Import Skip, Can we notify!!!

Hi,

Material Number is a Key field in matching records in Import map.

It is Set to Skip (For New materials in Incoming file)

and Update all Mapped Fields (For existing materials in MDM & Incoming file)

Now, My incoming file has some Materials which are currently not existing in MDM.

So, this import map skips that records to be imported.

How will i know that a Particular record in a file is Not imported??

in MDM server Log folder there is a File called 'Import@...'

This log says 1 record ''skiped'' and 20 records are ''Update all mapped fields''.

But is not giving any key information like this particular record or this particular material is Skipped!!!

Firstly is it possible to find it in MDM, or is there any workaround to find.

Please Help..

Kind Regards

Eva

SAP Managed Tags:
SAP NetWeaver Master Data Management

You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.

Comment

Accepted Solutions (0)

Answers (3)

Hi,

Alternate solution:

Create a new port and a new table in MDM. Then create a new import map with Create New.

Replicate the incoming file and push it to the Ready folder. Use the time stamp of the incoming file/import log and newly created records to find the newly created materials for any given day.

Thanks,

Shambhu.

You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.

Comment

Hey Shambhu,

Will that tell which records are skipped?

Can you detail the process flow.

Thanks,

Ravi

Yes, It will create the new records in the new table. So you can go in Data Manager and find the new records.

Create the table with some important fields like primary key and few important fields.

Thanks.

Hi Shambhu,

Good idea but i guess it will demand for creation of new record in MDM.That is where it wont suffice,it will have to be stored somewhere else.

Otherwise the flow should go like PI will have to send 2 files to each of these ports simultaneously.

On one hand you will know which are being updated by 1st.

On other you will know which are being updated by 2nd.

But one has to take care that fields updated have to be different as you mentioned.

I see a challenge here MDM does not give you a functionality to search updated records based on fields.

So for that we will have to create fields which track updation for 1st scenario and 2nd scenario separately.

Catch is it has to be run daily.Once this is set up a java code can give you daily report too.

what do you say.

Thanks,

Ravi

Yes, thats why I suggested to create a new Main table"A" with a field referenced to the Material master main table primary key in the MDM Repository.

So only the new records gets created in table A. Then you can use the Data Manager to view the new records and prepare your report or complete your analysis.

You can use syndication map to generate report instead of additional java code for reporting. All records getting created in table A are the records that were present in the incoming file but skipped in the Material Master Main table for any particular given date.

I think this will work with not much performance impact on the repository, if the fields and number of records are manageable.

Thanks.

Hi Shambhu,

Yes I think this should workout but only concern is its a big development

Hey Eva,

Please let us know what do you think of this.

Thanks,

Ravi

I think this is very easy development and a quick way to fulfill Eva's requirement.

Eva, mail me or call me if required.

Thanks,

Shambhu.

I agree PI config,MDM dev and port and map creation as against ini parameter change that was the point i was refering to.

Thanks Shambhu.

Ravi

Hi Shambhu , Ravi,

Sorry was on leave for Couple of days.

Thats more than a help you guys have done..

2 import maps, One Skips and second one creates.

Created ones will be pointed to Dummy 2nd main table.

Skipped ones will update the records in main table.

That sounds like a plan.

Does this mean that Skip Unchanged Records=True will never give the Skipped records in the import log!!!

I tried in sandbox, there is no difference in the Import log, I requested Basis To do it in Dev!!

Kind Regards

Eva

Hey Eva,

Thanks to Shambhu for coming up with this idea.

There should be a entries in Import logs.Exactly where are you looking for entries?

Thanks,

Ravi

Hi,

The Skip Unchanged Records=True in MDS.ini file.

I have an Import map..

Material Number is the key in matching Tab.

The import map settings are...

>Skip the records which has New material numbers

>Update all mapped Fields for the Existing material Numbers.

I have dropped a File with New Material Number (Which is not existing in MDM).

The Import server Skipped that Record.

In the File Server .. /mdm/distributions/Dww_ORCL/MAT/Inbound/MAINFRAME/MF_TO_MDM_DATA_PORT/Log

Folder i can see a Import log which says...

****************************************************

<?xml version="1.0" encoding="utf-8"?>

<?xml-stylesheet type="text/xsl" href="MDM_Log.xsl"?>

<MDM_Log version="1.0" path="/usr/sap/DMD/MDIS01/_20110831174532.xml"

<Version ts="2011/08/31 17:45:32.355 GMT" tid="1163262272" entry-no="0"><Server>7.1.07.239</Server><Build>Built on 2011-Jul-30</Build></Version>

<Trace ts="2011/08/31 17:45:32.363 GMT" tid="1163262272" entry-no="1">[MDS=dsgmdcsd101 Repository=xxx ClientSystem=MAINFRAME Port=MF_TO_MDM__DATA_PORT]: Chunk size/parallel[2000/40]: Import Task Started.</Trace>

<Trace ts="2011/08/31 17:45:32.415 GMT" tid="1163262272" entry-no="9">1 Source file(s) retrieval + Repository Caching: 0.061603 milli-seconds.<LF/>-- INPUT_FILE_MF_TO_MDM.xml</Trace>

<Trace ts="2011/08/31 17:45:32.416 GMT" tid="1173752128" entry-no="10">xStructuralTransformer: Thread started.</Trace>

<Trace ts="2011/08/31 17:45:32.416 GMT" tid="1184241984" entry-no="11">xValueTransformer: Thread started.</Trace>

<Trace ts="2011/08/31 17:45:32.419 GMT" tid="1194731840" entry-no="12">xImporter: Thread started.</Trace>

<Trace ts="2011/08/31 17:45:32.452 GMT" tid="1205221696" entry-no="13">xXmlSaxParser: Thread started.</Trace>

