cancel
Showing results for 
Search instead for 
Did you mean: 

Delete Master data from SID Table

chanda
Contributor
0 Kudos

Dear Guru's,

scenario : I loaded some master data to an existing info object for a new comp code( CC01) . I found the master data is not correct . I want to delete it .

1) I tried to delete using se38 --> RSDMDD_DELETE_BATCH  & also from Maintain masterdata

No Master data deleted

2) I went to SLG1 &  found that log like below ;

Master data SNBCB1000135 (SID                                          21554042) is used in /BIC/DYYY_C_032; is not deleted

Also some SID values found in some cube for CC01

3) I checked in SID Table & found an entry with the Comp Code CC01 & SID.

My Question is :

1) Can i delete the data from SID ,if so how to do selective deletion for only CC01 data , i found in some posts that deleting SID data might give problems when i reload the same master data.

2) How to delete the SID from cube .

Thank you for your time.

Regards,

Sudhir.

Accepted Solutions (1)

Accepted Solutions (1)

former_member976
Active Contributor

Hi janardhan,

) RSA1 --> find your Info object --> right click --> maintain master data

2) Select rows you want to delete and click 'Delete' icon. Then SAVE.

3) If system prompt you message "some master data cannot be deleted" ... continue with this message.

4) Then goto tcode SLG1 --> enter Object: RSDMD and Subobject: MD_DEL and User: your userID and press F8.

5) In the next screen Expand node and double click 'Problem class additional information'

6) This will give you which ever objects using SIDs of this master data.

7) You should delete this SID according to that list.

😎 Then U should delete particular data from DSO/Cubes.

Once data deleted from cube, then

9) Use Tcode RSRV -> Tests in Transaction RSRV -> All Elementary Tests -> Transaction Data -> Entries Not Used in the dimension of an InfoCube (Double Click)

10) From the right side window, expand “Entries Not Used in the Dimension of an InfoCube” Node & Enter InfoCube (like 0SD_C01), click “Transfer”

11) Now, click “Execute” (Toolbar)… and the results displayed in the right side window..

12) Now, click “Correct error” (Toolbar)

Hope it will help.

Thanks,

Phani.

chanda
Contributor
0 Kudos

Hi Phani,

Thanks for the detailed steps.

Actually i am struck at step - 7 .  7) You should delete this SID according to that list.

what are the affects of its deletion , as some posts say that there might be problem in loading the same master data again.

If the SID table data can be deleted then , can we do selective deletion from SID table and how?

Also how the delete the SID data from the cube ? ( as shown in the screen ).

Regards,

Sudhir

RamanKorrapati
Active Contributor
0 Kudos

Hi,

If your company code data was wrong, then why you want delete?

Just reload whole master data to company code. it overwrites the old values.

Master data deletion is not easy task, if those master data objects have been used in another targets it won't allow you delete.

Selective deletion not possible for master data objects.

Thanks

former_member976
Active Contributor
0 Kudos

Hi Janardhan,

Please fallow the Raman Steps, it is useful for you.

deleting of master data not easy task, before loading the TD should aware about MD-  attr, text, hier.

Thanks.

Phani

chanda
Contributor
0 Kudos

Hi Ram,

It is the master data , which is wrong . I deleted the requests from the attributes ( Manage requests ), but still the data is available .

Even after doing a full load again to the master data it still has the wrong values.

I have deleted the data from all the targets like DSO & Cubes.

After deleting from the targets i tried to delete from the info object using the program RSDMDD_DELETE_BATCH  , but unable to delete as there are some SID existing on the infoobject & the cube .

Selective deletion of master data is possible using the program RSDMDD_DELETE_BATCH & selecting the filter .

Thanks & regards,

Sudhir

RamanKorrapati
Active Contributor
0 Kudos

Hi,

You master data was wrong. fine.

With out deleting we can rectify the master data. if your source have correct entries then just reload the master data to info object. its overwrites the correct values. in that case why you need delete and make a complex issues of sids.

if you have wrong data at cube/dso, do the selective deletion and reload selectively.

Thanks

anshu_lilhori
Active Contributor
0 Kudos

Hi,

You need to delete the data from all targets in which that particular infoobject is being used.

You can check the same with the help of where used list of that infoobject.

Once you are through with it then it will allow to delete the master data and you can do it with SID also as you have deleted all its reference.

Regards,
AL

chanda
Contributor
0 Kudos

Hi Phani,

Yes, i am aware of it, but now the problem is we have flat file coming from source system, and this is uploaded to the info object . A file was given to me and i uploaded, because of which there is huge data loss in transaction data .

upon investigation we found the discrepancy in the master data and hence we want to rectify it.

Thanks & Regards,

chanda.

chanda
Contributor
0 Kudos

Hi AL,

I have deleted the data from the targets. Now how to delete it from the SID table?

below is the concern i have:

according to the link  :

Deleting Attributes and Texts for a Characteristic - Business Intelligence - SAP Library

If you delete the SID table entry for a particular characteristic value, the SID value assigned to the characteristic value is lost. If you reload attributes for this characteristic value later, a new SID value has to be created for the characteristic value. This has a negative effect on the runtime required for loading. In some cases, deleting entries from the SID table can also lead to serious data inconsistencies. This occurs if the list of used SID values generated from the where-used list is not comprehensive.

