Problem detecting tables and views in Fabric warehouse

[AnHell]

Hello everyone,

I am creating a dataset that reads from a table within a Fabric warehouse, in a specific schema.

This warehouse has other schemas with their own tables and views.

I have created a ‘Microsoft Fabric Warehouse’ linked service that works correctly.

I use this linked service in the dataset, but when I look at the available tables, some of them do not appear.

Instead of selecting it, I have tried the option of entering the name manually, but, as with the drop-down list, some tables work correctly, while others are said to not exist.

Do you know what could be causing the adf dataset to be able to see some tables and not others, given that they are all in the same Fabric warehouse?

Thanks in advance!

[AnHell]

Hi @tayloramy , 

There is one difference that I forgot to mention.

The tables that have been created for a long time and that I can see are in various schemas that were created for these tables some time ago.
The tables that I cannot see are in a different schema.

I did not mention this because I understood that in Fabric, you either have permission for the entire database or you do not, but you cannot have a user who can see some schemas and not others, as was the case in SQL Server on premise.

Could that be a clue?

Thank you!

Translated with DeepL.com (free version)

[tayloramy]

Hi @AnHell, 

Actually in Fabric databases, you can have granular item level permissions - it's not just an all or nothing access. 

See SQL Granular Permissions - Microsoft Fabric | Microsoft Learn

and Lakehouse sharing and permission management - Microsoft Fabric | Microsoft Learn

If you found this helpful, consider giving some Kudos. If I answered your question or solved your problem, mark this post as the solution.

[AnHell]

Thank you very much!

When I was reviewing how to grant permissions in tables or schemas, I think I was fixated on doing it by schemas (as I had always done) and the answer was that it couldn't be done.

I'm going to take a good look at these links, but at first glance they already look very promising.

Thanks again @tayloramy !

[tayloramy · Accepted solution]

Hi @AnHell, 

Are you connecting to the warehouse using a SQL endpoint, or directly through OneLake with a Warehouse connection type?

The SQL Endpoints can take some time to sync. 

If you found this helpful, consider giving some Kudos. If I answered your question or solved your problem, mark this post as the solution. 

[AnHell]

Hi,

I'm not sure what the problem is, but at least I've been able to do what I need to do.

In this scenario, there are three groups of tables.

Group 1. Tables that were created more than a year ago. Yesterday, I could see them correctly.

Group 2. Tables that were created two months ago. Yesterday I could NOT see them, and today I still can't.

Group 3. Tables that I created yesterday, which are the ones I need. Yesterday I could NOT see them, and today I CAN see them.

I think there is something about the tables in Group 2 that I don't know about, and that the tables in Group 3 did need some time for permissions.

Thank you!

[tayloramy]

Hi @AnHell, 

THis is facinating. 

Is there anything different about these tables? How were they created? Are there any shortcuts around?

If you found this helpful, consider giving some Kudos. If I answered your question or solved your problem, mark this post as the solution.