Sap system copy

SAP System Landscape Copy
for SAP NetWeaver and mySAP Solutions
Best Practice for Solution Management
Version Date: February 2006
The newest version of this Best Practice can always be
obtained through the SAP Solution Manager
Contents
Applicability, Goals, and Requirements....................................................................................................2
Best Practice Procedure and Verification.................................................................................................4
Planning the System Landscape Copy ..............................................................................................4
Reasons for Copying SAP System Landscapes..........................................................................4
Challenges when Copying an SAP System Landscape ..............................................................4
Define Scope of System Landscape............................................................................................5
Copy Scenarios............................................................................................................................6
System Copy Procedure – Main Phases .....................................................................................8
Copy Technologies.......................................................................................................................8
Potential Risks............................................................................................................................11
System Landscape Copy Procedure................................................................................................17
Overview ....................................................................................................................................17
Phase 1: Preliminary Tasks........................................................................................................21
Phase 2: Preparations on Target Environment ..........................................................................23
Phase 3: Preparations on Source Environment.........................................................................26
Phase 4: Copy Process..............................................................................................................31
Phase 5: Final Activities on Source Environment ......................................................................39
Phase 6: Final Activities on Target Environment........................................................................40
Further Information.................................................................................................................................51

Best Practice: SAP System Landscape Copy 2
Applicability, Goals, and Requirements
To ensure that this Best Practice is the one you need, consider the following goals and requirements.
Goal of Using this Service
This Best Practice provides the general information for copying an SAP System Landscape composed
of SAP NetWeaver and other mySAP Solutions. It provides the background information for planning
the system copy of an SAP landscape and lists the required steps for executing the copy process. Any
additional information that is required to execute these steps for a specific SAP solution or component
is provided in complementary notes available from SAP Service Marketplace.
Applicability
This Best Practice applies to copying complete components of an SAP system landscape (for example
copy of a complete SAP Web Application Server (SAP WebAS) based system). It does not deal with
the procedure or the post-customizing of a client copy within or between SAP components.
Alternative Practices
SAP’s optimization service “SAP Continuity Management” offers support with planning a system copy
strategy for SAP system landscapes. For more information please see the SAP Service Catalog at
http://service.sap.com/servicecat.
Staff and Skills Requirements
To apply this Best Practice, you must have experience in copying systems and be familiar with the
administration of the operating system, the database, the ABAP Dictionary and the SAP middleware
components used by your SAP solutions.
A technical consultant who is specifically certified for system copy should carry out the system copy
procedure onsite. This holds irrespective of whether the system to be migrated is a development
system, a test system, or a production system. This ensures that sufficient know-how is available to
handle the complexity of the procedure.
You should have a contact person available onsite who can process messages on SAP Service
Marketplace in case of problems with system landscape copy.
Duration and Timing
When copying or refreshing a system containing production data, choose a well-defined starting time
for executing the copy. The following things should be considered:
• System landscape copy may require downtime for the production (source) systems
(depending on the copy strategy selected)
• The target landscape will not be available until the complete copy process is finished
• When refreshing an existing target landscape, data in the target landscape will be lost. This
affects any work or projects done on the target landscape.
• Since many different projects or groups may be working on the target landscape, a good
coordination is required to find an acceptable time for doing a refresh.
How to Use this Best Practice
This Best Practice provides the general information for copying an SAP System Landscape (or rather:
copying all components of an SAP System Landscape). It depicts the copy procedure, which is
common to all SAP solutions. In addition to these common tasks, specific tasks may be required to
copy a specific SAP solution and all its components. This solution-specific information is provided in
complementary SAP notes, in the following called [SAP System Landscape Copy Notes], which will be
available for each SAP solution.
© 2006 SAP AG

An overview of currently available solution-specific [SAP System Landscape Copy Notes] can be
found in SAP note 885343, which also contains supplementary information like recent updates and
corrections to this Best Practice.
In addition to the above documents you will also require the standard SAP Guide “Homogeneous and
Heterogeneous System Copy for SAP Systems Based on <SAP release>” (in the following called
[System Copy Guide]) to get information on copying SAP WebAS-based components. This
documentation was written on the basis of [System Copy Guide] for WebAS version 6.40 SR1 and for
NetWeaver 2004s. For other releases of SAP Web Application Server or SAP Basis, slight differences
may occur for some detail descriptions.
For copying SAP Solutions which deploy the JAVA part of SAP WebAS on release 6.40 (JAVA system
or JAVA Add-in), you also need to refer to the SAP Guide “Homogeneous and Heterogeneous System
Copy for SAP Systems Based on SAP Web Application Server JAVA”. When referring specifically to
this guide, we use the reference [System Copy Guide - JAVA for 6.40]. For NetWeaver 2004s, system
copy for the JAVA part is included in the [System Copy Guide].
This Best Practice document consists of two main parts: a planning section and a description of the
practical steps.
The planning section provides the background information you need to design a copy strategy suitable
for your environment. It describes the different strategies and reasons for doing a system copy and
discusses the different technologies available for doing the copy. It also discusses the potential risks
that may be involved with different copy strategies.
The practical part starts with an overview of the general phases and steps that are required for
executing the copy process and then provides a detailed description of each individual step.
© 2006 SAP AG

Best Practice Procedure and Verification
Planning the System Landscape Copy
Reasons for Copying SAP System Landscapes
SAP customers use clones of their production systems for various purposes: Upgrades should first be
performed in a test system. Software development and modifications performed in a development
system should first be tested in a quality assurance system before being transported to the production
system. This helps to identify any problems that may impact the production system. Integration testing
is used to verify the interaction of processes and components in more complex system landscapes
before putting new functionality into operations.
For example, using system copies, you can use system copy to set up a landscape for:
a) Development
b) Quality assurance tests
c) Integration tests
d) Upgrade Tests
e) Backup and Restore Tests
f) Training Systems
g) Demonstration Systems
h) Operating system migration
i) Database migration
j) Change of hardware
k) Unicode conversion
While reasons a) - g) on this list imply the creation of a cloned system or system landscape (where the
source systems are duplicated), the last four items h) – k) refer to changes on the production
environment. In this latter case, the production landscape is not duplicated but moved to another
hardware or software environment instead. This procedure is also included in the terms “SAP System
Copy” or “SAP System Landscape Copy” although it does not deliver a second instance of the system
or system landscape. Unless noted otherwise, when speaking of system copy we will usually refer to a
“real” second copy created from a productive source system.
Instead of using the standard procedures for installing, setting up, and customizing SAP systems, the
SAP system copy procedure can be used to create a new target system landscape or to refresh the
data of an already existing target system landscape. The advantage of a system copy is that it enables
you to perform many system setup activities at once, because the whole system environment with its
customizing, support packages, modifications, corrections, plug-ins and other technical settings are all
copied to the target system (instead of being manually recreated). After the copy, these settings need
only to be adjusted.
Challenges when Copying an SAP System
Landscape
SAP implementations today usually exploit the capabilities of several different SAP solutions using
different SAP systems and components, all installed on separate servers. Business data is distributed
among these components and thus imposing new requirements on system copy procedures.
Providing cross-component data consistency in the target landscape is a topic to consider when
planning for system landscape copies.
© 2006 SAP AG

From that point of view, the copy of an SAP System Landscape is different from a standard SAP R/3
system copy. In addition to the standard copy tasks for SAP R/3 or SAP WebAS, each SAP solution
may require specific tasks to be executed during system copy. These need to be incorporated when
planning a landscape copy strategy for a specific customer system landscape.
When copying a landscape of connected systems, it is vital to protect the source systems from any
damage they may be exposed to from the target systems. The most illustrative example is the system
connections for exchanging messages between the systems. If a target system still has a connection
to any of the source systems, it is possible that some test activity in the target system will trigger
messages that will go into production and do incalculable damage to the production data. Thus, most
importantly, all system connections must be carefully adjusted in the complete target environment.
So when defining a copy strategy for an SAP system landscape, the following questions arise:
- Which components make up my SAP solution?
- Which systems are to be copied, which systems are required in the target landscape?
- How can each individual component be provided in the target landscape, is a system copy
required or is an installation and configuration sufficient?
- What are the consistency requirements of the target landscape?
- Which copy technology is adequate to achieve these consistency requirements? Which
components need to be copied together?
- How can I reduce or avoid the impact the system copy has on productive operation?
- How do I copy a “live” system landscape that is constantly exchanging messages? How do I
handle data that is in the message queues?
- How can I protect the production environment from any possible damage?
- What parameters do I have to change in the target environment?
- What interfaces do the components have, how do I handle them?
The following sections will discuss these questions in more detail to provide the required information to
determine on a suitable copy strategy.
Normally, to perform a system landscape copy you should perform the following steps:
• Read the planning section of this Best Practice
• Define the scope of your system landscape to be copied
• Define the requirements and potential risks in your particular case and choose the copy
procedure to be implemented
• If required, register to get a migration key from SAP for changing the operating system or
database platform
• Before starting a system copy, create a detailed project plan describing your system
landscape copy and including the systems and hardware involved, methods of copy,
milestones, test procedures for important transactions, people responsible for each phase,
and so on
• Perform the practical steps of the system landscape copy in accordance with your project plan
and the copy procedure described in this document
Define Scope of System Landscape
Copying SAP NetWeaver or a mySAP solution often requires more than just copying a single SAP
Web Application Server and its database. To plan the copy process, you need a clear understanding
what components make up your landscape and whether you also need them in your target landscape.
This usually depends on the designated purpose of the target landscape. For local functionality
testing, for example of SAP BW info cubes and SAP BW reporting tools, it may be sufficient to have
© 2006 SAP AG

an isolated copy of the production SAP BW system. Whereas, for example, for extended integration
testing before go-live of complex business processes it is mandatory to provide a test landscape of
multiple components and to make sure that data consistency between these components is given to
provide sufficient test accuracy.
So if you require copying multiple components of an SAP system landscape, you need to consider the
dependencies (data exchange) between different components and decide what level of data
consistency is required between the components of the target landscape. This again is dictated by the
application department’s demands. “Strict consistency” needs to make sure that the copy procedure
does not introduce any new inconsistencies that were not present in the source landscape. Only some
selected copy methodologies can ensure this strict consistency. But it may as well be sufficient for a
test landscape to reflect a “nearly consistent” image of the production landscape where users have to
keep in mind that some inconsistencies might have been caused by the copy procedure itself. If the
system usage does not rely on dependencies between the systems, you might not have any
consistency demands and could copy the components separately.
The above decisions influence what components of a landscape need to be copied together to
achieve the required level of consistency in the target landscape.
To decide on the appropriate copy method, you then need to know what type of data these
components are using.
• Components without application data just require to be installed once during an initial copy,
which means the initial creation of a new target landscape. For a refresh of an already existing
landscape there generally is no further action required for this type of components.
• Components with application data need to be installed during initial copy. The data can be
provided to a non-productive target landscape by two means:
o Data copy
This is the usual method for SAP WebAS based components.
o Data initialization
This is the usual method for components which only hold replicated data from other
components. The initialization is done at a later time, when the copy of the other
“main” components is finished.
Copy Scenarios
As mentioned before, the SAP system copy procedure applies both for creating a non-productive
clone of a productive landscape and for moving or migrating the production landscape itself to another
platform. When creating a non-productive system (NPS) landscape, we further have to distinguish if
the target NPS landscape has to be newly installed or if the landscape already exists and only its data
is to be refreshed by the most current data from production. For each of these different demands,
different actions may be required for executing the copy process. In order to incorporate these
distinctions into the copy process described below, we introduce different “copy scenarios”.
• PRD PRD
The target landscape will replace the previous production (source) landscape and will become
the new production landscape. This applies for example if the operating system, the database
system or some main hardware components of a production system are to be changed.
Using this scenario you would typically only copy one component of a system landscape at a
time and not the complete landscape in one. Since the system to be copied must be down,
data consistency is no issue. Similar necessities as with a complete landscape copy need to
be considered, for example regarding maintenance of system connections.
• PRD NPS
The production system landscape is copied (cloned) onto a non-productive landscape for
example for creating a test environment.
• New Installation
The target NPS landscape does not exist yet and needs to be installed during the copy
process.
© 2006 SAP AG

