UNICODE PREPARATION PHASE:

Preparation

1. Make ABAP programs Unicode-enabled:

Programs: transactions UCCHECK and SCOV

UCCHECK: Performed for Y* and Z* programs

1
 2. Unicode Enhanced Syntax Check
The system profile parameter abap/unicode_check=on can be used to
enforce the enhanced syntax check for all objects in non-Unicode systems.
The mentioned parameter should be set to the value "on" only, if all customer
programs have been enabled according to transaction UCCHECK.
If you have problems with regeneration of reports RK31*, RK32*, RK33* or
RK34*, create a customer message under component CO-PA.

3. Pre- Conversion Correction of Table Data

Before starting the Unicode conversion you need to maintain data with value
“SODI” in field PERS_GROUP in table ADRP according to the SAP Note 712619
Run report UMG_ADD_PREP_STEP. This report consists of several reports
which were created for maintaining corrupt or obsolete table data before a
Unicode conversion.
Delete entries for
SDBI_POOL_CLEANER TCP00, TCP01 or 920138
TTDCF from ATAB
Delete entries from
STSTCP_REPAIR 897002
TSTCP
Delete extracts for
GRIX_DELETE_RW_EXTRACTS 892607
Report Writer
Delete content of
RSSPAM_PREPARE_UC 889596
OCSCMPLOBJ
Correct characters in
SIAC_DELETE_INVALID IACSERVP and 862196
IACS_C
RS_FILL_TRDIR_RLOAD Fill REPOSRC-RLOAD 684332
Delete screen source
codes with invalid
RS_SCRP_D020S_CLEAN keys from tables 870601
D020S and
DYNPSOURCE
Delete entries from
RSDB4PREP_UCPC 932034
SQ* tables
Delete entries from
RGX_SETS_TO_40 tables T800S, T800X, 928859
T800Y and SETDATA
Delete entries from
CY_TCYD_UNICODE 899931
table TCYD
Delete entries from
RPU470ADJUST_T52C5 679275
T52C5
Delete entries from
RSTXCLEARTTDTG 992857
table TTDTG
Delete entries from
INDX_SH_DELETE 989070
table INDX

2
Call transaction SE38. Enter report UMG_ADD_PREP_STEP and choose . On
the next screen the included reports are displayed. Choose Background. All
reports will be executed automatically in sequential order.

If you need to rerun one or more selected reports later, you can use button .
Note: Do not use this option before UMG_ADD_PREP_STEPS has been run
once using the background option.

3
We recommend that you maintain corrupt data in tables T52C5, TTDTG and
in tables ADRC and ADRP, according to the SAP Notes mentioned below.
Tables: T52C5; TTDTG; ADRC and ADRP tables
Documentation: SAP Notes 679275; 794966; 673941
Single Code Page: recommended

Delete QCM tables from the source system as described in SAP Note 9385.
Tables: QCM <table_name>
Documentation: System Copy Guide, section 2.2 Technical Preparations";
SAP Note 9385
Single Code Page: recommended

Excute tcode SE14

PA-PA-GB: Special handling for table T5GEFDATA required.

Table T5GEFDATA is a container table, i.e. structured data (text and binary
data) are stored in the character like field T5GEFDATA-DATA1. For the
Unicode preparation a special handling has to be done for table T5GEFDATA,
taking the substructure of the container field into account.
Follow the description in SAP Note 1239557.
Table: T5GEFDATA
Documentation: SAP Note 1239557
Single Code Page: required

4
 4. Text Language for Customer Objects
All LANG fields in Data Dictionary (DDIC) objects which are delivered by SAP
have an attribute Text Lang. (indicator for a language field). This indicator
specifies that the LANG field determines the code page of the character data
in that table. Text Lang. flags are evaluated during RFC communication
between Unicode and non-Unicode systems to ensure that the correct

5
character conversion is used. In Single Code Page conversions the flags are
evaluated by R3trans.
For more information, read SAP Note 480671.
To set this DDIC attribute for customer objects, run report RADNTLANG.
Transports into a system can cause the indicator to be overwritten. It is
therefore necessary to (re)run RADNTLANG directly before beginning the
Unicode preconversion in order to actualize the flags.
After you have run RADNTLANG please read SAP Note 672835 carefully. You
will find information about how to maintain special tables which could be
affected by wrong data conversion after the Text lang. flag has been set.
Text lang. flag maintenance must be finished before the Unicode
Preconversion Phase. If you do any changes as described in this section after
or between the SPUMG scans, you must reset the changed tables and rerun
the scan(s)

6
 5. Consistency Check for the ABAP Dictionary
SAP recommends running report RUTCHKNT prior to the SPUMG scans to
check and clean up inconsistencies between DDIC objects and runtime
objects.
In transaction SE38 enter report RUTCHKNT and choose . On the following
screen checkbox Check against inactive nametab must be deselected! Press
again to access the next screen.

