Issue with Mirrored Azure Databricks Unity Catalog Tables: Data Preview Unavailable

[BhaskarMishra]

Hi everyone,

I'm running into a persistent issue with the Mirrored Azure Databricks Unity Catalog feature in Microsoft Fabric and was wondering if anyone else has experienced the same.

Here's the situation:

• I mirrored an Azure Databricks Unity Catalog into Fabric.

• The Unity Catalog contains around 10 schemas, each with 2–3 tables.

• Everything works fine initially – for the first 1–2 days, I'm able to preview the data from all the mirrored tables directly in Fabric.

• But after 2–3 days, the data preview stops working for some of the schemas  – I can no longer see the table contents inside Fabric.

I’ve double-checked:

• Permissions – everything looks good on both Databricks and Fabric sides.

• Networking configurations – no issues identified there either.

Despite that, the issue continues. The mirrored tables show up, but data preview fails consistently after a few days.

[v-prasare]

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.


If we don’t hear back, we’ll go ahead and close this thread. For any further discussions or questions, please start a new thread in the Microsoft Fabric Community Forum we’ll be happy to assist.
Thank you for being part of the Microsoft Fabric Community.

[v-prasare]

Hi @BhaskarMishra,

We are following up once again regarding your query. Could you please confirm if the issue has been resolved through the support ticket with Microsoft?
If the issue has been resolved, we kindly request you to share the resolution or key insights here to help others in the community.

Thanks,

Prashanth Are

[preshah]

Hello everyone, I wanted to check if the issue still persists where data preview fails intermittently. Appreciate if you can share the support tickets raised and I am happy to follow up. We are also in parallel actively trying to reproduce and look at the issue. Thank you.

[v-prasare]

Hi @BhaskarMishra,

We would like to follow up to see if you have submitted support ticket as our community member suggested. let us know if you need any further assistance.

@burakkaragoz,@preshah&@Mahens  thanks for your prompt response.

Thanks,

Prashanth Are

MS Fabric community support

If our super user response resolved your issue, mark it as "Accept as solution" and click "Yes" if you found it helpful.

[preshah]

Apologies for the inconvenience caused. Does this issue persist? Can you please raise a support ticket and share the ticket #? Do you see data in the SQL Endpoint?

[Mahens]

For every Mirrored Databricks Artifact, there is SQL Artifact that gets created. Can you please navigate to SQL Endpoint and run SQL Query. Let us know error message you see

[burakkaragoz]

Hi @BhaskarMishra ,

Yeah, I’ve seen this kind of thing pop up with mirrored Unity Catalog tables. It’s super frustrating when everything works fine at first and then just stops for no clear reason.

A couple of things you might wanna try:

• Even though permissions look fine, try re-authenticating the Databricks connection in Fabric. Sometimes token expiration or silent auth failures can mess with preview access.
• Check if the affected schemas have any changes in table structure or metadata after the initial sync. Even small changes can cause the preview to break if Fabric doesn’t re-sync properly.
• Also, try doing a manual refresh of the mirrored catalog from the Fabric side. That sometimes kicks things back into shape.

If none of that helps, it might be worth opening a support ticket. This feels like something that could be happening behind the scenes with how Fabric caches or maintains the mirror.

Let me know if you want help testing a specific schema or table.

If my response resolved your query, kindly mark it as the Accepted Solution to assist others. Additionally, I would be grateful for a 'Kudos' if you found my response helpful.

Translation and formatting supported by AI