<Trace ts="2011/08/31 17:45:32.453 GMT" tid="1205221696" entry-no="15">Parsed and populated Chunk No: 1 w/ 1 segments.</Trace>

<Trace ts="2011/08/31 17:45:32.466 GMT" tid="1205221696" entry-no="16">xXmlSaxParser: Thread finished; Start -> End: 0.000000000 milli-seconds.</Trace>

<Trace ts="2011/08/31 17:45:32.476 GMT" tid="1173752128" entry-no="18">Structural Transformation completed for Chunk No: 1 w/ 1 segments.</Trace>

<Trace ts="2011/08/31 17:45:32.516 GMT" tid="1184241984" entry-no="24">Value Transformation completed for Chunk No: 1 w/ 1 segments.</Trace>

<Trace ts="2011/08/31 17:45:32.561 GMT" tid="1194731840" entry-no="33">Importing: 'INPUT_FILE_MF_TO_MDM.xml.MDM__DATA' Table --> 'xxx.xxx' Table.</Trace>

<Trace ts="2011/08/31 17:45:32.566 GMT" tid="1194731840" entry-no="36">Import action:<LF/>~~Skip: 0<LF/>~~Create: 0<LF/>~~Updated (NULL fields only): 0<LF/>~~Updated (all mapped fields): 0<LF/>~~Replace: 0<LF/>~~Delete [destination]: 0<LF/>~~Value Exceptions: 0<LF/>~~Import Exceptions: 0</Trace>

<Trace ts="2011/08/31 17:45:32.567 GMT" tid="1194731840" entry-no="38">xImporter: Total chunk import time: 0.000000000 milli-seconds.</Trace>

<Trace ts="2011/08/31 17:45:32.567 GMT" tid="1194731840" entry-no="39">1 Segments were Successfully imported.</Trace>

<Trace ts="2011/08/31 17:45:32.581 GMT" tid="1173752128" entry-no="40">xStructuralTransformer: : Start -> End: 0.000000000 milli-seconds.</Trace>

<Trace ts="2011/08/31 17:45:32.581 GMT" tid="1173752128" entry-no="41">xStructuralTransformer: Thread signalled to stop: RC = 0x0</Trace>

<Trace ts="2011/08/31 17:45:33.455 GMT" tid="1184241984" entry-no="42">xValueTransformer: : Start -> End: 0.000000000 milli-seconds.</Trace>

<Trace ts="2011/08/31 17:45:34.458 GMT" tid="1194731840" entry-no="43">xImporter: : Start -> End: 0.000000000 milli-seconds.</Trace>

<Trace ts="2011/08/31 17:45:34.478 GMT" tid="1163262272" entry-no="44">[MDS=dsgmdcsd101 Repository=xxx ClientSystem=MAINFRAME Port=MF_TO_MDM__DATA_PORT]: Chunk size/parallel[2000/40]: Import Task Finished.</Trace>

</Open>

</MDM_Log>

******************************************************

This log did not give me the Skipped Record Info.

However it said me that '<Trace ts="2011/08/31 17:45:32.567 GMT" tid="1194731840" entry-no="39">1 Segments were Successfully imported.</Trace> '

Which is very strange...???

Do you know any other Log which says about Record details...!!!

Kind Regards

Eva

Eva,

Not sure if any logs will show the missing material number. Even if it does, In real time scenario you cannot monitor the log files to find missing material numbers and then send out a report.

The alternate solution proposed by me will help you to automate this reporting.

Thanks,

Shambhu.

Hi Eva,

I understand this is not most user friendly format,but please check the directory which has been provided in MDIS.ini.

If that does not work,lets go ahead creating another main table as Shambhu mentioned.

Same source files for both the ports and teh Dummy table has Create for New and Update Mapped for the others so that the records can appear when searching for Created/Updated records.

Thanks,

Ravi

Hi,

I have seen a Log in Repository. Does not help.

However i have a question on the Workaround solution.

My Main table_1 has all the Material related data init.

Import map 1 -> When i load a File and if file has new material Number in it the setting is to Skip all New Materials and 'Update All Mapped' for existing materials. This is Mapped to My Main table1.

Now i have this Main table_2 (which is created for the sake of Work around).

This table will be Blank.. Now my question is, When i create a Import map, when i map to a Dummy Materail Number field , How will the Import map Know that whether the Incoming material Number is available in the main table 1 ??

But our target in the map is main table 1 !!!

Can you please help!!

Kind Regards

Eva

My Main table_1 has all the Material related data init.

Import map 1 -> When i load a File and if file has new material Number in it the setting is to Skip all New Materials and 'Update All Mapped' for existing materials. This is Mapped to My Main table1.

Now i have this Main table_2 (which is created for the sake of Work around).

But our target in the map is main table 1 !!!

Well starting condition for table 2 can be either same records for table1 and table 2 or table 2 empty.

Scenarios 1 - Table 1 and Table 2 having same data for Material no to start with.

lets say 20 records come for import of which 10 records are those already created in table 1,they are updated and 10 are new ,they are skipped.

For Table 2 these,10 new records will be created and 10 old will be updated.

You can create a additional field in table 2 Description,it will also be overwriiten in case of update.

Now you can search records coming in from table 2,it can tell you which records came for import to Table 1.They are the same which are created/updated today in Table 2 ,Export them out.

Table 1 gives you which records were updated today,Export them out.

Skipped records = Total record available for import (Exported from table 2) - Total record Updated (From table 1)

Scenario 2 - Table 2 is Empty.

Here every record coming in will be created.But after a certain duration of time record updation will also start.

So records Created/updated with todays date will form the search criteria and will tell you Total records that were available for import.And hence you can export this data out.

This data can also be syndicated based on Supress unchanged record criteria used in a WF which can has Threshold of 1 day.

Hope this helps.