Single Code Page System Conversion

to Unicode

In addition to the “Single Code Page Conversion to Unicode Guide” you need the System

Copy Guide for your release. The System Copy Guide is required for the Unicode Conversion

Phase which is in fact a System Copy procedure with special code page settings. You can

download all valid System Copy Guides from the SAP Service Marketplace at:

http://www.service.sap.com/installNW70

How to check if the system is a Single Code Page System

If you are not sure whether your system is a Single Code Page system or not, open database

table TCPDB in transaction SE16. If this table contains only one entry, it is a Single Code

Page system.

If TCPDB contains more than one entry, the system is an MDMP system.

Requirements

Software Requirements

1. For the conversion from non-Unicode to Unicode you need both the non-Unicode and

the Unicode Installation Package:

a. a non-Unicode CD containing the SAPinst tools for the export of the database

and

b. a Unicode Kernel CD for the import of the database

If you use SAP Office functionality in your SAP system, read SAP Notes 691407 and
690074 for further information on Unicode conversion.

If you use SAP Easy Document Management (EasyDMS), follow the instructions in SAP
Note 1056170.

Conversion of CRM systems: Note that Business Transaction duration data must be

saved before the Unicode Conversion. Follow SAP Note 1122968.

General Description

To convert a non-Unicode system to Unicode, all character data (i.e. text data) in the non-

Unicode database must be converted to Unicode. The overall conversion procedure is a

System Copy with special code page settings. The actual data conversion to Unicode is done

during the export. For further information about system copy optimization, go to

www.service.sap.com/systemcopy → optimization.

1. Preparation

1.1 Consistency Check for Cluster Tables

Programs: Report SDBI_CLUSTER_CHECK

Documentation: SAP Notes 1348055; 1077403

Status: mandatory

You use report SDBI_CLUSTER_CHECK to search for initial cluster records in cluster tables.

Initial cluster records must be removed before the database export; otherwise the export

might cancel. SAP Note 1348055 provides a description of the report and explains how to

proceed when initial cluster records are found. For information about the latest version of the

report, read SAP Note 1077403.

SDBI_CLUSTER_CHECK is not restricted to systems which will be converted to Unicode

afterwards. You can use it as a regular maintenance task in all systems. You can execute the

report anytime and as often as needed independent from SPUMG.

Output of the report

Remote Code Page Check

Programs: UMG_CHECK_RFC

Documentation: online documentation in transaction SE38

Status: recommended.

Before the Unicode Conversion it is useful to check your own TCP/IP-connections and adjust

the code page information if required. Run report UMG_CHECK_RFC in SE38 as described
in the online documentation to check if the connection between the local system where you

run UMG_RFC_CHECK and a remote TCP/IP destination is consistent regarding the

involved local and remote code pages.

Output of the report

1.3 Make ABAP programs Unicode-compliant

Programs: transactions UCCHECK and SCOV

Documentation: transaction ABAPHELP, search with keyword "Unicode"

System Documentation in UCCHECK and SCOV

SAP Notes: 367676; 549143; 548016

Asian (especially Korean) customers: SAP Note 1007073

Status: mandatory.

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. When setting this parameter,

only Unicode-enabled objects (objects with the Unicode flag) are executable. Note that after

setting the Unicode flag, automatically generated programs might need to be regenerated.

The mentioned parameter should be set to the value "on" only, if all customer programs have

been enabled according to transaction UCCHECK.

Special handling for table DBTABLOG required.

DBTABLOG is a container table, i.e. structured data (text and binary data) are stored in the

character like field DBTABLOG-LOGKEY. For the Unicode preparation a special handling

has to be done for table DBTABLOG, taking the substructure of the container field into

account.

Follow the description in SAP Note 1255556.

Table: DBTABLOG

Documentation: SAP Note 1255556

Status: mandatory

2.1 SPUMG

2.1.1 Restart SPUMG

It is mandatory that all data are completely deleted before starting the database check! Go to

Scanner → Restart SPUMG.

After a restart, you will be displayed the entrance screen of SPUMG. All data which might
have been already entered in SPUMG are now deleted.

2.1.2 Welcome to SPUMG

The entrance screen is divided into two sections. In the upper section you will find general
information and a couple of links to Unicode information sites.
In the lower section there are three pushbuttons with additional Info buttons. Each
pushbutton triggers a mandatory action to be performed before the worklist for the
Consistency Check can be initialized. The checkbox Would you like to replace code page
1100 with 1160 in the Language List can be optionally selected in case the frontend code
page 1100 has been used with system code page 1100.
The last pushbutton Continue with transaction SPUMG will trigger the automatical
initialization of the Language List and lead to the Consistency Check screen.

