☁️ ExtraNet.GlobalAzureKit

Open-source IoT workshop kit for Global Azure Torino 2026. Stream real-time sensor telemetry from an ESP32-C3 to Azure IoT Hub and visualize it in a live dashboard — all deployed with one Terraform command.

Engineered by ExtraNet, built in collaboration with GitHub Copilot — code, infrastructure and documentation crafted through human-AI pair programming.

About This Project

This isn't a branded gadget you forget in a drawer.

ExtraNet.GlobalAzureKit is a complete, working IoT pipeline — from a physical sensor on your desk to a real-time dashboard powered by Azure. We built it because we believe the best way to learn cloud isn't reading slides — it's getting your hands on real infrastructure, real telemetry, and real architecture decisions.

The kit reflects how we work every day: Terraform for repeatable deploys, Event Hub for scalable ingestion, Log Analytics for observability, least-privilege auth everywhere. The same patterns you see here are the ones that run in production — scaled up, hardened, and monitored.

GitHub Copilot was our accelerator: we directed the architecture, made the design choices, and Copilot turned them into code at speed. The expertise is ours; the velocity is AI.

What's in the Kit

Component	Description
ESP32-C3 Super Mini	RISC-V single-core microcontroller, Wi-Fi 4, USB-C, I²C
BMP280 sensor	Temperature + atmospheric pressure (I²C, Bosch Sensortec)
BME280 support	Temperature + pressure + humidity — pin-compatible drop-in replacement
LED	Standard LED (any color) — controlled via Cloud-to-Device commands from the dashboard
220Ω resistor	Current-limiting resistor for the LED (GPIO2 → 220Ω → LED → GND)
Arduino firmware	Template-based `.ino` with Azure SDK for Embedded C, SAS auth, MQTT/TLS
configure-sketch.ps1	Auto-generates the firmware from Terraform outputs — zero manual credential copy
Terraform IaC	One-command deploy: IoT Hub, Event Hub, Log Analytics, monitoring, alerts
Monitor dashboard	Flask + Chart.js + Socket.IO — dark theme, real-time charts, CSV export, LED control
Workshop guides	Step-by-step ESP32-C3 wiring, flashing, and troubleshooting (§1–§8)

Architecture

graph TB
    subgraph EDGE["🔌 Edge — ESP32-C3 + Sensors"]
        ESP["ESP32-C3 Super Mini\nRISC-V · Wi-Fi 4 · USB-C"]
        SENSOR["BMP280 / BME280\nTemperature · Pressure · Humidity*"]
        LED["LED + 220Ω Resistor\nGPIO2 · Active HIGH"]
        ESP -->|"I²C — SDA=GPIO8 · SCL=GPIO9"| SENSOR
        ESP -->|"GPIO2"| LED
    end

    subgraph AZURE["☁️ Microsoft Azure"]
        subgraph DM["📡 Device Management"]
            IOT["IoT Hub · F1 Free Tier\nMQTT over TLS 1.2"]
        end
        subgraph SP["📊 Stream Processing"]
            EH["Event Hub · Standard\n2 partitions · 1-day retention"]
            CG["Consumer Group\nmonitor"]
        end
        subgraph OBS["🔍 Observability"]
            LA["Log Analytics\n30-day retention"]
            ALERT["Metric Alert\ndropped telemetry > 10"]
        end
        subgraph IAC["🏗️ Infrastructure as Code"]
            TF["Terraform\nazurerm ~> 4.14"]
            STATE["Remote State\nAzure Storage"]
        end
    end

    subgraph LOCAL["🖥️ Local Dashboard"]
        DASH["Flask · Chart.js · Socket.IO\nReal-time visualization"]
    end

    IOT -->|"C2D · LED commands"| ESP
    ESP ==>|"D2C · MQTT/TLS"| IOT
    IOT -->|"Message Route"| EH
    EH --> CG -->|"AMQP"| DASH
    IOT -.->|"Diagnostics"| LA
    EH -.->|"Diagnostics"| LA
    LA -.-> ALERT
    TF -.->|"manages"| IOT
    TF -.->|"manages"| EH
    TF -.->|"manages"| LA
    TF -->|"state"| STATE

    style AZURE fill:#e8f1fb,stroke:#0078d4,stroke-width:3px
    style EDGE fill:#e8f5e9,stroke:#2e7d32,stroke-width:2px
    style LOCAL fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px

