What can go wrong with fire-and-forget work created by asyncio.create_task()?

Explains task references, exceptions, caller lifetime, shutdown, and durable-work boundaries.

1Say itAnswer out loud first

2CompareCompare with Core answer

3RefineRefine with Stronger and Advanced

TL;DR

create_task() schedules a coroutine and returns immediately, so the caller no longer waits for its result. If nobody owns the task, its exception may go unobserved, shutdown may cancel it, and request-scoped resources may be closed before it finishes. The event loop keeps weak task references, so detached tasks also need a strong reference until completion. Reliable work needs explicit ownership, not a discarded return value.

Full answer below

Practice focus

Explains task references, exceptions, caller lifetime, shutdown, and durable-work boundaries.

Answer first

Answer path

Core answer first, then checkpoints toward a stronger answer.

Start
Core answer
Start here. This is the direct spoken answer to practice first.
Overview
Explains task references, exceptions, caller lifetime, shutdown, and durable-work boundaries.
create_task() schedules a coroutine and returns immediately, so the caller no longer waits for its result. If nobody owns the task, its exception may go unobserved, shutdown may cancel it, and request-scoped resources may be closed before it finishes. The event loop keeps weak task references, so detached tasks also need a strong reference until completion. Reliable work needs explicit ownership, not a discarded return value.
Checkpoint
Stronger answer
Optional upgrade with examples, trade-offs, and sharper phrasing.
For related in-process work I prefer a TaskGroup or an application-owned task supervisor that retains tasks, observes completion, records failures, and participates in shutdown. A small best-effort notification may be acceptable only when losing it is an explicit product decision. Email, billing, and state changes usually need a durable queue or persisted operation because process memory cannot guarantee eventual execution.

Review signals

Use these to catch weak phrasing, shallow reasoning, and risky claims.

Common mistakesCandidateCommon mistakesMistakes that weaken the answer or signal shallow understanding.Open details

Details

Discarding the task reference
Passing request resources into later work
Calling in-memory work durable

What can go wrong with fire-and-forget work created by asyncio.create_task()?

Practice focus

Answer path

Overview

Review signals

Follow-up prompts

Continue practice

Try a variant

Skill areas

Suggest an improvement

Found this answer useful?