2.1.3 Exception List

The Exception List contains SAP delivered tables which store data from only one language,
for example tables with only Japanese or only German data.
The exception list is provided as XML file attached to SAP Note 996990. You must upload
this file once. Select Download Exception List from note 996990 and save the XML file to
your local PC.Then select Update the SPUMG exception tables and upload the XML file.
You can display the content via Exception List → Default Exception List after the
Consistency Check has been executed.
This list of tables is predefined by SAP. In principle it is possible to add or remove tables from
the Exception List, but it is not recommended to change the default content in Single Code
Page systems.
You can proceed after the confirmation message Upload of exceptions successfully finished
is displayed.

You cannot proceed with SPUMG until you have executed UMG_ADD_PREP_STEPS once

using the background option

Report output in foreground

2.1.7 Initializing the Worklist

Select Scanner 􀃆 Initialize 􀃆 Worklist. Depending on the data volume the initialization job

may take several minutes. Select . You have generated the worklist for the Consistency

Check. The worklist of the Consistency Check (CC worklist) contains a list of all tables taken from the

database’s nametab. By default, only 200 entries are displayed. To increase the number of

entries that are visible in the ALV Grid, select Selection from the toolbar and increase the
number of lines.

SPUMG Main Log

Check the main log regularly. Select Scanner → Main Log from the menu bar.

2.1.8 Consistency Check

Select Schedule Worker Jobs from the toolbar, and then set the time and date for the

worker job. You can run multiple jobs simultaneously, but you must add a suffix to the

second and all subsequent jobs.

Remark : one error found in Consistency check.

Read SAP Note 33814 for general information about inconsistencies between database

and data dictionary

2.2.2 Last Preparation Reports

1. Delete matchcode IDs:

Run report TWTOOL01 [page 37]

Note:

Do not start the export before all matchcode IDs have been deleted! Run this report until

message “Check successful, no action necessary” is shown!

4. Create the nametabs.
Caution

Do not import new DDIC objects or modify existing DDIC objects after you have created the

Unicode nametabs and before the export of the database! Otherwise the nametabs will

become invalid!

In SPUMG select Extras → Unicode Nametabs → Create Unicode Nametabs from the

menu. A report is automatically started in the background with variant UNICODE-02-CRE.

This variant creates the Unicode nametab.

If you try to repeat this process in SPUMG, you will receive a system warning. If you ignore

this warning, SPUMG will automatically use variant UNICODE-02.

After having created the nametabs select Extras → Unicode Nametabs → View log of

RADCUCNT from the menu. Check the log carefully. If you ignore errors, the import will not

succeed and you will have to repeat the conversion procedure.

Nametab creation may produce messages due to old DDIC inconsistencies. Check the

following SAP Notes for help:

After the nametabs have been successfully created, you should update the

database statistics for the alternative active nametabs DDNTF_CONV_UC and

DDNTT_CONV_UC in order to improve the performance of the R3load export

procedure.

2.2.4 Tables with obsolete data

DDSHPVAL50
Personal Help
Values

You cannot use personal value lists from table DDSHPVAL50 and generated selection reports

(namespace /1BCF41/) of the input help (F4) after a Unicode conversion. You must delete them before
the conversion and create them again in the Unicode system during runtime. Use report
UMG_F4_CLEANUP according to SAP

Note 1275149.

Post Implementation Reprot

You have to perform this manual post-implementation step manually and separately in each system
after you have imported the Note to implement.

You must manually create the missing report texts.

To do this, proceed as follows:
Call transaction SE38 or SE80 for report UMG_F4_CLEANUP.
Call the text elements by selecting "Text elements".
Select the "Selection texts" tab. Add the following:

CHECK Check
DELETE Delete
GEN_REPO Generated reports
PERS_VAL Personal values
UMG_MODE in UMG mode

Select the "Text symbols" tab. Choose "Compare text symbols" and select the checkbox "Text symbols to
be added to the text pool." Use the default value.
Then save and activate the texts.

GREP

Report Writer: File ( not completed)

of Stored Reports

After a Unicode conversion, extracts for Report Writer reports can not be used anymore. Extracts
should be deleted before the conversion to avoid converting data unnecessarily.

SAP Note : 837184; 892607

Solution :

Reason and Prerequisites

 Extracts (the table GREP)

Extracts are not intended for the long-term archiving of reports, are lost during release
upgrades (see Note 67893) and may be rendered inconsistent when corrections are
implemented (see Note 78109), as they depend on a program that is specially generated for
the saved data. A Unicode conversion changes the data so significantly that the generated
program is no longer suitable for this data.

 Generated Report Writer programs (the table GRRT)

As is also the case after an upgrade and when you import certain Support Packages, the
Report Writer report groups must be regenerated after the Unicode conversion.

