Retries and delivery logs

FluxDigest retries network failures, 408, 409, 425, 429, and 5xx responses with exponential backoff and jitter. Ordinary 4xx responses are treated as permanent failures.

Retries continue for up to 72 hours. Because delivery is at least once, the same event ID can arrive more than once.

Respond quickly

Verify, persist, and enqueue the event, then return a 2xx response. Perform slow CRM calls, email processing, or analytics work outside the webhook request.

Deduplicate

Store processed event IDs with a unique constraint:

insert into processed_webhook_events (event_id, processed_at)
values ('evt_01JEXAMPLE', now())
on conflict (event_id) do nothing;

Inspect deliveries

The developer dashboard shows the response status, attempt number, duration, next retry, and sanitized response metadata. You can redeliver an event after fixing an endpoint.

FluxDigest may disable an endpoint after sustained failure and will notify workspace Admins. Re-enable it only after confirming the endpoint returns a successful response.