Long-running operations: 202 and polling

API design: REST, HTTP, GraphQL and gRPC Mid-level Python 3.13+ PostgreSQL 17+ Reviewed 31 Oct 2026 By Best Latest Apps

Order the steps

Order the lifecycle of an export request.

Client POSTs /exports with an Idempotency-Key
Worker picks the job and updates status to running
Worker writes the file and sets status succeeded with a result link
Client polling /exports/{id} sees succeeded and downloads
Server stores a pending operation, enqueues a job, returns 202 + Location

Choose an answer, then open Show answer below to check it.

Show answer

Answer: A E B C D

Client POSTs /exports with an Idempotency-Key Retries can't start two exports.
Server stores a pending operation, enqueues a job, returns 202 + Location Request ends in milliseconds.
Worker picks the job and updates status to running Progress is visible on the operation.
Worker writes the file and sets status succeeded with a result link The link can be a short-lived presigned URL.
Client polling /exports/{id} sees succeeded and downloads Polling used Retry-After as its interval.

30-second answer

Don't hold the request open. Accept it with 202 Accepted and a Location header pointing to an operation resource (/exports/{id}) that reports status (pending, running, succeeded, failed) and, when done, a link to the result. Clients poll that URL with backoff, or you notify them by webhook. The job runs in a worker from a queue, and creating it should take an idempotency key so retries don't start two exports.

Practise saying it · 30 s. No recording, just a timer

How do you expose a slow operation?

Say your answer out loud, then check which of these points you covered:

202 + Location
Operation status resource
Worker from a queue
Polling or webhook
Idempotent create

Deep dive

Return Retry-After on the status resource to suggest a polling interval.

Store progress and errors on the operation so clients can show them.

Results can be a presigned URL with a short expiry.

Make cancel an explicit action on the operation.

Common mistakes

Holding an HTTP request for minutes behind load balancers with 60 s timeouts.
Polling every 100 ms.
No way to find the result after the client restarts.

Senior take & follow-ups

Describe 202 + status resource + worker + result link, with idempotency and polling hints.

Follow-up questions

Think of your answer first, then open the hint.

Polling or webhooks?
Show hint
Polling is simplest for clients; webhooks for servers that can receive them.
Where does the job state live?
Show hint
In a database row the worker updates.

Suggest a fix if something here is wrong or out of date.