Errors in OLAP Storage Engine “The Attribute Key cannot be found when Processing”

1. If you are facing this error, Try first to Full Process all your Dimensions and then reprocess your cube.

2. When the Key value is present in the fact table but not present in the Dimension table (The keys on which Dim and Fact tables are linked See key relation in Dimension Usage tab.)

Dim Table				Fact Table
ProductKey	Product Name	Product Status		ProductKey	Sale Price	Qty.
-1	Unknown	unknown		1	200	20
1	Shirt	A		2	300	15
2	cap	A		3	20	100
3	Pen	A		4	20	100
				5	10	15

Workaround: 

• If you face this issue select distinct key values from Dim table and Select distinct Key values from fact table , compare and check which key value that is present in Fact table and not present in Dim table . Investigate and plan your next course of action.
• If above point is true in your scenarios, you can insert missing Key value in Dim table that is present in fact table but not present in Dim table.
• You can restrict Fact table records for the key value that is not present in the dim table.

(This is just an example you have to plan your course of action what to do according your requirement. But I would like to suggest you investigate in your fact table and dim table populating logic why this key value is present in fact table and not present in Dim table.) 

3. When there is a null value present in your fact table key column value.

Some times while populating the fact table from different sources, we use many joins (left, right etc). Due to lack of data in any source tables, Key value can be null in fact table. Then you can face this issue.

Dim Table				Fact Table
ProductKey	Product Name	Product Status		ProductKey	Sale Price	Qty.
-1	Unknown	unknown		1	200	20
1	Shirt	A		2	300	15
2	cap	A		3	20	100
3	Pen	A		null	10	15

See above fact table, there is null value in Key column  and there is no matching Key value is in Dim table .In this Case, you might experience error “Errors in OLAP Storage Engine The attribute key cannot be found when processing.”

Workaround:

      While designing the dim table you should insert one record with -1 for key value.

Let say you have product table for dimension with columns [Product ID], [Product Name],[Product status]  you should populate it as

Product  ID	Product  Name	Product Status
-1	unknown	unknown

While populating the dim table your SQL should be like below

SELECT

               -1  as ProductID,

               ‘unknown’ As ProductName,

               ‘unknown’  AS ProductStatus

 Union ALL

Select

                ProductID,

                ProductName,

                ProductStatus

From your

Use following logic to populate Fact Table key columns

Select ISNULL( ProductID  ,-1) AS ProductID

This practice you have to follow for other key column values also. Doing so, when due to any reason if key value column have null from source tables then it will convert to -1 key column value in fact table .Now you have -1 as key value in both Fact table and Dim table. In this case there will be no chance of key mismatches and all the record will be aggregated under unknown or -1 and your cube will not fail during processing.

Now see you have populated -1 for the null key value column of fact table. It will match with -1 key value of Dim table.

Dim Table				Fact Table
ProductKey	Product Name	Product Status		ProductKey	Sale Price	Qty.
-1	Unknown	Unknown		1	200	20
1	Shirt	A		2	300	15
2	cap	A		3	20	100
3	Pen	A		-1	10	15

 
4. Sometimes keys are present in the fact and Dim table but still you get this error.

 

Workaround:

·         In this case it might be the reason that you have used Named Query or View to build your dimension in cube where you have made some restriction/Filter etc.

·         If your Dim table and fact table key values are of Varchar/String data type then there may be chance of blank spaces. Then you should take care of blank spaces using Ltrim/RTrim etc