7
Go to section Obj. types for checking. Select checkboxTables.
Go to section Select Objects by name, field table category. Select and
choose tabstrip Exclude Single Values. Select Multiple selection. Select table
categories “INTTAB” (Structure) and “VIEW” (General View Structure). Choose
and then . Afterwards execute the report.
Afterwards check all log messages ("Check activation of object was
successful" must be checked too!).
Check DDIC objects in transaction se11 (Utilities → Runtime Object → Check)
and activate them.

8
9
 6. Consistency Check for Cluster Tables
SAP recommends checking inconsistencies in cluster tables prior to the
SPUMG preparation steps.
Programs: report UMG_R3CHECK_FILES; program R3check
Documentation: SAP Note 89384
Single Code Page: recommended.
Run report UMG_R3CHECK_FILES in transaction SE38 order to create control
files for R3check. Read the report documentation for details. Run R3check on
database level as described in SAP Note 89384.

10
 7. Translation Environment
If you are using the SAP translation environment (transaction SE63) you must
export the Proposal Pool in the non-Unicode system. When the Unicode
Conversion is finished, you must import the Proposal Pool into the Unicode
system.
Follow the instructions in SAP Note 1055820.

11
Unicode Preconversion Phase

Transaction SPUMG
SPUMG is used to prepare the database tables of a non-Unicode system for
the conversion to Unicode. All steps in the Unicode Preconversion Phase are
designed to be run at system uptime (i.e. when users are logged on to the
system). The system conversion is carried out by the separate tool R3load
during the database export. R3load uses the results of SPUMG which are
stored in the Export Control Table, the Vocabulary and the Language Lists.
Tool: transaction SPUMG
Documentation: this document; SAP Note 548016;
Single Code Page: required.

1. Preconversion Analysis: SPUMG Scans

During the Unicode Preconversion the database tables have to run through
several consecutive scans which are listed below.

12
Download umgexceptions_ver_00_011.xml to
c:\AstUpg\UC\umgexceptions_ver_00_011.xml and upload it

If the system code page is 1100 you should check whether the frontend code
page (or SAP GUI code page) is also 1100. Run report RSCP0013 in
transaction SE38 and check, if 1100 is displayed in field Character set of
SAPgui. If yes, users are able to input characters which are not covered by
the system code page. Such characters – for example Euro sign (€) and the
trademark sign (™) - will get lost during the database conversion because
they are not stored on the database. You can prevent this at least for
characters in transparent tables by switching to 1160 in the SPUMG Language
List.

Not necessary to change because RSCP0013 shows Character set of SAPgui =

1160:
But for older versions perhaps SAPgui codepage is 1100. Therefore do change
to 1160.

2. Database Check
Start SPUMG again: Press button Continue with transaction SPUMG

Reset all Scans

Select Scanner – Reset All Scans

13
Select Yes
Select Scanner – Initialize Worklist
This starts job UMG_PMIG_STOCK_INIT
Press button Selection, fill in limit 80000 -> list is shown: 72856 entries

3. SPUMG Settings
Select Scanner – Settings
Changed from:

14
to:

and Save!
The following settings are changed from their defaults:
• Global Fallback Codepage 1100 -> 1160
• Minimum Work Length 1 -> 2
• Collision Resolution Character based (fine) -> Codepage based
(coarse)

15
4. Initialize Worklist

A worklist contains all tables of a system that have to be prepared by SPUMG

for the
Unicode conversion. In SPUMG, the tables have to run through several scan
levels, each of
which has its own worklist.
The worklists must be generated as follows:
1. Select Scanner - Initialize - Worklist (F2). Depending on the data
volume the
initialization job may take several minutes. Press pushbutton . You have
generated
the worklist for the first scan (Consistency Check).
This starts job UMG_PMIG_STOCK_UPDAT which takes a few minutes to finish.

