It looks like Microsoft recently added a new feature to the Power BI service that retries scheduled semantic model refreshes up to 4 times when they timeout. I can find no documentation on this feature.
This seems like a nice feature, but in my use case it ties up my database and causes a cascade of failues. It also causes the database to be tied up so nothing else can be done.
I have searched in vain to find a way to adjust the number of retries. Any help you can provide on this would be greatly appreciated.
Here are some details on my environment:
The impacted workspaces are not Premium capacity.
The data source for all of the reports are on a single Azure SQL Database.
Most of semantic model refreshes are scheduled in the service. The others are triggered via a Logic App that calls the api.
Up until recently, timeouts happened infrequently. In the last week or so, the number of timeouts has spiked.
