Loading of flat file (csv) into PSA – no data loaded

Hi BW-gurus,
We have an issue regarding loading a flat file (csv) into PSA using an infopackage u2013 (BI 7.0)
The infopackage has been used for a while. Prior the consultants with SAP_ALL-profile have run the infopackage. Now we want a few super users to run the infopackage.
We have created a role for the super users, including authorization objects:
Data Warehousing objects: S_RS_ADMWB
Activity: 03, 16, 23, 63, 66
Data Warehousing Workbench obj: INFOAREA, INFOOBJECT, INFOPACKAG, MONITOR, SOURCESYS, WORKBENCH
Data Warehousing Workbench u2013 datasource (version > BW 3.x): S_RS_DS
Activity: All
Datasource: All
Subobject for New DataSource: All
Sourcesystem: FILE
Data Warehousing Workbench u2013 infosource (flex update): S_RS_ISOUR
Activity: Display, Maintain, Request
Application Component: All
InfoSource: All
InfoSource Subobject: All values
As mentioned, the infopackage in question, has been used by consultants with SAP_ALL-profile for some time, and been working just fine. When the super users with the new role are executing the infopackage, the records are found, but not loaded into PSA. The load seems to be stuck, but no error message occurs. The file we are trying to load contains only 15 records.
Details monitor:
Overall status: Missing messages or warnings (yellow)
Requests (messages): Everything ok (green)
-> Data request arranged (green)
-> Confirmed with: OK (green)
Extraction (messages): Errors occurred (yellow)
-> Data request received (green)
-> Data selection scheduled (green)
-> 15 Records sent (0 Records received) (yellow)
-> Data selection ended (green)
Transfer (IDocs and TRFC): Missing messages (yellow)
Processing (data packet): Warnings received (yellow)
-> Data package 1 (? Records): Missing messages (yellow)
     -> Inbound processing (0 records): Missing messages (yellow)
     -> Update PSA (0 Records posted): Missing messages (yellow)
     -> Processing end: Missing messages (yellow)
Have we forgotten something? Any assistance will be highly appreciated!
Cheers,
Anne Therese S. Johannessen

Hi,
Try to use the transaction ST01 to trace the authorization of the upload with the SAP_ALL.
And the enhance your Profile for the super user.
Best regards
Matthias

Similar Messages

Translating Excel query into flat file(.CSV)

Hi,
I have a requirement where i need the Excel Queries to be displayed in Flat file(.CSV) format.
To be more clear i have an excel query but my client requires this query to be displayed in a Flat file format. Can someone explain me how i can acheive this.
Thanks

Hi All,
Thanks for all the information provided by i missed mentioning one point, actually thse queries are to be automatically schedule. the requirement is like this i need to bring to extract data and build queries export it as a .CSV file these queries are not end user queries so they have to be automatically scheduled and will be picked up by an interface so they do some mapping with the .CSV file we provide.
So we decided to do it in Reporting Agent in Excel format and automate it and send the file to a specific location from where the Interface team can pick it. which doesnt need an user to run the query.
so can someone tell me if i can automate a query and as well as have the query in .CSV format?? I hope i was clear
Thanks.

Extract work order data from r/3 system in flat file(csv)and export to BI

Hi,
I am new in interface.
I need to extract data regarding actual cost and quantities of work assigned to Service Providers from SAP system and send it to BI for reporting purposes.
This interface will extract Master data as well as transactional data. Extraction of Master data will be a full extract and that of transactional data will be Delta Extraction.
Custom development will extract the data from different fields and will export it to flat files (CSV format). This program will amalgamate all the flat files created into one big file and export it to BI system.
Export of data to BI system will be done by schedule program Control M, which will export the data from SAP system to BI system in batches. Control M is expected to run daily at night.
Please provide the step-by-step proces to do this.
Thanks
Suchi
Moderator message: anything else? please work yourself first on your requirement.
Edited by: Thomas Zloch on Mar 25, 2011 1:21 PM

Hi Ravi,
you've got to set up the message type MDMRECEIPT for the Idoc distribution from R/3 to XI. Check chapter 5.2 in the IT configuration guide available on <a href="http://service.sap.com/installmdm">MDM Documentation Center</a>. It describes the necessary steps.
BR Michael

Extract data from Flat file CSV to SQL Server 2008 using SSIS 2008 (Data gets corrupted when extraction)

Hi,
I am trying to extract data from multiple CSV files to SQL into a single table. The data type of all the columns in SQL table is nvarchar(MAX). I am able to extract the data from the flat files but some of the data(on extraction) is
corrupt including question marks(?) and other invalid special characters. Also I tried selecting the UTF-8, 65001(Unicode) format but the problem still persists. Also I tried using data converter but no use.
I checked with the data in the flat file but there is no data with question mark(?) or any other special characters.
The separator in the flat file is Comma(,)
Please help.
Thanks in advace.

