cancel
Showing results for 
Search instead for 
Did you mean: 

Report on process chains using SEM-BPS

Former Member
0 Kudos

We are looking to develop a tool to:

1. Monitor data loads at the process chain level

2. Within in the same tool, enter error resolution /

action steps for all failed loads

3. Report out % of successful or failed loads, using the data captured by the tool.

We have studied the BW Tables rspcchain, rspcchainattr, rspclogchain, rspcprocesslog, rspclogs, rseventchain and

rspcchaint.

We are considering the following options:

1. Develop an ABAP report and allowing the process chain monitor to manually enter the error resolutions (Perhaps in a Z-Table)

2. Create an InfoCube in BW using the above tables and then write a query, but we can not allow the user to enter error resolutions in the query.

3. Create a transactional infocube in SEM-BPS using the above tables and use the input functionality of tansactional infocube to facilitate the error resolution report entry.

Can anyone suggest any alternate solution, or has anyone encountered similar requirement on project?

One more doubt to check: is there any way to create an infoobject with CHAR type to enter data with 1000 or more characters? Or is there any other alternative solution to it? The doubt is to take care of the text entry for the error resolution.

Milind Vad

Accepted Solutions (0)

Answers (1)

Answers (1)

klaus_werner
Active Participant
0 Kudos

Just some thoughts:

Have you considered using document service in BW or even KM (Portal) technology rather than CHAR 1000+ fields? This would give you more flexibility also on the type of document.

For loading there are monitors available (based on tables like RSMONMESS). If you are interested in loads only (and not other process types), maybe using information from the BW load monitor could be beneficial (like RSMONMESS) rather than the process chain stuff.

Maybe it's even worth having an InfoObjekt with a resolution code that links to a document describing the resolution with a key and whenever there is a new solution available create a new key with a new document. That would be most beneficial for reporting (like how many times has this problem / resolution occurred and the like).

Regards, Klaus

Former Member
0 Kudos

I would create a transactional InfoCube and fill it via ABAP. The ABAP should be triggered by a program in the process chain. You may even want to create your own process type for this because you will want to add it to all your chains. In a web template you can then create documents that are attached to InfoCube values and if you create a Chain number that is filled via number range then you can make it document relevant and have a documentation per process chain.

Best regards

Dirk

Former Member
0 Kudos

Hi Dirk,

Many thanks for your reply. We have been deliberating about the Pro's & Con's of using a transactional InfoCube V/s the traditional Z-table and writing of ABAP report.

We have noted that there will be an extra data staging, where we need to load data in an ODS, before loading it in the transaction infocube, where as, in the case of a pure ABAP solution, we can extract the data in a custom table and run the report on it. Can you advice us on any specific plus'es of using SEM-BPS for this purpose?

Regards,

Milind

Former Member
0 Kudos

There are SEM-BPS APIs, especially API_SEMBPS_SETDATA and API_SEMBPS_POST that allows you to write data directly into a transactional InfoCube. Then you don't need to do any data staging to run the reports, no jobs that run any five minutes or so and you have state-of-the-art reporting without any additional effort.

Best regards

Dirk