cancel
Showing results for 
Search instead for 
Did you mean: 

Key Fig mismatch.. source and target

Former Member
0 Kudos

Dear DP experts

A process chain loads a source file to update a key figure in the infocube of the planning area. The value in the target bucket is different from the value in the source. Data is loaded at the most granular level.

What could be reasons for such an error esp. if this has not happened before. What could be a temporary and permanent fix to this.

Guru.

Accepted Solutions (1)

Accepted Solutions (1)

rajkj
Active Contributor
0 Kudos

Hi Guru,

Could you also please answer the following questions to understand the context?

1. Have you verified your InfoCube data (t.code LISTCUBE) with that of your source file?

2. What is the magnitude of data difference between planning area and Info Cube data?

3. Have you attempted to update the data of the key figure in context at any level in the planning area?

4. What are the key figure settings maintained for this key figure in the planning area? t.code /SAPAPO/MSDP_ADMIN.

Thanks,

Rajesh

Former Member
0 Kudos

Thanks Rajesh

1. Have you verified your InfoCube data (t.code LISTCUBE) with that of your source file?

LISTCUBE shows the data for the bucket in question against more than one Request ID's on the same day. I have deleted the Request ID's and reconctructed the same. So I now have data against only one request ID. This happened because data was loaded multiple times in the day because of some errors observed in other steps in process chain. But I couldnt still figure out why esp. when the update mode is a "full update"

2. What is the magnitude of data difference between planning area and Info Cube data?

There is no difference between planning area and Infocube data. The difference is between source file and planning area because of reasons above.The qty in the infocube was sum of qty loaded in each request ID.

3. Have you attempted to update the data of the key figure in context at any level in the planning area?

Yes I did TSCUBE after 1) above  to keep infocube and planning area in sync.

4. What are the key figure settings maintained for this key figure in the planning area? t.code /SAPAPO/MSDP_ADMIN

This is a time series key figure, calc type S, time based disagg p, key fig semnatics 001

I have been able to fix the problem with data target reconstruction but what I could not understand is why this should happen when update mode is full update to data target (via PSA !). My understanding is next request ID should overwrite the values loaded in prev req ID. i.e. IF the same combinaiton is found in data target. I am yet to analyze the content of the source file that was the same in both instances.. once with some error in process chain and once without.

The proces chain error refers prior failure of missing CVC's.

I have tried several other things incl. deletion of indexes in infocube and rebuilding the same.

The problem as such is now fixed

Thanks for your attention.

Guru

rajkj
Active Contributor
0 Kudos

Hi Guru,

Good to know that your issue was resolved. With respect to InfoCube data load, any of your BI team members can guide you with appropriate data upload strategy.

Thanks,

Rajesh

former_member209769
Active Contributor
0 Kudos

Hi Guru,

What is your system version? At least in newer versions of APO using BI 7.0, the following would apply. (Most likely you would be using DTP and transformations in newer versions. I am not fully sure regarding how this looks if you are on older versions)

If your older Requests in the cube are supposed to be deleted, then you would see a step in your process chain whose header would have a "delete" sign and it would read "Request in Infocube".

In a process chain, it is this step that would specify whether deletion of older requests or should happen or not. Here you can also give different kind of options or write code for deletion of only specific requests.

Thanks - Pawan

Former Member
0 Kudos

Thanks Pawan

There is no step in process chain for deleting olders requests. This is probably because this load to target happens once a month "normally". So I manually deleted the same. There are other complicating issues esp. that of datasource valid records etc. I will publish the troubleshoot lessons.

Thanks all for your responses. This is a 4.1 system

Guru

Answers (0)