Workflow Error Classification

Definition. Categorising failures as transient, permanent, or requiring human intervention, so that each is handled appropriately rather than uniformly retried. Uniform retry is the most common workflow defect, since it wastes effort on permanent failures and gives up on transient ones.

How it works. Define the categories and map each error type to one. Retry transient errors with backoff. Route permanent errors to a quarantine with the reason. Route ambiguous errors to human review. Log the classification, since the distribution indicates which dependency is unreliable.

Example. A workflow retries all errors three times. Analysis shows 62 percent of retried errors are permanent validation failures that can never succeed, consuming 18 percent of the workflow's capacity. Classifying errors and retrying only transient ones frees that capacity, and the permanent failures are routed to a correction queue where they are fixed and reprocessed.

Use case. Error classification is required for any workflow with retries, and it is the difference between efficient retry and wasted capacity. It is also a diagnostic that reveals unreliable dependencies.

Warnings. Retrying a permanent error wastes capacity and delays the correct path. Never treat all errors alike. And never classify an error as permanent without confirming it cannot succeed on retry.

Limitations. Classification requires distinguishable error types, which many APIs do not provide. It also requires maintenance as dependencies change their error behaviour.

← Browse all glossary terms