• Refresh
The target NPS landscape does already exist and only the data shall be refreshed.
Another distinction comes from the level of risk mitigation versus impact on productive operation
during the copy process. This leads to the following copy scenarios
• Minimum Risk
The copy procedure is mainly targeted at reducing any possible risks that may impact integrity
of any systems of the production source landscape. Various precautions can be taken that
could help to prevent messages from the target landscape reaching a production system.
Measures that can be taken include
o offline copy, which means taking the source systems offline while doing the technical
copy and/or while performing the final cleanup actions on the target systems
o stop of productive operations on source system
o locking of message queues on the source systems, so at startup, the queues on the
target systems will be disabled for message processing
o disabling RFC destinations already on the source systems, so the target systems will
start up with invalid communication destinations
o copy with empty message queues on the source systems. This eliminates the need
for processing pending messages on the target systems which may impose some
risks.
Although this scenario offers different alternatives for risk mitigation, not all possibilities for
reducing the risk may be desirable in each individual case. For further discussion of the risks
involved when doing without different precaution steps see section “Potential Risks”.
• Minimum Impact
The copy procedure aims to reduce any impact the copy process may have on productive
operation in the source systems. This procedure also prevents impacts to the source systems’
integrity but sticks to only the most important actions without taking all conceivable
precautions. With this procedure the danger of errors is higher if the vital steps are not
performed with extreme caution. This scenario should only be performed if you have a very
high level of experience and understanding of the dependencies and implications of the
complete landscape and copy process.
These above copy scenarios are not exclusive but rather need to be combined to form your individual
copy strategy. For example you may want to refresh an existing NPS landscape using online copy
procedures with minimum impact on productive operation. This means that you combine the copy
scenarios “PRD NPS”, “Refresh” and “Minimum Impact”. Concerning the measures to be executed,
the distinction between some scenarios may not even be fixed. For example you may come up with a
strategy located between “Minimum Risk“ and “Minimum Impact“ by executing only some steps of the
“Minimum Risk“ scenario while skipping others.
Depending on the target environment of the copy, we further need to distinguish between a copy to a
“compatible” environment or an environment where operating system or database are different.
• Homogeneous System Copy
Operating system and database system are the same for both source and target systems.
• Heterogeneous System Copy
The operating system or database system of the target system is different from the source
system (also called OS/DB Migration).
If you want to change the operating system or the database system during the copy of a
productive SAP WebAS based system, you need to order and use the SAP OS/DB Migration
Service. This service is also available for SAP NetWeaver and mySAP solutions.
Heterogeneous system copy requires a migration key that will be provided by SAP. The copy
must be done offline and allows using only the SAP R3load tool (see copy technologies
below). OS/DB Migration must be performed by consultants who have been specifically
trained for these types of migrations.
© 2006 SAP AG

For information concerning OS/DB migration, see the SAP OS/DB Migration page in SAP
Service Marketplace (http://service.sap.com/osdbmigration).
Since the copy procedure is the same for homogeneous and heterogeneous system copy, we do
not further distinguish both in the process descriptions to follow.
System Copy Procedure – Main Phases
The actual system copy process can be divided into six main phases. These are:
1. Preliminary tasks: Plan and prepare the system copy procedure. Check that all required
information is available and that the copy strategy is complete. Prepare or check the target
hardware and software environment.
2. Prepare the target environment for the system copy: This phase is mainly required for
refresh, when the target landscape already exists and shall be provided with new data from
the source (production) landscape. It includes saving configurations and data from the target
systems and preparing them for the refresh.
3. Prepare the source systems for the system copy: Perform general and technical
preparations in the source systems in accordance with your selected copy strategy. The
amount of actions required here depends on the copy scenario chosen. The copy scenario
also determines the impact (and most of all the downtime) imposed on the source landscape.
4. Perform the system copy: Run a homogeneous or heterogeneous system copy of the
source SAP WebAS systems using one of the copy technologies described below.
5. Final activities on source landscape: Depending on the selected copy scenario the
production systems must be put back to their original state so productive operations can be
resumed. This phase may be delayed until after phase 6 to reduce any possible risks to an
absolute minimum.
6. Final activities on target landscape: In the newly copied systems, subsequent technical
actions and post-installation steps must be performed -- most importantly to prevent the target
environment from interfering with the production systems. This includes changing system
connections and other cleanup activities but also renaming and reconfiguring to meet the
requirements of the target landscape. Installation of additional components that are not copied
is usually done in this phase as well.
More details on the actual steps to be executed during each phase are provided in section ‘System
Landscape Copy Procedure’ below.
The phases 2 – 6 must be done in parallel for all components of the landscape that are to be
copied together.
For every component you also need to observe the information from the solution-specific
[SAP System Landscape Copy Notes].
Copy Technologies
The actual copy of a component (the component’s data) can be done using different tools or
technologies.
R3load
R3load is a tool provided by SAP to export and import the database of an SAP system in a database-
independent format. R3load is integrated with the SAP installation tool SAPInst.
© 2006 SAP AG

SAPInst is first run on the source system to export the source database (option “Export ABAP
database Content”). On the target host SAPInst is run to install the target system and import the
previously exported data from the source system (using installation option “System Copy/Migration”).
R3load requires to shutdown the source SAP WebAS while doing the export, since otherwise no
consistent state can be exported. R3load can only export database objects or structural information
known to the ABAP dictionary. R3load takes care of renaming the SID and database schema while
doing the system copy.
R3load only exports the ABAP part of an SAP WebAS installation. For JAVA export, see the Jload tool
below.
For more information on R3load see [System Copy Guide].
To optimize the system copy and reduce downtime using R3load, several additional tools are
available, like Incremental Migration (see below), Migration Monitor (see [System Copy Guide]), Time
Analyzer (see SAP note 784118), Package Splitter (see SAP note 784118) and Distribution Monitor
(see SAP note 855772). Also see “http://service.sap.com/systemcopy Optimization”.
Using the R3load procedure is mandatory for Heterogeneous System Copy or
Unicode Conversion. For heterogeneous system copy of a productive system, the
SAP OS/DB Migration Check is mandatory, see http://service.sap.com/osdbmigration.
IMIG
The Incremental Migration (IMIG) is an extension to the R3load procedure that aims at reducing the
production downtime by doing part of the data export for large tables during uptime. It is thus
applicable for very large systems or large systems with few very large tables when downtime with
conventional procedures is not acceptable.
IMIG has not been generally released yet and can only be performed by selected
consultants. It is not available for all SAP solutions. See SAP note 693168 or
http://service.sap.com/systemcopy for more information.
Jload
Jload is a tool provided by SAP to export and import the JAVA components of an SAP WebAS
system. Jload is available starting with SAP WebAS 6.40 and SAP NetWeaver 2004s. Jload works for
both JAVA system and JAVA Add-in installations of SAP WebAS. It exports the JAVA database in a
database-independent format as well as the configuration and file system components of an SAP
WebAS JAVA system (the “JAVA central instance”). Jload is integrated with the SAP installation tool
SAPInst.
SAPInst is first run on the source system to export the JAVA source system. On the target host
SAPInst is run to install the JAVA target system and import the previously exported data from the
JAVA source system.
For copying an SAP WebAS ABAP + JAVA (Java Add-in), the ABAP and JAVA export with SAPInst
can be combined. The order and exact procedure depends on the actual release of the WebAS
system as described below.
See [System Copy Guide - JAVA for 6.40] (WebAS 6.40) or [System Copy Guide] (NetWeaver 2004s)
for more information on the Jload procedure.
© 2006 SAP AG

Please see SAP note 785848 respectively 870863 for possible restrictions and for a list
of JAVA components currently released for system copy with Jload.
Copy of WebAS - JAVA components
If not using the Jload tool to copy the JAVA database of an SAP WebAS JAVA or SAP WebAS
ABAP+JAVA system, but one of the methods described below (e.g. database restore), there is still
some JAVA-specific content in the filesystem which needs to be copied as well, both for New
Installation and Refresh. With WebAS 6.40 and NetWeaver 2004s, the SAPInst tool provides the
required functionality to export this filesystem content of a JAVA system (the JAVA central instance)
and import it into the target system. When choosing the SAPInst option “Copy the database by using
database tools”, SAPInst will skip the Jload database export and only export the source JAVA central
instance required to install or refresh the target JAVA central instance.
Database Restore
A system copy can be done by restoring an offline or online backup of the source database onto the
target system. The restore procedure must allow restoring the database with a different SID unless the
target system shall have the same SID as the production system.
The target system must already be installed before doing the restore and you must make sure that all
database and SAP profiles for the target system are correct. Especially when using an online backup,
analyze the risks given for copy scenario Minimum Impact.
The database restore method is supported by SAPInst for a New Installation of the target system.
See [System Copy Guide] for more information on the Database Restore method for the different
database systems.
Database-specific Methods
In addition to the database restore method, other database-specific tools or procedures may be
available. These methods include for example copy of the database files or special tools for unloading
a source database and loading the data into a target database.
See [System Copy Guide] for more information on such SAP supported database-specific methods.
Split-Mirror
Split-mirror technology allows creating a copy of the production disk volumes at storage system level.
An additional set of storage volumes to be used for the target system are synchronized with the
production disks. At a suitable time, this mirrored relation is discontinued (the mirror is split), causing
the target volumes to contain a point-in-time image of the source system. After connecting the target
host to these volumes and doing the final activities required for the target environment, the target
system can be brought up.
Split mirror technology allows for non-impact online copy and very fast refresh of an already existing
target system. All disk volumes can quickly be refreshed using a storage-based local or remote
mirroring technology. With the appropriate experience, it may also be possible to copy the complete
SAP WebAS and database instance (software) using split-mirror so it may not even be necessary to
install the target systems first. Please note that this requires a comprehensive understanding of the
implications and the necessary adjustment measures.
Because of its small impact and small runtime, Split-mirror is a solution especially suited to copy large
systems where other copy technologies cannot be accepted because of long downtime or runtime
requirements.
When combining split-mirror technology with Consistency Technology offered by various storage
system vendors, it becomes possible to even create a copy of multiple systems or of a complete
system landscape during online operation. Thanks to SAP’s fault-tolerant, stateful messaging
technology used for data exchange between systems (SAP asynchronous RFC and SAP XI
© 2006 SAP AG

messaging protocols), consistency technology will provide a data consistent image of the landscape
from an application point-of-view.
For more information on split-mirror copy see http://service.sap.com/split-mirror.
For more information on Consistency Technology see http://service.sap.com/atg Backup & Restore
Federated Backup.
Copy of non-WebAS components
For copying non-WebAS based components no SAP copy tools are available. Such components
usually are installed newly in the target landscape.
For non-WebAS based components that hold data of their own (like SAP liveCache), the data can
either be initialized after the copy of the main SAP WebAS based components (like SAP APO) or it
can be copied in parallel to the other systems using an appropriate technology (e.g. database copy).
What option is applicable for a specific component used in an SAP solution is described in the
solution-specific [SAP System Landscape Copy Notes].
Potential Risks
Depending on the copy scenario or the copy technology to be used, different risks may be implied.
Make sure that you consider these risks when designing your landscape copy strategy and that you
have answers how to deal with them.
Risk 1: Avoid interactions with productive environment
After the system copy, the new target systems must not be able to connect to the production systems.
Otherwise, at the first system startup, data might be transferred automatically to the production
system, thus destroying the systems integrity and making the production data inconsistent.
The following things are critical concerning a clean separation of source and target environment and
must be absolutely resolved on the target environment before bringing the target systems to operation:
• Disable all interfaces pointing to any components of the production landscape
This is vital to prevent any communication or data exchange of the target systems from
reaching the source systems. All interfaces of the systems need to be checked and adapted,
including interfaces to the “real” world to prevent the systems from sending as if they were
productive.
Possible interfaces of SAP systems that need to be checked and adapted are
RFC and R/3 connections, HTTP, R/2 connections, CPIC, File, Batch input, DB links,
registered programs, adapters, archive link, SAP phone, SAP connect, email
• Adapt transport settings (TMS)
This is vital to prevent any objects reaching the source systems via incorrect transport routes.
• Adapt NFS mounts
This is to prevent the target systems from unintentionally accessing or writing NFS mounts of
the production environment.
• Check batch jobs
This must prevent that any batch jobs in the target systems trigger production-like activities.
• Adapt printers, check print jobs
This must prevent the target system from creating output on productive printers which could
not be distinguished from real production print output.
• Modify users and authorizations
To prevent production users from accidentally logging on and working on the target systems,
we recommend using different users in an NPS landscape.
© 2006 SAP AG

