Greetings all,
Within the PBI service, my company has a pretty standard deployment pipeline (Dev > Beta > Prod). All of them draw their data from a single Azure data warehouse.
Recently we decided we wanted to reduce the amount of data in our Beta environment to a single year of data for select customers. To do this, we accessed the cube on the backend of the PBI report via Analysis services in SSMS. We selected the table we wanted to clear out, and did a Process Clear on that table.
The intent was to then apply some filters and then reprocess the table so it only had the limited data. However, when we attempt to reprocess the table, it fails within seconds giving the following error.
Failed to save modifications to the server. Error returned: '{"error":{"code":"DMTS_MonikerWithUnboundDataSources","pbi.error":{"code":"DMTS_MonikerWithUnboundDataSources","details":[{"code":"Server","detail":{"type":1,"value":"XXX"}},{"code":"Database","detail":{"type":1,"value":"XXX"}},{"code":"ConnectionType","detail":{"type":0,"value":"Sql"}}],"exceptionCulprit":1}}}
Here, the XXX has replaced the server and database names for security purposes, but I'll note that both are correct, so it doesn't appear to be mispointed in any way.
If it's relevant, the major tables are also partitioned. I have tried processing individual partitions but get the same error.
Other things we've tried include accessing the model through Tabular Editor and reapplying the Refresh Policy to see if something was broken with that as it controls the partitions. It was able to detect and add a new partition to the table, but did not solve the issue of the error. We have also tried scripting things out and running them to no effect.
I could really use some help with this as this is causing quite a bit of hold-up in our deployment pipeline for this report!
