briven· techmenuclose
status
briven, right now.
the check below ran when you loaded this page — no cached numbers and no dashboard someone forgot to update. it checks one endpoint, and that endpoint reads through to the database rather than only confirming a web server is alive. what it does not check is listed at the bottom, because a green tick for something nobody measured is the whole problem with status pages.
the checked endpoint is answering
checked at 2026-08-20T09:36:21.896Z · not cached
app + control plane
the dashboard, the trpc api, and the control-plane database behind them. this check reads through to the database.
answering200 · 121ms
- database
- reachable
- db latency
- 1ms
the probe gives up after 3000ms. a timeout, a refused connection and a dns failure all read the same here, because from where you are sitting they are the same thing.
what this page does not check
- your own database's availability — the check above reads the control plane's database, not yours.
- the connection pooler, object storage, and the embeddings service. none of them expose a public health endpoint yet.
- historical uptime. nothing here stores samples over time, so there is no percentage to quote.
- incident history. when something breaks, we write to affected accounts — this page shows only right now.
something looks wrong that this page says is fine? tell us — that gap is the most useful bug report we can get.