We also recommend using RFC users in NPS landscape that are different from RFC users
used in production. This can serve as additional protection from illegal execution of messages
in the wrong landscape.
• Adapt logon groups and application servers
This must prevent the message server of a target system from routing requests to application
servers of the productive system.
• Check external commands
This must prevent a target system from calling operating system commands intended only for
a production system.
All these actions should be done before or immediately after starting a target system.
If you cannot make absolutely sure that the production environment will be
unaffected, we recommend separating the network environments of source and target
systems.
SAPInst automatically starts up the target SAP system during installation.
If this imposes incalculable risks to your source environment due to the copy
scenario chosen, you should consider taking additional precautions or choosing
another copy scenario (at least for the initial installation of the target system).
To disable any automatic server connection before you are able to do the above cleanup tasks, you
should consider separating the network environment of source and target landscapes e.g. by using
one of the following methods:
• Install the target landscape in its own network environment.
• Deactivate the network routes between the target servers and the production servers.
• Prohibit the connections within the SAProuter (see SAP Note 30289).
• Remove the new server from the Name Server service (DNS / WINS) and delete entries for
the source systems from the hosts / lmhosts file, to prevent any network connection to the
source system.
Risk 2: Online Copy
Online copy means performing the system copy without taking the source systems out of operation.
Online copy is possible by restoring an online backup of the source systems or by using split-mirror
technology.
Online copy imposes special risks because you have no possibility of already taking some precautions
in the source systems before the copy that aim to prevent data exchange being triggered by the newly
copied target system. Using online copy you cannot for example disable RFC destinations or lock
message queues beforehand in the source systems. It is also not possible to empty message queues.
Since the source systems stay online, the target systems could initiate some interactions with the
source environment right after startup and thus destroy the source systems integrity.
So using online copy requires careful preparations for avoiding any such interactions at startup of the
target systems. In addition to the measures described under “Risk 1”, any possible measures should
© 2006 SAP AG

be taken that can prevent SAP processes from running unintentionally in the target systems. On the
target systems you can:
• Prevent automatic start of batch jobs so these cannot trigger any unintended activities.
• Disable unintentional user logon or user activity by locking the system and/or by locking SAP
users.
• Disable system connections / RFC destinations per SQL statement directly on the database,
before starting the SAP WebAS target system. This step can help to prevent any messages
from being transferred from the target to the source environment. The actual adjustment of the
destinations to their final values (the real servers of the target environment) can be done later
using the official SAP tools.
Online copy for filesystem components of a system (e.g. the JAVA central instance) bears the risk of
inconsistent files and inconsistent data which may leave the copy unusable!
Risk 3: Changing the Logical System Names
In any system landscape where data is distributed and exchanged between different systems, there is
a potential risk when changing the configuration parameters responsible for data exchange. These
parameters include RFC destination, host names, and logical system names. Among these
parameters, changing the logical system names might have the strongest impact on a systems
internal integrity and its ability to communicate.
Definition of logical systems
A logical system is an application system in which the applications work together in a common
database. In SAP terms, a logical system is a client in the database of one system of an SAP
landscape. A logical system can be for example a client in an SAP R/3 system, a client in an SAP
CRM system or even a non-SAP system. Messages are exchanged between logical systems.
To identify a system as communication partner, participating systems in a distributed environment
must each have a unique ID, the logical system name. This name is assigned explicitly to one client in
an SAP System.
A logical system may be compared with an address on a letter. To ensure that data is sent to the
correct recipient (or to the correct address), logical system names must be unique among the systems
belonging to an SAP landscape exchanging data.
Possible issues when changing logical system names
To change a logical system name that is already used or defined in a system, transaction BDLS can
be used. This program searches all relevant tables and converts the corresponding entries referring to
that logical system name. But changing logical system names bears the risk of some data in unknown
tables not being converted by BDLS. This would lead to unexpected data inconsistency or improper
application work. The results are unpredictable and may have a big impact on business processes.
Transaction BDLS dynamically identifies these relevant tables by certain domains (LOGSYS and
EDI_PARNUM). But if some application uses logical system names within tables that do not reference
to these standard domains, the data cannot be converted by BDLS. Such data can also not be
converted if some references to logical systems are saved as part of fields in cluster or pool tables.
For SAP standard solutions, such critical tables should be described in the [SAP System Landscape
Copy Notes]. A possible risk mostly remains for customer-owned developments.
Another issue when changing logical system names is the generally long runtime of transaction BDLS.
This extends the duration of the system copy procedure and thus the unavailability of the systems.
Is it necessary to change the logical system names?
Because of the two reasons mentioned above (long runtime of transaction BDLS and possible
inconsistencies), it may be desirable, to avoid the renaming of logical systems during system copy.
Whether logical systems names need to be changed at all after a system copy depends on the type of
the target system:
© 2006 SAP AG

Scenario “PRD PRD”
The main requirement for the system copy procedure is that the new production system is identical to
the old one. For communication to work properly after the system copy, the new production system
must have the same logical system names as before, a renaming is unnecessary.
Scenario “PRD NPS”
From a purely technical point-of-view, after a system copy it is sufficient to adapt all RFC destinations
so that there is no longer a connection to the original system landscape. Messages to an “old” logical
system will then be routed to the new server given in the RFC destination. On the other hand,
customers may want to rename the logical system names in their NPS landscape because these often
include the SID of the production system.
So generally, renaming of logical systems can be avoided by using non-unique logical system names
throughout a landscape of production, quality assurance and development systems. To prevent any
confusion in that case, the logical system names should not include the SID in their names.
But whether it is possible to use non-unique logical system names requires careful evaluation since
some central components of an SAP system landscape like SAP Central User Administration or SAP
Solution Manager do not only connect to the production landscape but also to quality assurance and
development systems. For these systems, different logical system names may be mandatory and the
renaming during system copy cannot be avoided.
Conclusion
• Never change the logical system names in production systems!
• For non-production systems, it may be possible to avoid the renaming of logical systems by
using non-unique logical system names.
Risk 4: System Copy with Nonempty Message Queues
When allowing production downtime for the system copy (scenario “Minimum Risk“), one could try to
empty the message queues by stopping all system activity and suspending all external interfaces
some time before the copy. If it was thus possible to prevent new messages from reaching the
systems to be copied, it would be possible to wait until all message queue entries are processed. This
ideal situation, where all queues would be fully processed and contain no entries when doing the copy
will usually not be achievable in reality. Even with offline copy, this could extend downtime beyond the
tolerated limits. For example, there may be a high load in the system 24 by 7 and it may take too long
for all the queues to be fully processed. So to minimize downtime you may need to copy the systems
with entries in the queues and process them after the system copy.
Also with scenario “Minimum Impact“, stopping the message queues in production would be too
restrictive.
If the queues are not empty when doing the copy, they will contain valid data in the target systems
after the copy. Processing these queues will be important for application data consistency in the target
landscape (also see the discussion of data consistency in “Risk 5”).
A potential problem arises for such message from before the system copy if the logical system names
need to be changed after the system copy. Queue entries often contain a reference to the old logical
system names as a part of their fields. If the logical system names are changed, these entries can no
longer be processed in the target landscape. This would lead to data loss or inconsistencies.
Currently, no automatic tools or methods are available that can process entries in the queues after
changing logical system names. So the question of how to proceed in case there are some entries in
the outbound or inbound queues before system copy is critical.
Let us consider the different copy scenarios from above:
In this case, because the logical system names are not changed after the system copy, the newly
copied systems are identical to the source ones and the queues can be processed regularly after
system startup.
© 2006 SAP AG

In this case, to process the queues, the old logical system names are required. A possible workaround
is to initially start the newly copied systems with the old (unchanged) logical system names. After
adjusting the RFC destinations and starting the inbound and outbound queues, these messages can
be processed since there are no differences between source and target systems from the
communication point of view. After all entries have been processed successfully, the logical system
names can be changed. Since the target systems are not released to operations yet, no new
messages will be created and it is possible to wait for the queues to become empty.
During this initial procedure, before the logical system names are changed, you need to make sure
that no messages can accidentally reach the source (production) landscape. To guarantee this, the
newly copied systems should run in their own network environment that is separated from the source
systems. This prevents any messages from being sent to the source systems, for example in case
some system connections were not adjusted accidentally.
It is vital that the RFC Destinations were adjusted to the target environment before
processing these messages!
Otherwise the messages would go into the production landscape.
System copy with non-empty message queues is critical with SAP Exchange
Infrastructure (SAP XI) systems!
On the target system, you may not process messages in status “scheduled for
outbound processing”!
Since these messages already went through receiver determination, the receiver hostname
is already contained in the message. Such messages would always go into the production
landscape when processed on the target SAP XI system.
With SAP XI you need to ensure that no such messages exist in the system when doing the
copy.
Risk 5: Data inconsistencies in target landscape
When copying a landscape of multiple connected systems, you need to consider data consistency
between the systems. Data consistency in this context means that application data that is exchanged
between the systems has a matching state regarding data modifications transferred by messages.
The consistency requirements when doing system landscape copies may vary widely, mainly
depending on the type of applications and the scope of projects to be run on the target landscape. You
need to decide in coordination with the designated users of the target landscape which systems need
to be included in the copy and what level of consistency is required between the different systems
(also see section “Define Scope of System Landscape” above).
If the target system will become the new production system, data must be consistent and identical in
the target and source systems. This can always be achieved since the copy must be done offline. In
case the copy is done with nonempty message queues, these messages must not be deleted.
For scenario “PRD NPS”, the consistency requirement may range from
• No consistency required
In this case, there are no dependencies between different systems or components of your
target landscape. This for example suffices if you want to run local functionality tests or
technical upgrade tests that are not related to any other connected systems.
© 2006 SAP AG

You can copy or refresh a single component independently of other components of the target
landscape. Any pending messages in the queues of a component should be deleted after the
copy since they cannot reasonably be processed if the other systems have a different state.
• Only near consistency (but not 100% consistency) required
This applies if you need to achieve a high level of data consistency between the systems, but
it is not relevant if some inconsistencies are caused by the copy procedure. This option may
be sufficient for example for demo or training systems. This level of consistency is interesting
since it can be achieved without any special consistency technology even with online copy.
In this case, all systems must be copied at the same time. Pending messages in the queues of
the systems may be processed or deleted, providing either some more or some less data
consistency.
Technologies to achieve this level of consistency are:
o Point-in-time restores of an online backup of each database to a similar point in time
o Online split-mirror copy without using consistency technology options
• Strict consistency (100% consistency) required
Strict consistency will usually be demanded in a test landscape to be used for integration or
functional upgrade tests, but also training or demo systems may have this level of
requirement.
For this, all systems of the source landscape in scope need to be copied at the same time. To
ensure full data consistency in the target landscape, all systems must either be shut down in
order to execute an offline copy or, when online copy is needed, some special consistency
technology must be used.
Any pending messages in the queues of the systems must be processed in the target
landscape after the copy was done.
Options to achieve this level of consistency are:
o Offline copy (shut down all systems during the copy)
o Online split-mirror copy using Consistency Technology offered by multiple storage
vendors (see http://service.sap.com/atg for more information)
o Online copy exploiting the Coordinated Suspend methodology (see
http://service.sap.com/atg for more information)
Conclusion
To provide a consistent copy of a source landscape, all components need to be copied together at the
same time using an appropriate technology. The copy phases and steps as described later in this
guide must be executed in parallel for all components of the landscape.
In case no consistency is required for a target system or target landscape, each component can be
copied separately without regards to copying any other components. If you are still using a common
copy procedure (for example because an appropriate technology is in place), you can renounce from
processing any pending messages and just delete them.
© 2006 SAP AG

System Landscape Copy Procedure
Overview
The following table provides an overview of all steps to be executed during the system landscape copy
process. These steps are grouped into 6 main phases. On the right side, this table shows whether a
step is applicable for the different copy scenarios introduced above.
For executing these steps you will need the detail information given below and also need to check the
solution-specific [SAP System Landscape Copy Notes] for additional details to be observed during
execution of each step for a specific SAP solution or component.
The phases 2 – 6 must be done in parallel for all components of the landscape that are to be copied
together.
PRD NPSPRD PRD
Installation Refresh
Step
Whereto
execute
Description System
Type
Min
Risk
Min
Imp.
Min
Risk
Min
Imp.
Min
Risk
Min
Imp.
Phase 1: Preliminary Tasks ↓
1.1
Get required
documentation
X X X X X X
1.2
Determine a clear copy
strategy
X X X X X X
1.3
S Build up desired initial
state of source system
Any X X X X
1.4
T Prepare the hardware
and environment for the
target systems
Any X X X X
1.5
T Check the hardware and
environment for the
target systems
Any X X
1.6
S,T User Information X X X X X X
Phase 2: Preparations on Target Environment ↑ ↓
2.1
T Save important
information of
components of the
target landscape
WebAS X X
2.2
T Finish operations on
target systems
Any X X
2.3
T Do solution-specific
preparations
Any X X
2.4
T Stop components of the
target landscape
Any X X
2.5
T Save other configuration
data of the target
landscape
Any X X
© 2006 SAP AG

PRD NPSPRD PRD
Step
Whereto
execute
Description System
Type
Min
Risk
Min
Imp.
Min
Risk
Min
Imp.
Min
Risk
Min
Imp.
2.6
T Backup target landscape Any (X) (X)
2.7
T Prepare any other
systems outside the
target landscape
Any X X X X
Phase 3: Preparations on Source Environment ↑ ↓
3.1
S Prepare source systems X X X X X X
3.2
S Save configuration
information
X X
3.3
S Stop productive
operation in source
systems
X X X X
3.4
S Complete message
processing in source
landscape
WebAS X X X * X X *
3.5
S Disable communication
between source systems
WebAS X X X
3.6
S Shutdown source
landscape
Any X X X X
3.7
S Backup source
landscape
Any X X (X) (X) (X) (X)
3.8
S Prepare any other
systems outside the
source landscape
Any X X X X X X
Phase 4: Copy Process ↑ ↓
4.1
S Capture data of SAP
WebAS source system
WebAS X X X X X X
4.2
T Start installation of SAP
WebAS target system
WebAS X X X X
4.3
T Copy the data using the
selected method
WebAS X X X X X X
4.4
T Complete installation of
SAP WebAS target
system
WebAS X X X X
Phase 5: Final Activities on Source Environment ↑ ↓
5.1
S Startup production Any X X
5.2
S Re-enable
communication between
source systems
WebAS X X
5.3
S Perform follow-up
activities from [System
Copy Guide]
WebAS X X X X
© 2006 SAP AG

