Ashby connector: use syncToken-based incremental sync instead of full re-import every sync
Which connector?:
Ashby (link to actual connector)
Additional details:
Our Ashby connector currently performs a full re-import of every table on every sync, rather than syncing incrementally. Per your support team (tickets #371936 and #387567), this is because "Ashby's API doesn't expose a timestamp field required for incremental updates."
However, Ashby's own API documentation describes support for incremental sync via a syncToken parameter, used to fetch only records changed since the last sync: developers.ashbyhq.com/docs/pagination-and-incremental-sync#incremental-sync. This is referenced in several of Ashby's list endpoints (e.g. job.list, offer.list).
If the connector isn't using this mechanism, that would explain why every sync — regardless of actual data volume — behaves like a full historical resync. In practice this has meant:
- Initial sync took ~3 weeks to complete
- Subsequent syncs were taking ~1 week each
- A recent sync triggered by a schema change on our end failed after 12 days and has been re-attempting for several days since
This is currently blocking us from reliably getting recruiting/hiring data into our data warehouse. Given the impact, we'd appreciate this being prioritized as a performance/reliability fix rather than a lower-priority enhancement — happy to provide additional details as needed.
Thanks
Please sign in to leave a comment.
Comments
0 comments