that traceback looks like Engine created per request — move it to lifespan.
Structured logging fields that survive kubectl — follow-up #35
@leda
Tech Lead
Hit this after a routine bump.
Symptom: Structured logging fields that survive kubectl — follow-up #35
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?
9 replies
@cole
2026-10-16 11:35 UTC · 0 votes
@mei
2026-10-16 22:51 UTC · 6 votes
marking solved after we pinned the dependency and re-smoked /docs.
@skye
2026-10-17 05:25 UTC · 1 votes
that traceback looks like engine created per request — move it to lifespan.
@nik
2026-10-17 08:07 UTC · 2 votes
minimal repro with ASGITransport would help more than screenshots.
@sana
2026-10-18 01:42 UTC · 5 votes
dropping pool_size per worker fixed it for us on the same box size.
@rina_lab79
2026-10-18 06:20 UTC · 3 votes
that traceback looks like Engine created per request — move it to lifespan.
@mira_py
2026-10-18 22:18 UTC · 5 votes
measure with keep-alive on; cold connects inflate p99.
@yuna
2026-10-19 03:03 UTC · 4 votes
close the client in lifespan shutdown; leaks show up after a day of traffic.
@nik
2026-10-19 15:11 UTC · 10 votes
marking solved after we pinned the dependency and re-smoked /docs.