PRD NPSPRD PRD
Step
Whereto
execute
Description System
Type
Min
Risk
Min
Imp.
Min
Risk
Min
Imp.
Min
Risk
Min
Imp.
5.4
S Restart productive
operation in source
systems
Any X X
Phase 6: Final Activities on Target Environment ↑
6.1
T Execute follow-up
activities on OS and DB
level
WebAS X X X X X X
6.2
T Prevent any
unintentional actions on
target system
WebAS X X X X X X
6.3
T Start target system WebAS X X X X X X
6.4
T Adjust system
connections
WebAS X X X X X X
6.5
T Execute follow-up
activities from [System
Copy Guide]
WebAS X X X X X X
6.6
T Process pending
messages
WebAS (X) X (X) X (X) X
6.7
T Convert logical system
names
WebAS X X X X
6.8
T Do solution-specific
adjustments
Any X X X X X X
6.9
T Import saved information
from step 2.1
WebAS X X
6.10
T Register queues WebAS X X X X X X
6.11
T Install new SAP license WebAS X X X X X X
6.12
T Install and configure
additional components
Other X X
6.13
T Check and adjust
integration to other
components
Other X X X X X X
6.14
T Load or initialize data in
additional components
Other X X X X
6.15
T Check operability and
data consistency
Any X X X X X X
6.16
T Release target
landscape for operations
Any X X X X X X
© 2006 SAP AG

Notes: The column “Where to execute” indicates whether a step is executed on the source (S) or target (T)
systems. The column “System type” denotes for what kind of system components a steps is applicable:
• “WebAS” stands for steps that are done for SAP WebAS based components.
• “Other” stands for non-WebAS based components
• “Any” stands for WebAS and non-WebAS based components
* required for SAP XI, see description below
The following list summarizes some of the most important aspects discussed in the first part of
this Best Practice:
The copy procedure for WebAS-based systems consists of similar steps. Therefore, the
following procedures are written to be executed in parallel for each system or component of
your landscape. For specific SAP solutions, solution-specific steps may be required. These
are described in separate notes. You can follow the procedure step by step, checking for
each step whether a solution-specific action is required.
To avoid data inconsistencies between the systems of the target landscape, do not copy an
individual system but the entire system landscape.
A system copy procedure may require additional actions that are special only for your
implementation. During the system copy planning phase, identify and document any such
additional actions not described in the SAP documentations.
When moving or migrating a production system (scenario “PRD PRD”), no changes may
be done to the SID, logical system names or RFC destination names.
When building up a new system through system copy (scenario “PRD NPS”), the new host
name and SAP System name can be chosen freely.
Implement a sound strategy for logical system naming in an NPS landscape that, whenever
possible, spares renaming logical systems after a system copy
During system copy, all non-WebAS components of a landscape should generally be
installed manually using the standard Installation Guides. This can be done either before or
after copying the WebAS-based system.
After you have executed the technical installation and database copy, the new target systems
and components must be adjusted to the new environment as described in this document.
After system copy, the target systems must not be able to connect to the production systems
to avoid any damage due to misguided messages or function calls.
We recommend that you document all the actions performed.
Thoroughly test the selected copy technology and copy procedure including all required post-
processing activities
Check for any possible restrictions with the selected copy technology or SAP solution to be
copied
Take care when copying an SAP XI system with non-empty message queues, never process
messages with status “scheduled for outbound processing” in the target landscape.
Executing the system copy for SAP WebAS JAVA is only required if the source system
actually includes a J2EE engine. If a source system was for example upgraded from a
previous release, it often only contains an ABAP stack without J2EE engine and there is no
need for the JAVA copy even if this is offered during installation of the target system.
© 2006 SAP AG

Phase 1: Preliminary Tasks
Step 1.1 Get required documentation ↑
Task of step
Ensure that you have the current version of all required documentation available for doing the system
landscape copy.
What to do
Check if all documentation is available and up-to-date (also refer to section “How to use this Best
Practice”)
• This Best Practice, which is available from http://service.sap.com/solutionmanagerbp.
• SAP note 885343 providing updates and corrections on this Best Practice
• SAP Guide “Homogeneous and Heterogeneous System Copy for SAP Systems based on
SAP Web Application Server ABAP” [System Copy Guide], available from
http://service.sap.com/instguides. Choose the release corresponding to your SAP WebAS
installations.
• When copying SAP components based on SAP WebAS JAVA:
SAP Guide “Homogeneous and Heterogeneous System Copy for SAP Systems based on
SAP Web Application Server JAVA” [System Copy Guide - JAVA for 6.40], available from
http://service.sap.com/instguides. Choose the release corresponding to your SAP WebAS
installations.
• Additional documentation and SAP notes referred to by the [System Copy Guide] and [System
Copy Guide - JAVA for 6.40]
• [SAP System Landscape Copy Notes] for all solutions included in the scope of your
landscape. An overview of those notes currently available is contained in SAP note 885343.
• Possible additional SAP notes mentioned in these [SAP System Landscape Copy Notes].
Step 1.2 Determine a clear copy strategy ↑
Task of step
Before starting the system copy, you need to understand clearly how it will be performed in your
particular case. The information given in the first part of this best practice will assist you in determining
an appropriate landscape copy strategy.
What to do
Work out a sound copy strategy by performing the following actions:
• Define the key requirements that the newly copied systems should meet.
• Define scope of landscape to be copied (which systems belong to the landscape to be copied)
• See solution specific SAP notes for an overview of components belonging to a SAP solution
landscape. If available you can also check your business process descriptions or swim-lane
diagrams to find out which components are involved in the business processes you require in
the target landscape.
• Define the system copy procedure based on the key requirements and potential risks.
• Decide on the copy technology to be used
• Define the exact changes that will be made during the system copy to hardware, host names,
operating system, database, SID, logical system names, and so on. Decide how logical
systems from production will be mapped to logical systems in the target environment.
© 2006 SAP AG

• Estimate the potential risks created by these changes: analyze the possible workarounds and
ask if it is really necessary to make the changes.
• Prepare the detailed project plan for system copy, including the verification procedures after
the system copy.
• Contact SAP if you are not sure about parts of the system copy procedure or potential risks.
Step 1.3 Source Build up desired initial state of source system ↑
Applicable for scenario “PRD NPS“.
Task of step
If desired, apply any lacking changes to the source systems which should be present in the target
systems as well. This step can spare from applying required modifications twice.
What to do
• Make sure that all desired transports were imported into production.
• Apply all SAP recommendations from SAP Service Sessions and any SAP recommendations
resulting from customer problem messages in the source systems
• Implement all currently available SAP Support Packages in the source systems
Step 1.4 Target Prepare the hardware and environment for the target systems ↑
Not applicable for Refresh of the target landscape.
Task of step
As the system environment fundamentally influences the installation and system copy procedures, it is
important to have a clear understanding of the source and target landscape before you start the
system copy. Before you can begin with the practical tasks of installing or copying the components of
the target landscape, you need to plan the setup of the landscape and prepare the environment
accordingly.
What to do
Decide which components you need in your target system landscape, work out how these
components must be distributed to hosts and calculate the sizing of the hardware involved.
• Define and size the target hardware environment
• Determine SID of target systems
• Get required installation CDs, check the required software releases, patches, …
• Check if you have the appropriate tool versions for your current SAP kernel
• Prepare the target host for the SAP installation
o Do the preparations described in [System Copy Guide]
o When copying an SAP WebAS 6.40 JAVA system, also prepare the target host
according to the preparations listed in [System Copy Guide - JAVA for 6.40]
• Get Migration Key when doing a heterogeneous system copy (migration), see
http://service.sap.com/osdbmigration
• When using SAPCRYPTOLIB for encrypting data, it may be required to order a new license
ticket for the target system (see SAP note 662340)
For more information, refer to the installation and configuration guides. Normally, on the basis of
information about the expected workload, applications to be implemented and number of users, an
SAP hardware partner can recommend a feasible configuration.
© 2006 SAP AG

Step 1.5 Target Check the hardware and environment for the target systems ↑
Applicable for Refresh of the target landscape.
Task of step
Check if the target environment still fits to serve as target for copying the source system because the
source environment may have changed since the last copy or refresh was done.
What to do
The following things should be done before each refresh of an existing target landscape:
• Define scope of landscape to be refreshed
• Compare components used in source and target landscapes: Are there any new components
used in the source or target landscape?
• Check target disk space: Is it still sufficient to host the current source database size?
• Compare the software level of source and target components. If the source versions have
changed, the target components may need to be adapted correspondingly. Check operating
system, database and SAP software versions. Check versions / patch levels of SAP WebAS
and non-WebAS based SAP components.
Step 1.6 S / T User Information ↑
Task of step
Inform users about the planned copy or refresh. Users of the target system need to be aware of the
refresh since some information in a NPS will get lost. They need to coordinate their projects and know
when the system is not available. Users of the production systems need to be informed about the
downtime when choosing an offline copy method.
What to do
• Email time schedules to users and project owners working on the target landscape
• Communicate scheduled downtime of production systems (if applicable)
Phase 2: Preparations on Target Environment
Step 2.1 Target Save important information of components of the target landscape ↑
Task of step
Since the refresh will delete any specific data created in the target systems, save any data or
configuration information that needs to be preserved before refreshing the target systems. This
information can later be re-entered into the target systems.
What to do
Make notes of any important information that is required to repeat the target landscape configuration
after the refresh. Export any data that needs to be preserved, for example using the SAP transport
mechanism, or take screenshots of the current configurations.
Information you may want to preserve can include for example:
• Users
If you want to preserve users, user settings and authorizations of your target system for later
© 2006 SAP AG

re-import after the refresh (for example if you are using different users than in production), you
can export them to a transport using transaction SCC8, parameter SAP_USER. After the
refresh this transport can be reimported into the target system using transaction STMS,
followed by import post-processing with transaction SCC7. For more information on SCC8 and
SCC7 see topic “client copy” in the SAP online documentation.
If your target system is connected to a system for central user administration (CUA), the target
system users can also be redistributed from the CUA system after the copy.
• Printers
Export to a transport can be done with transaction SPAD Output devices Edit
Transport, also see SAP note 64337 or the Online Documentation.
• Pending transports which were not imported to production systems
• Current transport number
Note the current transport number given in field ‚trkorr’ of table E070L. This information might
be required to reset to an appropriate number after the copy, also see SAP note 12799.
• Logon groups
Take screenshot of current settings in transaction SMLG.
• List of logical system names
Take screenshot of current settings from transaction BD54 or table V_TBDLS.
• RFC destinations
Take screenshots of settings in transaction SM59.
• Connections to legacy systems
• ALE partner profiles
Export can be done with transaction BD64.
• List of outbound destinations and registered inbound queues
Make screenshots or save the list output of transactions SMQS and SMQR.
• List of RFC users
A list of RFC users can be obtained with transaction SUIM User By logon date and
password change.
• Operation modes
Note the operation mode definitions in transaction RZ04.
• System messages
If required, note the current system messages as displayed by transaction SM02.
• Tests results
• If the system copy is going to overwrite the current TMS domain controller, the domain
controller must be moved to another system. After the copy, it can be moved back if desired.
See SAP online documentation for more information.
Check if you need to preserve any other information not listed above. Consider if there’s also
information on non-WebAS based components that needs to be noted or preserved.
Also check the solution-specific [SAP System Landscape Copy Notes] for possible additional objects
or activities.
Step 2.2 Target Finish operations on target systems ↑
Task of step
Finish operations on the target systems. Ensure consistency to systems outside the target landscape
that are not refreshed during this procedure.
This step is also important to provide a useful state for backup for the fallback scenario (see step 2.6).
What to do
• Finish any work in the target systems.
• Lock users except DDIC and maybe other special users
© 2006 SAP AG

• Check and stop batch jobs. This can be done using transaction SM37 or report BTCTRNS1.
• Complete queue processing on all communication channels to systems of the NPS landscape
that are not refreshed
Step 2.3 Target Do solution-specific preparations ↑
Task of step
Some SAP solutions may require additional preparations to be done before a system refresh.
What to do
Check the solution-specific [SAP System Landscape Copy Notes] for the solutions you are copying
and execute any actions listed there.
Step 2.4 Target Stop components of the target landscape ↑
Task of step
The target landscape must be shut down before the data can be copied.
What to do
Shutdown all components of the target landscape. See specific [SAP System Landscape Copy Notes]
for a list of components. If more information is required (for example concerning the order of the
components’ shutdown), also check the solution-specific operation guides.
Step 2.5 Target Save other configuration data of the target landscape ↑
Task of step
Save any other configuration information of the target environment that could be overwritten by the
copy procedure (like operating system or database profiles). Such data is stable after the components’
shutdown.
What to do
On target landscape save:
• Database profiles
• SAP profiles
• Operating system user profiles / home directories
Check if you need to preserve also other information not listed above. Consider if there’s also
information on non-WebAS based components that needs to be noted or preserved.
Step 2.6 Target Backup target landscape ↑
Only required if you need a fallback option for the target landscape
© 2006 SAP AG