Thanks & Regards,

Sudhir

chanda
Contributor
0 Kudos

Hi Ram,

As mentioned, i have deleted the new requests of master data from info object attributes.

when i checked the data is not deleted .

I uploaded the new file . But still the wrong values are shown, with new ones.

So then i thought to delete every thing . but struck with SID.

Thanks & regards,

Sudhir

anshu_lilhori
Active Contributor
0 Kudos

Hi ,


Please refer the document master data clean up activity.which explains the complete matrix of sid table for particualr infoobject..

You may find the same in my bookmarks else you can search it as well.

Regards,

AL

RamanKorrapati
Active Contributor
0 Kudos

Hi,

have you tried to deletes sids?

have you got any error? are you trying delete whole sid table data or particular objects data?

if possible try to implement custom abap program and use sid table,delete the selective sids.

if not possible as suggested above and your on bw 7.3 then

delete the whole master data(with client approval) with sid(there is doc on Google written by Krishna Tangudu - BW 7.3 different ways to delete md data) for your info object if client accepts and reload the whole data.

Thanks

former_member976
Active Contributor
0 Kudos

Hi janardhan,

Master data have overwrite function

1.already you loaded master data previously  and you want delete existing master data

2. No need to delete master data

3. now load new file with master data and do activate master data ( common should be one object)

     select master data target -> right click -> active master data.

4. after that cross check some records.

Thanks,

Phani.

former_member185132
Active Contributor
0 Kudos

Hi,


A file was given to me and i uploaded, because of which there is huge data loss in transaction data .

upon investigation we found the discrepancy in the master data and hence we want to rectify it.

This part is not clear. How could a master data load cause any loss in the transaction data?

And even if you have the old (junk) entries along with the new (good) entries in the master data it shouldn't matter as long as the transaction data refers to the new entries. You don't need to delete any master data at all.

Regards,

Suhas

chanda
Contributor
0 Kudos

Hi Suhas,

sorry for reverting late.

I think i did not put it properly .

Please look at  the screen shot , from there i explain my situation

below is the master data in the info object

In transfer routine the TO field is constant --> 31.12.9999. I do not understand how is 05.12.2013 is added.

I deleted the psa and reloaded again, still it shows me the 2 records as above.

I should be seeing only 2nd record.

In my routine to cube i am fetching CAMPAIGN field passing  member card (191 ) & Valid TO is GE 05.12.2013 & Valid FROM is LE 05.12.2013.

with the above logic its fetching 1st entry for which CAMPAIGN is blank .

I am concerned only with values where CAMPAIGN code is not blank. hence my transaction data is affected.

Kindly help to solve this.

Thanks & regards,

Sudhir

former_member976
Active Contributor
0 Kudos

Hi Sudhir,

In transfer routine the TO field is constant --> 31.12.9999. I do not understand how is 05.12.2013 is added.

A. constant value Valid To- 31.12.9999 but showing 05.12.2013 these data format please check the one more time transformation level constant value.

you wrote the routine in cube transformation level and check the one more time select statement.

select * from /BI0/xxxxxx into CORRESPONDING FIELDS OF TABLE
       lt_xxxxx
       FOR ALL ENTRIES IN RESULT_PACKAGE
       WHERE CAMPAIGN = RESULT_PACKAGE-HE_CAMPAIGN
           and DATETO = '99991231'.

try hits code.


Hope it will help.


Thanks,

Phani.

chanda
Contributor
0 Kudos

Thank you for your reply phani,

I did check again --> Date TO is being forced to 31.12.9999.

Is there any means to find out how an entry is created in the info object?

Or is it that if transaction data is created with out master data , the masterdata is created ...

I cannot change the cube logic as it is being used for other comapany codes also.

regards,

Sudhir.

former_member976
Active Contributor
0 Kudos

Hi Sudhir,

I am working on Higher education application, this is student master

please check the records and we are looking only latest record only means 32.12.9999 record


select * from /BI0/xxxxxx into CORRESPONDING FIELDS OF TABLE
       lt_xxxxx
       FOR ALL ENTRIES IN RESULT_PACKAGE
       WHERE CAMPAIGN = RESULT_PACKAGE-HE_CAMPAIGN
           and DATETO = '99991231'.


in your routine put only DATETO = '99991231'. it will work. now you see only latest record in target level


Thanks,

Phani.


chanda
Contributor
0 Kudos

Hi Phani,

Big sorry .. i did not check that the info object has the tick checked for TEXTS ARE TIME DEPENDENT.

Hence a new record is added having the TO data lower than the creation/ modified date of the masterdata entry.

Yes as you mentioned, the change to the code in Cube is inevitable . An entry with DATETO = 99991231 should be picked .

Thanks to all of you .

regards,

sudhir

former_member976
Active Contributor
0 Kudos

Hi Sudhir,

Now you understood problem.

always welcome

Thanks,

Phani.

Answers (1)

Answers (1)

former_member199945
Active Contributor
0 Kudos

Try with this Program & Check any Inconsistency in Sid's For that IOBJ.

RSDMD_CHECKPRG_ALL.

chanda
Contributor
0 Kudos

Hi Seshu,

I tried to run it but taking very long time .. still running.