How to read IBM Cognos TM1 Version Numbers

Breaking this one down:
10.1.00100.24600

"10.1.0" = TM1 10.1.0

"00100" = FP1

"24600" = Build number, useful for IBM Support.

The important numbers are the middle ones (00100).

The first is part of the major release (aka 10.1.10000 = TM1 10.1.1)

The next 2 numbers are for the FP.

The last 2 numbers are for the IF (aka Hotfix).


Examples:

10.1.00100.24600 = TM1 10.1.0 FP1

10.1.10000.26473 = TM1 10.1.1 Base Release

10.1.10100.31864 = TM1 10.1.1 FP1    (For 10.1.1 FP2 this would be 10.1.10200.xxxxx)

10.1.10104.34644 = TM1 10.1.1 FP1 IF4

TM1/CDM Best Practices from IBM Supportlink

Check out some new TM1 and CDM best practices on IBM Supportlink:

https://www-304.ibm.com/connections/blogs/basupportlink/entry/new_documents_of_interest_for_ibm_cognos_tm1_and_fpm_users?lang=en_us

IBM Cognos TM1 10.1.0 and 9.5.2 object names cannot contain the following reserved characters

Goes with the best practice guidelines for using underscores in model names:

http://www-01.ibm.com/support/docview.wss?uid=swg27024713&myns=swgimgmt&mynp=OCSS9RXT&mync=E


Reserved Characters

The following characters are reserved. You cannot use these reserved characters when you name TM1 applications, cubes, views, dimensions, subsets, processes, and chores:

Character​	Description​
’​	apostrophe​
*​	asterisk​
@​	at sign​
\​	back-slash​
:​	colon​
,​	comma​
"​	double-quote​
>​	greater-than​
<​	less-than​
|​	pipe ​
?​	question-mark​
;​	semicolon​
/​	slash​

Note: The right curly brace (}) should not be the first character in any user-created TM1 object name. TM1 and control object names always begin with the right curly brace. If a user-created object name begins with a right curly brace, the object will be hidden to end-users if Display Control Objects is turned off.

Active Form Calculation mode

I would recommend reading this article on IBM if you work with active forms on a regular basis:

https://www-304.ibm.com/support/docview.wss?mynp=OCSS9RXT&mync=R&uid=swg21500429&myns=swgimgmt

Controlling the calculation mode will improve performance and prevent some screen flickering.

Some Active Form quirks in Cognos Express (9.0) TM1 (941)

-          In data validations, if you want to display a message make sure you have a title otherwise it will display null at the front of the message

-          If you have action buttons that navigate between sheets and leave the workbook name blank then the navigation will work but the workbook name will be lost from the tab bar. You can set the workbook name to be a named range if this is an issue

-          In the action button process options, don’t delete the messages, only untick them to not display anything

Active Form Top Ten Tips

In no particular order but hopefully a useful list for people to check through to ensure good performance:

1. Delete any Named references with #Ref in them, do an Excel find to see if there are any #Ref in cell formula, check for external links and change/delete
2. Delete any unused columns or rows. Use the slider as an indicator as to how many rows/columns are saved with the sheet
3. Keep SUBNM, ELCOMP formula to a minimum
4. Before publishing the Active Form delete all rows below the tm1rptrow formula
5. Any lookups to cube that are used in other formula’s need to be a DBR rather than DBRW
6. Workbooks perform best when uploaded to TM1 Server (copied to TM1 server through Add File route), subsequent navigation options via action buttons should then have “Look In: TM1 Applications” selected
7. Try to minimise the number of worksheets in a workbook. Definitely less than 8
8. Try not to use Action buttons formatted as hyperlinks
9. Don’t use dynamic named ranges e.g. ranges that are not visible to objects within the workbook
10. Use named ranges in action button properties rather than cell references e.g. =rngTM1Server not =’Sheet1’$A$1

TM1 and virus scan software

Link to official recommendations:

https://www-304.ibm.com/support/docview.wss?mynp=OCSS9RXT&mync=R&uid=swg21458762&myns=swgimgmt

Virus scan software can negatively impact the performance of TM1 and can cause corruption if it is checks TM1 files while TM1 is running. For that reason, virus scan software should be set to skip checking the following:

• The TM1 data directory
• The TM1 logging directory (if not in the data directory)
• The directory where the tm1s.cfg file is located (if not in the data directory)
• The directory where any Excel sheets (or other docs) are located and uploaded to the TM1 server as a reference
• The directory where any flat files may be written to (output from a TI process, for example) or pulled from (import into a TI process, for example)
• Any other databases being used by a TM1 process (Oracle, SQL Server, MSAS, etc.)
• The TM1Web/TM1WebEx files - by default located at: C:\Inetpub\wwwroot\

TM1 Active Reports - Alternative Row shading

Alternative row shading is a good techniques for visual reporting.
With Active forms you can easily achieve it by using a formula in column A such as:

=IF(TM1RPTELISCONSOLIDATED($B$20,$B20),"C",MOD(ROW(),2))

This formula will then return a result of C, 1 or 0 and your format range just needs to cater for these with 1 and 0 being different shading.