Task of step
Create a backup of the target systems in case you need a fallback option should the refresh or copy
procedure fail.
What to do
Backup all components that are to be refreshed. This backup should be consistent in relation to any
other components of the target landscape that are not refreshed during the copy procedure.
Step 2.7 Target Prepare any other systems outside the target landscape ↑
Task of step
You might need to take any precautions on the components of the target landscape that are not
copied or refreshed during this procedure. For example you might need to disable connections to or
from other systems.
What to do
Check if there are any such actions required for your landscape.
Check if there are any actions described in the solution-specific [SAP System Landscape Copy
Notes].
Phase 3: Preparations on Source Environment
Step 3.1 Source Prepare source systems ↑
Task of step
Prepare all components of the source landscape for the copy.
What to do
Execute the source system preparations listed in the following places:
• For SAP WebAS based components, do the system copy preparations as described in
[System Copy Guide].
• When copying SAP WebAS 6.40 JAVA based components, also do the preparations of the
source system as described in [System Copy Guide - JAVA for 6.40].
• Check solution-specific [SAP System Landscape Copy Notes] to see if any other tasks are
required for a component (either WebAS or non-WebAS based components)
• For a Unicode conversion, also check the Unicode Conversion Guide for special preparations
(see SAP note 548016)
Some of the actions described there are mainly important for scenario “PRD PRD”.
They may not be mandatory for creating or refreshing a non-productive system and
especially not for scenario “Minimum Impact”.
Step 3.2 Source Save configuration information ↑
Required for scenario “PRD PRD”.
© 2006 SAP AG

Task of step
Save any configuration information of the source environment that may be needed for configuring the
landscape after the copy / migration or in case a fallback to the original environment is inevitable.
Having the original configuration settings at hand can be helpful to compare source and target
systems.
What to do
You can save the screenshots of transactions SMQR, SMQS, SM59, BD54 as images or text files. Do
this on all systems to be copied.
Step 3.3 Source Stop productive operation in source systems ↑
Applicable for scenario “PRD PRD”
Applicable for scenario “Minimum Risk”
Task of step
Before shutting down the source systems in a later step, stop any user or batch job activity to allow for
the communication queues to empty and for additional preparations to be taken.
What to do
Stop productive operation in source systems by
• Locking the production systems for user access.
This can be done either by using “tp locksys” to lock the complete system or by locking
users selectively. The second option may be required for example if RFC users need to stay
unlocked to finish processing of pending messages but dialog users shall be locked so no
new RFCs can be created. When all local RFCs are processed (see next step), the system
can be locked completely to prevent from newly incoming messages from systems not
belonging to the landscape to be copied. To prevent newly incoming messages, it might
also be necessary to disable communication paths in other systems outside the landscape
to be copied.
• disconnecting all current users
• stopping and disabling all batch jobs. This can be done using report BTCTRNS1.
Step 3.4 Source Complete message processing in source landscape ↑
Only applicable for scenario “Minimum Risk”
Required with SAP XI for all messages in status „scheduled for outbound processing“
Task of step
Complete message queue processing in all source systems so the target systems will not try to
exchange messages after startup.
For discussion of system copy with non-empty message queues see Risk 4 above.
When copying a landscape including an SAP XI system, all messages with status
“scheduled for outbound processing” must be processed before doing the copy!
Messages with this status already went through receiver determination and contain the
receiver hostname. Since the receiver determination cannot be repeated after the system
copy, such messages would be incorrectly routed to the production source landscape.
© 2006 SAP AG

What to do
Check that
• RFC queues are empty
• All IDOCS are processed (transaction WE05)
• SAP XI messages queues are empty
• Make sure that no SAP XI messages are in status “scheduled for outbound processing”
If the queues are not empty, start the queues and wait until all the messages are processed.
Please note that some message queues may not be supposed to be processed or may not
need to be emptied (for example some SAP CRM Mobile Client queues).
Step 3.5 Source Disable communication between source systems ↑
Task of step
Stop all communication between components of the source landscape and invalidate all
communication paths. The intention is to disable all communication already on the source landscape
thus the target systems are instantly unable to communicate at startup. This can help to reduce any
risks for the source environment in case the target systems are started unintentionally before other
necessary precautions have been taken or if these precautions are insufficient.
Especially when doing an online copy using online backup or split mirror technology, it is
desirable to skip this step to avoid an impact on productive operation. In this case you need
to ensure by other means, that Risk 1 and Risk 2 discussed above are eliminated. Most of all
you must ensure that during startup of the target landscape (whether intentionally or
accidentally) no communication can be established with the production (source) systems.
If you execute a copy with empty message queues (which means that you executed step
3.4), the actions described here may not be required. If no messages exist, nothing can be
exchanged between the target systems. The only thing to make sure is that no new
messages are created on the target systems before adaptation of the system connections
was completed.
What to do
Execute those actions from the list below which are appropriate for your copy scenario and your risk
mitigation analysis:
1. Save the initial configuration of all parameters you are changing below.
Save queue status, RFC destinations, and so on before changing them. This information about
the parameter values before the copy is needed to reset the parameters to their original values
after the copy was done. This queue status is also important so you later know which queues
were already unregistered even before the copy.
You can save screenshots of the transactions SMQR, SMQS, SM59 as images or text files.
2. Lock the outbound queues
Outbound queues can be locked by creating a stopmark. This can be done with transaction
SMQ1 or report RSTRFCQ1. Generic queue names like ‘queue*’ are supported whereas the
© 2006 SAP AG

destination must be fully specified, which means that the lock needs to be set for each
destination you want to be locked.
When the outbound queues are locked, no queued RFC (qRFC) using this queue and
destination can be processed. This provides control over when pending qRFCs will be
transmitted in the target landscape.
Note: This does not include transactional RFCs (tRFCs) because these do not use queues.
3. Unregister the outbound destinations
Unregistering outbound destinations at the outbound scheduler offers the advantage that it
works for both qRFC and tRFC, since usually also tRFCs are processed by the outbound
scheduler.
Destinations can be unregistered using transaction SMQS. All the queues and destinations
should then have type ‘U’. All queue entries for these destinations remain in the “recorded” state
until either the destination has been registered or the entries have been manually activated.
Note: Unregistering destinations cannot provide complete protection against RFC execution
because
• manual activation is not prevented. You should make sure that only few administrators
are permitted to activate queues (transaction SM58 for tRFC and SMQ1 for qRFC).
• some application programs might activate their destinations automatically.
• it is not possible to deregister generic destinations. Thus you have to make sure that
you unregistered the outbound destinations for all connected systems. When
executing an RFC call for a non-listed destination, this destination is automatically
registered by the QOUT-scheduler.
Note: Locking the outbound queues or unregistering the outbound destinations before changing
the RFC destinations prevents communication errors during the processing of queue
entries. But this is not critical since the queues can be restarted in either case.
4. Lock the inbound queues
Locking the inbound queues cannot prevent communication in the target landscape since
incoming messages are still accepted and only their execution is halted. But it may still be
helpful for a controlled start of the target landscape since the execution of inbound messages
may again trigger new messages to be sent to other systems, which may not yet be desired
then.
Inbound queues for qRFCs can be locked by creating a stopmark with transaction SMQ2 or
report RSTRFCI1. Generic queue names are supported.
5. Unregister the inbound queues
Unregistering inbound queues at the inbound scheduler has the same effect as locking inbound
queues. Inbound queues can be unregistered with transaction SMQR. You can select all queues
and unregister them in one go.
Note: It is also possible to unregister a generic destination like ‘*’, but you must be aware that
this will automatically delete all other destinations comprised by the generic term! So
make sure that you have a written copy of the original SMQR settings so you are able to
reconstruct the original state of all inbound queues listed in SMQR.
6. Disable the RFC destinations and connections to other systems
To change RFC destinations pointing to other productive systems, use transaction SM59. For
example, you can change the host name by adding a string like “.syscopy” or a prefix like “#”.
Then the host name points to a nonexistent server and communication is impossible via this
destination.
© 2006 SAP AG

Remember to also execute appropriate actions on the non-WebAS based components of the
source landscape if required.
Step 3.6 Source Shutdown source landscape ↑
Required for scenario “PRD PRD”
Task of step
Shutdown the source systems to provide a stable and consistent state for copying.
This step can also protect against any interference from the target landscape before the cleanup is
done there. But this additionally requires that the startup of the source systems is delayed until the
cleanup on the target systems will be done, which means that phase 5 will be executed after phase 6.
Offline copy is mandatory for scenario “PRD PRD”.
What to do
Shutdown all components in the correct order. See specific [SAP System Landscape Copy Notes] for
a list of components. If more information is required (for example concerning the order of the
components’ shutdown), also check the solution-specific operation guides.
Step 3.7 Source Backup source landscape ↑
Required for scenario “PRD PRD”.
Optional for scenario “PRD NPS”.
Task of step
Before migrating or moving the source landscape, perform a full backup of all components involved in
the copy. This is important to have a valid backup in case migration or relocation of the productive
system (scenario “PRD PRD”) fails. This step is not required if you copy or refresh to a NPS without
affecting the production systems.
What to do
Backup application data and filesystems, including the software and configuration data for all
components to be copied.
Backup database profiles and SAP profiles.
Step 3.8 Source Prepare any other systems outside the source landscape ↑
Task of step
You might need to take any precautions on the components of the source landscape that are not
copied during this procedure. For example you might need to disable connections from other systems.
What to do
Check if there are any such actions required for your landscape.
© 2006 SAP AG

Check if there are any actions described in the solution-specific [SAP System Landscape Copy
Notes].
Phase 4: Copy Process
The copy process describes the actual phase of providing (installing) the target system and
transferring (copying) the source data to the target database system. Several methods are available
for this. For an overview of applicable methods see section “Copy Technologies”. The following table
provides an overview of the general steps to be executed using the different copy technologies.
With an SAP WebAS ABAP system, it is sufficient to only copy the ABAP database from the source to
the target system. With an SAP WebAS ABAP+JAVA system, the ABAP and JAVA database as well
as the JAVA Central Instance (JAVA CI) must be copied. The JAVA CI includes for example the JAVA
filesystem structure, the SDM and possible application-specific filesystem content. With an SAP
WebAS JAVA system, the JAVA database and the JAVA CI must be copied.
R3load / Jload Database
Restore
File Copy
(not available for all
databases)
Split-Mirror
4.1 Capture
source data
Export source
database and
JAVA CI with
SAPInst
Create backup of
source database;
Export JAVA CI
with SAPInst
not required * not required *
4.2 Install target
system
Run SAPInst Run SAPInst Run SAPInst Run SAPInst
4.3 Transfer
source data
into target
database
Import is done
automatically by
SAPInst
Restore database Copy database
files
Split mirror disks
4.4 Complete
installation of
target system
SAPInst SAPInst SAPInst SAPInst
* For a WebAS ABAP+JAVA or WebAS JAVA system, the export of the JAVA CI needs
to be factored in
The highlighted fields indicate the period when the preparation of the source system must be
finished according to phase 3 from above, for example when the source system must be
offline.
With the File Copy and Split-Mirror method, you may consider running the first part of the
installation with SAPInst (step 4.2) before executing the steps of phase 3, to shorten any
impact on the source systems imposed by phase 3.
With R3load / Jload and Database Restore method, steps 4.1 and 4.2 are completely
decoupled.
When copying a landscape of multiple systems, cross-system data consistency between the
different systems will only be determined by step 4.1 or 4.3 respectively! All other steps of
the copy procedure can be done in parallel - but independently - for each of the systems
belonging to the landscape.
If the data of the target landscape must be consistent between the systems, an appropriate
© 2006 SAP AG

consistency technology must be used in step 4.1 or 4.3 (see discussion of Risk 5 above).
When copying an SAP solution that is installed on top of an SAP WebAS, this solution is
completely copied by the above steps. An additional call of SAPInst, which is required during
a regular installation of such an SAP solution, is obsolete when doing a system copy.
The detailed proceeding depends on the different WebAS deployment options and WebAS releases.
The following tables summarize the general copy procedure for Web AS ABAP, WebAS ABAP+JAVA
and WebAS JAVA on Oracle using
• R3load / Jload method for SAP WebAS 6.40
• R3load / Jload method for SAP NetWeaver 2004s
• Database Restore method for SAP WebAS 6.40 SR1
• Database Restore method for SAP NetWeaver 2004s
More details on these steps and on the other copy methods can then be found in the remainder of this
section.
Depending on the installation type and system setup, additional installation steps (installation
services) which are not shown in the tables below may be required, for example for installing
a high availability system. For database systems other than Oracle, some installation
services listed below may be obsolete or may have to be run on the central instance host
instead of the database host (e.g. with MaxDB or DB2 z/OS).
See the installation and system copy guides for WebAS for more specific information.
The following tables use the following notation:
S = source system
T = target system
DB = database host (host where the database is installed)
CI = central instance host (host where the SAP central instance is installed)
SAPInst(DB) = run SAPInst on the database host
SAPInst(CI) = run SAPInst on the central instance host in case of a distributed system installation, otherwise
run SAPInst on the database host. This step is required even in case of a central system
installation unless explicitly marked by “Distributed System only”!
© 2006 SAP AG

