Quantcast
Channel: SQL Server Analysis Services forum
Viewing all 14337 articles
Browse latest View live

Time Dimension Not Appearing When Adding Dimension to Cube

$
0
0

I've asked a question similar to this before, but didn't quite find an answer the first time around, so hopefully the second attempt will garner some results!

I have a cube that was essentially built using views/named queries as the "tables". The time dimension was generated using the create time dimension on the server option.

Now we are moving to a 4-4-5 accounting cycle. In order to accommodate this change, I need to add several new columns to the existing time dimension for the new FY columns. Since the time dimension was created on the server, that's not possible.

I went through the process of creating a new time dimension using the wizard and created a physical table in the underlying data source. The wizard also created a data source view for the newly generated table. The new dimension appears in the list of dimensions. The new dimension is called ICF FY Calendar. The dimension is also configured properly:

Once the physical table is generated, I then add it to the DSV which houses all the other views/queries for the cube and its dimensions.

However, when trying to add this new dimension it fails. It doesn't actually fail, the dimension does not appear in the list of available dimensions to add to a cube:

The dimension processed successfully and I'm able to browse it no problem.

Why wouldn't a dimension appear in the available list??

Have no idea what the scoop is! If anyone has any insight that would be fantastic!\

Thanks!!

 


A. M. Robinson


Connect to DB

$
0
0
Can anyone tell me why my Adventure Works DB will not connect? When I click the Database Name drop down nothing is there? Thank You 

JohnB

Dynamic calculated dimension member for "new customers"

$
0
0

Hi!

I've got a dynamic set = show customers where sales in this month, but not in previous 12 months. I'm using it as a Pivot Table filter. This works fine in Excel, unfortunately sets are not supported in Power BI. How to I create a calculated member to my Customer Dimension, that does the same thing in Power BI? This needs to be a dynamic dimension member so I could use it as a filter in Power BI not as a measure. It should also work with ohter related dimensions, like product groups.


Thanks,
Kaarel.

Errors when using CROSSJOIN command against SQL Server 2008 R2 from MSOLAP.4 provider.

$
0
0

I am trying to run a query containing the CROSSJOIN command against a SSAS cube from the client-side application, and get the below error:

-2147467259: Microsoft OLE DB Provider for Analysis Services 2008 R2.: Internal error: An unexpected exception occurred.:0

The issue is isolated to queries using a CROSSJOIN between filtered sets and normal measures. Other queries run fine.

Also, the same CROSSJOIN query runs perfectly fine if I run it from SSMS instead.

The client is a legacy VB application. It creates a System.Data.OleDb.Connection object and uses the below connection string

Provider=MSOLAP.4;Integrated Security=SSPI;Data Source=<ServerName>\<InstanceName>

SSAS version is  SQL Server 2008 R2 SP3 (10.50.6000.34)

A similar issue was highlighted for SQL 2005 and SQL 2008 here. But its supposed to be fixed in SQL 2008 R2 SP3.

Any help would be appreciated.

Thanks!!


Model Filtering Not Working Properly

$
0
0

Hi,

I built the following model in order to analyse transfer statuses changes over time. 

The tables are connected using Dim_Transfers.Transfer_ID = Fact_Transfers_Status_Hisotry.Transfer_ID

For example lets take the following data:

I have two serious issues when using the model via excel pivot table:

1. When i place Dim_Transfers.Transfer_ID in Filters and Fact_Transfers_Status_Hisotry.Transfer_ID on Rows and i try to use the filter in order to filer out some transfers, the filtering doesn't work and i see all 3 transfers. 
The filtering only works if i drag one of my two distinct count measures form the fact table.

2. when i place Dim_Transfers.Transfer_ID and Fact_Transfers_Status_Hisotry.Transfer_Status in Rows i get for every Transfer_ID all the statuses and not only the statuses relevant for that transfer.
The issue is solved when if i drag one of my two distinct count measures form the fact table.

Is that expected behavior or am i missing something?

Thanks in advance,
Udi

SSAS Hangs when processing a particular database

$
0
0

We have a multi-tenant app. Each tenant has several SQL and one OLAP (MD) databases associated with it. The tenants run mostly independently, receiving and processing data, updating databases and occasionally processing an OLAP database. Since OLAP processing is very resource intensive on the machine, we serialize access to OLAP so that we will never have two tenants trying to process OLAP at the same time.

The past weekend we had an incident in which one of the tenants went to process OLAP and encountered an error (bad foreign key value).  Several hours later, this tenant went to process OLAP again but this time OLAP processing froze.  Never finished and did not respond to cancellation attempts.  In the time between the two attempts from this tenant, several other tenants successfully processed their respective OLAP databases.  Since we serialize access to OLAP, over the course of a few hours we eventually ended up in a situation where all tenants were blocked waiting for OLAP.

We restarted our service - a few tenants successfully processed OLAP, but once it came around to the tenant that blocked before, that tenant tried to process OLAP and again blocked with no progress.  Again, after a few hours all tenants were blocked waiting for OLAP.

