"Cannot establish connection with provided parameters." when adding tables from Redshift (build succeeds, table list is empty)
Applies to: Sisense · Native connectors (Amazon Redshift) · Add Data / Select Data
Category: Native Connectors
Symptoms
You can successfully run a Build on the data model using an existing Redshift connection.
But when you use Add Data on that same connection and reach the Select Data step, the table list stays empty and this error appears:
Cannot establish connection with provided parameters.
Steps that reproduce it:
Open the data model and click Add Data.
Select the existing Redshift connection.
Proceed to the Select Data step.
The error appears and no tables are displayed.
The contradiction — build works, but browsing tables fails — is the key signal for this case.
Cause
The database user configured in the connection does not have read permission on one of the objects (a view) in the schema it is trying to list. When Sisense enumerates all tables and views in the schema at the Select Data step, the permission error on that single object causes the whole listing to fail, surfacing as the generic "Cannot establish connection with provided parameters." message.
A Build keeps working because it only reads the specific tables already in the model, whereas Add Data attempts to read metadata for every object in the schema — including the view the connection user cannot access.
This is a database-side permissions configuration issue, not a Sisense connectivity defect. The connection parameters themselves (host, port, credentials) are valid.
How to confirm you've hit this specific case
Builds on the existing model succeed, but Add Data → Select Data fails with the error above and an empty table list.
The connection uses a restricted database user that was granted access to only a subset of objects (common when a dedicated reporting user is created for Sisense).
On the Redshift side, there is at least one view in the same schema that the connection user cannot read.
Resolution
Work with your database team to resolve the permission gap. Either option fixes it:
Grant the connection user read access to the view(s) in the schema it is enumerating:
(In Redshift, ALL TABLES IN SCHEMA includes views.)
Or move the inaccessible view to a different schema or database that the connection user is not pointed at, so it is no longer part of the object list Sisense reads.
After granting the permission (or relocating the view), reopen Add Data → Select Data — the table list populates and you can add tables normally.
Prevention
When creating a dedicated, least-privilege database user for a Sisense connection, make sure it has SELECT on all objects (tables and views) in the schema you expose to Sisense — not only the specific tables in the current model. The Add Data flow reads the entire schema's object list.
If you intentionally keep certain objects off-limits to the connection user, place them in a separate schema so they are not enumerated during table selection.
