GlassFlow CLI

Local development environment for GlassFlow ETL

The GlassFlow CLI provides a quick way to set up a local development environment for exploring and testing GlassFlow - an open-source ETL tool for real-time data processing from Kafka to ClickHouse.

Note: This CLI is designed for local testing, demos, and exploration only. For production deployments, use the official GlassFlow Helm charts.

⚡️ Quick Start

Prerequisites

• Docker (or compatible runtime like Docker Desktop, OrbStack, Colima, or Podman)
• Helm (v3) – used to install charts (installed automatically via Homebrew, or install manually)
• kubectl (installed automatically via Homebrew, or install manually)

Give Docker enough resources (e.g. 6–8 GB RAM, 4 CPUs) so all pods can schedule. If pods stay Pending, increase memory/CPU in Docker Desktop → Settings → Resources.

Installation

Install via Homebrew (Recommended)

brew tap glassflow/tap
brew install glassflow

Install from GitHub Releases

Download the latest release for your platform from GitHub Releases.

Usage

Step 1: Start GlassFlow

glassflow up

This creates a Kind cluster and installs GlassFlow. Once ready, you can access:

• GlassFlow UI: http://localhost:30080
• GlassFlow API: http://localhost:30180

From the UI, you can connect to your own Kafka and ClickHouse instances.

Step 2 (optional): Run the demo

To see data flowing end-to-end with a local Kafka and ClickHouse:

glassflow setup-demo

This installs Kafka + ClickHouse, creates a demo pipeline, and starts a Kafka producer sending sample events.

All-in-one (GlassFlow + Kafka + ClickHouse + demo pipeline):

glassflow up --demo

Stop the environment:

glassflow down

What Gets Installed

glassflow up creates a Kind cluster and installs:

• GlassFlow ETL (API, UI, operator, NATS, PostgreSQL) in the glassflow namespace
• Port forwarding for the GlassFlow UI and API

glassflow setup-demo adds:

• Kafka (kafka namespace) and ClickHouse (clickhouse namespace)
• A demo pipeline with a Kafka producer sending sample events
• Port forwarding for ClickHouse HTTP (http://localhost:30090)

glassflow up --demo does both in one step.

Commands

# Start GlassFlow (Kind cluster + GlassFlow only)
glassflow up

# Set up demo (installs Kafka + ClickHouse + demo pipeline)
glassflow setup-demo

# All-in-one (GlassFlow + Kafka + ClickHouse + demo)
glassflow up --demo

# Stop and clean up
glassflow down

# Force stop (skip Helm uninstall, delete cluster directly)
glassflow down --force

# Show version
glassflow version

# Use a custom config file
glassflow up -c /path/to/config.yaml
glassflow --help

Production Deployment

For production use, deploy GlassFlow using the official Helm charts:

• Helm Charts Repository: github.com/glassflow/charts
• Installation Guide: docs.glassflow.dev/installation/kubernetes

Resources

• GlassFlow ETL: github.com/glassflow/clickhouse-etl
• Documentation: docs.glassflow.dev
• Slack Community: Join GlassFlow Hub
• Helm Charts: github.com/glassflow/charts