Mirrored database error

[jpthoma7]

I keep getting this error on csv uploads to start my mirror.  I have tried multiple CSVs.



What's worse is there are no MS support options related to database mirroring.

[v-ssriganesh]

Hi @jpthoma7,

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

Thank you.

[jpthoma7]

A little more context:  I am trying to get open mirroring set up with Airtable following these instructions: https://learn.microsoft.com/en-us/fabric/database/mirrored-database/open-mirroring

As described in the instructions, you load the initial CSV/Parquet file into the mirrored database.  Once I get the first one, then I send changes to the endpoint from Airtable or an intermediate.

[v-ssriganesh]

Hi @jpthoma7,
Thank you for the clarification and it’s great to hear you’re exploring Open Mirroring with Airtable.

You're absolutely right that the initial CSV/Parquet file is needed to bootstrap the mirrored table, as outlined in the Open Mirroring documentation.

Consider below steps:

• Ensure you're uploading the file via the correct Open Mirroring entry point use the UI or Open Mirroring API that supports initial file ingestion and the file should typically be in Parquet or CSV format and follow the naming/schema conventions expected by your mirrored table definition.
• Avoid uploading via staging paths or from general storage folders (e.g: .../Staging/...), as these are not supported for mirrored DB initial loads.
• Double-check the table schema you created via the Open Mirroring setup it should match your CSV column names exactly.

If this helps, please mark it as “Accept as solution” and feel free to give a “Kudos” to help others in the community as well.
Thank you.

[jpthoma7]

Thank you for recognizing what I am trying to do.   The error was happening in the initial steps and setup, but I just tried again and didn't get it.

  This is the screen right after creating the Mirrored database from step 1.  

When I click Upload files as instructed in steps 2 and 3, I get this

It doesn't ask me where to store the folders.

This process creates the first folder with the first file and the associated metadata.json file.

Now that its not throwing an error in the initial wizard, the next issue is that it doesn't seem to be automatically creating nor updating a Delta Table.  

[jpthoma7 · Accepted solution]

Turns out Parquet works fine and as documented.  It gets hosed with the CSV.  Maybe its formatting?

[burakkaragoz]

Hi @jpthoma7 ,

This error usually pops up when there's a leftover identity or cached credential from a previous Outlook profile or account setup attempt. Even if the account was removed, some traces might still be stored locally.

Here’s what I’d try:

1. Remove any existing Outlook profiles:

   • Go to Applications > Utilities > Outlook Profile Manager
   • Delete any profiles that might be linked to the same Outlook.com account

2. Clear cached credentials:

   • Open Keychain Access
   • Search for anything related to Outlook, Microsoft, or your email address
   • Delete those entries (make sure you know your credentials before doing this)

3. Reboot your Mac, then try adding the account again.

If that doesn’t work, try creating a new macOS user profile and add the account there. If it works in the new profile, then the issue is definitely local to your user environment.

Let me know how it goes.

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.

[v-ssriganesh]

Hello @jpthoma7,
Thank you for reaching out to the Microsoft Fabric Forum Community and thanks to @bao_phan for the helpful insights.

To add on:

The error you’re encountering “OperationNotAllowedOnThePath – BadRequest”

typically occurs when attempting to upload flat files (like CSVs) into a Mirrored Database, which is not supported. Mirrored databases are designed to sync data from external source systems (e.g., Azure SQL DB, Cosmos DB, Snowflake), and do not allow direct file uploads.

Recommended Approaches:

• If your goal is to work with flat files (like CSVs) please use a Lakehouse instead of a Mirrored DB and upload the CSV under the Files section of your Lakehouse.
• If you're trying to mirror a data source, could you please clarify what source system are you trying to mirror (e.g., Azure SQL, Snowflake)? And Are you using any specific tool or method to initiate the mirror?

If this information is helpful, please “Accept as solution” and give a "kudos" to assist other community members in resolving similar issues more efficiently.
Thank you.

[bao_phan]

if you are trying to upload flat files, you should upload it to lakehouse, or you can create a pipeline/data flow to connect flat files to a lakehouse or warehouse. Can you provide more specific context (like where you are upload the flat files ? and what source are you trying to do mirroring ?)