stevienova.com

Crazy error today. In Analysis Services, (SSAS), some jobs got hung, sync command got hung, and what ended up happening is a cube on the default instance was totally unusable. Try to delete, rename, resync, redeploy, anything, and I would get the error.. Errors in the metadata manager. The dimension with ID of.. Couldn’t do [...]

Ok, I just spent a few hours debugging a cube that wasn’t bringing back any data. Checking all the relationships, the source data, the marts, etc, etc. And lo and behold, find out that a colleague commented out all the calculations, including the CALCULATE statement which says: The CALCULATE command controls the aggregation of leaf [...]

Being that we are now in the year 2009, you would think that most people are currently using or have been using the Office 2007 suite for a couple of years now. The truth is that there are many businesses “in the wild” that are still standardizing on the Office 2003 suite. Why? Well there [...]

It is the year 2008, we are half way through. Excel 2003 is 5 years old. Stop using it please. Why? Excel 2003 has the old “limits” – 65,000 rows, 256 columns, memory limits etc. Excel 2007 on the other hand, 1 million row limit, etc etc. That coupled with the way pivot tables work [...]

In SQL2005, you can create SQL Agent jobs. They can be scheduled, have multiple steps, alert, notify, etc. Pretty great setup. There are some downfalls though. Like in order to call a job from a job, you need to execute the second job with T-SQL. Thing is, it doesn’t run synchronously. It runs asynchronously, which [...]

In SQL Server Analysis Services 2005, you can create "perspectives" on cubes. What a perspective can do, is allow you hide different dimensions, measure groups and attributes. This works great but I still think there are a few things that are missing. The first thing is that the cube itself is a perspective, one that [...]

Ugh. Found this one out the hard way. Usually when you change underlying object id’s in SQL Server Analysis Services, it shouldn’t cause any harm. You might have some XMLA to process dim’s and measure groups, if so you would have to change those, etc. But all reporting services reports and excel 2007 pivot tables, [...]

Ok.. if you are using SSAS 2005 and using the deployment wizard (not using BIDS) – probably because you want to retain your partitions or something. But anyways, if you use it and choose the option to retain your roles, you will deploy, and no one will be able to get to your SSAS DB [...]

Ok, add this one to the annoying bin. If you are writing SSRS (SQL Server Reporting Services) reports of a SSAS (SQL Server Analysis Services) cube, and there is a time dim, you need to format the date parameters in the SSRS report to match the SSAS date format, as a string. I started out [...]

So, you have a SSAS 2005 (SQL Server Analysis Services) Cube, and its not performing how you want it to. What do you do? The first thing to look at is your attribute relationships (http://technet.microsoft.com/en-us/library/ms174557.aspx). Make sure you don’t have redundant relationships (AS should warn you as well) By default when you create a cube, [...]