> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cimento.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Delivery and retries

> What Cimento does when your webhook endpoint fails, how to see what was sent, and how to send events again.

A delivery is one event being sent to one endpoint. A delivery succeeds when your endpoint answers with a `2xx` status within 10 seconds. Otherwise Cimento tries it again.

## Retries

Cimento makes up to 8 attempts over about 32 hours. After a failed attempt it waits, then tries again:

| Attempt | Wait before it |
| - | - |
| 1 | None |
| 2 | 1 minute |
| 3 | 5 minutes |
| 4 | 30 minutes |
| 5 | 2 hours |
| 6 | 5 hours |
| 7 | 10 hours |
| 8 | 14 hours |

A retry can start up to a minute after it is due. If the eighth attempt fails, the delivery is marked **Failed** and Cimento stops trying. You can [send it again](#send-deliveries-again) yourself.

Every attempt sends the same event with the same `webhook-id`. The `webhook-timestamp` and the signature are new each time.

## When an endpoint keeps failing

Retries apply to one delivery. Cimento also watches each endpoint as a whole, so that an endpoint that is down isn't sent every event eight times.

* **Paused after 5 failures in a row.** Cimento stops sending to the endpoint for 1 minute. Each further failure doubles the pause, up to 60 minutes. Deliveries wait during a pause, and the wait doesn't use up one of their 8 attempts.
* **Back to normal on the first success.** Any `2xx` answer clears the pause and the failure count.
* **Disabled after 5 days.** If an endpoint has failed continuously for 5 days, with at least 20 failures in a row, Cimento disables it and cancels the deliveries still waiting for it.

On the **Webhooks** page, an endpoint with recent failures shows a **Failing** badge beside **Active**. Hover over it to see how many deliveries failed, since when, and when a pause ends. An endpoint Cimento disabled shows **Disabled** with the reason **Disabled automatically after 5 days of failed deliveries**.

Cimento does not email you when an endpoint is failing or disabled. Check the Webhooks page, or watch for the gap on your side.

### After you fix it

**1. Enable the endpoint**

If the endpoint was disabled, open its **⋯** menu and choose **Enable**. Its failure count starts again from zero.

**2. Send a test event**

Choose **Send test event** and check it succeeds in the delivery log.

**3. Send what was missed**

Open the endpoint's deliveries and use **Retry failed deliveries** for the days it was down. See [Send deliveries again](#send-deliveries-again).

Events that happened while the endpoint was disabled were never recorded for it, so there is nothing to send again for that time.

## The delivery log

Each endpoint has a log of what Cimento sent it. Open it by clicking the endpoint's URL on the **Webhooks** page, or **View deliveries** in its **⋯** menu.

![The Deliveries page for an endpoint, with a table of deliveries behind a Delivery details panel showing a failed delivery's status, the event as JSON and two attempts](https://mintcdn.com/cimento/n6nGSMtgFNGbPpHJ/images/webhooks/delivery-log.png?fit=max\&auto=format\&n=n6nGSMtgFNGbPpHJ\&q=85\&s=0e6459e5b06b5e13461e531112a4285b)

An endpoint's deliveries, with one delivery's details open

Each row is one delivery, with its event type, status, number of attempts, the last response code, when it was created and, while it is waiting, when the next attempt is due.

| Status | Meaning |
| - | - |
| **Pending** | Waiting to be sent, or waiting for its next attempt. |
| **Succeeded** | Your endpoint answered `2xx`. |
| **Failed** | All 8 attempts failed. |
| **Cancelled** | The endpoint was disabled or deleted before the delivery was sent. |

Filter the log by **Status**, **Event** and a **From** and **To** date. Dates are whole days in your browser's time zone. **Load more** shows older deliveries.

Click a delivery's event, or **View details** in its menu, to see:

* the **Event**: its `id`, `type`, `occurredAt` and `data`;
* each **Attempt**: when it was made, the response code, how long your endpoint took, the error if there was one, and the start of your endpoint's response.

Deliveries are kept for 30 days.

## Send deliveries again

You can retry a delivery that is **Failed** or **Cancelled**, as long as the endpoint is active. A retried delivery goes back to **Pending**, starts a fresh set of 8 attempts, and is sent within about a minute.

* **One delivery.** Open the delivery's details and click **Retry delivery**, or choose **Retry** in its row's menu.
* **A range of days.** Click **Retry failed deliveries**, choose the **First day** and **Last day**, and click **Retry deliveries**. Cimento sends every Failed and Cancelled delivery created in that range again and tells you how many it queued. A range covers up to 30 days. Deliveries that succeeded are left alone.

Retrying sends the same event with the same `webhook-id`, so a receiver that [recognizes duplicates](https://docs.cimento.ai/webhooks/receiving-events.md#duplicates) handles a retry of something it already has without trouble.

The log doesn't refresh by itself. Reload the page to see how a retry went.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.