Errors in the metadata manager while running "SQL Server 2005 Data Mining Configuration Wiz

Hi,

When I'm running Getting Started "Microsoft SQL Server 2005 Data Mining Add-Ins for Office 2005" for Excel and

1. choose "Use an existing instance of Microsoft SQL Server 2005 Analysis Services that I administer"

2. Click "Next"

3. To run the Server Configuration Utility, Click on link provided:

C:\Program Files\Microsoft SQL Server 2005 DM Add-Ins\Microsoft.SqlServer.DataMining.Office.ServerConfiguration.exe

4.This opens a new configuration wizard - SQL Server 2005 Data Mining Configuration Wizard

i. Click Next

ii. Select Server Name "localhost" and Clicked Next

iii. Checked "Allow creating temporary mining models" and Clicked Next

vi. In Step 3, By default radio button "Create New Database" is selected and Database Name is "DMAddinsDB"

Here, I've got following error in the message box :

Errors in the metadata manager. The dimension with ID of 'Dim Employee', Name of 'Dim Employee' referenced by the 'Adventure Works DW' cube, does not exist. Errors in the metadata manager. An error occurred when loading the Adventure Works DW cube, from the file, '\\?\C:\Program Files\Microsoft SQL Server\MSSQL.2\OLAP\Data\Analysis Services Project1.0.db\Adventure Works DW.5.cub.xml'.

I've droped existing "Adventureworks" and "AdventureworksDW" databases and recreated again but still i'm getting the same problem.

Please help to resolve this problem

Thanks!

Regards,

K. Sandeep

It seems that the "Analysis Services Project1" database on your server is corrupted. Could you try to remove that database, then re-run the ...ServerConfiguration.exe application?

Errors exporting to excel

Report works fine in HTML and exporting to PDF but when exporting to excel and opening the file we receive a notification that errors were detected and a repair was attempted. The repair notification states that the damage to the file was too extensive to repair and the file can be opened with only the raw tabular data (no formatting or charts).
This only happens with certain parameter values?
--
Message posted via http://www.sqlmonster.comThere are a couple of fixes out there for Excel. Take a look at
Microsoft.com for Excel\Reporting services issues.
If you have a good repro I can do in house, please send it to me.
--
| From: "William Nichols via SQLMonster.com" <forum@.SQLMonster.com>
| Subject: Errors exporting to excel
| Date: Fri, 28 Jan 2005 13:07:35 GMT
| Organization: http://www.SQLMonster.com
| Message-ID: <b4e260ddb47c48b7bf16a89e152b3412@.SQLMonster.com>
| X-Abuse-Report: http://www.SQLMonster.com/Uwe/NB/Abuse.aspx
| Newsgroups: microsoft.public.sqlserver.reportingsvcs
| NNTP-Posting-Host: 178.67-18-207.reverse.theplanet.com 67.18.207.178
| Lines: 1
| Path:
cpmsftngxa10.phx.gbl!TK2MSFTNGXS01.phx.gbl!cpmsftngxa06.phx.gbl!TK2MSFTNGP08
.phx.gbl!TK2MSFTNGP12.phx.gbl
| Xref: cpmsftngxa10.phx.gbl microsoft.public.sqlserver.reportingsvcs:41174
| X-Tomcat-NG: microsoft.public.sqlserver.reportingsvcs
|
| Report works fine in HTML and exporting to PDF but when exporting to
excel and opening the file we receive a notification that errors were
detected and a repair was attempted. The repair notification states that
the damage to the file was too extensive to repair and the file can be
opened with only the raw tabular data (no formatting or charts).
|
| This only happens with certain parameter values?
|
| --
| Message posted via http://www.sqlmonster.com
||||I've got the same problem!
I turned off the legend on the chart, and now it works perfectly!
Any ideas'
Try turning off the legend and let me know if that worked.
- Joel
"William Nichols via SQLMonster.com" <forum@.SQLMonster.com> wrote in message
news:b4e260ddb47c48b7bf16a89e152b3412@.SQLMonster.com...
> Report works fine in HTML and exporting to PDF but when exporting to excel
> and opening the file we receive a notification that errors were detected
> and a repair was attempted. The repair notification states that the
> damage to the file was too extensive to repair and the file can be opened
> with only the raw tabular data (no formatting or charts).
> This only happens with certain parameter values?
> --
> Message posted via http://www.sqlmonster.com

ErrorCode and ErrorColumn in Excel Destination

I am exporting records with errors to Excel using the Excel Destination tool. The ErrorColumn is a numeric. How do I find out which column it is?

