Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowData Days is here! Join us now for 60+ days of learning, challenges, and connection. Learn more
Hi everyone,
I’ve been testing OneLake Integration for semantic models, and I’ve encountered an issue when trying to query a semantic model (based on Excel import) from a Lakehouse shortcut.
✅What I did:
❌The issue:
When trying to preview the table or run a SQL query via the Lakehouse analytics endpoint, I get the following error:
Table 'Sheet1' was invalid in Lakehouse with ID in Workspace with ID due to 'Required property 'storageDescriptor' not found in JSON. Path '', line 1, position 356.'.
🧩 Expected behavior
I was expecting to be able to:
💡 Context
❓Questions
Any insight or fix would be appreciated. Thanks in advance!
Solved! Go to Solution.
Hi @AntoineW ,
Thanks for reaching out ot MS Fabric community support
Excel-derived semantic models exported via OneLake Integration don’t fully support Lakehouse/SQL consumption due to missing metadata (specifically, the storageDescriptor property). This is an issue with OneLake’s export pipeline when the source is supported by Import mode semantics but lacks structured metadata, as Excel does.
Thanks,
Prashanth Are
MS Fabric community support
Hi @AntoineW ,
Thanks for reaching out ot MS Fabric community support
Excel-derived semantic models exported via OneLake Integration don’t fully support Lakehouse/SQL consumption due to missing metadata (specifically, the storageDescriptor property). This is an issue with OneLake’s export pipeline when the source is supported by Import mode semantics but lacks structured metadata, as Excel does.
Thanks,
Prashanth Are
MS Fabric community support