PAKET_TOO_BIG error after increasing the PACKET SIZE IN DTP

[former_member596753]

Dear All,

I am facing one problem ,while transferring the data to ODS ,I have maintain one End routine between ODS and PSA .the end routine compare the refreance doc on each line and and put DEBTORS in each row

For example

DATA IN PSA

Doc Number      Customer      Ref Doc

100000                100023         555555

100001                  ---------         555555

100002                  -------           555555

After END routine

Doc Number      Customer      Ref Doc

100000                100023         555555

100001                100023         555555

100002                100023         555555

Now My problem is that ,somtime Document number 100001 and 100002 is coming in different packages so the conversion will not happen,

So i increased the size to 4000000 as the data in PSA is 3600000 but now following error come

Short text of error message:

Unexpected program status

Long text of error message:

Technical information about the message:

Message class....... "RSBK"

Number.............. 899

Variable 1.......... "PAKET_TOO_BIG"

Variable 2.......... " "

Variable 3.......... " "

Variable 4.......... " "

[Former Member]

Just a note that SAP removed this limit in the following note:

SAP Note 2055056 - PAKET_TOO_BIG when reading from PSA

[Former Member]

You should use the semantic group functionality also found in the DTP. It should ensure that records with the same values in the semantic group field (reference doc) are kept together.

[Former Member]

Thanks man,,

But i cant do use the semantic ,as it may reside in another package,but any how the pbm is sorted out ,by filtering the value

[jagadeesh_mandepudi2]

Hi

Try to increase background processes and for testing you trigger only that loading so that all the processes will used by current job.

Also keep DTP parellel jobs to 1 so that will not raise any issue while loading to DSO.

Jagadeesh

[Former Member]

Try the semantic grouping! It will sort your data package by the semantic key so that records with the same key all goes into the DTP package. It's the only "real" solution all others are work arounds.