If you want to do this manually (i.e. not programmatically), you can open the Advanced UI for the Excel Dest, and go to the "Input and Output Properties" tab. There, under "Excel Destination Input", you can look through the colums. As you click on each one, look for the "ID" property on the right that has the value you found in the ErrorColumn field of the error row.

Let me know if this helps or not.

Thanks,
Mark|||Hi

How would you do this programmatically? Ideally, I'd like to include the actual column name as another output column in my error destination table. Also, is it possible to obtain the error description as well as the ErrorCode (similar to what you might see in the Execution Results pane when debugging)?

Thanks

Rob|||Hi Rob

I want to do something similar but write out the column name to a table along with the data when rejected. Did you ever figure out how to do this?

Thanks

Marcus|||While it's fairly easy to add an ErrorDescription column on top of the ErrorCode, it's NOT so easy to add an ErrorColumnName on top of the ErrorColumn numeric ID value, because of all the ifs ands and buts that determine whether the column name is readily available or not. (For example, did the preceding component have synchronous or asynchronous outputs, and so forth.)

You can run your error output through a Script Component and try looping through the columns in its InputColumnCollection and VirtualInputColumnCollection looking for the ID value that you've got in the ErrorColumn column to see that this information is usually unavailable.

-Doug

Error: The MDX function ROOT failed because...

I'm having problem calculating sales for all customer minus certain occupations.

This Adventureworks query is something like what Excel 2003 produces.

Code Snippet

WITH MEMBER [Customer].[Occupation].[Exclude Occupations] AS

'AGGREGATE({

[Customer].[Occupation].&[Clerical],

[Customer].[Occupation].&[Management]

})'

MEMBER Measures.[Other Occupation Sales] as (ROOT(Customer), [Measures].[Internet Sales Amount]) - [Measures].[Internet Sales Amount]

SELECT

{Measures.[Other Occupation Sales]} ON COLUMNS,

[Product].[Category].[Category].MEMBERS ON ROWS

FROM [Adventure Works]

WHERE

([Customer].[Occupation].[Exclude Occupations])

Error: The MDX function ROOT failed because the coordinate for the 'Occupation' attribute contains a set.

If you delete this line, the query works!

[Customer].[Occupation].&[Clerical],

Does anyone know what this means? Is there another way to get Excel to formulate an equivalent query?

The problem is because there is a set in the current coordinate, and Root has problem dealing with it (although it shouldn't!). The best solution is to replace Root(Customer) with [All Customers] - which will have exactly same functionality, better performance, and won't generate the error that started this thread.|||

Yes, I replaced Root(Customer) with [Customer].[Occupation].[All Customers] and it works. The problem is that I wanted to create a general purpose calculated measure that would override every attribute in the customer dimension with "All". I thought Root did precisely that. If I can't use Root() then I will need to remember to update this calculation every time an attribute is added or removed from the customer dimension.

Thanks for the input.

|||

> The problem is that I wanted to create a general purpose calculated measure that would override every attribute in the customer dimension with "All"

[All Customers] does exactly that ! There is no need to update your calculation every time new attribute is added, [All Customers] (in any hierarchy of customers dimension) will automatically move coordinate to All member in every attribute.

|||

Try my AdventureWorks repro with [All Customers] instead of Root(). It doesn't work. It doesn't replace the aggregate set of occupations with the All tuple. I end up with zeros in the results because [All Customers] is doing absolutely nothing for me.

The only way I got it to work was with [Customer].[Occupation].[All Customers]. This member does, in fact, replace my aggregate set of occupations.

Put that in your pipe and smoke it!

I'm curious how [All Customers] is defined to work (vs. Root()).

|||Of course - you are right. I don't know what I was thinking. What I said is only true when dimension has single hierarchy which includes all attributes, which is obviously not the case for most of the dimensions, especially Customer.|||Thanks for the help. At least I have a work-around for now. It would be nice if Root(dimension) did the same thing as change coordinates to [All Customers] on every attribute. Maybe in the next version...