Hello,

I need help understanding how to address the following error: (see image also)
Error: Prolog procedure line (21): Dimension element "2025-01" not found in dimension "Time" 

I think it is triggered because my time dimension format is MonthFiscalYear  (see image also)

but the current Process Variable Formula is as follows:
vTime=subst(Today(1),1,5)|subst(Today(1),6,2);
so it produces 2025-01.  (see image also)

To fix this, do I change the formula to produce something compatible to my time dimension so TM1 recognizes it? If so, I have edited the variable formula as follows...vTime=subst(Today(1),6,2)|'-'|subst(Today(1),1,4); to return 01-2025

Or, am I completely missing the mark and need to focus on an issue somewhere else

Thank you, in advance!

------------------------------
Lakeisha Warren
------------------------------

Hi Lakesha,

In short, open the }ElementAttributes_Time cube (you may need to select View > Display Control Objects in Architect to see this cube) and check if any attributes are missing. The indication is that the attribute "YYYY-MM" has not been populated for FY25 periods. Once these are populated then your system should operate as has been setup and the TI process should work.

To get a better feel about how this works read/ research alias attributes and how they differ from text and numeric attributes. 

Another is to investigate what the TODAY function returns and how SUBST works, not significant in this example (as it works) but the vTime variable could be updated to a single SUBST

------------------------------
Edward Stuart
Solutions Director
Spitfire Analytics Limited
Manchester
------------------------------

Hi Edward,

I appreciate the response. After I submitted this posting, I did more research and determined the cause of the error log is due to wrong formatting in a dynamic subset. I made another post about it but, in short, all of 2025 under the alias should be formatted as YYYY-MM but it is currently listed as Jan 2025, Feb 2025 and so on. I need change it to 2025-01 and 2025-02...

I don't know how to make that change and I get a warning message that the subset was configured with expression and may become static if changed, so I know I need make the changes a specific way. 

------------------------------
Lakeisha Warren
------------------------------

Original Message:
Sent: Thu January 09, 2025 10:34 AM
From: Edward Stuart
Subject: Error Log Issue

Hi Lakesha,

In short, open the }ElementAttributes_Time cube (you may need to select View > Display Control Objects in Architect to see this cube) and check if any attributes are missing. The indication is that the attribute "YYYY-MM" has not been populated for FY25 periods. Once these are populated then your system should operate as has been setup and the TI process should work.

To get a better feel about how this works read/ research alias attributes and how they differ from text and numeric attributes. 

Another is to investigate what the TODAY function returns and how SUBST works, not significant in this example (as it works) but the vTime variable could be updated to a single SUBST

------------------------------
Edward Stuart
Solutions Director
Spitfire Analytics Limited
Manchester
------------------------------

If you want to transform the data inside the TI process

vTime = TIMST( Now(), '\M \Y' );         will give you JAN 2025  and that should work to load to Jan 2025

vTime = CAPIT(LOWER(TIMST( Now(), '\M' ))) | ' ' | TIMST( Now(), '\Y' );      will give you Jan 2025 if you want it visually exact too, though case sensitivity is not an issue.

Now() returns a numeric serial date and TIMST casts it as strings. TODAY(1) generates a string of date components that are numbers so you get 01 2025

------------------------------
Walter Coffen
Technology Manager
QueBIT Consulting, LLC
------------------------------

Hi Walter,

I appreciate the response and I am sure that formula will be needed one day.

After I submitted this posting, I did more research and determined the cause of the error log is due to wrong formatting in a dynamic subset. I made another post about it but, in short, all of 2025 under the alias should be formatted as YYYY-MM but it is currently listed as Jan 2025, Feb 2025 and so on. I need change it to 2025-01 and 2025-02...

I don't know how to make that change and I get a warning message that the subset was configured with expression and may become static if changed, so I know I need make the changes a specific way. 

------------------------------
Lakeisha Warren
------------------------------

Original Message:
Sent: Thu January 09, 2025 10:56 AM
From: Walter Coffen
Subject: Error Log Issue

If you want to transform the data inside the TI process

vTime = TIMST( Now(), '\M \Y' );         will give you JAN 2025  and that should work to load to Jan 2025

vTime = CAPIT(LOWER(TIMST( Now(), '\M' ))) | ' ' | TIMST( Now(), '\Y' );      will give you Jan 2025 if you want it visually exact too, though case sensitivity is not an issue.

Now() returns a numeric serial date and TIMST casts it as strings. TODAY(1) generates a string of date components that are numbers so you get 01 2025

------------------------------
Walter Coffen
Technology Manager
QueBIT Consulting, LLC
------------------------------

Hi Lakeisha,

I believe it will be more robust to create a temporary subset in your process instead of using an existing one.

You could use the following function to get it done: SubsetCreateByMDX - IBM Documentation Just make sure that the subsetname does not exists and that you set the temporary flag to 1.

Kind regards,

Luc

------------------------------
Luc Cusiel
------------------------------

Original Message:
Sent: Thu January 09, 2025 11:50 AM
From: Kay Warr
Subject: Error Log Issue

Hi Walter,

I appreciate the response and I am sure that formula will be needed one day.

After I submitted this posting, I did more research and determined the cause of the error log is due to wrong formatting in a dynamic subset. I made another post about it but, in short, all of 2025 under the alias should be formatted as YYYY-MM but it is currently listed as Jan 2025, Feb 2025 and so on. I need change it to 2025-01 and 2025-02...

I don't know how to make that change and I get a warning message that the subset was configured with expression and may become static if changed, so I know I need make the changes a specific way. 

------------------------------
Lakeisha Warren
------------------------------


Original Message:
Sent: Thu January 09, 2025 10:56 AM
From: Walter Coffen
Subject: Error Log Issue

If you want to transform the data inside the TI process

vTime = TIMST( Now(), '\M \Y' );         will give you JAN 2025  and that should work to load to Jan 2025

vTime = CAPIT(LOWER(TIMST( Now(), '\M' ))) | ' ' | TIMST( Now(), '\Y' );      will give you Jan 2025 if you want it visually exact too, though case sensitivity is not an issue.

Now() returns a numeric serial date and TIMST casts it as strings. TODAY(1) generates a string of date components that are numbers so you get 01 2025

------------------------------
Walter Coffen
Technology Manager
QueBIT Consulting, LLC

Hi Lakeisha,

Unless you have an explicit element named "2025-01" or an explicit alias to an element in the dimension, named "2025-01", that name won't be recognized.

Regards.

------------------------------
Jose Luis Senas
Data Analyst
Kimberly-Clark Mexico
Mexico City
------------------------------