Show HN: Glintlog – Self-hosted log aggregation in a single binary

8 points by caioricciuti 7 hours ago

Hello all!

I built Glintlog because I wanted a simple way to aggregate logs without spinning up Elasticsearch, Loki, or a SaaS that costs my kidney…

What it is: A self-hosted log aggregation tool that runs as a single binary. No Docker, no dependencies, no config files required.

How it works: ``` curl -fsSL https://raw.githubusercontent.com/ibero-data/glintlog/main/s... | bash ```

That's it. Open localhost:8080 and you have a log viewer with search, filtering, and live tail.

Tech: - Go backend with embedded DuckDB for storage - Native OTLP support (gRPC on 4317, HTTP on 4318) - Works with any OpenTelemetry SDK out of the box

Why I built it:

I run small side projects and wanted observability without the complexity. Existing solutions either require too much infrastructure (ELK), are SaaS-only, or don't support OpenTelemetr natively. Glintlog is meant to be the SQLite of log aggregation – simple, embedded, good enough for most use cases. I use a lot for ETL’s now… (I’m a data engineer too)…

It's not meant to replace SaaS that has a 500-person company. It's for small teams, and enthusiasts who want logs without the overhead, also covering all we need for it…

Docs: https://glintlog.com/docs GitHub: https://github.com/ibero-data/glintlog

Would love feedback. What features would make this useful for you? Would you pay per year?