Master data in process chain

I have to create pocess chain for updating master data for following objects:
0Vendor - full load
0GL_account - Delta
0pur_group - Delta
any guideline how i will proceed?

SAP BI
Usually we create seperate process chains for Delta and Full loads. Since you have only three objects you could create just one process chain for all three. First create the start process and load all these three objects in parallel and one after other depends on the volume of records.
Hope this helps
Thnaks
sat

Similar Messages

Loading master data through process chain

Hi Gurus
I am designing the process chain , I know that I have to start by loading master data first for that I am planning to load attribute first and then to load text and then Hierarchy. Is that the right sequence?
Also how can I make sure that once the master data is loaded , the transaction data loading start automatically?
I am planning to put attribute chage run as last process in every master data loading . Is that correct?
Also i would appreciate if you can tell me the very first step I should follow to start my design, I have already identified the master data and transaction data and their respective dependancies.what would be my next step?
Thanks for your help in advance.
Kris

Hi Kris,
Make two seperate Process Chains for Master Data and Transaction Data
For Master Data:
Your sequence is absolutly correct, Identify what all Master Data Objects and make a list of InfoPackages and Start creating Process Chain as
Start --> Attribute --> Text --> Hier --> Attr Change Run
For Transactional Data:
Identify the Data Targets and the Sequence of loading, then create the Process Chain
NOTE: Always Master Data should be loaded before Transaction Data
After Creating different Chains for Master Data and Transaction Data, Make a one Super Chain called Meta Chain where you can put all these process chains,
Meta Chain --> Process Chain 1( Master Data Chain) --> Process Chain 2( Transaction Data)
also check these threads for process chain creation
Re: Process Chain : Master Data
Re: process chains ?
Hope this helps,
Sudhakar.

Delete Master Data using Process Chain?

Hi Friends,
How do I delete Master Data using Process Chain?
Thanks,
Pradeep

Hi Bhanu,
I am not sure why we would want to delete MD. But that is the task given to me. I actually know it is not possible. But is there a way to delete MD using chains?
Thanks,
Pradeep

How to delete Master data using Process Chain

I need to delete data in respective data targets such as Infocubes, DSOs & Master data Infoobjects. It was fine with Infocubes & DSOs using respective process type. But the issue is with Master data Infoobjects.
Is there any process type by using which I can delete data in Master data Infoobjects using process chain?

Murali,
Master data object process type is not there because in BW, master data is conformed or is shared by other infoproviders. Due to this dependency, master data loading is a bit more complex. You need to drop all the data from all the infproviders which are using the master data you want to delete.
-Saket

Issue while loading Master Data through Process Chain in Production

Hi All,
We are getting an error in Process chain while loading Master Data
Non-updated Idocs found in Source System
Diagnosis
IDocs were found in the ALE inbox for Source System that are not updated.
Processing is overdue.
Error correction:
Attempt to process the IDocs manually. You can process the IDocs manually using the Wizard or by selecting the IDocs with incorrect status and processing them manually.
I had checked the PSA also but I could not find any record and the strange thing is, Job itself is not getting scheduled. Can any one help me out in order to resolve this issue.
Regards
Bhanumathi

Hi
This problem is not related to Process chain..
u can try this..
In RSMO, select the particular load you want to monitor.
In the menu bar, Environment >>> Transact. RFC >>> Select whichever is required, BW or Source System.
In the next screen you can select the Execute button and the IDOCS will be displayed.
Check Note 561880 - Requests hang because IDocs are not processed.
OR
Transact RFC - status running Yellow for long time (Transact RFC will be enabled in Status tab in RSMO).
Step 1: Goto Details, Status get the IDoc number,and go to BD87 in R/3,place the cursor in the RED IDoc entroes in tRFC
queue thats under outbound processing and click on display the IDOC which is on the menu bar.
Step 2: In the next screen click on Display tRFC calls (will take you to SM58 particular TRFC call)
place the cursor on the particular Transaction ID and go to EDIT in the menu bar --> press 'Execute LUW'
(Display tRFC calls (will take you to SM58 particular TRFC call) ---> select the TrasnID ---> EDIT ---> Execute LUW)
Rather than going to SM58 and executing LUW directly it is safer to go through BD87 giving the IDOC name as it will take you
to the particular TRFC request for that Idoc.
OR
Go into the JOB Overview of the Load there you should be able to find the Data Package ID.
(For this in RSMO Screen> Environment> there is a option for Job overview.)
This Data Package TID is Transaction ID in SM58.
OR
SM58 > Give * / user name or background (Aleremote) user name and execute.It will show you all the pending TRFC with
Transaction ID.
In the Status Text column you can see two status
Transation Recorded and Transaction Executing
Don't disturb, if the status is second one Transaction Executing. If the status is first one (Transation Recorded) manually
execute the "Execute LUWs"
OR
Directly go to SM58 > Give * / user name or background (Aleremote) user name and execute. It will show TRFCs to be executed
for that user. Find the particular TRFC (SM37 > Req name > TID from data packet with sysfail).select the TrasnID (SM58) --->
EDIT ---> Execute LUW
(from JituK)
Hope it helps
Darshan

