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

# Retry Schedule

> How and when Convoy retries failed deliveries.

When an endpoint is experiencing a temporary disruption of service, events sent to it might fail and require you to retry them. Convoy allows you to set the number of attempts to a particular endpoint and how to initiate the retry. Project defaults use `config.strategy.type`, `config.strategy.duration` (seconds), and `config.strategy.retry_count`. The duration is recorded in delivery metadata as `interval_seconds`. Convoy supports two retry strategies:

* `linear`: waits the configured `interval_seconds` between attempts.
* `exponential`: increases the delay using `interval_seconds * 2^n`, where `n` is the delivery's attempt count before the current dispatch. A random jitter of less than five seconds is added, and the final delay is capped by the instance's `max_retry_seconds` setting. There is no fixed sequence of retry times.

Retry settings include a `retry_count` that limits the total delivery attempts, including the initial attempt. For example, two attempts means one initial attempt and one retry. The delivery metadata records the attempt count as `num_trials` and the configured limit as `retry_limit`.

Retry defaults are configured per project, and subscriptions can override them. See [Subscriptions](https://getconvoy.io/docs/product-manual/subscriptions.md).


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