R3load / Jload method for WebAS 6.40
Execute
on
SAP WebAS ABAP SAP WebAS ABAP + JAVA
(JAVA Add-in)
SAP WebAS JAVA
(JAVA system)
Step
CI WebAS shutdown WebAS running WebAS running 4.1
CI
/
SAPInst (CI)
export JAVA CI and
JAVA DB
SAPInst (CI)
export JAVA CI and
JAVA DB
4.1
CI
/
Stop WebAS when requested on
display of the 'summary screen'
select option stop 'J2EE cluster'
to automatically shutdown the
WebAS
4.1
CI / Finish SAPInst (CI) / 4.1
DB SAPInst (DB)
export ABAP DB
SAPInst (DB)
export ABAP DB /
4.1
S
DB Transfer export files Transfer export files Transfer export files 4.1
DB SAPInst (CI)
install ABAP CI
SAPInst (CI)
install ABAP CI
Distributed System only:
SAPInst (CI)
prepare CI host
4.2
4.3
CI SAPInst (DB)
import ABAP DB
SAPInst (DB)
import ABAP DB /
4.2
DB
/
SAPInst (DB)
import JAVA DB
installs & imports JAVA CI
(central system)
SAPInst (DB)
import JAVA DB
(central system)
4.2
4.3
T
CI
/
SAPInst (CI)
install & import JAVA CI
SAPInst (CI)
4.4
R3load / Jload method for NetWeaver 2004s
Execute
on
(JAVA Add-in)
SAP WebAS JAVA
(JAVA system)
Step
CI WebAS shutdown WebAS shutdown WebAS shutdown 4.1
DB SAPInst (DB)
export ABAP DB
SAPInst (DB) *
export ABAP+JAVA DB
exports JAVA CI (central
system)
SAPInst (DB) *
export JAVA DB
exports JAVA CI (central
system)
4.1
CI
/
SAPInst (CI) *
export JAVA CI
SAPInst (CI) *
export JAVA CI
4.1
S
DB Transfer export files Transfer export files Transfer export files 4.1
CI Distributed System only:
SAPInst (CI)
prepare CI host
SAPInst (CI)
prepare CI host
SAPInst (CI)
prepare CI host
4.2
DB SAPInst (DB)
import ABAP DB
installs CI
(central system)
SAPInst (DB)
import ABAP+JAVA DB
installs CI & imports JAVA
CI (central system)
SAPInst (DB)
import JAVA DB
(central system)
4.2
4.3
T
SAPInst (CI)
install CI
SAPInst (CI)
install CI & import JAVA CI
SAPInst (CI)
4.4
* on DB2 z/OS: SAPInst is run only once on the CI host, this will export the database and the JAVA CI
© 2006 SAP AG

Database Restore method for WebAS 6.40 SR1
Note: For WebAS 6.40, the database restore method is supported with service release SR1. For more
information see the newest version of the [System Copy Guide - JAVA for 6.40]
Execute
on
(JAVA Add-in)
SAP WebAS JAVA
(JAVA system)
Step
CI WebAS running WebAS running 4.1
CI
/
SAPInst (CI)
export JAVA CI
SAPInst (CI)
export JAVA CI
4.1
CI
/
Stop WebAS when requested on
display of the 'summary screen'
select option stop 'J2EE cluster'
to automatically shutdown the
WebAS
4.1
CI / Finish SAPInst (CI) / 4.1
S
DB Create DB Backup,
Oracle: Generate controlfile
Create DB Backup,
Create DB Backup,
4.1
CI SAPInst (CI)
install ABAP CI
SAPInst (CI)
install ABAP CI
SAPInst (CI)
prepare CI host
4.2
DB SAPInst (DB) SAPInst (DB) SAPInst (DB) 4.2
DB Restore
ABAP DB
Restore
ABAP+JAVA DB
Restore
JAVA DB
4.3
DB Finish SAPInst (DB) Finish SAPInst (DB) Finish SAPInst (DB)
(central system)
4.4
T
CI
/
SAPInst (CI)
install and import JAVA CI
SAPInst (CI)
4.4
Database Restore method for NetWeaver 2004s
Execute
on
(JAVA Add-in)
SAP WebAS JAVA
(JAVA system)
Step
CI WebAS shutdown WebAS shutdown 4.1
CI
/
SAPInst (CI)
export JAVA CI
SAPInst (CI)
export JAVA CI
4.1
S
DB Create DB Backup,
Create DB Backup,
Create DB Backup,
4.1
SAPInst (CI)
prepare CI host
SAPInst (CI)
prepare CI host
SAPInst (CI)
prepare CI host
4.2
DB SAPInst (DB) SAPInst (DB) SAPInst (DB) 4.2
DB Restore
ABAP DB
Restore
ABAP+JAVA DB
Restore
JAVA DB
4.3
DB Finish SAPInst (DB)
installs CI
(central system)
Finish SAPInst (DB)
installs CI & imports JAVA
CI (central system)
Finish SAPInst (DB)
(central system)
4.4
T
SAPInst (CI)
install CI
SAPInst (CI)
install CI & import JAVA CI
SAPInst (CI)
4.4
© 2006 SAP AG

Step 4.1 Source Capture data of SAP WebAS source system ↑
Task of step
Capture the data of the source system for later import into the target system.
What to do
With R3load / Jload method:
SAPInst can be used to perform both the ABAP and JAVA export of an SAP WebAS system. The
appropriate steps of phase 3 must be completed before starting the export. SAPInst is run in the
source system to export the source database to a file in a database independent format. With a
WebAS ABAP+JAVA or WebAS JAVA installation, the JAVA filesystem (JAVA CI) needs to be
exported in addition to the database.
With WebAS 6.40, SAPInst must be run separately for ABAP and JAVA export. When copying a SAP
WebAS with ABAP and JAVA, the JAVA export of the source system is done before the ABAP export.
The JAVA export, which is started on the source central instance host exports the JAVA filesystem
and JAVA database. The WebAS system must be running when starting SAPInst for the JAVA export
and is shutdown during the process. With WebAS JAVA, SAPInst can automatically shut down the
system whereas with WebAS ABAP+Java, the shut down must be done manually. For the correct
order of the required export steps see the corresponding table above.
With NetWeaver 2004s, SAPInst can do both, the ABAP and the JAVA export. Only with a distributed
installation, SAPInst needs to be called separately on the central instance host to export the JAVA CI.
The WebAS system should be shutdown before starting SAPInst. For the correct order of the required
export steps see the corresponding table above.
For more information on the export with SAPInst see [System Copy Guide] and [System Copy Guide -
JAVA for 6.40].
For a Unicode conversion, also check the Unicode Conversion Guide for any special information (see
SAP note 548016).
When the export is finished, copy the export files to the target server.
With Database Restore method:
Create a database backup that will be used to restore the database onto the target system. According
to the copy scenario chosen (“Minimum Risk” or “Minimum Impact”), the appropriate steps of phase
3 must be completed before creating the backup.
For Oracle databases, a new controlfile must be created for the target database. This can be done as
described in section “Generating the Control File Structure on theSource System” in [System Copy
Guide].
For a WebAS JAVA or WebAS ABAP+JAVA system, an export of the JAVA central instance
(filesystem content like directory structure, SDM, etc.) is required to copy this information to the target
system. This export is done using the SAPInst tool and must be run on the central instance host in
case of a distributed installation. Choosing the SAPInst option “Copy the database by using database
tools”, will tell SAPInst to skip the Jload database export and only export the source JAVA central
instance. For the correct order of the required export steps see the corresponding table above. Also
see the descriptions about the source system export in [System Copy Guide - JAVA for 6.40] or
[System Copy Guide for NW 2004s].
With File Copy / Split-Mirror method:
For the database data nothing needs to be done here since the data will be provided to the target
system with step 4.3 of the procedure.
For a WebAS JAVA or ABAP+JAVA system, an export of the JAVA central instance (filesystem
content like directory structure, SDM, etc.) is also required to copy this information to the target
system. This is required for a New Installation as well as for Refresh. The export can be done using
© 2006 SAP AG

the SAPInst tool and must be run on the central instance host in case of a distributed installation.
Choosing the SAPInst option “Copy the database by using database tools”, will tell SAPInst to skip the
Jload database export and only export the source JAVA central instance. See the descriptions about
the source system export in [System Copy Guide - JAVA for 6.40] or [System Copy Guide for NW
2004s].
To ensure consistency of the JAVA CI export files and the database data, the JAVA CI export should
be done together with the data copy in step 4.3. The export should be done offline to ensure
consistency of the exported filesystem contents.
Step 4.2 Target Start installation of SAP WebAS target system ↑
Applicable for New Installation of the target landscape.
Task of step
Install the SAP WebAS based components of the target landscape. During installation, the regular
database load phase from installation CD will be replaced by the import of the data from the source
system using one of the technologies described above.
Changing the SID
With scenario “PRD NPS” it is often desired to assign a different SAP system-ID (SAPSID) to the
target system. This is usually not a technical requirement since two systems with the same SAPSID
can run on different hosts. But to avoid confusion of the users, a target system should have a SAPSID
different from the production system. But renaming the SAPSID also affects the SID used on the
database (DBSID), which often is the same as the SAPSID. The SAPSID is used in the directory
structures of the SAP and the database installation and for some SAP WebAS releases it was also
used for naming the database schema and tablespaces.
• SAP SID
During New installation with SAPInst, a new target-SAPSID can be chosen freely. During
Refresh, the SAPSID is not affected and stays the same.
• Database SID
With R3load, the DBSID can be changed.
With DB Restore method, the DBSID can be changed. The procedure and requirements of
changing the DBSID for doing the restore are different for each database system.
With split-mirror, the DBSID needs to be changed before starting the target database on the
mirror.
• Database schema name
The database schema name can only be changed when using the R3load tool. Using restore
or split-mirror (on Oracle), the database schema cannot be renamed. To avoid the
disadvantage of having the production SID in the target schema-ID, a new schema-ID was
introduced with SAP WebAS 6.20. This should be chosen neutrally as described in SAP note
617444.
What to do
Run SAPInst to install all SAP WebAS based components as described in the installation
documentations for the components of your target landscape. Proceed as described in the [System
Copy Guide] to install and copy all SAP WebAS based components of your landscape. Choose the
appropriate SAPInst option according to your copy method:
SAPInst is used to install the target system, choosing the installation option “System copy / Migration”.
SAPInst will then automatically start the import of the database into the target system, see step 4.3.
© 2006 SAP AG

With WebAS 6.40, SAPInst must be run separately for import of ABAP and JAVA database. With a
distributed intallation, SAPInst must be run additionally to import the JAVA CI. To prepare the import of
the JAVA CI in case of WebAS JAVA, SAPInst needs to be run on the CI before installing the
database instance. With WebAS ABAP or WebAS ABAP+JAVA this preparation is not required due to
the initial installation of the ABAP CI. For the correct order of the required import steps see the
corresponding table above.
With NetWeaver 2004s, SAPInst can import the ABAP and JAVA database as well as the JAVA CI in
one go. Only with a distributed installation, SAPInst must be run separately to install the CI and import
the JAVA CI. To prepare the installation of the CI on a separate host in that case, SAPInst additionally
needs to be run on the CI host before installing the database instance. For the correct order of the
required import steps see the corresponding table above.
With Database Restore / File Copy / Split-Mirror method:
With a distributed installation of WebAS, SAPInst must be run first to install the CI or to prepare the CI
host for the installation. Then SAPInst is run to install the database instance. For the correct order of
the required import steps see the corresponding table above.
When SAPInst is run to install the target system, choose the installation option “System copy /
Backup/Restore”. Proceed as described in the [System Copy Guide] for your database system type.
At some point, SAPInst will interrupt the installation to allow the data to be provided to the target
system, either by restore from the backup, by file copy or by split-mirror copy, as described in the next
step.
With some database systems, installation of the database and preparation of the required directory
structure must also be done manually at this point, before copying the data.
Step 4.3 Target Copy the data using the selected method ↑
Only applicable for components with data to be copied
Task of step
Provide the target system with the data copied from the source system. This step must be done for all
components that hold application data which needs to be copied from source to target components.
This is also the main step for a Refresh of the target system.
What to do
Execute the technical system copy or refresh using the selected method
The import of the data is started automatically by SAPInst. SAPInst will prompt for some input
parameters that allow you to control the import process for example by defining the degree of
parallelism during import.
For Refresh, a Reload procedure is available for WebAS 6.40 and Oracle databases (ABAP only),
see [System Copy Guide for WebAS 6.40]. With NetWeaver 2004s, the Reload for WebAS (ABAP
only) is integrated into SAPInst, see [System Copy Guide for NW 2004s].
With Database Restore method:
Restore a database backup onto the target database. On Oracle databases, a new controlfile must be
created for the target system. Rename the DBSID as determined by the restore procedure (depending
on the database system).
With File Copy method:
Copy the database files onto the target host. Consistency of the data files must be ensured, usually by
shutting down the source database for the duration of the copy. On Oracle databases, a new
controlfile must be created for the target system.
© 2006 SAP AG

