Pattern

Treat a timeout as an unknown outcome before repeating the action

No reply is not the same as no action.

About 1 min read

Before you use it

Use when
  • A send, create or update request times out after leaving your application.

Keep this limit in mind

  • Reconciliation against stale or incomplete data can also mislead. Do not invent an automatic retry rule for an undocumented interface.

An idea to explore.

Try this idea.

Inspect the example, try the idea in your situation, then check your next move.

Keep reading

What to look for

No reply is not the same as no action.

Why it matters

The receiver may have completed the operation while its response was lost. Preserve an unknown or pending state and reconcile using the operation's identity before creating a fresh request. Where a documented idempotent retry is available, follow that contract.

An example

An import request times out. Its run identifier reveals that the job is already processing, so a second import is not started.

Check your result

The status distinguishes confirmed failure from an unconfirmed outcome.

Save and practise this idea

My learning

Read. Recall. Repeat.

Save this idea and return to it on any device.

Use email
Keep exploring

Connected ideas

You are readingTreat a timeout as an unknown outcome before repeating the action

Evidence and sources
Supports

A request that receives no response may already have created its intended resource; blindly retrying can create additional effects.

Reconciliation requires an authoritative receiver view and enough identity information to distinguish this operation from another.

Making retries safe with idempotent APIs · Retrying and side effects

All sources (1)
Edition and review

Version 0.1.0

No review details are supplied for this edition.

Read this idea as JSON · Release details