After a lot of log reading and analysis, we opened up SSMS on the database tier and tried to process the OLAP database for the tenant that became stuck.  SSMS also became stuck - immediately.  It didn't make it far enough in the process to have written anything to the progress window other than the start time.

This seemed like clear evidence that the problem lay inside SSAS, so we stopped SSAS using the Computer Management UI.  The service did not stop timely (the SCM gave up), but it did eventually stop. 

We then disabled the tenant what was becoming blocked, and restarted SSAS.  Once SSAS was up, we processed the blocked tenants database manually with no error.  All of the other tenants awoke from their slumber and processed their OLAP databases without error.  We then re-enabled the previously blocked tenant and it too processed it's OLAP database without error.

So, it seems that all this trouble was caused by an OLAP database becoming stuck - all attempts to process that database resulted in a hung connection.  We could see when we restarted our app that other databases processed correctly, so it seems that a single database was stuck in SSAS while others processed successfully.

We're running on SQL 2008 R2 (ancient, I know).  I was unable to find anything that sounded similar to this using bingle or in Connect.

I'm curious if anyone else has seen anything like this, and especially if anything's been fixed since 2008 R2 that might explain this.  Meanwhile, we can only chalk it up to a random occurrence... until it happens again, of course.


-cd Mark the best replies as answers!



Cube process stoped forcefullly and now unbale to connect SSAS in SSMS

$
0
0

Hi ,

We are using the Sql server 2012 ver.

I have fourcefully stoped the cube process on SSMS, because the cube process is running from long time.  and after that unfortunately closed the ssms  application.

Now I am trying to re connect the SSAS but it is not connecting and also not showing any error message.

So could you please help me to connect the SSAS in SSMS

--Siva


Is it possible to load project data from PWA online?

$
0
0

I'm using SSAS Tabular Model, and I have tried to load data from PWA Online into PowerPivot via OData. It works. I can refresh it. But when I tried to do it in SSAS, it failed. If I imported this power pivot file into SSAS, I can't deploy it or refresh data. I think something wrong on connection. But I don't know the reason. The connection is the same, but not work in SSAS. what should I do? Do you have any suggestion?

Kim.


SSAS Method not found

$
0
0

Hello Everybody!

I hope you guys can help me with this, it's the dreaded situation where the google results page shows exactly one sponsored link but no help at all regardless of how I phrase the question.

I'm in the process of setting up a cube in SSAS. I've got an old cube project that used to work fine and a new one that should, in theory, also deploy instantly. Whenever I try to deploy the cube, I'm getting the following error message:

SeverityCodeDescriptionProjectFileLineSuppression State
ErrorMethod not found: 'Void Microsoft.AnalysisServices.DeploymentEngine.ConfigurationSettings.ExtractFromObjectModel(Microsoft.AnalysisServices.Database, Boolean)'.0

It looks a lot like there is just some method missing (obviously), but I haven't had the joy of tampering with Visual Studio before and have no idea where I'd have to start if I wanted to insert a (new) method.

Any help would be greatly appreciated!

Have a nice weekend![

Ron

Conditional (IIF/CASE) Calculated Measure ignores filter in Excel

$
0
0

Hello forum members. I've been trying to solve an issue with Calculated measures removing dimensional filters in Excel and need your help. Here is a simple scenario:

Two dimensions: DimCategory, DimColorand a Fact table

DimCategory:
CategoryA
CategoryB

DimColor
Blue
Green
Red

Fact
CatID          ColorID    Value
CategoryA    Blue         0
CategoryA    Green      0
CategoryB    Blue         0
CategoryB    Red          0

There is also a calculated measure in the cube, CalculatedValue:
IIF([Measures].[Value]=0, 0, [Measures].[Value] * 100)

All seems to be fine with browsing the Cube in Excel, applying filters to either Category or Color untilCalculatedValue is dropped in the list of browsed Values. This is when all filters get ignored.

Example: Cube is browsed in Excel, Category filter is applied to select only the list of colors in CategoryA (colors: Blue and Green). As soon as CalculatedValue is used as a fact in the pivot table, the Red color is added the list of colors (thus, the above CategoryA filter is ignored)

capt

Is there an explanation and a solution to this behavior?

Thank you, Dmitry

How to setup connection so that outsiders can access my cube

$
0
0

Hi

I have a cube, say ABC. The cube has 2 roles, R1, R2. I assign 2 partners, P1 and P2, to the 2 roles, respectfully.

From my machine, I can use login of the 2 partners to access the cube perfectly. This is my network.

Now, I want to set connection so that partners, who not in my network, can access the cube ABC.

In another work, how can I set connection so that outside people can access my cube, with login and password assigned to them ?

Thanks

How find out all the cubes in a SSAS Server using DMV Quries

$
0
0

Hi,

I have a requirment to find out all the cubes in a SSAS Server.. Lets see say we have five databases and each Database has five cubes..Now I want the list of all 25 cubes with database Name?