With Split-Mirror method:
Create a copy of the source system on a set of mirror disks intended for the target system.
Consistency of the copy must be ensured by either suspending the database, using Oracle hot backup
mode or using ‘consistent split’ technology. After splitting the mirror of the source system’s disks, the
volumes are mounted to the target host. If applicable, rename the SID in all relevant locations of the
mirror (mount points, directories, filesystem permissions, profile and configuration files, …). On Oracle
databases, a new controlfile must be created for the target system.
Step 4.4 Target Complete installation of SAP WebAS target system ↑
Task of step
Complete the installation of SAP WebAS based components of the target landscape. With some copy
methods, the installation was interrupted for copying the source data into the target system. The
installation must be restarted to finish completely. For JAVA systems, an additional installation / import
step may be required.
SAPInst automatically starts up the target SAP system during installation.
If this imposes incalculable risks to your source environment due to the copy
scenario chosen (also see Risk 1 and Risk 2 above), you should consider taking
additional precautions or choosing another copy scenario (at least for the initial
installation of the target system).
What to do
There is no need to restart SAPInst, since the export files from the source system are imported without
interruption.
For a distributed installation with JAVA, SAPInst no needs to be rerun on the central instance host to
import the JAVA CI from the export files created in step 4.1.
With Database Restore / File Copy / Split-Mirror method:
Since SAPInst was interrupted to allow for copying the data, SAPInst must now be continued as
described in the [System Copy Guide]. After that, the complete database (both ABAP and JAVA) will
have been copied to the target system.
For a WebAS JAVA or ABAP+JAVA system, the JAVA CI needs to be installed and imported into the
target system. Depending on the release of WebAS, for a central system this is done with the above
call of SAPInst. For WebAS 6.40 or for a distributed installation, SAPInst needs to be run on the
central instance host of the target system to install or import the JAVA CI. In case of a New
Installation, this will install all required JAVA parts and import the JAVA export from the source
system. In case of a Refresh, this will just import the JAVA export from the source system. See the
above tables for the correct order of the required import steps.
© 2006 SAP AG

Phase 5: Final Activities on Source Environment
With scenario “PRD PRD”, phase 5 is not applicable. Continue with phase 6 since the
target environment will be the new production environment.
Step 5.1 Source Startup production ↑
Applicable for scenario “Minimum Risk”.
Task of step
Resume operations in production environment.
To further minimize the risk of the target systems interfering with the production
environment, restart of production could be delayed until the most critical activities
on the target environment (steps 6.1 – 6.6) are completed.
Note: This additional protection prolongs downtime for production.
What to do
Restart all production components in the correct order. For more information see the solution-specific
operation guides.
Step 5.2 Source Re-enable communication between source systems ↑
Task of step
Re-enable the communication in the source landscape if it was disabled in step 3.5 to prevent
unintended interaction between target and source systems.
What to do
Revert any changes done in step 3.5 “Disable communication between source systems” by resetting
all changed parameters to their original state. The tasks are executed in opposite order using the
information saved before doing the modifications.
1. Reset the RFC destinations and connections to their original values
2. Register, unlock and start the inbound queues
Unregistered inbound queues can be registered with transaction SMQR. Locked inbound
queues can be unlocked using transaction SMQ2 or report RSTRFCI3. Processing of inbound
queues can be restarted with transaction SMQ2 or report RSQIWKEX.
3. Register, unlock and start the outbound queues
Unregistered outbound destinations can be registered with transaction SMQS. Locked
outbound queues can be unlocked using transaction SMQ1 or report RSTRFCQ3. Processing
of outbound queues can be restarted with transactions SMQ1 (qRFC) and SM58 (tRFC) or
report RSQOWKEX.
© 2006 SAP AG

Step 5.3 Source Perform follow-up activities from [System Copy Guide] ↑
Task of step
Reverse any actions done on production system before the copy was done.
What to do
If applicable, perform the actions described in the [System Copy Guide] in section “Performing Follow-
up Activities in the Source System”, for example
• Reschedule cancelled batch jobs (transaction SM37 or report BTCTRNS2)
• Adapt operation modes to original values
When copying a SAP WebAS system with JAVA components, perform any actions described in the
[System Copy Guide - JAVA for 6.40] in section “Post-Installation Activities in the Source System”.
Step 5.4 Source Restart productive operation in source systems ↑
Task of step
Re-enable productive operation by reversing the changes done in step 3.3.
What to do
• Unlock the production systems for user access
• Reschedule or restart batch jobs
Phase 6: Final Activities on Target Environment
With scenario “PRD PRD”, this will be the new production environment. Note that not all
steps or actions of phase 6 may be applicable to scenario “PRD PRD”.
Step 6.1 Target Execute follow-up activities on OS and DB level ↑
Task of step
Prepare the target system on operating system and database level for startup. This includes for
example adaptation of profiles and users.
What to do
Execute the follow-up activities described in the [System Copy Guide] for operating system and
database level.
In case of a target system refresh, restore profiles saved before the refresh (see step 2.5).
For adapting the profiles in case of a Unicode conversion see SAP note 790099.
© 2006 SAP AG

Step 6.2 Target Prevent any unintentional actions on target system ↑
Task of step
Prevent that any activities that might interfere with production are triggered when starting the systems
of the target landscape. This especially means avoiding immediate communication between the
systems after the first startup.
What to do
Take the following actions to prevent any processes from starting on the target system. Optimally,
these actions should already be done before starting the SAP instance (e.g. on database level). Some
actions that cannot be executed before start of the SAP WebAS instance should be done immediately
after the startup in step 6.3.
Please note that during installation of the target system with SAPInst, SAPInst will automatically start
up the SAP system.
• Set number of batch processes to 0 in the SAP instance profile on operating system level to
prevent any batch jobs from running after startup (parameter rdisp/wp_no_btc).
• Lock system
Execute “tp locksys” to lock the system against user logins. This can be done before starting
the target SAP system, only the database must be running.
You can also use ”tp locksys” to temporarily lock users in the source (production) systems
while performing the most important cleanup activities in the target landscape. Since RFCs
cannot be processed if the corresponding RFC user needed for RFC execution in the
receiving system is locked, this can help to avoid any messages from accidentally reaching
the source systems. The same effect can be achieved with temporarily locking all RFC users
in the production systems while the target systems are started.
• Disable system connections / RFC destinations to prevent any communication from being
issued by the system. Execute an SQL statement which invalidates the hostnames
respectively the IP address in the main RFC connections table, for example by adding a suffix
‘.copy’ or a prefix ‘#’ to the hostname / IP address. The hostname / IP address is contained in
field RFCOPTIONS of table RFCDES.
This step can help to prevent any messages from being transferred from the target to the
source environment. The actual adjustment of the destinations to their final values (the real
servers of the target environment) is done later in step 6.4 using the official SAP transaction
SM59.
Step 6.3 Target Start target system ↑
Task of step
Start the SAP WebAS based systems of the target landscape so the follow-up activities can be done
in the SAP system.
What to do
Start the SAP WebAS server.
Step 6.4 Target Adjust system connections ↑
Task of step
The main purpose of this step is to adjust the parameters responsible for communication between the
systems of the target landscape.
© 2006 SAP AG

With scenario “Minimum Risk” the systems of the target landscape are unable to communicate with
each other because RFC destinations point to nonexistent servers and the inbound and outbound
queues are unregistered.
With scenario “PRD PRD” the communication settings need to be checked and adapted to the new
servers.
With all other scenarios it is vital to prevent the target system from communicating with the
productive environment.
What to do
Adapt, disable or delete all interfaces to any systems of the production environment.
For example, use transaction SM59 to adapt all RFC destinations and other communication entries
and make sure that the correct target host names of the NPS landscape and the corresponding user
logon data are used.
• The following table provides an overview of possible interfaces or connections that need to be
checked:
Connection type To be maintained with
RFC and R/3 connections SM59
HTTP SM59
TCP/IP SM59
Registered programs SM59
Java connector (JCO connection) SM59 and Visual administrator
Trusted/Trusting Systems SM59, SMT1, SMT2
R/2 connections SM59
CPIC SM59
SAP XI communication SM59 and Integration Directory
Adapters
File
Batch input
DB links
Archive link
SAP phone
SAP connect
Email
• Check or create RFC users and passwords to be used in the NPS environment.
We recommend defining different RFC users in the NPS landscape than in production. This
can help to protect against NPS messages being executed in production using the production
RFC users (e.g. in case it was forgotten to change the server name or IP address).
• Execute connection check in SM59 to verify that the new settings are valid
• If trusted/trusting system connections are used between SAP systems in your landscape,
these need to be deleted or adapted. Check transactions SMT1 and SMT2. Trusting relations
can be deleted be deleting the contents of tables RFCSYSACL and RFCTRUST.
• Restart the SAP system after making all the above changes
© 2006 SAP AG

Make sure that no communication entries point to hosts of the production landscape.
We recommend that you change RFC physical destinations (host names and IP addresses)
but not the RFC destination name. If you want to change the names of the RFC destinations,
check all the Configuration Guides corresponding to your scenarios and perform all the
required Customizing steps.
If ALE is used in the target landscape:
Do not create new RFC destinations and logical systems in the target landscape! ALE may
be used, for example, for exchanging IDocs between the systems, downloading
organizational units from the backend OLTP system, or for central user administration. If
new destinations are created, ALE distribution no longer works. If ALE is used, only the
values and parameters such as the host names, IP addresses, and logon data within the
existing RFC destinations should be modified to match the target environment.
Since the setups for SAP solution landscapes are different for each individual case, it is impossible to
give full recommendations on maintaining RFC destinations in this document. SAP Configuration
Guides provide detailed information on maintaining the RFC destinations for each individual solution.
Check the guides corresponding to your solutions and maintain the RFC destinations as
recommended.
Also see the solution-specific [SAP System Landscape Copy Notes] since they may also provide
information on the connections used be these solutions.
Step 6.5 Target Execute follow-up activities from [System Copy Guide] ↑
Task of step
Do further cleanup actions in SAP WebAS-based systems. Similar to the connections dealt with in the
previous step 6.4, some of these actions are critical concerning integrity of the production landscape.
Special care must be taken that these actions are executed to prevent damage to the production
environment.
Do not change logical system names yet if there are pending messages to be
processed. This will be done in step 6.7.
Some of the actions to be done below may be substituted by step 6.9 “Import saved
information from before the refresh” (also see step 2.1).
What to do
Execute the follow-up activities on SAP system level for ABAP and JAVA, as described in the [System
Copy Guide] and [System Copy Guide - JAVA for 6.40].
For scenario “PRD PRD”, most of those actions may be obsolete.
© 2006 SAP AG

For system copy “PRD NPS”, several of these actions are critical for avoiding any interference with
the production systems or to prevent the target system from behaving like production. Take special
care with executing the following actions:
• Adapt transport settings (TMS)
• Check batch jobs
• Adapt printers, check print jobs
• Modify users and authorizations
• Adapt logon groups and application servers
• Check external commands
Additionally execute the other actions for the ABAP stack as listed in [System Copy Guide], if
applicable to your system.
When copying a SAP WebAS system with JAVA stack, also perform the actions described in section
“Performing Follow-Up Activities for Java” of the [System Copy Guide] (NW2004s) or in section
“Follow-Up Activities on the Target System” of the [System Copy Guide - JAVA for 6.40] (WebAS
6.40). Also see SAP note 870863 (NW2004s) or 785848 (WebAS 6.40).
Some of these actions for the JAVA stack are critical for avoiding any interference with the production
systems as well, especially connections to source landscape components like:
• Configuration of JCo
• EP connections for TREX, Repository Managers, Collaboration
• BI-JAVA source system connection
For a Unicode conversion, also check the Unicode Conversion Guide for special follow-up activities
(see SAP note 548016)
Step 6.6 Target Process pending messages ↑
not required if the queues were empty on the source system before doing the copy
Task of step
If the source system was copied with non-empty queues, pending communication messages need to
be exchanged to ensure data consistency in the target landscape. To allow processing of these
messages, the logical system names in the target landscape may not have been changed yet because
the messages might rely on the old logical system names.
Do not start the queues before adjusting the RFC destinations (step 6.4).
In scenario “PRD NPS” make sure that the target systems are separated from the source
landscape.
The logical system names may not be changed before this step.
Otherwise pending messages cannot be processed.
© 2006 SAP AG

