May 1, 2009 at 1:26 pm
I'm trying to process a dimenion (ex. dimcustomer) that contains 2 columns - (CustomerKey,CustomerName). Some of the customer names are in Asian/Chinese characters. When I try to process, I get the following error:
Errors in the OLAP storage engine: The attribute key cannot be found when processing: Table: 'DimCustomer', Column: 'EndCustomerName', Value:[][][][][][][][][][][]. The attribute is 'CustomerName'.
The value is : '坂口テレビサービス株式会社 '. SSAS can't read this value so it can't find the corresponding attribute Key.
After the processing failed, I tried to query the CustomerID's related to this value on SSMS. I got the following result:
4655884 ?????????????
Where 4655884 is the CustomerID and ????????????? is the CustomerName.
I do not receive any error when I load this dimension on my DB via SSIS or when I apply Foreign Key relationship w/ the Fact tables.
Any ideas on how I can get this to work?? Current server collation is :SQL_Latin1_General_CP1_CI_AS
May 1, 2009 at 1:33 pm
If your data is in Chinese your column definition must be in Nvarchar and that column collation needs the version Chinese you are using, check with the person who owns the data and then check the BOL so you could get Chinese in your cubes. The reason it is not practical to expect to render Chinese, Japanese and Korean with Latin alphabet collation which is your current collation.
Post again if you still need help.
Kind regards,
Gift Peddie
May 21, 2009 at 1:04 pm
Thanks.
The collation for the column in source database is: Latin1_General_CI_AS.
I've tried the following fixes:
1) Since the table contains both English and non English data, I've changed the collation of the column to different Chinese\Japanese types anyway...and the processing failed.
2) I've also created a surrogate key w/ customer key column to see if it makes any difference. It doesn't.
3) Installed SSAS 2005 and tried processing the dimension there, not dice!
Since the collation of the source table is Latin1..., now I'm not sure if this is a collation problem. However, I can't figure out what is causing the dimension to fail over and over again.
May 21, 2009 at 1:50 pm
I am not sure you can use that collation for Chinese, Japanese or Korean languages, this is SSIS code page converter from Microsoft you could use it to move your data to a new database that can hold what you need. I don't know why you code is failing but that is not your current problem because you need an Nvarchar column with east Asian collation.
Kind regards,
Gift Peddie
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply