I set a log stream custom webhook to process events.
The events are received in batches through the my webhook API, but sometimes the processing of some events in the batch fail on my side.
I would like to return an error to Auth0 webhook so that these events logs would be retried, but instead of retrying the entire batch I would like only the failing events to be retried.
My questions are:
Is there a way to mark in the webhook API response that only some events failed, so that only they will be retried?
Is there an API documentation for the log stream webhook contract with the request and response payload?
I see in the docs that there are three attempts to retry the delivery. Then the operation will be retried for 7 days. What are the retry intervals between the 3 initial retries? And during the 7 days?
I see in the docs that there are three attempts to retry the delivery. Then the operation will be retried for 7 days. What are the retry intervals between the 3 initial retries? And during the 7 days?
As stated in the documentation regarding the Log stream, when the webhook fails to reach a server by the third attempt, you’ll see an error in the Log Stream Heath view. This process of 3 attempts restarts for any errored event for the 3 consecutive days. If after 7 days, webhooks won’t have a successful deliver log stream will be paused.
When an event is triggered:
We attempt to deliver the event to your server up to three times.
If we fail to reach your server by the third attempt, we log an error visible in the Health view for the log stream.
We restart this process for any errored events until the problem is resolved.
If we fail to reach your server for 7 consecutive days, we’ll automatically pause the stream. You’ll need to resolve the issue, and then resume the stream manually.
Regarding the payload preview - is there a specification for the response of the custom webhook?
i.e. we get a webhook with a batch of events - if we return 200 status code, it means that we succeeded to process the batch.
If we return non-200 status code it means that we failed to process the batch. Therefore Auth0 will retry to send those events.
Is it possible to mark somehow in the response that only some events in the batch failed, so that only some events will be retried?