Strange behaviour of DG filling process

[Former Member]

Hi,

I want to check if you also have the same  DG-filling behavior as I will describe now. First I will describe how we have it configurated: we have an automated process for starting the filling transaction. In case a change in the transport classification takes place of a DG-classification (which are linked to real substances), for each linked real substance the DG-filling report (DGPDATCO) will be submitted separately. Also we don't use change pointers (SUBMAS is not activated).

According to SAP the setting without change pointer and with explicitly selected specific data records can be used for correcting incorrect data. So to my opinion it should work. When checking the changes and the result after filling it is indeed working, but not for all scenarios. When I add or change a transport regulation, after running the DG-filling process,  the DG-masters are created/changed accordingly. When clearing the fields of a specific transport regulation, the relevant DG-master has been removed after DG-filling. However when clearing all transport regulations fields (or even by deleting the transport classification instance), the DG-filling process can't find any relevant change and isn't doing anything. I see the same behavior when doing this manually (instead of the automated process).

I can't find any related topic in OSS. I am wondering if this is a bug in SAP or implicit in the design of the DG-filling process. Can this maybe be solved by using the change pointer approach? Do you see the same behavior in your system?

Regards,

Paul

[christoph_bergemann]

Dear Paul

do you use data model as REAL_SUB = DG_CL_SUB and delete "transport classification instance" on DG_CL_SUB level?

I might imagine that this is not detected (but e.g. if you would remove DG_CL_SUB it could be detected). Never ever had the need to dig such deep in this process as this "maintenance" scenario is not used on our side.

C.B.

PS: may be try it different: do a "experiment" and maintain the data direct on REAL_SUB level (without reference) and delete them afterwards. Here I would assume it might work