TM1 952 Parallel Interaction

TM1 952 introduces and exciting new feature called Parallel Interaction which massively improves the locking around multiple writers to cubes.

There is a good IBM tech note here to read:

https://www-304.ibm.com/support/docview.wss?mynp=OCSS9RXT&mync=R&uid=swg21460595&myns=swgimgmt

TM1 Cube Dependency

Cube dependency has become an important aspect admins should be aware of within their models.
There is a good overview describing what it is and how to cater for it here:

http://publib.boulder.ibm.com/infocenter/ctm1/v9r5m0/index.jsp?topic=/com.ibm.swg.im.cognos.tm1_op.9.5.2.doc/tm1_op_id3833CubeDepenedcy_MN4002.html

Does TM1 support clustering /load balancing?

Interesting link here about whether TM1 supports clustering/load balancing:

https://www-304.ibm.com/support/docview.wss?mynp=OCSS9RXT&mync=R&uid=swg21420705&myns=swgimgmt

Answer

None of the shipped versions of the TM1 software support load balancing or clustering. The TM1 software does not possess any integrated load balancing or cluster functionality. Nor does it support and use load balancing or cluster standards and interfaces used by some external load balancing or clustering software.

For instance:
Cluster-aware software run by WCS (Windows Cluster Services) can execute a hot failover. If the master node of the WCS cluster fails, its slave node becomes the new master node. Cluster-aware software can directly switch from the failed old master node to the new master node. Due to the hot failover, from a user point of view nothing has happened, the user sessions survive the hot failover, the users are not forced to log on again after the switch of the cluster-aware application from the old failed master node to the new running master node.

But the TM1 Server software is not cluster-aware, thus it cannot perform a hot failover. In contrast to a cluster-aware application a non cluster-aware application must be started on the new master node. And because the user sessions have been terminated by the failure of the old master node, after the non cluster-aware application is up and running on the new master node, the users do have to log on again.

TM1 952 - New Rule and TI functions

The following link outlines the new rule and TI functions available in TM1 952.
The ability to do Averages in rules is something very exciting

TM1web export to pdf not working

Sometimes you may receive an error when installing TM1 that the pdfcamp did not install correctly or on an existing TM1web or contributor application you may find the "Export to PDF" button is not working.

You should then test this by going into the TM1\Bin\PDFCamp directory and running the appropriate setup.exe (32 bit or 64 bit depending on your OS).

If you receive a message such as:

"enummonitors_failed error (1722)"

then it's likely the your print spooler service has not been started and you need to manually start it as per the instructions here:

http://www-01.ibm.com/support/docview.wss?uid=swg21439516&myns=swgimgmt&mynp=OCSS9RXT&mync=R

I think this same issue can happen when installing Cognos Express too

Rename TM1 Elements

One of the biggest time savers I have used recently is the SwapAliasWithPrincipalName Turbo Integrator command.


It isn't well documented but allows you to swap an alias with the element name, thus allowing you to rename TM1 elements.


binhints.com has an excellent tutorial here:


http://www.bihints.com/renaming_elements

Automaticalling Deleting Log Files

2 great posts here on olapforums for how to automatically delete old log files in TM1 or Cognos Express:

http://forums.olapforums.com/viewtopic.php?f=3&t=4108

IBM Cognos Solution Implementation Methodology

If you are about to implement a new cognos product or a new project within an existing install then its worth checking out the CSIM roadmaps available:

 http://www-01.ibm.com/software/data/cognos/services/implementation-roadmaps.html


There are roadmaps for Cognos BI and TM1 available 

TM1 TI process SQL preview killing

If your TI process datasource consists of a sql statement that takes a long time to run then it will become annoying when you want to make changes to that TI and the sql executes every time you open the TI process.

There is though a simple workaround.

At the end of your sql add the to your WHERE condition “1” =’?pCondition?’

 Then in your Prolog tab at the very start add a line  pCondition = '1';

 By having the sql where clause not being populated until run time (execution of the TI) it means your TI will now open without running the SQL statement

Set sort order in tree prompts based on TM1 datasource

Nice little article available here on how to set a customized sort order in hierarchy prompts based off a TM1 datasource in report studio:

https://www-304.ibm.com/support/docview.wss?mynp=OCSS9RTN&mync=R&uid=swg21457443&myns=swgimgmt

HR Cubes

When doing detailed bottom up budgeting, people like to be able to budget for staff individually per department. There is often a difference between the level of departments that a person can see for GL purposes and what they can see for HR budgeting.

e.g. an accountant may be able to see the whole department structure for GL but should only budget for one department when it comes to budgeting salary costs per person.

I find the best architecture is to build the HR cube with a separate dimension called HR_Department. This dimension is a direct replica of the normal Department dimension but it can have entirely separate security applied to it

Now its just a case of creating some extra groups to handle clients access to the HR_Departments.

Automatically load TM1

If you want to ensure that when a user loads an Excel report that TM1 is loaded, you can add the tm1p.xla as a reference to that file.
That way Excel will always check TM1 is loaded.
In the VBA go Tools > References > Browse
and browse to where tm1p.xla is installed.
This will then appear as a reference in your project.
Save the file, close down excel and open that file.
The file will load tm1 automatically.