Initial full load of Master data using process chain

Hi All,
Could you please help me regarding, initial master data load to characteristics with attributes and text. I need to load master data to 23 info objects, by using process chain can I do full load of master data to all info objects at a time. And one more doubt is, as per my knowledge we can't maintain more than one variant in an info package, is that right ? or we can ?
Means Start Variant -> Info Package (0Customer_Text, 0Customer_Attr,0BILL_TYPE_TEXT, BILL_CAT_TEXT) -> DTP ( ", ", ", ") -> ACR.
Your Help will be appreciated.
Thanks & Regards
Sunil

Hi,
"I need to load master data to 23 info objects, by using process chain can I do full load of master data to all info objects at a time."
if there is no dependency between attributes then you add you can create process chains and trigger them at a time. No issues.
we can't maintain more than one variant in an info package, is that right ? or we can ?
With one info pack you can't load data to all 23 psa. because each data source have own psa. you need to sue 23 info packs.
in general start variant--> info pack --> dtp (assuming as your bw 7.x)---> attribute change run.
like that you need to create 23 chains
Or create one two big chains.
one is for attribute and another for text.
In attribute
start varaint--> info pack(info bject 1)--DTP(infoobject 1))--> info pack(infoo bject 2)-->dtp(infoobject 2).
Like that way you can create in series and parallel chains to load attributes data into info objects. at end you add change run for 6 info objects each. SAme you can do for text loads also.
Thanks

How to delete master data via process chain

Hi All,
I have to delete master data before loading it again via a process chain.
In the process chain I chose the process " complete deletion of the data target content" but here I did not find infoobject as the data target option to delete the data.
Please suggest how to solve this.
Thanks,
Harini

Hi Harini,
P_CHABNM - It is for the reference characteristic.
I think if you dont have a ref characteristic then give the info-object name.
This is what I fetched fom help.
Reference characteristic
The reference characteristic has the technical properties of a characteristic such as data type and length, as well as the master data (attributes, texts and hierarchies). The characteristic itself also has buinesses semantics. Several characteristics can refer to the same reference characteristic. Such characteristics automatically have the same technical properties and master data.
Example:
The characteristics sender cost center and receiver cost center use the reference characteristiccost center and with it the same values and the same texts.
Bye
Dinesh

Usage of IO (read master data) in Process Chain

Hi experts,
I have a little bit complicated the task.
i have some InfoObject ( as input) and i need to find out the list of Process Chains which read data from this IO, directly by using update rules/transformations or by using function modules.
It means to check all steps in Process Chain and check all routines, selection conditions, update rules, transformations, functions ( function can call next function) , ... and find out if we read informations from IO (master data). I think it is complicated.
Maybe next solution - is existing any log of Process Chain which contains informations about "select" over IO (master data) ?
Thanks in advance
Martin

There is something, what can make your work easier
1. Go to table RSAABAP and put in field LINE value 'yourInfoobject'
you will get list of all places, where this IO is read
2. Go to table RSTRANSTEPROUT and paste all CODEID from previous table, and you will get list of all transformations.
Having transformation you can find all DTPs and then process chains
Regards,
DL

Deleting Master Data via Process Chain

I am in the midst of creating a process chain and am able to create process for deleting ODS Contents, but am not able to do the same for Master Data. Is this possible?