* Humidity available only with BME280 sensor

The Data Journey — Edge to Cloud

#	Step	Component	Protocol / Detail
1	Sense	BMP280 / BME280	I²C @ 0x76 — temperature, pressure, humidity*
2	Serialize	ESP32-C3 firmware	`{"temperature": 22.5, "pressure": 1013.2, "humidity": 65.3}`
3	Transmit	ESP32-C3 Wi-Fi	MQTT over TLS 1.2 → IoT Hub
4	Authenticate	Azure IoT Hub	Per-device SAS token (connection string)
5	Route	IoT Hub Message Route	All D2C messages → Event Hub endpoint
6	Buffer	Event Hub	Standard, 2 partitions, 1-day retention
7	Consume	Monitor dashboard	AMQP via `monitor` consumer group
8	Visualize	Browser	Chart.js + WebSocket push via Socket.IO
9	Control	Dashboard → IoT Hub	C2D messages — LED on/off via `{"action":"led","seconds":N}`
10	Observe	Log Analytics	Diagnostic logs → KQL queries + metric alerts

Quick Start

Goal: from zero to live dashboard in ~30 minutes.

Platform: the wizards and start.bat target Windows + PowerShell. On Linux/macOS you can still run the manual steps (Option B) — Terraform, Azure CLI and Python are all cross-platform — but the guided experience is Windows-first.

Option A — Interactive Wizard (recommended)

Windows — double-click start.bat (or run .\wizard-gui.ps1 in PowerShell).

A graphical window opens and guides you through every step with buttons, text fields, and a live log:

✅ Checks prerequisites (Terraform, Azure CLI, Python) with download links
🔐 Login to Azure (one click)
☁️ Deploy ~12 Azure resources (one click + confirmation)
📡 Configure firmware (type WiFi SSID/password, choose sensor, click Generate)
📊 Launch the monitor dashboard (one click → opens browser)

Every step enables the next. You can close and re-run anytime — it skips what's already done.

Terminal user? Run .\wizard.ps1 instead for a CLI-based wizard with the same steps.

Option B — Manual step-by-step

If you prefer to run each command yourself, expand the section below.

Every command shows what you should see. If the output doesn't match, a troubleshooting tip is right below.

Click to expand the manual procedure

Overview — The 4-Step Journey

Step	What you'll do	Time
1	Deploy Azure infrastructure (IoT Hub, Event Hub, monitoring)	~5 min
2	Wire ESP32-C3 + BMP280 sensor + LED with resistor and test with Arduino IDE	~15 min
3	Flash the IoT Hub firmware (credentials auto-injected)	~5 min
4	Launch the real-time dashboard and see live data	~3 min

Prerequisites

Install these tools before starting. Click the link to download, then verify with the command:

#	Tool	Install	Verify command
1	Terraform ≥ 1.5	Download	`terraform -version`
2	Azure CLI ≥ 2.50	Download	`az version`
3	Python ≥ 3.10	Download	`python --version`
4	Arduino IDE 2.x	Download	Open the app
5	Azure subscription	Create free account	—

⚠️ IoT Hub Free (F1) limit: Azure allows only one F1 IoT Hub per subscription. If you already have one (even empty or from an old experiment), the deploy will fail with MaxNumberOfIotHubsExceeded. Fix: delete the existing F1 hub, or set iothub_sku = "S1" in terraform.auto.tfvars (~€21/month pro-rata — pennies for a workshop).

Open PowerShell and verify everything is installed:

terraform -version

✅ You should see Terraform v1.x.x. If it says "not recognized" → install Terraform and restart PowerShell.

az version

✅ You should see "azure-cli": "2.x.x". If it says "not recognized" → install Azure CLI and restart PowerShell.

python --version

✅ You should see Python 3.1x.x. If it says "not recognized" → install Python (check "Add to PATH" during setup!) and restart PowerShell.

