Short dump error during data extraction in ECC for classification datasource - CONVT_TO_NUMBER

[bhat_vaidya2]

Hi,

I am getting short dump error when running extraction checker with RSA3 .

I found below SAP NOTE with correction but it is no longer valid in our system as we are on a much higher support level.

Does any know how I can resolve the issue or know of any recent sap note.

Bhat

[karthik_vasudevan · Accepted solution]

Hi Bhat

The short dump shows some information about a value 'SP' cannot be interpreted as number which seems to be reasonable.

Would you be able to check the underlying tables for this datasource to see if any issue with the data.

Could you also please check and post the how to correct the error part and information on where the error occurred from your dump analysis page?

If possible, restrict the extract checker by giving some small selection and see if it works, just to narrow down the reason.

Regards

Karthik

[bhat_vaidya2 · Accepted solution]

Hi, I am doing a test case for a particular sales document and getting this error. My datasource is classification datasource 1CL_OLIS003 where I have added a new field. I don't know which table stores the data. I have had a look at sales order table VBAK, the document is present but there is no information related to classification. Thanks, Bhat

[karthik_vasudevan · Accepted solution]

Hi

The below link gives you the information on where the data is being fetched for the datasource.

https://scn.sap.com/thread/1791432

You have mentioned that you have added a new field. which table does that field belong to. Do you have any codes written to get the data populated for the new field?

Was this data extraction working fine before you added the field?

Try giving some other sales document as well, just to check if the problem is for all sales documents. Get in touch with your functional team as well to see if they could help something with the information provided in the link

Regards

Karthik

[bhat_vaidya2 · Accepted solution]

Hi,

This new field is a characteristics created with CT04. Automatically, it creates a new field in the extract structure and text data source for the characteristics. I am new to working with Classification datasource and it is not the same way as other datasource.

I have looked at the function module CTBW_GET_KONFIG_DATA and the tables that I found didn't help much table. In CTBW, the source table is table VBPS and I cannot find this table in ECC.

VBPS TRAN 1CL_OLIS003

I have asked functional team to make the change using as existing document number. By the way, The value "SP" is the unit, this I found in VA03 transaction.

- Bhat

[karthik_vasudevan · Accepted solution]

Hi

In some cases, when the number and unit comes concatenated in the field, you will get this error where you need to split the unit and number.

Once your functional team comes back to you, try to extract a different data and see if it works. Then we could confirm that the issue is with the data.

Please keep the things posted.

In addition, you could also try debugging the extractor. I have seen some posts giving that suggestion. You could debug and find out where that SP comes from and the sales document corresponding to it. It will be better in this way

Regards

Karthik

[bhat_vaidya2 · Accepted solution]

Hi, Thanks for the tip. I have done further analysis and check documents without unit and those I could extract successfully. I will wait for functional team to create a new test case. - Bhat

[karthik_vasudevan · Accepted solution]

Good . Its unit which is the problem then. Atleast we came to this stage.

Once you have got the solution, please update it here so that others will use it.

Regards

Karthik

[bhat_vaidya2 · Accepted solution]

Hi, We finally, found the rootcause. It was because on inconsistency in data types for a particular field that has the unit. The unit field in question does not exist in DEV so it inherited to data type property of characteristic. this field unit has now been created and datasource regenerated. thanks