Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

The FabCon + SQLCon recap series starts April 14th at 8am Pacific. If you’re tracking where AI is going inside Fabric, this first session is a can't miss. Register now

Reply
jflo08
Frequent Visitor

Microsoft Fabric Automatic Semantic Model in Lakehouse creation

Hi,

 

I have seen that semantic models are no longer being created automatically when creating a new lakehouse. As posted here:
https://blog.fabric.microsoft.com/en-us/blog/sunsetting-default-semantic-models-microsoft-fabric?ft=...

 

I have a workspace, which contains a lakehouse, and the lakehouse contains tables which are shortcuts from my "Gold" table in my production environment.

This is a newly created workspace (one week ago), and there is no default semantic model. When I create a semantic model in the lakehouse and build a report off this model, and then publish and share, anyone but myself, will recieve an error saying "Something's wrong with one or more fields" in {table}. 

I have older workspaces created before this new update, and they all have default semantic models attached to the lakehouses created, and this error does not appear in any of the reports.

 

What is the fix for this? Clearly there is an issue with people accessing the data due to not having this default semantic model, and the lakehouse having shortcut only tables. How do you now create a semantic model, connect to it and then share successfully?

I have attached an image below to show you the set-up of my new workspace and how the semantic model sits outside the lakehouse, and there is no default lakehouse created now.

jflo08_0-1760959796952.png

 

3 ACCEPTED SOLUTIONS
AntoineW
Super User
Super User

Hello @jflo08,

 

When you manually create a model (+ New semantic model → From Lakehouse), it becomes an independent Fabric item.
That model does not automatically inherit Lakehouse permissions, even if it uses the same tables.

Result: You can see the data (because you’re the owner), but others cannot — their identities don’t propagate through the model to the underlying Lakehouse shortcuts.

2. Shortcuts add an extra layer of authorization

Your Lakehouse tables are shortcuts to the “Gold” tables in your production workspace.
When users query them through a separate model, Fabric performs a cross-workspace OneLake read, which requires either:

  • “Viewer” or “Contributor” access on the source Lakehouse (Gold), or

  • The shortcut Lakehouse owner to enable cross-workspace access for everyone (currently limited preview).

If users lack this access, the model returns field-level binding errors.

 

Sources : 

https://learn.microsoft.com/en-us/fabric/security/permission-model

https://learn.microsoft.com/en-us/fabric/onelake/onelake-shortcuts

https://learn.microsoft.com/en-us/fabric/onelake/onelake-shortcut-security

 

Hope it can help you ! 

Best regards,

Antoine

View solution in original post

suparnababu8
Super User
Super User

Hi @jflo08 

 

Yes, It's correct. As per the blog, default semantic models are no longer autocreated in newly created lakehouses. After creation of sematic model manually, then you built the report and you are getting the error. 

 

Please check whether your sematic model has onelake integration enabled, make sure you are conencting with the sematic model you can use Direct lake mode. S

 

Still if you are facing the issue, please share the screenshot of error to look into issue.

 

Thank you!

 

Did I answer your question? Mark my post as a solution!

Proud to be a Super User!

View solution in original post

V-yubandi-msft
Community Support
Community Support

Hi @jflo08 ,

Could you let us know if your issue has been resolved, or if you need any additional information or clarification? We are happy to help.

 

Thank you.

View solution in original post

6 REPLIES 6
jflo08
Frequent Visitor

Thanks for the help guys. Managed to resolve the issue. Appreciate the support from everyone 👍

v-veshwara-msft
Community Support
Community Support

Hi @jflo08 ,
Just wanted to check if the responses provided were helpful. If further assistance is needed, please reach out.
Thank you.

V-yubandi-msft
Community Support
Community Support

Hi @jflo08 ,

Could you let us know if your issue has been resolved, or if you need any additional information or clarification? We are happy to help.

 

Thank you.

rohit1991
Super User
Super User

Hi @jflo08 

 

1. What changed:

  • New Lakehouses in Microsoft Fabric no longer create a default semantic model automatically.

  • Old workspaces still have it, so they don’t show this issue.

2. Why the error appears:

  • Reports connected directly to the Lakehouse fail because there’s no semantic model layer to read data from.

3. How to fix it:

  • In your workspace, click New - Semantic model.

  • Choose your Lakehouse as the source and select the required tables.

4. Set up the model:

  • Open the new model and create relationships or measures if needed.

5. Connect your reports:

  • Build or edit reports to use this semantic model instead of the Lakehouse.

  • Everyone will now be able to view the reports without field errors.

6. For shortcut tables:

  • Ensure proper permissions and that the data in the original Lakehouse is accessible.

 


Did it work? ✔ Give a Kudo • Mark as Solution – help others too!
suparnababu8
Super User
Super User

Hi @jflo08 

 

Yes, It's correct. As per the blog, default semantic models are no longer autocreated in newly created lakehouses. After creation of sematic model manually, then you built the report and you are getting the error. 

 

Please check whether your sematic model has onelake integration enabled, make sure you are conencting with the sematic model you can use Direct lake mode. S

 

Still if you are facing the issue, please share the screenshot of error to look into issue.

 

Thank you!

 

Did I answer your question? Mark my post as a solution!

Proud to be a Super User!

AntoineW
Super User
Super User

Hello @jflo08,

 

When you manually create a model (+ New semantic model → From Lakehouse), it becomes an independent Fabric item.
That model does not automatically inherit Lakehouse permissions, even if it uses the same tables.

Result: You can see the data (because you’re the owner), but others cannot — their identities don’t propagate through the model to the underlying Lakehouse shortcuts.

2. Shortcuts add an extra layer of authorization

Your Lakehouse tables are shortcuts to the “Gold” tables in your production workspace.
When users query them through a separate model, Fabric performs a cross-workspace OneLake read, which requires either:

  • “Viewer” or “Contributor” access on the source Lakehouse (Gold), or

  • The shortcut Lakehouse owner to enable cross-workspace access for everyone (currently limited preview).

If users lack this access, the model returns field-level binding errors.

 

Sources : 

https://learn.microsoft.com/en-us/fabric/security/permission-model

https://learn.microsoft.com/en-us/fabric/onelake/onelake-shortcuts

https://learn.microsoft.com/en-us/fabric/onelake/onelake-shortcut-security

 

Hope it can help you ! 

Best regards,

Antoine

Helpful resources

Announcements
FabCon and SQLCon Highlights Carousel

FabCon &SQLCon Highlights

Experience the highlights from FabCon & SQLCon, available live and on-demand starting April 14th.

New to Fabric survey Carousel

New to Fabric Survey

If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.

Join our Fabric User Panel

Join our Fabric User Panel

Share feedback directly with Fabric product managers, participate in targeted research studies and influence the Fabric roadmap.

March Fabric Update Carousel

Fabric Monthly Update - March 2026

Check out the March 2026 Fabric update to learn about new features.