UNPKG

@lobehub/chat

Version:

Lobe Chat - an open-source, high-performance chatbot framework that supports speech synthesis, multimodal, and extensible Function Call plugin system. Supports one-click free deployment of your private ChatGPT/LLM web application.

72 lines (50 loc) 2.64 kB
--- title: Observability with Grafana, Prometheus, and Tempo description: >- Monitor and analyze your LobeChat instance using Grafana dashboards, Prometheus metrics, and Tempo traces. This guide covers setup, configuration for self-hosted deployments. tags: - Observability - Grafana - Prometheus - Tempo --- # Observability with [Grafana](https://grafana.com/), [Prometheus](https://prometheus.io/), and [Tempo](https://grafana.com/docs/tempo/latest/) LobeChat supports advanced observability for self-hosted deployments using open-source tools: - **Grafana** for dashboards and visualization - **Prometheus** for metrics collection - **Tempo** for distributed tracing - **otel-collector** ingesting other OpenTelemetry supported data We provided Docker Compose (`docker-compose`) file presets to bootstrap the observability stack with advanced self-hosting features. This guide will help you set up and use these tools to monitor your LobeChat instance. ## Prerequisites - `docker` CLI - OrbStack (macOS), or Docker Desktop Windows - `docker-compose` plugin enabled (check through `docker compose version`) ## 1. Deploy ```bash curl -O https://raw.githubusercontent.com/lobehub/lobe-chat/HEAD/docker-compose/production/grafana/docker-compose.yml curl -O https://raw.githubusercontent.com/lobehub/lobe-chat/HEAD/docker-compose/production/grafana/.env.example mv .env.example .env ``` 1. Update the password & secrets in the `.env` file as needed. 2. Start the stack with the correct profile: ```sh docker compose up -d ``` This will launch Grafana, Prometheus, Tempo, and the otel-collector alongside LobeChat with Casdoor, Minio, and other advanced services. ## 2. Access Grafana Dashboards - Open Grafana in your browser at [http://localhost:3000](http://localhost:3000) (or your server's IP). - Default login (if required): - **Username:** admin - **Password:** (see your environment or Docker Compose file) ## 3. Explore traces & metrics Click on "Explore" in the left sidebar to access the query editor to run ad-hoc queries against your Prometheus and Tempo data sources. ## 4. Troubleshooting - Ensure all containers are running: `docker compose ps` - Check logs for any service: `docker compose logs <service-name>` - Verify Prometheus and Tempo are scraping and receiving data from LobeChat and otel-collector. ## See Also - [Langfuse Observability](https://lobehub.com/docs/self-hosting/advanced/observability/langfuse) - [Self-Hosting Overview](https://lobehub.com/docs/self-hosting/start) --- For questions or feedback, open an issue on GitHub or join our community discussions.