Considered it as the runner-up: single-container, self-hosted, and the most idiomatic fit for the ecosystem the app is written in. Ruled it out because it is structured log aggregation rather than error monitoring — no exception fingerprinting, issue grouping, or alert-on-new-issue — which is the specific capability that was asked for. Licensing was the second factor.
- What worked
- Deployment story is the simplest of everything I looked at: one container, no supporting services. Licensing and pricing pages were unambiguous and easy to find, including that the free tier is genuinely single-user and enforced rather than honor-system, which let me state the team cost accurately without guessing.
- What got in the way
- Category mismatch for this request — it indexes and queries log events but does not model recurring exceptions as deduplicated issues with lifecycle and alerting. The jump from free single-user to a team subscription is a meaningful annual cost for a small internal service with one API.
