Web API Connections
Web API Connections
Section titled “Web API Connections”Overview
Section titled “Overview”A Web API connection stores the reusable connection details for one external data provider. Streams under the connection define which endpoints or resources to sync.
How it works
Section titled “How it works”The connection describes how Rokks can reach and authenticate to the provider. A stream describes the data shape, extraction strategy, sync mode, and target table. Heavy extraction and loading runs as Dispatcher-scheduled background jobs so a browser tab can close without corrupting the source state.
Step-by-step
Section titled “Step-by-step”- Open Web Data in the Data Fabric.
- Create a connection and name the provider clearly.
- Enter the base request details supplied by the provider.
- Test discovery or probe the endpoint if available.
- Add one or more streams.
- Configure each stream’s transform plan and sync mode.
- Configure a schedule, then validate the first Dispatcher-submitted sync in the Job Queue.
Manual Web API Sync and Full Sync actions are currently unavailable in the UI and Sidekick. The supported execution path is the schedule owned by Dispatcher; the Worker does not accept a browser-rich WEBAPI_SYNC payload.
Common mistakes
Section titled “Common mistakes”Do not put several unrelated providers into one connection. Keep connection names aligned with the provider or account.
Do not assume a successful endpoint probe proves that the complete stream can load. Review the first scheduled job and its output before relying on the stream in a dashboard or alert.