@brick
2026-10-01 14:35 UTC · 9 votes
dropping pool_size per worker fixed it for us on the same box size.
Hit this after a routine bump.
Symptom: CORS preflight storm from SPA — fix? — follow-up #36
TimeoutError / pool exhausted / upstream reset
(versions in next reply if needed)
Env: Python 3.12, FastAPI recent, uvicorn workers≥2.
What should I check first?
dropping pool_size per worker fixed it for us on the same box size.
minimal repro with ASGITransport would help more than screenshots.
paste engine kwargs and max_connections. worker×pool is the usual culprit.
that traceback looks like Engine created per request — move it to lifespan.
marking solved after we pinned the dependency and re-smoked /docs.
dropping pool_size per worker fixed it for us on the same box size.
orjson helped CPU a few percent; DB still the ceiling.