Microsoft Fabric Lakehouse Rename – SQL Endpoint Sync Failure (Error 911)

I think Error 911 points to a missing database name, not necessarily a table metadata issue. Check whether lh_silver is the old Lakehouse name and whether the SQL endpoint and saved connections still reference it.

Try SQL analytics endpoint → Explorer → Refresh. This syncs table metadata, not the database name. If Refresh itself still fails with 911, send the old/new names, workspace/Lakehouse/endpoint IDs, UTC time, and error details to Fabric support. No need to recreate the Lakehouse yet.

Hi Murtaza_Ghafoor​ 

I just tested this and am encountering the same problem. 

I'm wondering if perhaps this is a timing thing and will correct itself in the next 15-30 minutes. If waiting has not corrected it for you, I'd suggest to open a support ticket with Microsoft. 

Hi Murtaza_Ghafoor​,

Given the error details, I would treat this as more than a normal metadata-delay issue.

The important clue is:

Database 'lh_test_rename' does not exist
Code=911

while the Lakehouse itself still exists and the SQL analytics endpoint shows both Table sync status = Failure and SQL sync status = Failure.

Microsoft’s SQL analytics endpoint metadata sync guidance confirms that Fabric normally keeps the SQL endpoint synchronized automatically, and that you can also trigger metadata synchronization manually from the Fabric portal, REST API or T-SQL.

I would first try:

Lakehouse
→ SQL analytics endpoint
→ Metadata sync / Refresh

and then retry the failed table sync.

Microsoft’s Lakehouse troubleshooting guidance also calls out renamed Lakehouses and stale or incorrect references as possible causes of path/artifact lookup failures.

So if the endpoint continues to reference the renamed database incorrectly after a manual sync, I would suspect stale endpoint metadata rather than the Delta table itself.

One additional check I would make before recreating anything is OneLake security.

Microsoft’s OneLake security troubleshooting guidance notes that broken references to renamed or deleted objects can block synchronization until those references are corrected.

So I would check:

• OneLake security roles
• views or stored procedures in the SQL endpoint
• semantic models / shortcuts / connections that may still reference the old name
• any code using the old Lakehouse name instead of IDs

There are also community reports where renaming the Lakehouse back to its original name restored the SQL endpoint, after which the rename could be attempted again.

I would treat that only as a workaround rather than the documented fix, but it may be useful if you need to recover quickly.

So my troubleshooting order would be:

manual metadata sync
→ verify no security or SQL objects still reference the old name
→ test whether reverting to the original name restores sync
→ if the endpoint still remains in Failure, raise a Fabric support case with the sync timestamp, HResult and Lakehouse/endpoint IDs

I would not recreate the Lakehouse yet, because Microsoft provisions exactly one SQL analytics endpoint alongside the Lakehouse, and this looks more like the paired endpoint has retained stale metadata after the rename.

AI-assisted drafting: AI was used to help structure and phrase this response. I reviewed and validated the technical content before posting.

Thanks everyone for their response, i encountered that issue and now it is confirmed, currently re naming option of lakehouse in fabric creates this error.
Error message:
Database 'lh_silver' does not exist. Make sure that the name is entered correctly. (Code 911, State 1)