1ms | Managed PostgreSQL 15 database with persistent storage |

### Components

| Component | Type | Role |
|---|---|---|
| `taskflow-web` | Docker container on App Service | Python 3.11 / Flask app served by Gunicorn |
| `postgres` | Azure managed database | PostgreSQL 15 — stores all tasks |
| `taskflow-registry` | Azure Container Registry | Private Docker image registry |

### How Traffic Flows

```
Browser (HTTPS)
  → Azure App Service (built-in TLS, azurewebsites.net cert)
    → Flask container (Gunicorn :5000)
      → Azure PostgreSQL Flexible Server (:5432, SSL required)
        → Managed persistent storage (Azure-managed disk)
```

### HTTPS / TLS
Azure App Service provides a **free, automatic, browser-trusted TLS certificate** for all `*.azurewebsites.net` domains. HTTPS-only mode is enforced — HTTP requests are automatically redirected to HTTPS.

### Persistent Storage
Data is stored in **Azure Database for PostgreSQL Flexible Server**. This is a fully managed service — Azure handles backups, high availability, and storage automatically. Data persists independently of the application container.

### Automatic Restart
Azure App Service automatically restarts the container if it crashes or exits. The platform monitors the `/health` endpoint and restarts unhealthy instances without manual intervention.

### Configuration
All secrets (database host, user, password) are stored as **App Service Application Settings** — Azure's equivalent of environment variables. They are injected into the container at runtime and never stored in code or GIT.

---

## 3. Cost Analysis — 1,000 Users/Day, 50 GB Data (Azure, 1 Year)

| Resource | Specification | Unit Price | Billing | Annual Cost |
|---|---|---|---|---|
| App Service Plan | Linux B1 (1 vCPU, 1.75 GB RAM) | ~$13.14/month | Monthly | **~$157.68** |
| PostgreSQL Flexible Server | Burstable B1ms (1 vCPU, 2 GB) | ~$12.41/month | Monthly | **~$148.92** |
| PostgreSQL Storage | 50 GB | ~$5.76/month | Monthly | **~$69.12** |
| Azure Container Registry | Basic tier | ~$5.00/month | Monthly | **~$60.00** |
| Outbound bandwidth | First 100 GB free, ~60 GB/month at 1k users | $0 | Per GB | **~$0** |
| HTTPS certificate | Free (azurewebsites.net) | $0 | — | **$0** |
| **Total** | | | | **≈ $435.72 / year** |

> For the exam demo period the actual cost is ~$35/month (B1 plan + B1ms PostgreSQL). Delete resources after the exam with `remove-app.sh` to stop billing.

---

## 4. Uploaded Files and Their Content

```
sk1/
├── app/
│   ├── app.py              Flask app — routes, DB connection, Kanban UI, /health endpoint
│   ├── Dockerfile          Builds taskflow-web image (Python 3.11, Gunicorn, non-root)
│   └── requirements.txt    Python deps: flask, gunicorn, psycopg2-binary
├── db/
│   └── init.sql            SQL schema — tasks table, trigger, sample data
├── setup.sh                Full deployment — writes files, creates all Azure resources
├── prepare-app.sh          Rebuilds and redeploys container after code changes
├── remove-app.sh           Deletes entire Azure resource group
├── backup.sh               pg_dump backup and restore via psql
├── .env.example            Template for required environment variables (no secrets)
├── .gitignore              Excludes .env and backups from GIT
└── README.md               This file
```

---

## 5. Configuration Description

### Secrets and Environment Variables

Secrets are stored as **Azure App Service Application Settings** — never in GIT.

| Variable | Where stored | Used by |
|---|---|---|
| `POSTGRES_HOST` | App Service settings | Flask app |
| `POSTGRES_USER` | App Service settings | Flask app |
| `POSTGRES_PASSWORD` | App Service settings | Flask app |
| `POSTGRES_DB` | App Service settings | Flask app |
| `POSTGRES_PORT` | App Service settings | Flask app |
| `FLASK_ENV` | App Service settings | Gunicorn |
| `WEBSITES_PORT` | App Service settings | Azure (tells it app runs on 5000) |
| `DB_PASSWORD` | Shell env var at deploy time | `setup.sh` only, not stored |

A local `.env` file is written during deployment for use by `backup.sh`. It is excluded from GIT by `.gitignore`.

