We often get asked what happens with the data if a webhook wasn't received on our client's end. Is there a way to recover that data?
To answer this question, we must understand how our webhook works.
ROOK makes periodic requests to our data sources, and when it obtains data from a summary or event, it sends them to our client through the webhook that we have registered in our system. The events will be sent as soon as they are obtained, while the summaries will be sent through this means based on the schedule specified by the client. This is made possible thanks to our recent TimeZone feature.
Now, in the event that a summary is not delivered, recently, with the Rejected Information Project, a retry function was added to improve data delivery reliability.
The retry logic works as follows:
Failed deliveries trigger retries at the following intervals: 2 hours, 24 hours, and 48 hours.
In case all retry attempts fail, the data is stored in temporary buckets for client-side retrieval:
Sandbox environment: Data is retained for 3 days.
Production environment: Data is retained for 10 days.
However, in order to retrieve information that was not delivered via webhook, it is recommended that the client use our endpoints located in the ROOK API, as the Webhooks system is not intended for historical polling or re-delivery on demand.