Firstly, why do you want to delete master data.
You cannot delete master if the same is used across other transaction and master ( attr )data.
You may use :
Prg - > RSDMD_DEL_BACKGROUND
FM - >RSDMD_DEL_MASTER_ DATA.
To my knwldge you cannot delete MD using process chains.
You can delete data in PSA periodically..is that you are looking for ?

Loading master data via process chain

hi experts,
I have to load data in infoObject via process chain, my infopackage has update mode as full update. Now my question is do I have to delete the previously added data from infoObject and then run this full update infopackage in process chain or system will only fetch the newly added records from source.
regards,
ray

You can not delete the data from master data once the transcational data is posted against the master data value. Attribute value can be deleted, but you can not delete the value of the main characteristics from master data table.
Moreover, you dont need to delete the data when loading master data. Master data always overwritten with the new reocrds.
As described, first load the full load and then on daily basis you can load the delta.
DO NOT FOGET TO SCHEDULE ATTRIBUTE CHANGE RUN to activate the newly loaded data into master data.
Make sure that your master data is not enhanced, otherwise any change in the enhanced field will not generate delta record. If this is the case, then load in full mode everyday.
- Danny

Activation of Master Data In process

Hello,
I am creating a process chain for master data load. I want to know what is the procedure to activate master data in process chain .
Regards
Baljit Singh

Hi Baljit
After master data infopackages in the end of process chain, please keep a process of type Attribute Change Run. This will take care of it.
In tcode RSPC-> Process chain-> Maintenance screen-> click on process types icon-> here select attribute change run under "Other BW Processes"-> drag & drop process "attribute change run".
Hope it helps
Regards
Pradip

How to design data load process chain?

Hello,
I am designing data load process chains for the first time and would like to get some general information on best practicies in that area.
My situation is as follows:
I have 3 source systems (R3 and two for which I use flat files).
How do you suggest, should I define one big chain for all my loading process (I have about 20 InfoSources) or define a few shorter e.g.
1. Master data R3
2. Master data flat file system 1
3. Master data flat file system 2
4. Transaction data R3
5. Transaction data file sys 1
... and execute one after another succesful end?
Could you also suggest me any links or manuals on that topic?
Thank you
Andrzej

Andrzej,
My advise is to make separate chains for master & transaction data (always load in this order!) and afterwards make a 'master chain' where you insert these 2 chains one after the other (so: Start process -> Master data chain -> Transaction data chain).
Regarding the separate chains; paralellize as much as possible (if functionally allowed). Normally, the number of parallel ('vertical') chains equals the nr of CPU's available (check with basis-person).
Hope this provides you with enough info to start off with!
Regards,
Marco

Master data Reorganization Process for a single InfoObject

Hallo Experts,
mayby someone else had this problem bevor us?
SAP-BWProducion: Master data Reorganization Process for a single InfoObject (selfmade), other InfoObjects worked fine.
SAP BW 7.0 SP20 or 21
This element has daily attributes and the Q-Table has actual 180Mio entries. We startet reorganisation with 250Mio. entries.
Because of the daily loading, we cancel the batchprocess in the evening and started it again in the morning. This was fine for about 5 days. We elliminate 80Mio entries.
Our destination would be 80Mio entries left.
Whenever we run this feature now - in a process chain - no entries were deleted after 30 hours working, and there were no counting-stastics in SM50
     Sequentielles Lesen
     Insert
     Update
     Delete
The five days before, we saw Mio's of entries in this statistics.
We have checked the Q-table and there exist a number of records which have the same attribute values across several records and the date from - date to values of these records are in a row.
RSRV is fine and we checked note 1234411 Master data reorganization - Performance improvement.
There are no other notes helping us.
Please give us a solution-idea.
Thanks
Santra

Hi Santra,
Could you please provide the solution for this?
Thanks
Vinod

Deletion of changelog data using process chain

Hi All,
I have a requirement where i have to delete the changelog data for particular set of data using process chain.we have an option of deleting the entire data from changelog but how to delete only a particular set of data?????