Solution

 Extracts (the table GREP)

The extracts cannot be saved by means of the Unicode conversion. You can use transaction
GRE0 to delete the extracts before the Unicode conversion (by doing this you prevent the
large datasets of the table GREP being converted unnecessarily).

 Generated Report Writer programs (the table GRRT)

To avoid converting data of the table GRRT unnecessarily, the generated programs can be
deleted with the report RGRDLINC as in Note 850818.  See also Note 998892.
After the Unicode conversion, an automatic regeneration that writes the new entries to the
table GRRT occurs when a report group is called for the first time.
The report groups can also be regenerated with the mass generation (see Note 78109).

GRRT

Report Writer: Runtime information for generated programs

To avoid converting data of table GRRT unnecessarily, the generated programs can be deleted with
report RGRDLINC
SAP note : 837184 , 850818 ( no report or T. code found in sap solution manger)

VBDATA, VBHDR, VBMOD

Update Requests should be checked and maintained before the database export to avoid converting data
unnecessarily. Run report RSM13002 according to the documentation of the report.

4.2Generating DDL Statements

To migrate nonstandard database objects, you need to generate DDL statements using the ABAP
report SMIGR_CREATE_DDL.
You must perform this procedure before starting SAPinst.

Procedure

Log on to the system as a system administrator in a productive client.

Call transaction SE38 and run the program SMIGR_CREATE_DDL.

Select the target database. Depending on the database manufacturer, you might need to select the
database version. The value help supports you in selecting the database version. You should only
enter a database version that is available in the value help.

Select Unicode Migration if you also wish to perform a Unicode system copy (from Unicode to
Unicode) or a Unicode conversion (from non-Unicode to Unicode).

Specify an empty working directory to which the files generated by the report are to be written.

If required, you can restrict the generation of DDL statements to specific table types or individual
tables.

Execute the program.

The DDL statements are generated and are written to the specified directory.
no database-specific objects exist in the database, no SQL files will be generated. As long as the
report terminates with status successfully, this is not an error.

8.
Before starting the export, the generated SQL files have to be copied from the specified directory to
the <instdir> directory of the export.

For additional database-specific information, see SAP Note 888210.

To be able to perform the conversion from non-Unicode to Unicode,

you require a non-Unicode Kernel CD for the export and a Unicode
Kernel CD for the import. As a result, you need both the non-
Unicode and the Unicode installation packages.

Error during export

(EXP) ERROR: DbSlPrepare/BegRead failed

rc = 103, table "/BI0/PTCTUSERNM"

(SQL error 208)

error message returned by DbSl:

[Microsoft][SQL Server Native Client 10.0][SQL Server]Invalid object

name '/BI0/PTCTUSERNM'.

(DB) INFO: disconnected from DB

Solution : table /BI0/PTCTUSERNM is not in database. So create the

table in database to use SE14.

Error during IMPORT

ERROR: 2010-06-21 17:01:14 com.sap.inst.migmon.LoadTask run

Loading of 'SAPAPPL0' import package is interrupted with R3load

error.

Process 'D:\usr\sap\TSS\SYS\exe\uc\NTAMD64\R3load.exe -i
SAPAPPL0.cmd -dbcodepage 4103 -l SAPAPPL0.log -loadprocedure
fast' exited with return code 2.

For mode details see 'SAPAPPL0.log' file.

Standard error output:

sapparam: sapargv( argc, argv) has not been called.

sapparam(1c): No Profile used.

sapparam: SAPSYSTEMNAME neither in Profile nor in Commandline

Solution : maintain the environment variable SAYSYSTEMNAME.

(DB) INFO: ZPA0002 deleted/truncated

Interface access functions from dynamic library dbmssslib.dll loaded.

(IMP) INFO: EndFastLoad failed with <2: BCP commit failed:

[Microsoft][SQL Server Native Client 10.0][SQL Server]Violation of
PRIMARY KEY constraint 'ZPA0002~0'. Cannot>

(IMP) ERROR: EndFastload: rc = 2

(DB) INFO: disconnected from DB

Solution : Remove the primary key in the table ZPA0002.

Spau
Spdd
SAP UPGRADER Procedure
To make sure that all requirements are met, run the PREPARE [page 207] program
as soon as possible. You can reset and repeat the program as often as you need
before you start the actual upgrade. The successful execution of PREPARE is a
prerequisite for starting the upgrade

The SAP upgrade technology offers two upgrade strategies:downtime-

minimized and resource-minimized. Choose the strategy that is best suited to
your SAP system and to your requirements concerning system availability. Your
decision depends on two factors:

Multiple Components in One Database System (MCOD)

As of SAP NetWeaver ’04, the SAP Internet Transaction Server (SAP ITS) is
integrated into the SAP NetWeaver component SAP Web Application Server as
an Internet Communication Framework (ICF) service, which you can access, like
other services, through the Internet Communication Manager (ICM). With the
SAP ITS integrated in the SAP Web Application Server, the Web browser now

communicates directly with the SAP system. Furthermore, all SAP ITS-related
sources, such as service files, HTML templates, or MIME files, are now stored in
the database of the system.