Finally, install the Azure IoT extension (safe to re-run if already installed):

az extension add --name azure-iot

Step 1 — Deploy Azure Infrastructure

Open PowerShell in the repository folder (where you cloned/unzipped this project).

1a. Login to Azure

az login

What happens: your default browser opens. Sign in with your Azure account.

When done, the terminal shows your subscription info:

[
  {
    "name": "Your Subscription Name",
    "id": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
    ...
  }
]

❌ Browser doesn't open? Try az login --use-device-code instead — it gives you a code to enter at https://microsoft.com/devicelogin

1b. Navigate to the terraform folder

cd terraform

No output? That's normal — it just changed directory.

1c. Create the Terraform state backend

.\scripts\setup-tfstate-backend.ps1

What you'll see: the script creates an Azure Storage Account (takes ~1 minute):

Creating resource group 'globalazure-tfstate-rg'...
Creating storage account '...'...
Creating container 'tfstate'...
✅ Backend configuration written to backend.tfvars

❌ Error "subscription not found"? Go back to step 1a and make sure az login succeeded.

1d. Create and edit the variables file

Copy-Item terraform.auto.tfvars.example terraform.auto.tfvars

No output? That's correct — the file was copied silently.

Now find your Subscription ID:

az account show --query id -o tsv

What you'll see: a string like xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx. Select it and copy (Ctrl+C).

Open the variables file in Notepad:

notepad terraform.auto.tfvars

Find this line:

subscription_id = "ENTER-YOUR-SUBSCRIPTION-ID-HERE"

Replace the placeholder with your Subscription ID so it looks like:

subscription_id = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

Save the file (Ctrl+S) and close Notepad.

1e. Initialize Terraform

terraform init -backend-config=backend.tfvars

What you'll see (takes a few seconds):

Initializing the backend...
...
Terraform has been successfully initialized!

❌ "Error configuring the backend"? Check that backend.tfvars exists (step 1c). ❌ "Backend configuration changed"? Type yes to migrate.

1f. Deploy!

terraform apply

What you'll see: Terraform shows a plan listing ~12 Azure resources. At the bottom:

Do you want to perform these actions?
  Enter a value:

Type yes and press Enter.

Wait 2–3 minutes. When it finishes:

Apply complete! Resources: 12 added, 0 changed, 0 destroyed.

🎉 Azure infrastructure is ready!

❌ Error about IoT Hub region? Open terraform.auto.tfvars again and change iothub_location = "westeurope" to "northeurope".

Full details and troubleshooting: terraform/README.md

Step 2 — Wire and Test the ESP32-C3

Follow the hardware setup guide (§1–§7):

Guide	When to use
esp32c3_quickstart.md	Happy path — everything works first try
esp32c3_setup_guide.md	Troubleshooting — detailed explanations for each step
esp32c3_super_mini_pinout.md	Pinout reference — wiring diagram

✅ Done when you see sensor readings in the Serial Monitor:

Temp: 24.35 C   Pressure: 1013.25 hPa

Step 3 — Flash the IoT Hub Firmware

3a. Install Arduino libraries

In Arduino IDE: Sketch → Include Library → Manage Libraries...

Search and install these (one at a time):

Search for	Click "Install" on	Notes
`azure-sdk-for-c`	azure-sdk-for-c by Microsoft	Required
`Adafruit BMP280`	Adafruit BMP280 Library by Adafruit	Install if you have a BMP280
`Adafruit BME280`	Adafruit BME280 Library by Adafruit	Install if you have a BME280

3b. Generate the configured sketch

Go back to PowerShell (you should still be in the terraform folder from Step 1):

cd ../firmware

$pass = Read-Host "WiFi password" -AsSecureString
.\configure-sketch.ps1 -WifiSsid "YOUR_SSID" -WifiPassword $pass -SensorType BME280 -EnableLed

Replace BME280 with BMP280 if you have a BMP280 sensor.

