Forum Discussion
Redshift and Gateway issue
Then the pbix file is so big that it cant be published.You can bootstrap the incremental refresh partitions to keep the original size under 10 GB. Then use XMLA refresh to fill each partition separately. This will then allow you to load datasets that are half the size of your capacity, and supposedly even beyond.
Data duplication issues are usually caused by incorrect application of the RangeStart and RangeEnd filters. One of them must be inclusive, the other must be exclusive, and they cannot be overlapping.
Also keep in mind that incremental refresh is really bad at data changes post fact. If your data is transactional in nature then you will need to mix the incremental refresh with occasional full refreshes.
Thanks for the info. My data do not change post fact.
As for the RangeStart and RangeEnd filters I'm using the following:
RangeStart
1/1/0001 12:00:00 AM
RangeEnd
12/31/9999 11:59:59 PM
and I add a between filter. The pbix file has correct data. After publishing a filtered version and following the proccess with ALM Toolkit and SSMS and the removal of the filter, the refreshed partions, especially those of the last month, return wrong data. If it was a parameter issue, shouldn't it affect the entire date range?
Thanks again for your time, patience and quality support.
As for the RangeStart and RangeEnd filters I'm using the following: RangeStart 1/1/0001 12:00:00 AM RangeEnd 12/31/9999 11:59:59 PM and I add a between filter.Please don't do that. Do not use fake dates, and do not use a between filter (as that would make both dates inclusive).
Instead, use a subset of your dates (just enough for development)
RangeStart
1/1/2022 12:00:00 AM
RangeEnd
1/1/2023 12:00:00 AMand then in your Power Query use a filter
[Date]>=RangeStart and [Date]<RangeEnd