Error when deleting tables or when modifying column names in the tabular model

Question

Hi,

I am using tabular model level 1500. I have a postgres database that is connected to the azure analysis server through an on prem gateway. I have used powerquery to fetch data from the postgres database (via an odbc driver) to the model. I was able to succesfully create and deploy my tabular model. However recently one of the column names changed in the database. So I had to refresh my metadata and publish my model again. This when I ran into an issue. It does not allow me to do this. In fact I cannot even delete the table from my model and publish it. The errors I get are as follows:

Failed to save modifications to the server. Error returned: 'An unexpected error occurred (file 'Error.cpp', line 116, function 'err::BaseError::ThrowFromTLS').Technical Details:RootActivityId: f5e21bb9-5dca-4014-ac2c-2f8ddf45bbefDate (UTC): 9/6/2024 11:53:33 AM'.

I have attached screenshots here as well for reference to show what happens when trying to delete a table. I wonder if this is a permissions issue? The error no where reveals that. It's very obscure and comes from within some source code. Don't know if it's visual studio 2022, or the SSDT extension which is on 2019 version or if its from the Azure Analysis Services itself.

Answer 1

When column names change in the source database, it's essential to ensure that the metadata in your Power Query and tabular model reflects this change.

Try to manually refresh your Power Query connection to make sure it's fetching the latest metadata from PostgreSQL.

After the metadata refresh, try to delete or modify the column/table and redeploy the model.

You're using SSDT 2019 with Visual Studio 2022, and there could be version mismatches between your extensions and the Visual Studio IDE itself. Consider upgrading SSDT to the latest version compatible with Visual Studio 2022.

Sometimes, cached data in the model or temporary files may cause such errors. Try clearing any local caches (in both Power Query and Azure Analysis Services) and restart the Analysis Services server.