⚠️ Replace YOUR_SSID with your Wi-Fi network name. You'll be prompted for the password securely. The Wi-Fi must be 2.4 GHz — ESP32-C3 doesn't support 5 GHz networks. Add -EnableLed to enable LED control from the dashboard (the kit includes an LED + 220Ω resistor on GPIO2).

What you'll see:

Reading Terraform outputs...
Reading device key from Azure CLI...
Generating sketch for BMP280...
✅ Sketch written to output\GlobalAzureKit\GlobalAzureKit.ino

❌ "terraform: not recognized"? You moved to a different terminal. Open a new PowerShell in the repository folder and cd firmware. ❌ "Cannot find device"? Step 1f (terraform apply) must have completed successfully first.

3c. Upload to ESP32-C3

In Arduino IDE, open firmware\output\GlobalAzureKit\GlobalAzureKit.ino
Check Tools menu settings:

Setting Value

Board ESP32C3 Dev Module

Port COMx (the port you found in §2)

USB CDC On Boot Enabled ⚠️ critical!
Close the Serial Monitor if it's open (can't upload while it's using the port)
Press Upload (→ button) — wait for Done uploading.
Open Tools → Serial Monitor — set baud rate to 115200

What you'll see:

========================================
 ExtraNet.GlobalAzureKit
 Global Azure Torino 2026
========================================
[INFO] BMP280 sensor detected at 0x76
[INFO] Connecting to WiFi ...
.....
[INFO] WiFi connected, IP address: 192.168.x.x
[INFO] MQTT event MQTT_EVENT_CONNECTED
[INFO] Message published: {"temperature":24.35,"pressure":1013.25,"has_led":true}

If you see MQTT_EVENT_CONNECTED — the device is talking to Azure! 🎉

❌ Serial Monitor blank? → Tools → USB CDC On Boot → Enabled, then re-upload. ❌ "WiFi not connected"? → Check SSID/password. Must be 2.4 GHz (not 5 GHz). ❌ MQTT error? → Check error details in Serial Monitor. Try running configure-sketch.ps1 again.

Full troubleshooting: firmware/README.md

Step 4 — Launch the Monitor Dashboard

4a. Generate the .env configuration file

Go back to PowerShell:

cd ../terraform
terraform output -raw dot_env_content | Out-File -FilePath ../monitor/.env -Encoding UTF8

No output? That's correct — the .env file was created silently.

4b. Set up and start the dashboard

cd ../monitor
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt

You'll see pip downloading packages — wait until it finishes (takes ~30 seconds).

python app.py

What you'll see:

 * Running on http://127.0.0.1:5000

❌ "EVENTHUB_CONNECTION_STRING not set"? → Step 4a failed. Go back and re-run the terraform output command. ❌ "No module named flask"? → Run .\.venv\Scripts\Activate.ps1 first, then pip install -r requirements.txt.

4c. Open the dashboard

Open your browser and go to: http://localhost:5000

What you'll see:

A dark-themed dashboard with the Azure color palette
Temperature and pressure charts updating in real-time
A device table showing your ESP32-C3 with its latest readings
A telemetry log at the bottom with live messages

Done! 🎉

Congratulations — the kit is working end-to-end!

The complete data journey is live:

Sensor → ESP32-C3 → Wi-Fi → Azure IoT Hub → Event Hub → Dashboard

When you're done, remember to destroy the Azure resources to avoid ongoing costs:
cd terraform
terraform destroy

Supported Sensors

Sensor	Temperature	Pressure	Humidity	I²C Address	Notes
BMP280	✅	✅	—	0x76 / 0x77	Included in the kit
BME280	✅	✅	✅	0x76 / 0x77	Drop-in replacement, same pinout

The dashboard auto-detects the sensor type from the telemetry payload:

BMP280: {"temperature": 22.5, "pressure": 1013.2, "has_led": true}
BME280: {"temperature": 22.5, "pressure": 1013.2, "humidity": 65.3, "has_led": true}

When humidity data arrives, the humidity chart and stats appear automatically.

