I self-host OpenObserve to collect logs and traces from my projects, and the biggest win for me was how easy it was to get running. It ships as a single binary, and I had it ingesting data within minutes—without having to stand up Elasticsearch or maintain separate Prometheus and Jaeger stacks. It also accepts OpenTelemetry natively, so I simply pointed my existing OTel exporter at it and the data appeared right away.

Having logs, metrics, and traces in one UI saves me a lot of time when debugging, because I can jump from an error log to the related trace in just a couple of clicks. Queries are plain SQL as well, so there’s no new query language to learn. Storage efficiency is another plus: the data is kept as compressed Parquet, which really matters when you’re paying for your own infrastructure. Review collected by and hosted on G2.com.

The documentation doesn’t always keep pace with how quickly new versions ship, so I sometimes had to dig through GitHub issues or release notes just to understand a particular setting. Building dashboards also takes more effort than in Grafana, and there are fewer ready-made templates to use as a starting point. On top of that, some of the more advanced features are limited to the Enterprise edition. That said, Enterprise is free below a daily ingestion limit, which feels fair for smaller setups. Review collected by and hosted on G2.com.