Forum Discussion
semantic model Another user operation is already running. Wait for a few minutes, then refresh
According to Microsoft support it is part of a bigger known issue with Direct Lake.
Other symptoms are when you do get the new model created tables that are in lakehouse do not appear for 24 hours and you can only add a few at a time if they do exist. Of you try to add more you get this error again regardless of running pipelines, notebooks etc.
Basically you have to do it bit by bit by bit adding tables over a number of days to build your new direct lake semantic model. Crazy.
Thanks for confirming, this has been driving me crazy and consuming much time that i do not have.
Fabric is great, but it is very very rough around the edges. This is dogfooding at its worst.
- Sureshot1 year ago
Helper I
Rough around the edges is an understatement.
Have you seen the Latency issues between tables queried in PySPark versus the SQL Endpoint? We have had delays of upto 20 hours on that where data existing in PySpark has not updated in SQL Endpoint. Add to this that Fabric support is terrible and you have a fantastic product that really is not fit for purpose yet.
- wilson_smyth1 year ago
Post Patron
Tried to log a support request but it demands a premier support contract.
This could mean us backing away from Fabric for 12 months until its base functionality is at least somewhat stable.- Sureshot1 year ago
Helper I
DataBricks and Synapse SQL Serverless don't do these things but then again alot of functionality is not in those products.
We have spent alot of money on Fabric and our pilot projects are causing us to have to come up with lots of inhouse workarounds which only adds to frustration.