Wrong/redudant data after refresh

[Anonymous]

Hi,

I have a question regarding data refreshing. I have two views on an Oracle database on which I have build a report. I have published this report to PowerBI.com and configured a scheduled refresh once a day.

Most of the times this works as a charm and I have nothing to complain. But, for the second time now, I all of sudden see really unusual data in my reports. When I force refresh the data source, this unusual data dissappeares and the right data is shown.

It really annoys me that this is going wrong. I edited the views in such way that they ares orderd by date, which should allow PowerBI to see the new records better, but it seems that that didn't solve the issue.

Is there anybody who may know what causes this issue? I'm sure that the views are OK, so I think there is a only a tiny chance that it's caused by them.

Thanks in advance!

Wessel van Erp

[Anonymous]

@Anonymous ,

How do you design the Power BI report? Do you apply any filter or slicer in the report? Could you please post an example about "redudant data"? We need to know that if additional columns or additional rows are added in Power BI visuals.

Regards,
Lydia

[Anonymous]

Hi @Anonymous,

Thank you for your response.

I'm going to try to give you an explanation on how the report has been build. I have two views: One with budgets and one with actuals. I have connected these views based on the projects for which they are meant. In order to do so, I used the summarize function to create a new table that includes all unique projects inside the budget view and then connected the two views to this table.

Maybe it's worth mentioning that the budget view is based on months and the actuals view is based on dates. I found out that, when it goes wrong, the data for the actuals give really high numbers.

Furthermore, I have created a simple date table with DAX and a small custom table that allows me to connect projects to certain departments.

I indeed do use several filters and slicers. Most noticable are the dates, departments and projects.

My statement about redundant data is not based on facts. Since the values in my report were heavily inflated while something went wrong, I assumed that it had something to do with redundant data. But to be honest I'm not completely sure that the data was in fact redudant.

Hopefully some things are more clear right now and you may be able to help me further to find the cause of the problem.