Problem scheduling processing

When I manually run the IS packages to load the data warehouse and then process the dimensions and cubes, everything works fine.

When I schedule the packages as jobs, eventually the data goes wrong. The jobs are scheduled in the correct order.

Any ideas?

Thanks.

If I get your description right, the SSIS packages are not executed in the correct order. Is that so?

For one you can use serveral other ways to process your dimensions and cubes: You can use ascmd command line utility to process exactly the cubes and dimensions. Then wou can schedule a SQL agent job to run ascmd.

You can also use SQL Agent jobs directly to send processing commands to Analysis Serivces.

Edward.
--
This posting is provided "AS IS" with no warranties, and confers no rights.

|||

eventually the data goes wrong.

What do you mean by "goes wrong" is it possible that it is just not updated, or not fully updated?

One common trap when scheduling jobs is that usually the scheduled instance runs under a different user account. You should double check that this account has all the appropriate priviledges. Sometimes logging on to your workstation with the same account that the SQL Agent is running under and running the packages interactively can help to highlight any issues. You might find things like network drive mappings are not the same, or privileges to access other databases are not the same your own account, or even that the account in question does not have rights to process the SSAS database.

Problem processing new cube

All,
When I try to "Design Storage" for a new cube, I get
a "Fact table size can not be 0" message. (I am in the
Cube Editor...) If I focus on the fact table and
select "Browse Data", the data is displayed as it should
be. So, the fact table size clearly is not "0".
From Query Analyzer, I can select rows. The table has over
one million rows in it.
I can't see anything obvious, such as "Refresh connection"
that would seemingly rectify this problem.
Any help appreciated.
TIA,
MikeIn the cube editor, set a value for the Fact Table Size property in the
Advanced tab of the property panel. This value along with the sizes of
dimensions is used during the aggregation design process.
- Matt Carroll
--
This posting is provided "AS IS" with no warranties, and confers no rights.
"Mike" <anonymous@.discussions.microsoft.com> wrote in message
news:013901c3c58c$86ab1c40$a301280a@.phx.gbl...

quote:

---

> All,
> When I try to "Design Storage" for a new cube, I get
> a "Fact table size can not be 0" message. (I am in the
> Cube Editor...) If I focus on the fact table and
> select "Browse Data", the data is displayed as it should
> be. So, the fact table size clearly is not "0".
> From Query Analyzer, I can select rows. The table has over
> one million rows in it.
> I can't see anything obvious, such as "Refresh connection"
> that would seemingly rectify this problem.
> Any help appreciated.
> TIA,
> Mike

---

Problem processing cubes in example Adventure Works DW

Hi,

I've installed the sample databases, and I am trying to process the cubes in the sample.

I keep getting errors like this, for cube "adventure Works":

Errors in the OLAP storage engine: The attribute key cannot be found: Table: dbo_DimCustomer, Column: AddressLine1, Value: 8011 Mcnutt Ave. Errors in the OLAP storage engine: The record was skipped because the attribute key was not found. Attribute: Customer of Dimension: Customer from Database: Adventure Works DW, Record: 1024.

I have set processing to ignore errors, but the cube is never successfully processed.

What can I do to get the cube processed?

Hi John,

Could you please check that you have installed the right AdventureWorksDW relational DB, and that your AdventureWorksDW AS DB's datasource points to it.

YL

|||

Yan,

The "Adventure Works DW" AS DB datasource points to the "AdventureWorksDW" database installed as sample (on the same machine), going by the connection string.

How can I check that I have the "right" AdventureWorksDW?

John

|||

John,

If you installed the sample Adventure Works DW successfully, and verified that the relational DB is also installed and no records in it were removed, the processing should work.

Yan

|||

I've fixed the problem. I had SS 2000 installed as my default instance and 2005 as a named instance and it appears that was the source of the problem. I've had to uninstall both, then reinstall only 2005, but the sample is now working.

Problem processing cubes in example Adventure Works DW

Hi,

I've installed the sample databases, and I am trying to process the cubes in the sample.

I keep getting errors like this, for cube "adventure Works":

Errors in the OLAP storage engine: The attribute key cannot be found: Table: dbo_DimCustomer, Column: AddressLine1, Value: 8011 Mcnutt Ave. Errors in the OLAP storage engine: The record was skipped because the attribute key was not found. Attribute: Customer of Dimension: Customer from Database: Adventure Works DW, Record: 1024.

I have set processing to ignore errors, but the cube is never successfully processed.

What can I do to get the cube processed?

Hi John,

Could you please check that you have installed the right AdventureWorksDW relational DB, and that your AdventureWorksDW AS DB's datasource points to it.