The source system and application determines the code page and encoding. Is it Windows, Unix, Mainframe or some other type?
Unicode files sometimes begin with a byte order mark (2 bytes) to indicate little or big endian. If you open the file in notepad and then select save as, the encoding in the dialog will show the encoding notepad detected based on the BOM. If
that is ANSI instead of Unicode or UTF-8, you will need to know the code page the source system used when the file was created.
Dan Guzman, SQL Server MVP, http://www.dbdelta.com

Creating flat files (.csv) in user defined directory using PL/SQL

I need to extract data into a flat file (.csv) and store the flat file directly into a directory on local drive as specified by the user at runtime. I do not intend to use the UTL_FILE package. Is there a way put to do this?
Thanks
Amit.

Hi Amit,
Your users have any shared drive which every one can access? If so follow these steps.
1> Allow users to select their destination location to an APEX item.
2> On submit, save the user IP, destination location and user name to a database table.
3> On submit, using JS, open XLS file template (+kept this template file in shared location+) The template should contain all the macros that will do the needful. I mean they should connect to database and should fill it with data and saving it to user specified location etc.
Hope it helps.
Cheers,
Hari

Error while loading table from flat file (.csv)

I have a flat file which i am loading into a Target Table in Oracle Warehouse Builder. It uses SQL Loader Internally to load the data from flat file, I am facing an issue. Please find the following error ( This is an extract from the error log generated)
SQL*Loader-500: Unable to open file (D:\MY CURRENT PROJECTS\GEIP-IHSS-Santa Clara\CDI-OWB\Source_Systems\Acquisition.csv)
SQL*Loader-552: insufficient privilege to open file
SQL*Loader-509: System error: The data is invalid.
SQL*Loader-2026: the load was aborted because SQL Loader cannot continue.
I believe that this is related to SQL * Loader error.
ACtually the flat file resides in my system ( D:\MY CURRENT PROJECTS\GEIP-IHSS-Santa Clara\CDI-OWB\Source_Systems\Acquisition.csv). I am connecting to a oracle server.
Please suggest
Is it required that i need to place the flat file in Oracle Server System ??
Regards,
Ashoka BL

Hi
I am getting an error as well which is similar to that described above except that I get
SQL*Loader-500: Unable to open file (/u21/oracle/owb_staging/WHITEST/source_depot/Durham_Inventory_Labels.csv)
SQL*Loader-553: file not found
SQL*Loader-509: System error: The system cannot find the file specified.
SQL*Loader-2026: the load was aborted because SQL Loader cannot continue.
The difference is that Ashoka was getting
SQL*Loader-552: insufficient privilege to open file
and I get
SQL*Loader-553: file not found
The initial thought is that the file does not exist in the directory specified or I have spelt the filename incorrectly it but this has been checked and double checked. The unix directory also has permission to read and write.
Also in the error message is
Control File: C:\u21\oracle\owb_staging\WHITEST\source_depot\INV_LOAD_LABEL_INVENTORY.ctl
Character Set WE8MSWIN1252 specified for all input.
Data File: /u21/oracle/owb_staging/WHITEST/source_depot/Durham_Inventory_Labels.csv
Bad File: C:\u21\oracle\owb_staging\WHITEST\source_depot\Durham_Inventory_Labels.bad
As can be seen from the above it seems to be trying to create the ctl and bad file on my c drive instead of on the server in the same directory as the .csv file. The location is registered to the server directory /u21/oracle/owb_staging/WHITEST/source_depot
I am at a lost as this works fine in development and I have just promoted all the development work to a systest environment using OMBPlus.
The directory structure in development is the same as systest except that the data file is /u21/oracle/owb_staging/WHITED/source_depot/Durham_Inventory_Labels.csv and everything works fine - .ctl and .bad created in the same directory and the data sucessfully loads into a oracle table.
Have I missed a setting in OWB during the promotion to systest or is there something wrong in the way the repository in the systest database is setup?
The systest and development databases are on the same box.
Any help would be much appreciated
Thanks
Edwin