---

## 6. How to View and Use the Application

### Prerequisites
- Azure CLI installed and `az login` completed
- Docker installed and running
- `psql` installed (for backup/restore and schema init)
- `DB_PASSWORD` environment variable set

### Deploy (first time — creates all Azure resources)
```bash
export DB_PASSWORD="YourStrongPassword123!"
chmod +x setup.sh && ./setup.sh
```
Takes approximately **5–8 minutes**.

### Open in Browser
```
https://taskflow-14802.azurewebsites.net
```

### Redeploy after code changes
```bash
./prepare-app.sh
```

### Remove all Azure resources
```bash
./remove-app.sh
```

---

## 7. How to Perform a Data Backup

Install psql if needed: `sudo apt install postgresql-client`

```bash
# Create a timestamped backup
source .env && ./backup.sh

# List backups
ls -lh backups/

# Restore
source .env && ./backup.sh --restore backups/taskflow-20250601-120000.sql
```

---

## 8. How to View Access Logs from the Internet

Azure App Service logs all HTTP requests. View them via Azure CLI:

```bash
# Stream live access logs (all internet requests to the app)
az webapp log tail --resource-group taskflow-rg --name taskflow-14802

# Enable detailed logging first if needed
az webapp log config   --resource-group taskflow-rg   --name taskflow-14802   --application-logging filesystem   --web-server-logging filesystem   --level information

# Download logs as zip
az webapp log download --resource-group taskflow-rg --name taskflow-14802
```

Each log entry contains: timestamp, client IP, HTTP method, path, status code, response time.

---

## 9. Conditions for Running prepare-app.sh and remove-app.sh

### prepare-app.sh
- Azure CLI installed and `az login` completed
- Docker installed and running
- `.env` file must exist in the same directory (created by `setup.sh`)
- ACR name and App name are hardcoded from the initial deployment — re-run `setup.sh` to create fresh resources

### remove-app.sh
- Azure CLI installed and `az login` completed
- Resource group `taskflow-rg` must exist
- Prompts for confirmation before deleting
- All data will be lost unless `backup.sh` was run first

---

## 10. External Resources and Use of Generative AI

### External Resources

| Resource | Type | How Used |
|---|---|---|
| [Azure App Service Documentation](https://learn.microsoft.com/en-us/azure/app-service/) | Official docs | Web app creation, container config, app settings, logging |
| [Azure PostgreSQL Flexible Server](https://learn.microsoft.com/en-us/azure/postgresql/flexible-server/) | Official docs | Server creation, firewall, connection strings |
| [Azure Container Registry](https://learn.microsoft.com/en-us/azure/container-registry/) | Official docs | Image push, admin credentials |
| [Flask Documentation](https://flask.palletsprojects.com/) | Official docs | Routing, templates |
| [PostgreSQL pg_dump](https://www.postgresql.org/docs/current/app-pgdump.html) | Official docs | Backup commands |
| [Azure Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) | Tool | Cost analysis |
| Course lectures (azure, certbot, compose) | Lecture slides | Architecture, PaaS concepts, IaaS vs PaaS |

### Generative AI Usage
**Tool:** Claude by Anthropic (claude.ai)
**Type:** Productivity assistance — Azure CLI commands, shell scripts, Dockerfile, Flask app, README.
**Method:** Student described requirements, reviewed all output, verified understanding before submission. All architectural decisions directed by the student. AI accelerated writing; understanding was the student's own.