Hi guys,
I have an ETL script to process my cube
1) Process update on dimensions that are linked with the fact table SALES
Process full on other dimensions
2) Process full on all measure groups apart from Sales
3) Process Sales by script (partitioned over years from 2011 to 2015). The script is made so that only 2015 partition gets processed
4) Process index on all partitions
However, since I changed my process dimensions from full to update the cube crashed on error : cannot find SK_PROD XXX in partition SALES 2015.
When I query the datamart on sales joined with the date 2015 the SK_PROD XXX is found...
Anyone could imagine this scenario why cube crashes?