Dashboard Features

Dark theme with Azure-inspired color palette
5 stat cards — active devices, messages, temperature, pressure, humidity
Real-time charts — temperature, pressure, and humidity (Chart.js)
Device filter pills — click to show/hide individual devices on charts and table
Time range buttons — 1m, 5m, 15m, 1h, All
CSV export — per device or all devices combined
Device table — sensor type (BMP280/BME280), status, readings, message count
LED indicator — shows LED status per device (grey/off/glowing) when firmware has LED enabled
LED shortcut — one-click 💡 ON button for devices with LED support
C2D messaging — send any JSON command to a device via Cloud-to-Device
Collapsible telemetry log — color-coded entries with timestamps
Auto-detection — humidity chart appears only when BME280 data arrives
Live connection indicator — shows WebSocket status

See monitor/README.md for setup and API reference.

Project Structure

.
├── .gitignore                          # Git ignore rules
├── README.md                           # This file
├── start.bat                           # Double-click launcher for the GUI wizard
├── wizard-gui.ps1                      # GUI setup wizard (Windows Forms)
├── wizard.ps1                          # CLI setup wizard (terminal fallback)
├── LICENSE                             # Unlicense (public domain)
├── .editorconfig                       # Editor settings (indent, charset, EOL)
├── .github/
│   ├── copilot-instructions.md         # Copilot rules and conventions
│   └── prompts/
│       └── terraform-deploy.prompt.md  # Guided Azure deploy procedure
├── firmware/
│   ├── README.md                       # Firmware documentation
│   ├── configure-sketch.ps1            # Mustache template → configured sketch
│   ├── GlobalAzureKit/
│   │   └── GlobalAzureKit.mustache     # Mustache template (sensor blocks + credentials)
│   └── output/                         # Generated sketches (gitignored)
│       └── GlobalAzureKit/
│           └── GlobalAzureKit.ino      # Ready-to-flash sketch
├── monitor/
│   ├── app.py                          # Flask server + Event Hub consumer
│   ├── requirements.txt                # Python dependencies
│   ├── requirements-lock.txt            # Pinned dependency versions
│   ├── README.md                       # Monitor documentation
│   └── templates/
│       └── index.html                  # Dashboard (Chart.js + Socket.IO)
├── terraform/
│   ├── main.tf                         # Provider, backend, locals
│   ├── variables.tf                    # Input variables
│   ├── resources.tf                    # IoT Hub, Event Hub, devices
│   ├── monitoring.tf                   # Log Analytics, diagnostics, alerts
│   ├── outputs.tf                      # Connection strings, .env output
│   ├── backend.tfvars.example          # Backend config template
│   ├── terraform.auto.tfvars.example   # Variables template
│   ├── .gitignore                      # Excludes tfstate and sensitive files
│   ├── README.md                       # Terraform documentation
│   └── scripts/
│       └── setup-tfstate-backend.ps1   # Bootstrap storage for tfstate
├── esp32c3_setup_guide.md              # Full guide (§1–§8 + troubleshooting)
├── esp32c3_quickstart.md               # Quick-start guide (happy path)
└── esp32c3_super_mini_pinout.md        # ESP32-C3 pinout + wiring reference

Security

This kit is designed for workshop environments. Credentials are never committed to git:

Asset	Protection
Terraform state (`.tfstate`)	Remote Azure Storage backend, gitignored
Variable files (`.tfvars`)	`.gitignore` — only `.example` templates in repo
Event Hub connection string	`.env` file, gitignored
Device credentials	Mustache template with `{{&token}}` placeholders; `configure-sketch.ps1` injects at runtime
Generated firmware	`firmware/output/` is gitignored — never committed
Auth rules	Least privilege: send-only for devices, listen-only for monitor

Documentation

Document	Description
terraform/README.md	Terraform infrastructure — deploy, customize, destroy
monitor/README.md	Monitor dashboard — setup, API reference, features
firmware/README.md	Arduino firmware — libraries, configure-sketch, upload, troubleshoot
esp32c3_setup_guide.md	Full ESP32-C3 guide: wiring, flashing, troubleshooting (§1–§8)
esp32c3_quickstart.md	Quick-start ESP32-C3 guide (happy path)
esp32c3_super_mini_pinout.md	ESP32-C3 Super Mini pinout + sensor wiring

Estimated Azure Costs

Resource	Cost
IoT Hub F1	Free (8,000 msgs/day, 500 devices)
Event Hub Standard	~€13/month (1 TU)
Log Analytics	~€2.50/GB ingested
Storage (tfstate)	~€0.02/month
Estimated total	~€13–18/month

For a workshop lasting a few hours, the cost is negligible. Run terraform destroy when done!

