I face a problem with BI (Cubes) . I'm already install
analysis services and follow the link http://technet.microsoft.com/en-us/library/gg731892.aspx
to Configure Analysis Services and deploy cubes . but I had some
trouble like the figure below
Dimension [Company] : The 'Source' property references unknown 'Data
Source View' with ID='Dynamics AX
"warning Dimension Missing relationship [Geographic location].[State or
province]. The updater automatically adds one [Geographic location] ->
[State or province"
and when I run presses for cube from SQL analysis server I face
the error Below
in the OLAP storage engine: The attribute key cannot be found when processing:
Table: 'VENDTRANS', Column: 'DEFAULTDIMENSION', Value: '0'. The attribute is
'SETID'. Errors in the OLAP storage engine: The attribute key was converted to
an unknown member because the attribute key was not found. Attribute SETID of
Dimension: FrameworkDimensionsAttributeValuesSet from Database: Dynamics AX,
Cube: Accounts payable cube, Measure Group: Vendor transactions, Partition:
Vendor transactions, Record: 1."
I make search for this issue one of
them delete a partition and recreate again some partition process
with no error and others process with error like partition
for VENDTRANS at Accounts payable cube.-
The second scenario was process
all of dimensions and no error for it . but still the error
while process the cubes .-
N.B : as I do this scenario at My Test environment and I
do it at MICROSFT AX 2012 Virtual Hard drive For Services Industry
the same errors happen
Can you please investigate those errors and guide
me for the best scenario to deploy and configure Analysis Services for
Dynamics AX 2012 .
Except for the data source view id messages, all the other messages you are seeing are situations that are normal and should be ignored.
Info about these messages.
1. The missing data source view message is a serious error. It means the ssas project is somehow corrupted. either the data source view file is missing or its id somehow changed. If it got far enough to output a project take a look at the project and see if it has a file with a .dsv extension. If the file is present look at the ID value for the datasource view. If it's set to something other than "Dynamics AX" change it to that value.
That you were able to process the cubes, means either you've somehow fixed it, or else processed the original version of the cube project without configuring it. In most cases reconfigurating the project is optional. It's mainly there to update the cube project in cases where people have made schema changes to the views that are included in the cubes. If you are not getting a process error for the project, you don't need to reconfigure it.
2. The index warning messages in the first screen and the messages that the updater added a missing relationship are ignorable. You don't need to do anything about either of these messages.
3. The messages during processing that an attribute key is not found mean there are null values in a measure group view for some of the columns that define the relationship between a measure group and a related dimension. This is a normal situation. It just means the column allows nulls, and some of the records are nulls. The impact of this would be that when looking at the measures in that measure group by an attribute from a dimension where the linking columns have a null values, the values for the null rows will appear under a "Unknown" member of the dimension attribute.
For example, Suppose there are measures about Sales totals and you are going to look at these measures by Geographic location dimension's country attribute. Suppose also that some of the sales are to customers whose records don't have an address. In this case you'd see a process message like this one, and when looking at sales by geographic location, you'd see total sales for say France, Germany, etc. and at the end of the list you'd also see total sales for "Unknown", which would be the the sales to customers who are missing addresses.
To get records out of the Unknown category, add the missing data for the record.
For the case in the picture, the DefaultDimension column stores a link to a set of financial dimension attribute values that are associated to a vendor transaction record. Since it is not mandatory to assign transactions to financial dimension attributes, it is normal for many of the records to have null values for this column. You can ignore any null key warnings for that particular relationship.
I am facing the error point 3 while processing the cubes. But the message says that the Cube is processed with warnings. Is it okay if we proceed & start using the reports ?
Yes, it's OK to use the reports when there are process warnings.
yes, it will return data but it will take more time while processing, so to avoid the warning messages you can follow up this link support.microsoft.com/.../922673
As Saber pointed out, processing time will improve if the null key warnings are not emitted. Null key warnings can be removed by changing the error configuration setting for KeyNotFound is changed to IgnoreError. That will stop null key warnings from being generated in the processing log. I'd set that on every measure group.