Please let me know how to achive it using DMV Quries.

Apparent firewall blockage

$
0
0

When I am at Office Location A:  I can configure SSAS 2012 to run on port 443 or on port 80 and either way I ***CAN*** access it via an Excel pivot table connection.

Same spreadsheet that was working just fine at Office Location A, fails with an "error at transport later" at Office Location B.

I'm running out of ports.  If I cannot get port 443 or port 80 traffic through the firewall at Office Location B, what port am I supposed to use?  Or are firewalls sophisticated enough now to block certain traffic over port 80 but not others?

(I am also assuming that the firewall at Office Location B is the problem, although I suppose there could be other explanations.)

Any thoughts/comments appreciated, thanks.


Impersonation Information Page

$
0
0

Now I'm at the Impersonation Information Page (see below). The only credentials I can think to use is for my Windows Live account and that didn't work? If I chose Current User, would I be able to carry out the rest of the exercises? Thank you.



JohnB

KPID (Os thread) to map to SPID in SSAS Tabular

$
0
0

Hi,

We are running SSAS Tabular instance and I could figure out which thread in SSAS is consuming most of the memory by using Process Monitor but I am not sure how to map the thread ID (Os thread ID ) to the SPID so that I can track which query is making it slow. I tried master..sysprocess but that does not display SSAS tabular instances threads and I also tried OS_threads table but that also did not helped here 

Regards

Rajaniesh



DAX - Time Intelligence - Previous Year, Same week, Same day of week

$
0
0

Hi,

At a client we have a Tabular model in place having a Sales fact table and Date dimension table.

The client wants to know the sales quantity of the following period:[Previous Year] + [Same Week] + [Same Day Of Week].

The following measure works as expected when the "Date" attribute is selected in the context, but when "Month" or "Year" is selected as the context then you get an error that 'DimensionDate'[Date] can't be set with multiple values (returned by "VALUES").

MeasureSalesPY:=
SUMX
(
 FILTER
 (
  ALL('DimensionDate'[Date]);
  'DimensionDate'[Date] = VALUES('FactSales'[PreviousYearSameWeekSameDayOfWeek])
 );
 [MeasureSales]
)

The attribute 'FactSales'[PreviousYearSameWeekSameDayOfWeek] contains a calculated value of the current date in the previous year, same week, same day of the week.

How can this issue (you want to get the sales of the previous year on "Date" level, and want to aggregate it on "Month" or "Year" level) be solved?

Thank you in advance.

Regards,

Johan Machielse

A table full of duplications and we want to link it to other table full of duplications

$
0
0

Hello team,

A table full of duplications and we want to link it to other table full of duplications:

If I generate PK or composite pk on first table, how to populate the second table with those primary keys either autonumber of composite key that make sense?

Regards,

BittenApple


GGGGGNNNNN

numer of open projects at the end of a period

$
0
0

Hello,

I'm using Sql 2016 Standard, Analysis Services Tabular mode.

I have a tabular model with 3 tables:

-  A "fact" table Projects, with Project_ID, startdate, enddate, en Projecttype_ID. Several projects are still open, so they do not have an end date.

- A Date table with columns Date, month, quarter, year

- a Projecttype Table, with Projecttype_ID, type

I created a measue OpenProjects, which  should contains the number of projects at the end of a period.

But when I test it in excel with a pivot table, with Openprojects as value, year as row field, and projecttype as filter, the number of Openprojects is not correct. For example:

filter Projectype ABC

Year                count of openprojects    started                ended

2014                        75                        100                    25

2015                        62 (wrong)             80                     90      (openprojects 12-31-2015 should be  75 + 80 - 90 = 65)

2016                        68                         5                        2

I found several articles on Internet with formulas about countrows with filters like Startdate <= MAX(date) && Enddate > MAX(date), but nothing about how to do this when some enddates are empty.

DAX ROWNUMBER Measure

$
0
0

Hello,

I was trying to calculate a measure to get the same has the rownumber function... 

Basically I need a non repeatable and increasing number based on a certain sort of columns.

Is this possible to achieve with DAX? I tried RANKX but it does not work when I have repeating values on my ranking column... 

Thank you very much.


Luis Simões

history recordsin tabular model

$
0
0

we are trying to build a tabular model using historical data model. I've few challenge in representing historical data in model. PFB sample data.

Table1



 Table2

End result in tabular model is to  view historical data of both Table1 and table2.

Requirement is

when filter by hostid=1, they need to see the history for any timeline as below.

Approach is to join two history table in tabular model. We tried to get the result by expanding startdate and enddate of table 1 and table2 and join to DImdate table. But expanding date severely affects performance.

If a record is active between 2016-01-01 to 2016-12-31 then 365 records at end of year. If millions  of records, run out of volume.

Can anyone suggest whats the best method to show the result in tabular model. I'm having challenge on how to relate table1 and table2 to get above result without explode the data volume.


Viewing all 14337 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>