YL

|||

Yan,

The "Adventure Works DW" AS DB datasource points to the "AdventureWorksDW" database installed as sample (on the same machine), going by the connection string.

How can I check that I have the "right" AdventureWorksDW?

John

|||

John,

If you installed the sample Adventure Works DW successfully, and verified that the relational DB is also installed and no records in it were removed, the processing should work.

Yan

|||

I've fixed the problem. I had SS 2000 installed as my default instance and 2005 as a named instance and it appears that was the source of the problem. I've had to uninstall both, then reinstall only 2005, but the sample is now working.

Problem processing cube partition

I am getting the following error when trying to process a partition, as if
the process is trying to overwrite the database. How can I fix this?

TIA

Error 1 Errors in the metadata manager. The database with the name of
'ACRPROD_OLAP' already exists in the 'MBPENTSQL01R\ACR' server. 0 0

Are you deploying from BIDS? Did you rename a database that was previously deployed? The following post might shed some light. You might check the Name and ID properties of each deployed database and see whether any have the ID of ACRPROD_OLAP regardless of what Name they have.

http://geekswithblogs.net/darrengosbell/archive/2007/06/06/BIDSHelper-DeployMDXScript-feature-enhanced.aspx

Easy fix... delete the conflicting database off the server. More complex fix... instead of renaming a database on the server, backup the database, delete it, then restore it as the new name... that will make sure the new name matches the new ID.

sql

Problem processing cube

I am getting the error:

Warning 2 Errors in the OLAP storage engine: The record was skipped because the attribute key was not found. Attribute: B@. - key@. of Dimension: Brcledger from Database: Misys, Cube: Misys 1, Measure Group: Brcledger, Partition: Brcledger, Record: 24163. 0 0

I have read that this is caused by a referential integrity problem however, the fact table is the dimension table, and the joint primary key is definitely unique. There doesn't seem to be anything special about record number 24163 So why can i not process my cube?

Martin

Just incase it is any help.. the other error message which comes up I have pasted here:

Warning 1 Errors in the OLAP storage engine: The attribute key cannot be found: Table: dbo_brcledger, Column: B_x0040_, Value: 0; Table: dbo_brcledger, Column: key_x0040_, Value: 474142463032485130310000000C. 0 0

SELECT *
FROM brcledger
WHERE (key@. = '474142463032485130310000000C')

Returns 1 result

|||

You have a situation when Analysis Server cannot find a key in the dimension while processing partition.

I would guess you might have problems with dimension process that lead to partition processing errors.

Try re-process your Brcledger dimension change processing options to report and stop on every error.

After you fix dimension problems you should be able to process your partitions without a problem.

Edward.
--
This posting is provided "AS IS" with no warranties, and confers no rights.

|||

I have the same problem but it only occurs when i define my dimension as a time dimension. ]

Also my fact table is a view which gets it's surrgate keys by joining to the dimension tables in its definiton. This might be connected but the cube processes fine with the same data source, fact table, dimensions etc as long as i dont define the dimension as a time dimension.

Also the value of the attribute keys it cant find is always zero '0'. There are the same number of errors in the processing as there are nulls in the key column of the fact table. So i assume the nulls are being converted to 0 and then the referential integrity goes.....

Please help,

thanks,

Problem on Cube processing

Hi, all,

Thanks for your kind attention again.

I encountered a strange problem-whenever I process the cube(which is a small cube only, as I have already selected its partitions based on date dimension and the whole cube only with less than 1000 rows records), but I always got the problem which runs out the system drive space and of course resulted in the failure of the processing of the cube. I have actually manually allocated the storage locations which are not on the local system drive? Why did that problem happen and how can I figure it out?

That's really frustrated problem and please give me any advices and help on this if you have have any ideas what is going on for this.

I am looking forward to hearing from you and thanks a lot in advance again.

With best regards,

Yours sincerely,

What AS do you have AS2K or AS2K5?

Please, take attention to the queries that are send to data source sql server. I think you have somewhere cartesian product in your cube design that leads to the data explosion.

|||

Hi,

Thanks for your kind attention.

I am running AS2005. Yes, I have found the cause of the problem, as the fact table where the measure group is from is too big with over million records, thus even though my cube has been partitioned with only less than 1 thousand records, as the processing always need to query into the whole fact table still which thus resulting in the running out of the system drive space (as I am doing some test with my virtual machine at the moment only, which only has a poor system drive storage.)

I have then particularly partiton the fact table with only a small proportion of data for the cube, and now the processing is ok then.

Thanks again.

With best regards,

Yours sincerely,