Skip to main content

Frequently Asked Questions

Find answers to common questions about EdgeFlow. Can't find what you're looking for? Ask on Discord

General

What is EdgeFlow?

EdgeFlow is a lightweight, high-performance workflow automation platform designed specifically for IoT and edge computing. It provides a visual flow editor for building automations with 100+ pre-built nodes, all in a 10MB binary that runs on devices as small as Raspberry Pi Zero.

How does EdgeFlow compare to Node-RED?

EdgeFlow is inspired by Node-RED but optimized for resource-constrained devices:

10MB Binary size vs Node-RED's 80MB+
50MB RAM idle vs Node-RED's 150MB+
<1s Startup time vs Node-RED's ~5s
Go Language Native performance
Is EdgeFlow free?

Yes! EdgeFlow is 100% free and open source under the Apache 2.0 license. You can use it for personal projects, commercial applications, and modify it as needed.

What platforms are supported?
🍓Raspberry Pi (Zero, 3, 4, 5) 🐧Linux (x86_64, ARM64) 🪟Windows (x86_64) 🍎macOS (Intel, Apple Silicon) 🐳Docker (any platform)

Installation

What are the minimum requirements?
RAM 256MB min / 512MB recommended
Disk 50MB for binary + data
CPU Any ARM or x86 processor
How do I install EdgeFlow?

The quickest way is using the install script:

curl -fsSL https://edgx.cloud/install.sh | bash
Can I run EdgeFlow in Docker?

Yes! Docker is the recommended deployment method:

docker run -d -p 8080:8080 edgeflow/edgeflow:latest

Usage

How do I create my first flow?
1 Open EdgeFlow at http://localhost:8080
2 Drag an Inject node from the palette
3 Drag a Debug node
4 Connect them by dragging from output to input
5 Click Deploy
6 Click the Inject node button to trigger
Learn more in the Quick Start guide
Where are flows stored?

Flows are stored in SQLite database by default at:

./data/edgeflow.db

You can also export flows as JSON for backup or sharing.

Can I import Node-RED flows?

EdgeFlow has a Node-RED compatibility layer for basic flows. Complex flows may need some adjustments due to differences in the execution model.

Custom function nodes may require minor syntax changes

Hardware

How do I access GPIO pins?

Use the GPIO nodes in the palette. On Raspberry Pi, EdgeFlow automatically detects available pins. You may need to run with sudo or add your user to the gpio group:

sudo usermod -a -G gpio $USER
Which sensors are supported?

EdgeFlow includes native nodes for popular sensors:

DHT22 / DHT11 Temperature & Humidity
DS18B20 Temperature
BME280 / BME680 Environmental
PIR Motion
HC-SR04 Ultrasonic Distance
BH1750 Light

And more via I2C/SPI interfaces

Can I use I2C and SPI devices?

Yes! EdgeFlow has native I2C and SPI nodes. Enable the interfaces in your Pi's config using:

sudo raspi-config

Then navigate to Interface Options and enable I2C and/or SPI.

Networking

How do I send MQTT messages?

Use the MQTT Out node. Configure your broker address, topic, and optionally authentication. The node supports QoS 0, 1, and 2.

View MQTT node documentation
Can EdgeFlow act as an HTTP server?

Yes! Use the HTTP In node to create endpoints. EdgeFlow can handle:

  • Webhooks from external services
  • REST APIs for your applications
  • Dynamic content serving
How do I connect my Raspberry Pi to Wi-Fi?

Raspberry Pi OS Bookworm and later uses NetworkManager for all network configuration. The easiest method is nmtui:

  • Run sudo nmtui in the terminal
  • Select "Activate a connection"
  • Choose your Wi-Fi network and enter the password
  • A * next to the name confirms you're connected

Alternatively, use the command line: sudo nmcli dev wifi connect "SSID" password "pass"

View full Wi-Fi setup guide
Is there WebSocket support?

Yes, EdgeFlow supports both WebSocket modes:

  • Client mode - Connect to external WebSocket servers
  • Server mode - Accept incoming WebSocket connections

Perfect for real-time bidirectional communication!

AI / Machine Learning

Can I use GPT-4 or Claude?

Yes! EdgeFlow includes nodes for:

GPT-4, GPT-4 Turbo, ChatGPT
Claude 3.5 Sonnet, Claude 3 Opus

Just add your API key in the node configuration.

Can I run local LLMs?

Yes! Use the Ollama node to run local language models. This is great for:

  • Privacy-sensitive applications
  • Offline / air-gapped environments
  • Avoiding API costs
Requires sufficient RAM (4GB+ recommended)

Troubleshooting

EdgeFlow won't start. What should I check?
1
Check if port is in use
lsof -i :8080
2
Verify available RAM
free -m
3
Check logs (if using systemd)
journalctl -u edgeflow
4
Enable debug logging
EDGEFLOW_LOGGER_LEVEL=debug edgeflow serve
GPIO isn't working. How do I fix it?
1
Add user to gpio group
sudo usermod -a -G gpio $USER
2
Log out and back in Group changes require a new session
3
Check GPIO interface
ls /dev/gpio*
How do I report a bug?

Open an issue on our GitHub repository with:

  • Your system details (OS, architecture)
  • EdgeFlow version
  • Steps to reproduce the issue
  • Any error messages or logs

Still have questions?

Can't find what you're looking for? We're here to help!