discussion 2026-11-09 2084 views 4 votes

Prometheus histogram buckets that actually help

@jon_async
Senior Python Dev

Hit this after a routine bump.

Symptom: Prometheus histogram buckets that actually help

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?

6 replies

@gabe
2026-11-10 04:30 UTC · 8 votes

check proxy_read_timeout before blaming uvicorn resets.

@amira
2026-11-10 07:58 UTC · 7 votes

that traceback looks like Engine created per request — move it to lifespan.

@rine
2026-11-10 12:05 UTC · 3 votes

paste engine kwargs and max_connections. worker×pool is the usual culprit.

@oz
2026-11-10 17:27 UTC · 6 votes

minimal repro with ASGITransport would help more than screenshots.

@ina
2026-11-11 13:15 UTC · 10 votes

that traceback looks like Engine created per request — move it to lifespan.

@gabe
2026-11-12 02:05 UTC · 7 votes

paste engine kwargs and max_connections. worker×pool is the usual culprit.

Reply

Login or register to reply.