We have a number of dataflows in one Workspace (premium capacity) that started experiencing issues during refresh last week. These had been working for months without issue before. So far, I have seen 2 different error messages:
Error #1
Error: Compute engine failed: Insufficient memory. Please increase your compute engine/workload memory and try again. (8645). Details: A timeout occurred while waiting for memory resources to execute the query in resource pool 'default' (2). Rerun the query.
Error #2
There was a problem refreshing your dataflow. {"RootActivityId":"d8d04e68-36f7-4dd5-abeb-75ee885de2b1" "ErrorMessage":"Cannot acquire lock for model '797dbfa0-3988-499a-a807-e1b49a7c3377/1c1c7039-733f-4157-a2b0-387f536e5c22' because it is currently in use."}
For Error #1, from everything I have seen, dataflow memory should be managed by the service in premium capacity, and I don't see any settings I can change anyway.
For Error #2, the dataflow that produces this error does have several other flows that all feed into one Final dataflow for end user use, but there are no other refreshes occuring that should be locking the child dataflow when the error occurs.
I have other dataflows in another workspace using the same On-prem gateway and the same premium capacity that are refreshing without issues.
I have been able to work around both of these issues by removing the workspace from premium capacity then adding it back in, but as these are refreshing late at night, I would like to not need to do this anymore. Are there any issues going on that may be causing this?