docker compose up command starts the databases; a single pnpm dev command starts the web app, API, OTLP ingest proxy, and background workers. All of your telemetry stays on your own infrastructure.
Prerequisites
Before you begin, make sure you have the following installed:- Node.js 20+ — nodejs.org
- pnpm 9+ —
npm install -g pnpm - Docker (with the Compose plugin) — docker.com
Local service endpoints
Once running, the stack exposes three local endpoints:
Point your application’s OTLP exporter at
http://localhost:4101 to start sending data.
Setup
1
Clone the repository
Clone the Superlog monorepo from GitHub:
2
Install dependencies
Superlog uses pnpm workspaces. Install all dependencies from the repo root:
3
Start the databases
Use Docker Compose to start the database services in the background. The compose file also starts the OpenTelemetry Collector, which receives OTLP data and writes it to storage.Wait for all containers to report as healthy before continuing. You can check their status with:
4
Run database migrations
Apply the schema migrations to the database:You only need to run this once on a fresh install, and again after pulling commits that include new migrations.
5
Start all services
Start the web app, API, ingest proxy, and background workers together with a single command from the repo root:Open your browser at
http://localhost:5173. The onboarding wizard will guide you through creating your first organisation and API key.Editions
The Community edition you are running is fully open-source under the Apache License 2.0. You are free to use, modify, and distribute it for any purpose — including commercial use — without paying anything.The default community agent runner records a local incident summary for every detected incident. To enable AI-powered root-cause analysis and automated fix PRs, connect a GitHub integration — see the tip below.
Superlog Cloud is available at superlog.sh with a free tier and pay-as-you-go plans. Cloud provides managed infrastructure, automatic upgrades, and the full AI agent experience without any ops overhead.