Ready
Database and semantic index deep health check for uptime monitoring and alerting.
Not used as a Kubernetes probe so database failovers do not take down DB-free pages or cause pod restarts.
Readiness answers one question: can this pod serve requests? That is DB connectivity + the semantic index being recoverably in sync — genuine unservability, so both still return HTTP 503.
A FastEmbed embedder degraded to the hermetic hash-noise fallback is DEGRADED-BUT-SERVING, not unservable: the pod still answers routing requests (with a lower-fidelity embedder), so this endpoint reports it rather than failing on it. The Lane 3.D alarm/log (EmbeddingEngine._trigger_alarm) still fires on fallback engagement — only this readiness gate stopped treating "degraded" as "down".
Request
- Base URL: https://app.teetra.com (the document declares no server; this is the origin it was published from)
- URL: https://app.teetra.com/health/db
- Auth: none declared
Response
Successful Response
Changes
No recorded changes to this operation across all 2 revisions of this API.