Identifying Grouping Factor in Collect iDocs

[srikanth_srinivasan3]

Hello,

Is there a grouping factor or a correlation id for collected iDocs group that is a result of choosing Collect iDocs as the delivery method in Partner Profiles on SAP ECC? I had no luck searching which was perhaps due to incorrect keywords. Any pointers will be very helpful.

Thank you.

[engswee · Accepted solution]

Hi Srikanth

Grouping of IDocs are based on the key fields defined in the Partner Profile - so all the recipient partner fields and the message fields. The IDocs generated for the same values of these key fields are collected.

However, when triggering program RSEOUT00 to send the collected IDocs out, the message code and message function fields are not available. Meaning all collected IDocs for a particular message type (for the selected receiver) are sent out irrespective of message code/function.

Rgds

Eng Swee

[srikanth_srinivasan3 · Accepted solution]

Hello Eng Swee, Thanks for the note. I do know that most of the values of Control Record of iDocs in one collected bundle is same except few such as DOCNUM, SERIALNUM, as they are based on a specific Outbound Message type assigned for a Partner Number. What I am tring to achieve is identify the group based on an iDoc number.

[engswee · Accepted solution]

Hi Srikanth

Can you elaborate why you are trying to achieve identifying group based on IDoc number? Perhaps if you share the approach for the design and your intentions, then it might give a clearer picture.

As of now, I can't really picture what you are trying to do and as such can't really provide any further pointers.

Rgds

Eng Swee

[srikanth_srinivasan3 · Accepted solution]

Hello Eng Swee Yeoh, Thanks again. Here is some background information:

We receive a message with one or more iDocs from SAP ECC. Each iDoc contains a segment whose occurence is unbounded. Because the target system couldnt seem to handle such large message, we are splitting the message by segment for reasons that can't be explained here. This means that each split message will have an entry in our custom monitoring tool. Because there could be several entries in the monitoring tool for one business message, I was wondering if there is a standard parameter that I could use to correlate the split messages to the parent message in the monitoring tool. Note that, I can always create a correlation factor between the split messages and the parent message by assigning the parent message id to each of the split messages as an additional remark. The purpose of this discussion is only to find out, wheather or not a standard SAP-based correlation factor is available.

Note: We are not using PI here. The purpose of raising this question here is to seek advise based on a similar situation others might had seen.

[engswee · Accepted solution]

Hi Srikanth

The requirement is much clearer now and it was important to know that you explicitly mention that you are not using PI, because the general assumption in this space would be PI is used.

Back to your question, AFAIK, there is no standard way for correlation for a collected bunch of IDocs transmitted outbound. If the outbound transmission is via a tRFC port (which is the most common), there would be a tRFC transaction ID involved. However this ID is not stored anyway in the IDoc control record. I'm not sure what type of tool you have but if it is possible to capture the tRFC ID during transmission, then you can correlate all the IDocs in the same tRFC. Otherwise, I don't think you can correlate using any values in the IDoc control record.

Rgds

Eng Swee

[srikanth_srinivasan3 · Accepted solution]

Hello Eng Swee, Thanks for the note. Sadly, the transaction id isnt recorded for use. I'll have to implement a customized correlation mechanism. I appreciate your time and help.

Regards

[naveen_chichili]

Hi Srikanth,

This would help you just go through.

http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/30ea2fdf-f047-2a10-d3a2-955a634bd...

Thanks,

Naveen

[srikanth_srinivasan3]

Hello Naveen, Thanks for the note. I had gone through the resource you've supplied and dint quite serve the purpose. Is there a specific point that you would like to suggest from the article, in pertinance to the question?