How to load a long flat file field into BW

Hi all,
I have a field from flat file with 256 characters to be loaded into BW. As we all know the max length an info object can take is only 60. How can I split the flat file field so that I can load that data into BW and how can I then concatenate them in the BEx.
Can anyone suggest a way?
Thanks in advance.
Vinay

Hi Vinay,
we already had this discussion in this forum. In short:
1. create a number range object
2. create a table with a number field a key and a text field
3. while loading the data check your text against the table. If the text already exist, get the number out of the table. If the text doesn't exist, get a new number out of your number range object and post this number + the text into your table.
4. in your loaded data always work with the number you assigned to the text
5. create a infoobject for the number-field
6. create a variable that takes it's value out of the table.
regards
Siggi

Loading Flat files with into multiple tables using OWB

Hi,
How to implement the following logic in OWB.
LOAD DATA
INFILE 'myfile.txt'
BADFILE 'myfile.bad'
DISCARDFILE 'myfile.dsc'
APPEND
Into TABLE_Awhen (1:1) = 'A'
(Col1 Position(1:1) CHAR,
Col2 Position(2:5) CHAR)
Into TABLE_Bwhen (1:1) = 'B'
(Col1 Position(1:1) CHAR,
Col2 Position(2:20) EXTERNAL INTEGER)
Into TABLE_C
when (1:1) = 'C'
(Col1 Position(1:1) CHAR,
Col2 Position(2:20) EXTERNAL INTEGER)
I am using 10g version of OWB.I tried using the splitter operator.
I am getting the following error when i use the splitter.
An invalid combination of operators prevents the generation of code in a single implementation language (PL/SQL code, or SQL*Loader code, or ABAP code). For example, you may have included a SQL*Loader only operator such as the Data Generator in a mapping with a PL/SQL implementation type. If you designed the mapping to generate PL/SQL code, an invalid combination or sequence of operators prevents the generation of PL/SQL code in any of the operating modes(set based, row based, row based target only). If the mapping contains an operator that generates only PL/SQL output, all downstream dataflow operators must also be implementable by PL/SQL. You can use SQL operators in such a mapping only after loading the PL/SQL output to a target. Detail is as follows:
PL/SQL set based operating mode: Operator trailer_source_txt does not support SQL generation.
PL/SQL row based operating mode: Operator trailer_source_txt does not support SQL and PL/SQL generation.
PL/SQL row based (target only) operating mode: Operator trailer_source_txt does not support SQL and PL/SQL generation.
Both SQL and PL/SQL handlers are not supported by trailer_source_txt as output
SQL*Loader: Operator SPLIT does not support SQL*Loader generation.
ABAP: Operator trailer_source_txt does not support ABAP generation.
Thanks in advance,
VInay

Hi
Splitter can be used ib PL/SQL mappings, but if you use a flat file in a mapping, than it will be an SQLLoader mapping. So I suggest to you to create a mapping which load your flat file into a table, and from this table you load the data into the three table with the spillet in a PL/SQL mapping. Create two mappings.
Or you can use an external table in a mapping with a splitter.
Ott Karesz
http://www.trendo-kft.hu

How to upload the flat file records into internal table by position?

Hi
I have a flat file which has 7 records in each row and they are NOT provided with CSV or Tab demilited...
They are continous text without spaces....
but i know the fixed length of each field Eg : 1st field 7 char and seconc field 3 char and so on...
How can i upload this file into internal table by reading positions of each field...I know we can use GUI_UPLOAD and Read dataset and Open dataset...
But please let me know to read the file with the fixed postions and load into internal table...
Thanks in advance
MM

