discussion 2026-11-03 693 views 28 votes

How to cancel long SSE streams cleanly

@amira
Tech Lead

Hit this in staging after bumping workers.

Error / symptom: How to cancel long SSE streams cleanly

Traceback (most recent call last):
  File "app/main.py", line 42, in endpoint
    await db.execute(stmt)
TimeoutError: connection pool exhausted

Env: Python 3.12, FastAPI 0.115, uvicorn workers=4, Postgres 16.
What am I missing in pool settings?

5 replies

@raul
2026-11-03 14:37 UTC · 1 votes

paste your sqlalchemy engine kwargs. pool_size and max_overflow matter more than worker count here.

@raul
2026-11-03 20:04 UTC · 8 votes

try pool_timeout=30 and lower pool_size to 5 per worker first.

@ina
2026-11-04 05:40 UTC · 10 votes

orjson helps CPU a bit; it will not save a saturated DB.

@brick
2026-11-04 11:31 UTC · 4 votes

thanks — dropping pool_size fixed p99. marking notes in the thread.

@seth
2026-11-04 19:59 UTC · 10 votes

paste your sqlalchemy engine kwargs. pool_size and max_overflow matter more than worker count here.