How Discovery Works
Discovery Types
Common Questions
What’s the difference between initial and incremental discovery?
Initial discovery is a full catalog scan that runs once when you connect a database. Incremental discovery runs on your configured schedule (hourly, daily, or custom) and only records what changed since the last run - new assets, removed assets, schema diffs. Incremental is fast even on large databases because it queries the same metadata but diffs against the previous snapshot.Can discovery be triggered manually?
Yes. Click Run Discovery on the connection’s detail page to kick off an on-demand scan. This is useful after a deliberate schema change (migration, table addition) when you want the catalog to reflect the new state immediately instead of waiting for the next scheduled run.Does discovery pull my actual data?
No. Discovery queries only system catalogs (information_schema, pg_catalog, or equivalent) to read names, types, and metadata. It never runs SELECT * or reads row values. The Query Gateway enforces this.
Why didn’t a newly created table appear after I ran discovery?
Three common causes: (1) the table is in a schema excluded by your discovery include/exclude patterns; (2) your read-only database user doesn’t haveUSAGE on that schema or doesn’t see it in the system catalogs; (3) some databases delay catalog visibility for a few seconds after DDL. Check the connection’s discovery include list first.
Next Steps
Configure Discovery
Set up schedules and discovery options
Connect a Data Source
Add databases for discovery to scan