Hi
As per my knowledge i dont think thr is some function module or so to read with a fixed positions.
You can use the below method if you think this is the best way.
Suppose your file has
types : begin of ty_itab,
            field1 type char7,
            field2 type char3,
            field3 type chat3,
            field4 type char3,
            end of ty_itab.
types : begin of ty_upload,
             str type string,
            end of ty_upload.
data : it_itab type standatd table of ty_itab,
          it_upload type standard table ot ty_upload,
          wa_itab type ty_itab,
          wa_upload type ty_upload.
use gui_upload.get the data in it_upload.
here you know that u have first 16 charcters makes a first row n then next 16 charcters next row
Loop at it_upload into wa_upload.
   v_len = strlen ( wa_upload ).
   v_len = v_len / 16.
    You get number of rows per record ,if it is decimal value make it final value if 3.9 make to 4.
    do v_len times.
      wa_itab = wa_upload.
      By this only first 16 characters are moved and respective fields will get a value.or else u can use offset
       wa_upload+0(16).
      append wa_itab to it_itab.
      Now shift 16 characters using shift command in wa_upload.
    enddo.
endloop.
Hope this syntax help you to resolve your issue.May be something i have missed .Right now i don't have sap system to send you the full correct syntax code.
Cheers
Joginder

Flat file (csv) to oracle database - get dictionary data from oracle table

Hello,
I need to develop following scenario. I have flat csv file with some data, for example:
City;Address;Name
In oracle schema, there is table where data from imported csv should be written. Also, in the same schema, there is table named CityDictrionary. After ODI loads each row from csv, it should translate text representation of City to numerric ID got from CityDictionary table. And this numeric value should be placed to destination table, instead of direct text value from csv.
What is the simplest way to accomplish this task? Can you provide any tips?

You can achieve this easily with a single ODI interface. The flat file and the CityDictionary tables are your sources. Identify the field on the CityDictionary table and the corresponding field on the flat file that will be used to join the data sets (e.g. the city name). Create a join between the sources using these fields. Because you're using a flat file as one of your sources, the join logic will have to be performed on either the staging or the target.
The target is your destination table, and you should map the ID from the CityDictionary table to the appropriate field in the target, as well as any other required fields from the flat file.
This type of interface is a fairly typical method of populating a normalized table.
Alternatively, you can use a lookup - see the following blog for an example. In your case, the flat file will be your source and the CityDictionary table will be used for the lookup.
http://www.odigurus.com/2012/02/lookup-transformation-using-odi.html
Edited by: _Phil on Oct 1, 2012 11:52 PM
Edited by: _Phil on Oct 1, 2012 11:57 PM

Question on "Flat file upload into BW-BPS - How-to paper"

The How-to document for "How to Load a Flat file into BW-BPS using a Web Browser" discusses that the function modules "UPB_ALV_EXIT" and "UPB_ALV_EXIT_INIT" can be used as templates for copying the code to a zfunction module.
We are on BW 3.5 and cannot locate the above functions. I see there is a function "UPF_EXIT_INIT_EXAMPLE_01" that contains import,export and tables settings that matches one of the above functions but the code does not match to what is referred to in the document.
Does anyone know of a similar function that refers to the same code in the "UPB_ALV_EXIT" and "UPB_ALV_EXIT_INIT" FUNCTION MODULES" in the How-to-paper.
Thanks,
Kumar

Hi Kumar,
I think those functions mentioned in the "How to .." guide are replaced with the following two functions that will also provide the correct interface:
TEMPLATE_EXIT
TEMPLATE_EXIT_INIT
They are part of function group UPFX. This function group will also provide the rest of the function examples for functions of type 'Exit'.
Hope this helps.
Kind regards,
Dorothy

Error while loading data into PSA from data source

Hi Experts ,
I am loading purchase order history data from standerd datasource 0SRM_TD_CF into PSA using infopackage.However,whenevr I run infopackage,the call monitor shows the status as yellow.I monitored the process for half an hour but it never ended.I have verified following things from my side.
1.Checked whether datasource is properly replicated into BI system.
2.Checked if source system contains data.
3.Since i am loading transaction data,I have kept update as "initialize delta process"
Please let me know whether I am missing anything.
Regards,
Mandar.

