NextSign API
Events

Webhook Retry Logic

Understand how NextSign retries failed webhook deliveries and how to build duplicate-safe handlers.

Webhook Retry Logic

If a webhook delivery fails, NextSign retries automatically using exponential backoff.

What Counts as a Failed Delivery

A delivery is treated as failed when:

  • the destination cannot be reached
  • the request times out
  • your endpoint returns a non-2xx HTTP response

Delivery Schedule

NextSign attempts delivery immediately, then retries with increasing delay when the delivery fails.

Typical schedule:

  • Attempt 1: immediate delivery
  • Attempt 2: about 2 minutes later
  • Attempt 3: about 4 minutes later
  • Attempt 4: about 8 minutes later
  • Attempt 5: about 16 minutes later
  • Further attempts continue with exponential backoff

The current limit is up to 14 total delivery attempts for a webhook event.

Webhook delivery is at-least-once, not exactly-once. Your handler must tolerate duplicate deliveries.

  • return 2xx as soon as the payload is accepted
  • move slow work into a queue or background worker
  • deduplicate events before performing side effects
  • log failures with enough detail to replay safely

Example Deduplication Key

function getIdempotencyKey(payload) {
  const entity = payload.case || payload.flow;
  return [payload.trigger, entity?._id, entity?.updatedAt].filter(Boolean).join(':');
}

Operational Advice

  • do not download large files before acknowledging the webhook
  • do not send emails or sync external systems inline if the work can be queued
  • monitor repeated failures so you can fix broken endpoints before the retry window is exhausted

Webhooks

Receive real-time notifications when case and flow events occur in NextSign.