Problem with BSEG Cluster Table in Calculation View

[#1]

Hi,

When I try to create a calculation view type SCRIPT to do a reading on BSEG cluster table, the system answers me that the table does not exist in the schema.

My goal is to make a combined view of multiple tables where the main join is BKPF-> BSEG. As I do?.

Best regards,

/********* Begin Procedure Script ************/

BEGIN

  var_out = CE_COLUMN_TABLE("SAPHAD"."BSEG",

              ["MANDT", "BUKRS", "BELNR", "GJAHR", "BUZEI"]);

END /********* End Procedure Script ************/

Message :

  Internal deployment of object failed;Repository: Encountered an error in repository runtime extension;Internal Error:Deploy Calculation View: SQL: transaction rolled back by an internal error: invalid table name:  Could not find table/view BSEG in schema SAPHAD

[Former Member]

Hi,

Unable to read the BSEG table directly. The traditional solution is to take all secundary indices, do JOINS (BSIS-BSAS, BSIK-BSAK, BSAD BSID-) and then a final UNION to have all the information in the financial documents; but some types of position is left out, so it's not the solution.

Currently I have an OSS message processing.

Best regards,

[Former Member]

not sure how it would work in HANA, but in ECC you can't join BKPF on BSEG as it is a cluster table. i'm not aware of the cluster table format in HANA.

[Former Member]

Hi Gregory,

Exactly, in SAP ECC is not possible, this is handled at the level of ABAP internal tables. In the HANA documentation mentions that all tables CLUSTER / POOL were going to be able to use as transparent tables, but apparently not. What alternative do I have to be able to view data I need, a mixture of information from these two tables, plus additional ones.

Best regards,

[Ravi_Channe]

Hi Johnny,

Couple of things, CE_COLUMN_TABLE expects a column table as input. Hence your table has to be a column table and should be available in the query I posted earlier.

I am not very sure about the cluster table to transparent table conversion, but can you please check if that conversion is supposed to happen automatically and the note http://service.sap.com/sap/support/notes/1798054 is applicable to you (or something similar to this).

Regards,

Ravi

[lbreddemann]

Hi Johnny,

the most convenient way to consume the data from CLUSTER/POOL tables outside of the NetWeaver stack is to use the SLT replication for the data transfer.

- Lars

[Former Member]

Hi Lars,

Is there any performance issues/challenges in replicating Cluster Tables in SLT? It looks like SLT could solve reporting issues on CLUSTER tables especially CDPOS/CDHDR which are crucial in change log reporting for logistic data.

Regards,

Dominik

[lbreddemann]

Sorry,

can you please be more specific?

What kind of performance are you asking about?

Cluster/Pool tables will be decoded during the reading step of the replication and stored as transparent tables in SAP HANA.

Not really sure where your point regarding "change log" is going ...

- Lars

[Former Member]

Hi,

I mean change log as table(s) - CDPOS/CDHDR where you can find history of changed logistic objects (not only, like Purchase Order Item). Reports/Queries based on cluster tables are not performing well.

If you want to track changes in quantity or amount, CDPOS/CDHDR is only place to find it.

When I was talking about performance I was thinking how SLT is managing cluster tables - replication process from ERP (non-hana) to HANA (or BW on HANA). Is performance different from standard table?

Dominik

[lbreddemann]

The replication performance should not be noticeable be influenced by the declustering process.

It's the exact same thing that happens, when you read the table from ABAP.

- Lars

[Former Member]

thank you for info.

[Ravi_Channe]

Hi Johnny,

Can you please confirm if the column table exists in the database. You can check that in SQL editor with the following query.

select * from m_cs_tables where schema_name = 'SAPHAD' and table_name = 'BSEG';

Regards,

Ravi

[Former Member]

Hi Ravindra,

The table exists in the schema SAPHAD, but CLUSTER table type, not transparent type.

Best regards,