Hi mandarha,
Do the following steps
1)Check the job log in source systems.
2)Check the Data Source in RSA3(Extracter checker) on Source systems side.give DS name press enter then execute,If there is no issue with DS,it will give pop-up with records.If it has any errors first rectify it and then load it again.
3)In details tab of IP,check whether extraction completed or not.
4)Are you running the IP with option "Intialse with data trasfer" or "Without data transfer"
If you running with data transfer and without data selection,usually it will take time based on the volume of historical data.(to find how many records it picks do the 2 step)
Hope this helps
Regards,
Venkatesh.

Is it possible to read flat file .CSV place in Client Machine in network?

Hi All,
I want or read the flat file in the client machine place in network through external table.
I have created an directory called DATALOADNW for client machine path as \\velmurugand\temp
When i tried to read an flat file iam getting the below error
ERROR at line 1:
ORA-29913: error in executing ODCIEXTTABLEFETCH callout
ORA-29400: data cartridge error
KUP-04001: error opening file \\velmurugand\temp\Locfile.csv
ORA-06512: at "SYS.ORACLE_LOADER", line 52
ORA-06512: at "SRIRAM.LIGHTINGEVENT1", line 68
ORA-06512: at line 1
Please can any one help me i have shared and given full permission to access the .CSV file.

velu wrote:
I want or read the flat file in the client machine place in network through external table.You can't. External Tables only work when the flat file (and directory) are on the database server.
Please can any one help me i have shared and given full permission to access the .CSV file.Move the file to the database server.
Or use your client application (whatever it is) to read in the file.

Size of Infocube Export to a flat file (.CSV)

First off, I'm new to BW so I apologize if I use the wrong terminology. I was wondering if there was any way to estimate the size of a flat file infocube export. For example, if I look up all the associated DB tables for a cube, let's call it TESTCUBE, would the total size of those tables be about the size of the TESTCUBE export to a flat file. This assumes that the cube is not compressed in any way. So if the the total DB table size is 100 MB then is the flat file about 100 MB? If not, is there a way to estimate this.
Thanks!

Can't say for sure how all DBs that BW run on work, but with Oracle, space is consumed in blocks, not by rows/records.
Let's say you have a table where the average row width (record length) is 100 bytes. In a tablespace that has 8K blocks, in theory, you could fit 81 rows in the block.
However there is usually a certain amount of freespace left in a block when rows are added to accomodate subsequent inserts/updates. The BW tables I have looked at default this to 10%. So that means youe have over 800 bytes of the block that might not get filled. There are other things that can result in space being consumed, such as deletions that will leave dead space, and updates can cause a row to migrate to another block, etc. Also, space is allocated in chunks as a table grows, and you may add one row to a table that causes a 10MB chunk of storage to be allocated, but it is holding only 1 row.
So as you can begin to see. A flat export should require less space than the source table (unless you are using database table compression which could completely invalidate what I have said), but by how much depends on all the things I mentioned, and could change from day to day.
You could certainly use as a rough rule of thumb, that a flat file export will require the same space and you would be safe (except if you are using db compression).

FLAT FILE UPLOAD INTO BI

Hi,
What kind of questions are needed to be asked (from client) for requirements gathering for a flat file upload (from legacy system) to BI? Basically what are all the questions that i need to ask them to make sure I have all information for designing a flat file extractor?
Thanks

I would ask them what do they want to report on from the legacy system and is data already in R/3? I would also ask what kind of data is stored in the legacy system .... inventory, financial data, sales, etc ... Once I knew that, I would ask what do they want to report on .... stock quantities/costs, gl account data, revenue, deliveries, etc....
Then I would ask, do you want this to be seperated from the R/3 datastream? More than likely this will be yes, if not, fight to keep it seperated and possible in it's own cube/ODS and combine in a multiprovider (you need to use the same 0material, 0customer, 0vendor, 0gl_account info objects if you want to combine nicely on a multiprovider).
If the database the legacy system is on is either SQL Server or Informix or Oracle, you may want to consider DBConnect and connect directly to their tables.
In all cases, I try to avoid bringing everything in "just because we may need it". If the system is legacy and it's been a legacy for 2 years, see what people keep going back for and grab that data only. If someone claims to need data, but have not sought it out for 2 years, then the probably don't need it.
Also, check the granularity of the data. Do they need individual documents for legal reasons? Can you summarize the data? What are they going to use the data for?
Brian

Loading of flat file (csv) into PSA – no data loaded

Similar Messages

Maybe you are looking for