5. Monitor:
On each scan level the SPUMG monitor displays scan statistics and the
processing status of the tables. Select pushbutton Monitor from the toolbar.
Worklist Status (Consistency Check): Initialization of the CC Worklist:
Display of status and total number of initialized tables.
Overview number of tables: Here you get an overview of all initialized
tables sorted by table types (number of transparent, cluster, and pool tables
Worklist Overview: Here you can see how many tables are currently
processed on which scan level. Tables with warnings are inconsistent and
therefore need to be maintained before the conversion.
Tables in Process: Here you can see which tables are currently being
processed by which worker job
Select refresh button to update the monitor.
To see details about running worker jobs, select pushbutton Job Overview
from the toolbar.

16
6. Status
Select Scanner → Status to view a general overview of all current activities in
SPUMG
Via tab Status (not via Scanner – Status)

17
7. Scan 1: Consistency Check
This scan is mandatory for both Single Code Page and MDMP conversion.
Select Schedule Worker Jobs from the toolbar, and then set the time and date
for the worker job.
This starts job UMG_PMIG_MASTER_JOB that starts/monitors job
UMG_PMIG_WORKER_JOB_1_

No errors were found in the 72856 tables.

8. Update Database Statistics

Call transaction DB20 to update the database statistics for the tables
UMGCCTL; UMGPMIG; UMGPMDIC; UMGSTAT.

18
Unicode Conversion Phase

Program: R3load
The default conversion method is to export the entire database using R3load,
create a new
Unicode database, and then import the database using R3load again.

Go to the drive where ECC 6.0 Installation Master DVD is dump. In our case it
was dump in E drive.

E:\Installation_Master\IM_WINDOWS_I386\SAPINST\NT\I386 and click sapinst

19
Click on Software Life-Cycle Options  System Copy  Oracle  Central
System  Based on AS ABAP and select Export Preparation and click Next.

This will Prepare the System for Unicode conversion.

20
Give the profile directory path and click next

21
Password: Te4Leav3s and click next

22
Create a folder Export, in our case I had created it in E drive. Make sure you
have more than 20GB free space available on that drive.

Give the path of export folder and click next.

23
Select Target Database Type: Oracle and click next

24
Select Update statistics before the export starts and click next.

25
Click next

26
After this Export preparation step is over.

Again go to E:\Installation_Master\IM_WINDOWS_I386\SAPINST\NT\I386 and

click sapinst

27
Click on Software Life-Cycle Options  System Copy  Oracle  Central
System  Based on AS ABAP and select Database Instance Export and click
Next

28
Give the profile directory path and click next

29
Password: Te4Leav3s

30
Specify the Export folder path we created earlier and click next.

31
Select Target Database Type: Oracle and click next.

32
Click next

33
Tick on Unicode Target System check box. And select Little-Endian and click
next.

34
Select Update statistics before the export starts and click next.

35
Click next.

36
37
38
39
Export Phase is completed successful.

Now before you start the Import Phase backup the whole database. Once you
have successfully backup the database run sapinst and uninstall SAP and
Oracle Database, but don’t uninstall the Oracle software.

Then goto data drive where all the sapdata folder are created. In our case we
go to G drive and delete all the SAP Data folder, then goto H, I , J drive and
delete D01 folder.

Once you have deleted everything restart the system.

Login as D01adm and goto Installation Master DVD dump and start sapinst.

Select SAP ERP 6.0 Support Release 3  Software Life-Cycle Options 

System Copy  Oracle  Target System Installation  Central System 
Based on AS ABAP  Central System

Click Next

40
41
Select Parameter Mode: Custom and click next

42
SAP SID: D01

43
Select Standard System Copy/Migration (Load-Based)

44
Database SID: D01

45
Specify the path of the Export folder which we had created earlier.

46
DDIC Password: welcome1

47
 Give the path of Netweaver 7.0 SR3 Kernel & Oracle Client path and click
next.

48
Click next to start the Import

49
Enter Solution Manager Key

50
51
This Phase will take few hours to complete

52
Non-Unicode to Unicode conversion is finish successfully.

53