Direct Lake Model Outages

[parkergeis]

I am running into an issue on with my Direct Lake semantic model where, very randomly, tables seem to “lock out”. Then, some odd amount of time later (sometimes minutes for user A and hours for user B), all is well again.

Users will receive messages like “Please verify that the data source is available and your credentials are correct.” or “We cannot access column ‘XYZ’ of delta table ‘ABC’…either the column doesn’t exist or you don’t have permission…”.

Through troubleshooting, I have found none of the below items are responsible for this seemingly random outage:

• Capacity resource constraints

• Access issues

• Refreshing model

• ETL into lakehouse

• DAX Studio query activity reveals little to no activity

I feel as if I’ve exhausted all of my avenues of troubleshooting and would appreciate some feedback if anyone has experienced this as of late and has any suggestions. Thank you!

[Shahid12523 · Accepted solution]

Issue: Direct Lake semantic model randomly locks out tables, showing credential or column access errors.

Not Caused By: Capacity limits, access issues, refreshes, ETL, or query load.

Likely Causes:

- Silent fallback to DirectQuery due to unsupported features or metadata drift.
- Delta table schema changes or stale metadata.
- Workspace sync or permission misalignment.
- Backend Fabric service instability.

Fixes:
- Refresh metadata manually (Tabular Editor/Semantic Link).
- Monitor query mode with DAX Studio.
- Align lakehouse and semantic model workspaces.
- Consider logging a Microsoft support ticket.

[v-tejrama]

Hi  @parkergeis,

Thank you  @Shahid12523  for the response provided!

Has your issue been resolved? If the response provided by the community member addressed your query, could you please confirm? It helps us ensure that the solutions provided are effective and beneficial for everyone.

Thank you for your understanding!

[Shahid12523 · Accepted solution]

Issue: Direct Lake semantic model randomly locks out tables, showing credential or column access errors.

Not Caused By: Capacity limits, access issues, refreshes, ETL, or query load.

Likely Causes:

- Silent fallback to DirectQuery due to unsupported features or metadata drift.
- Delta table schema changes or stale metadata.
- Workspace sync or permission misalignment.
- Backend Fabric service instability.

Fixes:
- Refresh metadata manually (Tabular Editor/Semantic Link).
- Monitor query mode with DAX Studio.
- Align lakehouse and semantic model workspaces.
- Consider logging a Microsoft support ticket.

[parkergeis · Accepted solution]

The issue here is a bug that causes the SQL Analytics Endpoint to become out of sync with the semantic model. This can be resolved by refreshing the metadata in the SQL Endpoint. Microsoft is aware of the issue and has an ETA of mid-November for a fix to be released. They have not revealed what specifically causes the two to become out of sync.

[parkergeis]

Thank you @Shahid12523. I will move forward with opening a support ticket as all the other fixes you have proposed are items I have looked into.

[Gabry]

Hello,
Are you sure no refresh is running? Have you tried vacuuming and optimizing the delta table? Did you also check log consistency?