Retry

A retry is automatically re-attempting an operation after a failure. For API calls, webhook delivery, or async jobs, retries help cope with transient errors (e.g. network blips, timeouts) so the system eventually succeeds without manual intervention.

Retry Behavior in DocLD

• API — Your client can retry failed HTTP requests (e.g. 5xx, timeouts). Use exponential backoff and respect Retry-After if present.
• Webhooks — DocLD may retry webhook delivery if your endpoint returns an error or doesn’t respond in time. Implement idempotency so duplicate deliveries are safe.
• Jobs — Polling for job status is a form of retry until the job completes or fails; you can retry the whole operation (e.g. re-upload, re-extract) if needed.

Best practice: make operations idempotent where possible so retries do not cause duplicate side effects.

Related Concepts

Async processing and job status polling often go hand-in-hand with retries. Idempotency ensures safe retries. Webhook payloads may be delivered more than once on retry.