If your SAP system is a Unicode system, reports RUTTTYPACT and

UMG_POOL_TABLE must have been run before you start PREPARE.

If you have already executed PREPARE once or more than once, but do not want
to upgrade the system, then you must reset PREPARE.
We recommend that you replace the SAPup that has been delivered on DVD
with the latest SAPup from SAP Service Marketplace now. For more
information, see SAP Note 821032.

Phase UCMIG_DECISION
If you are upgrading a non-Unicode system, you are asked if you want to perform a Combined
Upgrade & Unicode Conversion (CUUC). If you plan to perform a Unicode conversion after the
upgrade, you must answer this question with "yes". In this case, several preparatory steps for the
Unicode conversion must already be performed on the source release or in the shadow system
of the upgrade.

If you upgrade a development system, you must create and release the transport request before
you perform the Unicode conversion of the SAP system. You cannot import a Unicode transport
request into a non-Unicode SAP system.

You decide whether to run the central instance and the shadow instance on the same host or on
different ones. If you install the shadow instance locally, PREPARE automatically creates the SAP
Service, required profiles, and directories, and copies programs and files for the shadow instance
in the Installation module. If you run the shadow instance remotely, the dialog instance that
already
exists is adjusted for use as the shadow instance.

SAP recommends that you install the shadow instance locally, provided that this does not cause a
bottleneck of resources when you run both instances at the same time.

If you want to run the shadow instance on a separate host, follow the instructions in SAP Note
430992 before you start the Installation module.

PREPARE writes its results to the CHECKS.LOG file. This file is located in the log subdirectory of
the upgrade directory. Each time you run PREPARE, it generates a header entry containing the text
Starting new execution of PREPARE modules <module name> at <timestamp> . This is followed by the PREPARE
results.
To make sure that all the requirements for the upgrade are met, you can repeat certain PREPARE
modules until their status is set to succeeded. If the CHECKS.LOG file already exists, it is
overwritten each time you execute a module again. Its contents are saved in the CHECKS.SAV file.
The entries in the CHECKS.LOG file can be error messages, information, or prompts for user action.

If you ignore open repairs, you could lose modifications.

Your SAP system contains locked SAP objects in open repairs or requests. The numbers of these
repairs or requests are listed.
Before you start the upgrade, you must release and confirm all the open repairs and requests
listed by PREPARE, since the objects in them are locked.
For more information, see Releasing and Confirming Open Repairs and Requests [page 203].

If open repairs that do not belong to this system appear, then this system was created by a
database copy. You cannot release these repairs in the normal way. For more information on
releasing these repairs, see SAP Note 62519.

To release and confirm these open repairs, you must log on to the system with the name of their
owner. The list in CHECKS.LOG contains these names.

The transport control program tp uses the table TRBAT to communicate with the SAP system.

To make sure that the configuration for SAPup is unchanged during the upgrade, the resource SAP-
R/3<SAPSID> is set offline the first time that the SAP system is stopped. This occurs with the
resource minimized upgrade strategy in the EU_IMPORT1 phase, and with the downtime-minimized
strategy in the MODPROF_TRANS
phase.

Starting the Upgrade

Prerequisites
You are logged on to the central instance as user <SAPSID>ADM.
You have completed the mandatory PREPARE modules.
You have not deleted any files in the upgrade directory. Deleting these files could cause
unpredictable inconsistencies.
You have made the preparations that are not supported by PREPARE.
You have decided on whether to use the Upgrade Assistant or scroll mode. We recommend that
you use the Upgrade Assistant.

A password is generally necessary if you want to ignore an error, or jump to a phase other than
the one currently being executed. Only enter this password when your troubleshooting has
clearly
identified the source of the error and you have checked with SAP. Support cannot be guaranteed
if you do not consult SAP. In only a very few cases is it a good idea to ignore an error with a
password.
A large number of cases are known where ignoring an error resulted in serious inconsistencies, at
worst making a recovery of the database necessary. Only in exceptional cases, such as in phases
ACT_<rel> and XPRAS_UPG, can you ignore errors with repair severe errors.
For more information on finding and correcting errors, see Troubleshooting [page 173] and
under Help in SAPup.

Some of the secondary indexes may not have been created because they were not unique. Use the
ABAP Dictionary tools (transaction SE14) to create these indexes in the database.

Make sure that your saposcol version matches your operating system, especially with respect to the
32-bit and 64-bit distinction.
For more information on how to get the latest saposcol, see SAP Note 19227.

If you want to increase the security of you message server, proceed as described in SAP Note
821875.