Docker with Langfuse Local Setup Guide

This guide provides steps to set up Langfuse locally using Docker and run a Python tracing example.

Prerequisites

Git
Docker
Python 3.x
pip

Setup Steps

Step 1: Clone the Langfuse Repository

First, clone the Langfuse repository and navigate into its directory:

git clone https://github.com/langfuse/langfuse.git
cd langfuse

Step 2: Generate NEXTAUTH_SECRET Key

Generate a secure secret for NEXTAUTH_SECRET. Open your terminal and run the following command:

openssl rand -base64 32

Copy the output, as you will need it for the .env file.

Step 3: Prepare Langfuse Keys

You will need to create LANGFUSE_SECRET_KEY and LANGFUSE_PUBLIC_KEY. These are typically generated by Langfuse itself or can be set to arbitrary values for local development. For local setup, you can use any string values.

Set LANGFUSE_HOST to http://localhost:3000.

Step 4: Configure GROQ_API_KEY for Groq Models

If you plan to use Groq models in your application, you can provide your API key using the GROQ_API_KEY environment variable.

Step 1: Get Your Groq API Key Sign up or log in to your Groq account at https://console.groq.com.

Navigate to your API section and generate a new API key.

Step 5: Configure Environment Variables

Create a .env file in the root of the langfuse directory (or modify the existing one) with the following content. Replace the placeholder values with your generated keys.

DATABASE_URL=postgresql://postgres:postgres@localhost:5432/langfuse
LANGFUSE_SECRET_KEY=your_langfuse_secret_key_here
LANGFUSE_PUBLIC_KEY=your_langfuse_public_key_here

LANGFUSE_HOST=http://localhost:3000
# NEXTAUTH_SECRET=$(openssl rand -base64 32)
NEXTAUTH_SECRET=your_generated_nextauth_secret_here
TELEMETRY_ENABLED=false
GROQ_API_KEY= # Optional: Your Groq API key if you plan to use Groq models

Note: Ensure you replace your_langfuse_secret_key_here, your_langfuse_public_key_here, and your_generated_nextauth_secret_here with actual values.

Step 6: Configure NEXTAUTH_SECRET in docker-compose.yml

To ensure secure authentication for the Langfuse web interface, you need to set the NEXTAUTH_SECRET environment variable in your docker-compose.yml file.

Reference it in your docker-compose.yml under the langfuse-web service:

langfuse-web: image: ghcr.io/langfuse/langfuse:latest restart: always depends_on: *langfuse-depends-on ports: - 3000:3000 environment: «: *langfuse-worker-env NEXTAUTH_SECRET: ${NEXTAUTH_SECRET}

Step 7: Start Langfuse Services with Docker Compose

From the langfuse directory, run Docker Compose to start the Langfuse backend and frontend services:

docker-compose up

Once the services are up and running, you can access the Langfuse UI in your browser at: http://localhost:3000

Step 8: Run the Python Tracing Example

Open a new terminal (while docker-compose up is still running in the first terminal). Navigate back to your project directory (where lfdockerlocal.py is located) and run the Python tracing script:

python lfdockerlocal.py

This script will send traces to your locally running Langfuse instance.

Step 9: View Langfuse Traces

After running the Python script, refresh the Langfuse UI at http://localhost:3000 to view the generated traces.