SAP XI System
You may not process messages in status “scheduled for outbound processing”!
Since these messages already went through receiver determination, the receiver hostname
is already contained in the message. Such messages would always go into the production
landscape when processed on the target SAP XI system.
This step is not required if you don’t require data consistency between the systems of
the non-productive target landscape.
In that case, you can simply delete pending messages from the queues in the target
systems.
What to do
This step consists of several actions to be executed:
1. Register and start the inbound and outbound queues on all copied systems
Unregistered outbound destinations can be registered with transaction SMQS. Locked outbound
queues can be unlocked using transaction SMQ1 or report RSTRFCQ3. Processing of outbound
queues can be started with transactions SMQ1 (qRFC) and SM58 (tRFC) or report RSQOWKEX.
Unregistered inbound queues can be registered with transaction SMQR. Locked inbound queues can
be unlocked using transaction SMQ2 or report RSTRFCI3. Processing of inbound queues can be
started with transaction SMQ2 or report RSQIWKEX.
To avoid starting queues that had been unregistered in the source system even before the copy, it
may be helpful to check the screenshots made for these transactions before the system copy in order
to get the correct status for each individual queue.
You only need to process those message queues which are relevant for the target landscape (for
example only few mobile client queues may be needed in an NPS environment).
2. Monitor the queues, correct the errors and wait until the queues are empty
Monitor if all queues are processed correctly and correct any errors.
Check the RFC queues using transactions SMQ1 for the outbound and SMQ2 for the inbound queues
or use any solution specific monitoring tools.
To prevent the creation of new entries, there must be no workload in the system. Make sure that
nobody is logged on and no background jobs are running.
3. Check that all queues are empty
Check that all inbound and outbound queues are empty in all systems of the target landscape.
Step 6.7 Target Convert logical system names ↑
not applicable for scenario “PRD PRD”
Task of step
You might want to change the logical system names used in the target environment, for example if the
logical system names contain the production SIDs. From a technical point-of-view there is no
requirement to change logical system names, it is only important to ensure that the underlying RFC
destinations are adapted for the target landscape.
© 2006 SAP AG

Please see the discussion of Risk 3 above when you consider converting logical system
names.
What to do
Make sure that the queues are empty and all the entries that refer to the old logical system names
were processed. We can now assume that the systems are in a consistent state.
When converting logical system names, there should be no other activities or communication going on
in the system. Following the system copy procedure described in this Best Practice, the target system
is still locked (step 6.2) and any pending messages were processed (step 6.6), so these prerequisites
should be ensured at this point in time. Otherwise, to prevent any data exchange during the
conversion, the connection between the systems should be disabled once again by executing the
actions as described in step 3.5.
Now execute the following actions to adapt the logical system names:
1. Convert the logical system names
Transaction BDLS converts the logical system names in all application and customizing tables.
See SAP note 121163 for more information. To obtain the latest version of the transaction BDLS, see
SAP Note 369758.
• Never change the logical system names in production systems!
• Before you run transaction BDLS, check the online documentation.
• During a run of the changing transaction, no other activities should be performed
anywhere in the system, and there should be no communication with other systems.
• The processing of all IDocs in the system must be completed, since the logical system
name may be contained in the IDoc data record and this is not modified by the change
transaction.
• Avoid making manual changes to logical system names in the relevant tables. Otherwise
the application documents can no longer be fully located - use only transaction BDLS!
Activities:
• Start transaction BDLS and
o enter the old and the new logical system names
o select “Conversion of Client-Dependent and Client-Independent Tables”
o select “Existence Check of New Names in Tables”
Note: You can start the transaction first as a test run. In this case, no conversion is performed
but only the list of relevant tables is displayed.
• Transaction BDLS must be run for each logical destination that is to be changed.
• Transaction BDLS must be run for every client where destinations are to be changed. The
logical system name is an attribute that can be set for every client.
• Transaction BDLS must be run on all connected clients on other systems of the landscape.
Example:
If you have for example two source R/3 clients R3PCLNT001, R3PCLNT002 and one source CRM
client CRPCLNT800 which are connected, you have to execute transaction BDLS in the target
landscape 9 times as follows:
© 2006 SAP AG

• In target client R3QCLNT001:
o Rename R3PCLNT001 to R3QCLNT001
o Rename CRPCLNT800 to CRQCLNT800
• In target client R3QCLNT002:
• In target client CRQCLNT800:
2. Delete the old logical system names
During the conversion of the logical system names, the old system names are not changed. Instead,
only the new system names are entered. The old logical system names must now be deleted.
Start transaction BD54 and delete all the old logical system names that were changed in the previous
step. This must also be done on all systems of the target landscape.
3. Check the new logical system names
Check that the automatic conversion assigned the new logical system names correctly to the
corresponding clients.
You can start the transactions for example via the following path:
SAP R/3: SPRO IMG Basis Components Distribution (ALE) Sending and Receiving
Systems Logical Systems Assign Client to Logical Systems
Please also see solution specific [SAP System Landscape Copy Notes] to find out if there are special
tools available.
Step 6.8 Target Do solution-specific adjustments ↑
Task of step
Most SAP solutions require additional actions to be done after a system copy. These solution-specific
actions are not described in the [System Copy Guide].
What to do
Execute the actions listed in the solution-specific [SAP System Landscape Copy Notes] for the SAP
solutions you are copying.
Step 6.9 Target Import saved information from step 2.1 ↑
Task of step
Provide the new target system with any information that was saved from the old target system before
doing the refresh.
What to do
Import or enter the information saved in step 2.1 into the new target system.
© 2006 SAP AG

Step 6.10 Target Register queues ↑
Task of step
If the message queues were not already enabled during step 6.6, the queues must now be registered
to allow message processing in the target environment.
What to do
Register the inbound and outbound queues as described in step 6.6, item 1. Starting the queues
should not be necessary since there should be no queue entries waiting to be processed.
Note that not all queues may need to be registered since some might have been in status unregistered
even in the source environment. Use the original screenshots of transactions SMQS and SMQR to get
the correct states.
Step 6.11 Target Install new SAP license ↑
Task of step
After migrating a SAP WebAS based system to another hardware (scenario PRD PRD), you need a
new SAP license for the system. You also need a new license for a new NPS system. When
refreshing an NPS system, the existing license needs to be reinstalled into the NPS system.
What to do
Install the correct license for your target system with transaction “slicense” or operating system
command “saplicense – install”. This command can also be automated using an input file “saplicense -
install ifile=<file>”. See SAP note 870871 for more information.
Step 6.12 Target Install and configure additional components ↑
Task of step
Currently, there is no specific procedure for copying non-WebAS components of a system landscape.
Such components should be newly installed in the target landscape and configured as required.
What to do
Install any required non-WebAS based components of the landscape as listed in the solution-specific
SAP notes. To obtain the detailed procedures for installing and configuring the components, refer to
the installation and configuration guides for the individual components. For configuration of these
components, you can also compare the configuration of a component in the target landscape with
configuration of the same component in the source landscape.
For some non-WebAS based components, it may be desired (or even required) to install
them in the target landscape before copying the SAP WebAS based systems. Shifting these
steps to the preparation phase 2 can help to reduce the time until the target landscape
becomes available for operations after the main copy phase. Please consult the specific
installation guides for information on the possible order of installations.
Even with scenario “Minimum Risk” and when postponing the restart of the source
landscape until the cleanup on the target is completed (which means executing phase 5 after
phase 6), the production (source) systems can be restarted before installing these
components in the target landscape. This will reduce downtime on production.
© 2006 SAP AG

Step 6.13 Target Check and adjust integration to other components ↑
Task of step
Any components of the target landscape may need to be adapted to the changes in the environment.
These usually are
• components that were not included in the copy process or
• already existing components that now shall be used together with some newly copied
(installed) components.
Components that are included in the scope of the copied landscape should already have been
adjusted in steps 6.4, 6.8 or 6.12.
What to do
Change configuration settings of these components as required. See solution specific SAP notes or
installation and configuration guides for these components.
Step 6.14 Target Load or initialize data in additional components ↑
Not applicable for scenario “PRD PRD”
Task of step
Non-WebAS based components of the landscape can also hold data. After installation or when
refreshing the target landscape, these components also need to be provided with data that is
consistent to the rest of the landscape. This can usually be done by (re-)loading the data from other
(usually SAP WebAS based) components.
What to do
See solution-specific [SAP System Landscape Copy Notes] for more information on the components
that need to be initialized and loaded during system landscape copy of an SAP Solution.
Step 6.15 Target Check operability and data consistency ↑
Task of step
After completing the system landscape copy you may want to verify that the copy is fully operational
and that the copied data is consistent.
What to do
• Check that each single component of your landscape operates in accordance with its tasks.
For each component you should check its availability, configuration and ability to connect to
other components. In case of problems, check the relevant configuration or installation guides
and compare the affected component with the same one in the source landscape.
• Cross-check the configuration guides to ensure that the Customizing in the new landscape is
fully completed.
• Especially with scenario “PRD PRD”, it is recommended to also do performance tests in the
new environment.
• Especially with scenario “PRD PRD”, update the database statistics in the new
environment.
• Run consistency checks as described in [System Copy Guide], [System Copy Guide - JAVA
for 6.40] and in the solution-specific [SAP System Landscape Copy Notes]
• Let test users check the functionality of main transactions and business scenarios.
To do so, you should prepare procedure templates for business processes and other normal
© 2006 SAP AG

tasks. These procedure templates should enable you to monitor the execution of each step
and verify the consistency of your data.
Step 6.16 Target Release target landscape for operations ↑
Task of step
Now the target landscape can be released to operations. Any precautions that were taken to prevent
user activities before finishing the complete copy procedure now need to be reverted.
What to do
Release the target systems by
• Unlocking those users permitted in the target system
• Reversing the actions taken in step 6.2 above
© 2006 SAP AG

Further Information
Information on System Copy tools and technologies:
• http://service.sap.com/systemcopy
• http://service.sap.com/instguides
• http://service.sap.com/split-mirror
• http://service.sap.com/atg
You can get further assistance with developing a strategy for System Landscape Copy through the
SAP Optimization Service “SAP Continuity Management”. Please see the SAP Service Catalog at
http://service.sap.com/servicecat for more information.
Feedback and Questions
To improve the quality of this Best Practice and the System Landscape Copy procedure, we are
depending on your feedback. Please send any feedback concerning errors or possible optimizations of
this document to atg@sap.com, Subject “System Copy Guide - Feedback”. Please note that we do
not provide support for problems with the system copy procedure at this email address!
Please use SAP’s customer support systems to get help with any issues during execution of the copy
procedure. Please use components BC-INS-UNX, BC-INS-NT, BC-INS-AS4 for homogeneous system
copy or BC-INS-MIG for heterogeneous system copy.
You can report general issues with SAP Best Practices by formulating an SAP customer message to
component SV-GST-SMC. You can do this at http://service.sap.com/message.
© Copyright 2006 SAP AG. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission
of SAP AG. The information contained herein may be changed without prior notice.
Some software products marketed by SAP AG and its distributors contain proprietary software components of other software
vendors.
Microsoft®
, WINDOWS®
, NT®
, EXCEL®
, Word®
, PowerPoint®
and SQL Server®
are registered trademarks of
Microsoft Corporation.
IBM®
, DB2®
, OS/2®
, DB2/6000®
, Parallel Sysplex®
, MVS/ESA®
, RS/6000®
, AIX®
, S/390®
, AS/400®
, OS/390®
, and
OS/400®
are registered trademarks of IBM Corporation.
ORACLE®
is a registered trademark of ORACLE Corporation.
INFORMIX®
-OnLine for SAP and Informix®
Dynamic Server
TM
are registered trademarks of Informix Software
Incorporated.
UNIX®
, X/Open®
, OSF/1®
, and Motif®
are registered trademarks of the Open Group.
HTML, DHTML, XML, XHTML are trademarks or registered trademarks of W3C®
, World Wide Web Consortium,
Massachusetts Institute of Technology.
JAVA®
is a registered trademark of Sun Microsystems, Inc. JAVASCRIPT®
is a registered trademark of Sun Microsystems,
Inc., used under license for technology invented and implemented by Netscape.
SAP, SAP Logo, R/2, RIVA, R/3, ABAP, SAP ArchiveLink, SAP Business Workflow, WebFlow, SAP EarlyWatch, BAPI,
SAPPHIRE, Management Cockpit, mySAP.com Logo and mySAP.com are trademarks or registered trademarks of SAP AG
in Germany and in several other countries all over the world. All other products mentioned are trademarks or registered
trademarks of their respective companies.
Disclaimer: SAP AG assumes no responsibility for errors or omissions in these materials. These materials are provided “as
is” without a warranty of any kind, either express or implied, including but not limited to, the implied warranties of
merchantability, fitness for a particular purpose, or non-infringement.
SAP shall not be liable for damages of any kind including without limitation direct, special, indirect, or consequential
damages that may result from the use of these materials. SAP does not warrant the accuracy or completeness of the
information, text, graphics, links or other items contained within these materials. SAP has no control over the information
that you may access through the use of hot links contained in these materials and does not endorse your use of third party
Web pages nor provide any warranty whatsoever relating to third party Web pages.
© 2006 SAP AG

Sap system copy

More Related Content

What's hot (20)

Similar to Sap system copy (20)

More from Raj p (14)

Recently uploaded (20)

Sap system copy