'%3e%3cpath%20d='M8%200C12.4183%200%2016%203.58172%2016%208C16%2012.4183%2012.4183%2016%208%2016C3.58172%2016%200%2012.4183%200%208C0%203.58172%203.58172%200%208%200ZM11.6162%204.38379C11.2257%203.99337%2010.5927%203.99338%2010.2021%204.38379L8%206.58594L5.79785%204.38379C5.40732%203.99334%204.77429%203.99329%204.38379%204.38379C3.99331%204.77429%203.99335%205.40733%204.38379%205.79785L6.58594%208L4.38379%2010.2021C3.99348%2010.5927%203.99341%2011.2257%204.38379%2011.6162C4.77426%2012.0066%205.40734%2012.0065%205.79785%2011.6162L8%209.41406L10.2021%2011.6162C10.5927%2012.0066%2011.2257%2012.0067%2011.6162%2011.6162C12.0067%2011.2257%2012.0066%2010.5927%2011.6162%2010.2021L9.41406%208L11.6162%205.79785C12.0066%205.40735%2012.0066%204.77429%2011.6162%204.38379Z'%20fill='%23080E17'%20fill-opacity='0.46'/%3e%3c/g%3e%3cdefs%3e%3cclipPath%20id='clip0_3761_713'%3e%3crect%20width='16'%20height='16'%20fill='white'/%3e%3c/clipPath%3e%3c/defs%3e%3c/svg%3e)
Fix Fabric Spark SQL Cannot Query ADLS Gen2 Shortcut Table
Question details
The user needs to resolve an issue where a Delta table created via an ADLS Gen2 shortcut is visible in the Fabric Lakehouse and SQL endpoint but returns a TABLE_OR_VIEW_NOT_FOUND error when queried in a Spark notebook.
- Product
- Microsoft Fabric
- Device & OS
- not provided
- Scenario
- Attempting to query an Azure Data Lake Storage Gen2 shortcut table using Spark SQL in a Fabric notebook.
- Observed behavior
- The Spark notebook returns a TABLE_OR_VIEW_NOT_FOUND error, even though the table is fully visible and accessible in the Fabric Lakehouse and SQL endpoint.
Verify that your Spark notebook is actively attached to the correct Lakehouse and that you have the necessary workspace permissions to read the underlying ADLS Gen2 data.
Troubleshoot Spark Catalog and Consult Microsoft Fabric Community
Since this is a specific configuration or catalog synchronization issue within Microsoft Fabric, verify your Spark context and seek specialized guidance from the official community.
The TABLE_OR_VIEW_NOT_FOUND error usually occurs when the Spark catalog metadata has not synced with the Lakehouse, or if the notebook is referencing the wrong database context.
Ensure your Spark notebook is correctly referencing the Lakehouse where the shortcut was created. Use the fully qualified name in your Spark SQL queries, formatted as `lakehouse_name.table_name`.
Because the Spark metadata catalog can sometimes take time to sync, restart your Spark session or use the `REFRESH TABLE` command to force an update of the metadata.
Collect your specific shortcut configuration, the exact notebook code you are running, the table location path, and the complete error message. Be sure to mask any sensitive storage keys or credentials.
Navigate to the official Microsoft Fabric Community (https://community.fabric.microsoft.com/) and post your gathered details in the Spark or Lakehouse forums for specialized support from Fabric engineers.
Need a Lightweight Tool for Your Data Engineering Documentation?
While you troubleshoot your Microsoft Fabric and Spark SQL environments, use WPS Office as your free, lightweight, and fully compatible alternative to Microsoft Office. It is perfect for writing technical documentation, analyzing exported CSV data, or creating data architecture presentations without expensive subscription costs.
- 1. Download the software: Visit the official WPS Office website to download the free installation package for your operating system.
- 2. Install WPS Office: Run the downloaded installer and follow the simple on-screen instructions to set up the suite.
- 3. Open and edit your files: Launch WPS Office to instantly open, edit, and save your existing Microsoft Office documents and data exports.
Frequently Asked Questions
Why does my ADLS Gen2 shortcut table show TABLE_OR_VIEW_NOT_FOUND in Fabric Spark?
This typically occurs if the Spark session is not connected to the correct Lakehouse context, the metadata catalog hasn't fully synchronized with OneLake, or the SQL query is not using the fully qualified table name.
How do I refresh the metadata for a Fabric Lakehouse table in a Spark notebook?
You can resolve many synchronization issues by restarting your active Spark session or by executing the `REFRESH TABLE [table_name]` command in a Spark SQL cell to force a catalog update.
Can WPS Office open Excel files exported from Microsoft Fabric query results?
Yes, WPS Office is fully compatible with Microsoft Excel (.xlsx) and CSV formats, making it an excellent free tool for reviewing and analyzing data extracts locally on your machine.