If you are on BI 7 than,
Choose "Deletion of Requests from the Change Log" from process category Further BI Processes by double-clicking.
Create a new process variant and in the variant maintenance screen select Type of Object as Change Log Table
Under Name of object select DSO name
Specify the requests that need to be deleted by determining the days or dates. You also have the option to specify whether you only want to delete successfully updated requests, and/or only incorrect requests that are no longer updated in an InfoProvider.

Time dep master data reorganization / process seems to do nothing

Hi,
Info:
I created a process chain just to reorganize time dep master data for 1 InfoObject (0UCINSTALLA). In this process variant I choose only to process time dep attributes - not text or time independent tables.
The InfoObject is flagged for change run (have som entries in M version)
The InfoObject with the properties:
/BI0/QUCINSTALLA = 9 393 539 rec (4 162 in M version) (same goes for Y-table). Those tables have just a few (3-4) fields.
The InfoObject 0UCINSTALLA exists in a few aggregates, but none of the time dep attributes does.
The system have these components:
SAP_ABA     700     0013
SAP_BASIS     700     0013
PI_BASIS     2005_1_700     0013
ST-PI     2005_1_700     0005
SAP_BW     700     0015
BI_CONT     703     0005
Running on Linux / Oracle 10.2.0.2.0
Problem:
The background job is started via process chain but checking the
process it seems to do nothing. I ran it for 3 days over the weekend, it doesn't time out or give any ABAP dumps or entries in SM21.
Checking the statement for the process in ST04 gives me:
SELECT
FROM
"/BI0/QUCINSTALLA" T_00
WHERE
EXISTS ( SELECT T_100 . "UCINSTALLA" FROM "/BI0/QUCINSTALLA" T_100
WHERE T_100 . "UCINSTALLA" = T_00 . "UCINSTALLA"
GROUP BY
T_100 . "UCINSTALLA" , T_100 . "OBJVERS" , T_100 . "CHANGED" ,
T_100 . "IND_CODE" , T_100 . "UCBILL_CLA" , T_100 . "UCRATE_CAT" ,
T_100 . "IND_NUMSYS" , T_100 . "IND_SEC"
HAVING
COUNT(*) >= :A0 )
ORDER BY
T_00 . "UCINSTALLA" , T_00 . "OBJVERS" , T_00 . "DATETO"
The Explain gives me:
SELECT STATEMENT ( Estimated Costs = 65 933 456 , Estimated #Rows = 0 )
       7 SORT ORDER BY
         ( Estim. Costs = 65 933 456 , Estim. #Rows = 4 )
         Estim. CPU-Costs = 101 899 203 438 143 Estim. IO-Costs = 56 497 939
           6 FILTER
             Filter Predicates
               1 TABLE ACCESS FULL /BI0/QUCINSTALLA
                 ( Estim. Costs = 20 562 , Estim. #Rows = 9 413 200 )
                 Estim. CPU-Costs = 3 825 943 517 Estim. IO-Costs = 20 208
               5 FILTER
                 Filter Predicates
                   4 HASH GROUP BY
                     ( Estim. Costs = 7 , Estim. #Rows = 1 )
                     Estim. CPU-Costs = 10 845 274 Estim. IO-Costs = 6
                       3 TABLE ACCESS BY INDEX ROWID /BI0/QUCINSTALLA
                         ( Estim. Costs = 6 , Estim. #Rows = 4 )
                         Estim. CPU-Costs = 45 379 Estim. IO-Costs = 6
                           2 INDEX RANGE SCAN /BI0/QUCINSTALLA~0
                             ( Estim. Costs = 3 , Estim. #Rows = 4 )
                             Search Columns: 1
                             Estim. CPU-Costs = 22 364 Estim. IO-Costs = 3
                             Access Predicates
Anybody have any idea what the problem could be - I have no idea how long a reorganization for the amount of records should take. But the process seems to do nothing.
Thanks,
Johan

A little update:
I could see that the background process (which seemed to do nothing) ran a oracle process and on the DB server instance this was no.1 in top cpu. So - something is happening. The DB only has 2 cpus. Is there any way to run this job with more then 1 process?

Master data in process chain

Similar Messages

Maybe you are looking for