Design Philosophy

This kit is designed for hands-on learning — not production use — but every architectural choice mirrors real-world practice:

Decision	Why
Event Hub instead of built-in IoT Hub endpoint	Decouples ingestion from consumption — the same pattern used when you add Stream Analytics, Azure Functions, or multiple consumers in production
Terraform with remote state	Infrastructure as Code from day one — not portal clicks you can't reproduce. Remote state on Azure Storage enables team collaboration
Separate auth rules (send / listen)	Least-privilege by default. The device can only send; the dashboard can only read. A compromised consumer can't inject messages
Log Analytics + metric alerts	Observability isn't an afterthought — if telemetry drops, you know immediately. This is the same stack that monitors production workloads
Mustache-based firmware template	Credentials never touch git. The generation script reads secrets at runtime from Terraform + Azure CLI — zero manual copy-paste
SAS token auth (not X.509)	Simpler for a workshop, but the architecture supports certificate-based auth with minimal changes when you move to production

Credits


ExtraNet Srl	Architecture, design decisions, and project direction. We build cloud solutions on Azure — this kit is a taste of how we work
GitHub Copilot	AI-accelerated development — code, infrastructure, and documentation generated through human-AI collaboration
Global Azure	The worldwide community event this kit was created for
Bosch Sensortec	BMP280 / BME280 sensor hardware
Espressif Systems	ESP32-C3 RISC-V microcontroller

License

This project is released into the public domain under the Unlicense. You can copy, modify, distribute, and use it for any purpose — no attribution required.

Engineered by ExtraNet, AI-accelerated with GitHub Copilot
Global Azure Torino 2026

Name		Name	Last commit message	Last commit date
.github		.github
firmware		firmware
monitor		monitor
terraform		terraform
.editorconfig		.editorconfig
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
esp32c3_quickstart.md		esp32c3_quickstart.md
esp32c3_setup_guide.md		esp32c3_setup_guide.md
esp32c3_super_mini_pinout.md		esp32c3_super_mini_pinout.md
start.bat		start.bat
wizard-gui.ps1		wizard-gui.ps1
wizard.ps1		wizard.ps1

Setting	Value
Board	ESP32C3 Dev Module
Port	COMx (the port you found in §2)
USB CDC On Boot	Enabled ⚠️ critical!

Latest commit

History

Folders and files

Repository files navigation

☁️ ExtraNet.GlobalAzureKit

About This Project

What's in the Kit

Architecture

The Data Journey — Edge to Cloud

Quick Start

Option A — Interactive Wizard (recommended)

Option B — Manual step-by-step

Overview — The 4-Step Journey

Prerequisites

Step 1 — Deploy Azure Infrastructure

1a. Login to Azure

1b. Navigate to the terraform folder

1c. Create the Terraform state backend

1d. Create and edit the variables file

1e. Initialize Terraform

1f. Deploy!

Step 2 — Wire and Test the ESP32-C3

Step 3 — Flash the IoT Hub Firmware

3a. Install Arduino libraries

3b. Generate the configured sketch

3c. Upload to ESP32-C3

Step 4 — Launch the Monitor Dashboard

4a. Generate the .env configuration file

4b. Set up and start the dashboard

4c. Open the dashboard

Done! 🎉

Supported Sensors

Dashboard Features

Project Structure

Security

Documentation

Estimated Azure Costs

Design Philosophy

Credits

License

About

Resources

Stars

Watchers

Forks

Releases

Packages

Used by

Contributors

Languages