Rapid Clone

Cloning Oracle Applications Release 11i ITPUB个人空间}6Upz2y4b])MM.baU
with Rapid Clone

August 2006

Cloning creates an identical copy of an existing Oracle Applications system. There are various reasons for cloning an Oracle Applications system such as:

• Creating a copy of the production system for testing updates.
• Migrating an existing system to new hardware.
• Creating a stage area to reduce patching downtime.

This document describes the process of cloning an Oracle Applications Release 11i system. The most current version of this note is document 230672.1 on OracleMetaLink. A FAQ is also available in document 230672.1 on OracleMetaLink. A FAQ is also available in document 216664.1 on OracleMetaLink.document 216664.1 on OracleMetaLink.

• Section 1: Prerequisites ITPUB个人空间ey;u f6p3k
  Tasks to perform before using Rapid Clone.
• Section 2: Clone Oracle Applications 11i ITPUB个人空间m"xZ+~0Qr
  Tasks to prepare, copy, configure, and verify a cloned Applications System.
• Section 3: Finishing Tasks ITPUB个人空间X
  yx&h-u$tV
  Tasks to complete the cloning process.
• Section 4: Advanced Cloning Methods ITPUB个人空间TN6r3NK WOO
  Tasks for advanced options such as refreshing and multi-node cloning.

Conventions

Section 1: Prerequisites

Before cloning, prepare the source system by applying patches and running AutoConfig.

1. Verify source and target nodes software versions
   6\:uB!z"h^J6I"j0In addition to the Oracle Applications software requirements (see Installing Oracle Applications Release 11i Part No. B10638-01), the following software component versions must exist on the source and/or target nodes. The location column indicates the node where the software component must be present.
   The following OS utilities must be in the PATH when running adcfgclone.pl:
   1M,z@PC j$p0   Unix:    make, ld, cc, ar
   ,?AT2[.u q-fU0   Windows:  gnumake, link, cc, cmd
   E[(F5mtc0 -->

   Software	Minimum Version	Location	Details
   Oracle Universal Installer	2.2.0.19	All source system nodes	Apply OUI22 patch 5035661 on every iAS and RDBMS ORACLE_HOME to be cloned. Do not apply the patch to your RDBMS ORACLE_HOME, if you have Oracle 10g or higher installed.
   Perl	5.005	Source and target database nodes	Use the Perl shipped with iAS1022 and RDBMS 9i if available or download it from Perl.com. Perl must be in the PATH before cloning. Windows users: do not use Perl from MKSToolKit.
   JRE	1.1.8	Source database node	If the RDBMS ORACLE_HOME was not installed using Rapid Install, install JRE 1.1.8 into the <RDBMS ORACLE_HOME>/jre/1.1.8 directory.
   JRE	1.3.1	Source database node	Windows users only: install JRE 1.3.1 into the <RDBMS ORACLE_HOME>/jre/1.3.1 directory. See Appendix A for installation instructions.
   JDK	1.3.1	Target middle-tier applications nodes	Refer to Upgrading to JDK 1.3 with Oracle Applications 11i on OracleMetaLink for instructions.
   Zip	2.3	All source nodes	Download from InfoZip. Zip must be in your PATH for cloning.
   OS utilities	N/A	All target nodes	OS utilities (see 11i Install Manual) must be in the the PATH when running adcfgclone.pl (Unix: make,ld,cc,ar - Windows: gnumake,link,cc,cmd)

2. Windows users only:ITPUB个人空间u6e4I%@G+Y4t7Q.X]:UO
   Apply patch 2237858 to enable long file names support.

3. Apply the latest AutoConfig Template patchITPUB个人空间5ks0u4w3X
   Update the Oracle Applications file system with the latest AutoConfig template files by applying the TXK AutoConfig Template rollup patch to all application tier server nodes. ITPUB个人空间*Q-d&hb$hi
   Please refer to document 165195.1 on OracleMetaLink to obtain the latest Autoconfig165195.1 on OracleMetaLink to obtain the latest Autoconfig Template Rollup Patch available in Metalink.

4. Apply the latest Rapid Clone patchesITPUB个人空间4O&J7OOv&Z/n
   Update the Oracle Applications file system with the latest Rapid Clone files by applying the following patches to all application tier server nodes :
   • 3453499 (11i.ADX.F)
   • 5225940 (Post ADX.F Fixes)

5. Setup Rapid Clone on the Applications Tier
   1. If the source Applications system was created with Rapid Install version 11.5.5 or earlier and has not been migrated to AutoConfig, follow the instructions Migrating to AutoConfig on the Applications Tier Migrating to AutoConfig on the Applications Tier in document 165195.1 on OracleMetaLink).

   2. All users must run AutoConfig on the Applications Tier (see section 5: Maintaining System Configuration Maintaining System Configuration in document 165195.1 on OracleMetaLink).

6. Setup Rapid Clone on the Database TierITPUB个人空间3@Z'_U*A@7_;T x
   Implement AutoConfig in the RDBMS ORACLE_HOME (Follow the instructions in section Migrating to AutoConfig on the Database Tier Migrating to AutoConfig on the Database Tier in document 165195.1 on OracleMetaLink). Follow steps 1 and 4 of Section 7 in the AutoConfig document for all versions of RapidInstall and every time you apply a new version of the Rapid Clone patch.

7. Maintain Snapshot informationITPUB个人空间kb9m8F }D'e;u X
   Log in to each Application Tier Node as the APPLMGR user and run Maintain Snapshot information in AD Administration.
   L)pO*}7s0See Oracle Applications Maintenance Utilities for more information.

Section 2: Clone Oracle Applications 11i

Use Rapid Clone to create template files for cloning on the source system. After the source system is copied to the target, Rapid Clone updates these templates to contain the new target system configuration settings. Rapid Clone will not change the source system configuration.

There are three phases to the cloning process:

1. Prepare the Source System
   6wp/m4|3K(l0Execute the following commands to prepare the source system for cloning.
   1. Prepare the source system database tier for cloning
      
      ^~.rD!m eE0Log on to the source system as the ORACLE user and run the following commands:
      nV ?Z#^Y7V(H/s-];Z0 cd <RDBMS ORACLE_HOME>/appsutil/scripts/<CONTEXT_NAME> perl adpreclone.pl dbTier

   2. Prepare the source system application tier for cloning ITPUB个人空间)?Y9Ej8nI.a.@
      Log on to the source system as the APPLMGR user and run the following commands on each node that contains an APPL_TOP:
      !P;xI2z0rx0 cd <COMMON_TOP>/admin/scripts/<CONTEXT_NAME> perl adpreclone.pl appsTier

2. Copy the Source System to the Target System
   (OK}$L uh6X9Ko0Ydq0Copy the application tier file system from the source Applications system to the target node by executing the following steps in the order listed. Ensure the application tier files copied to the target system are owned by the target APPLMGR user, and that the database tier files are owned by the ORACLE user.
   1. Copy the application tier file systemITPUB个人空间Ll.Q0| F+pev#Q
      Log on to the source system application tier nodes as the APPLMGR user.
      • Shut down the application tier server processes
      • Copy the following application tier directories from the source node to the target application tier node:
        • <APPL_TOP>
        • <OA_HTML>
        • <OA_JAVA>
        • <OA_JRE_TOP>
        • <COMMON_TOP>/util
        • <COMMON_TOP>/clone
        • <COMMON_TOP>/_pages (when this directory exists)
        • <806 ORACLE_HOME>
        • <iAS ORACLE_HOME>

   2. Copy the database tier file system
      (au&X6cWU)Q6T L0Log on to the source system database node as the ORACLE user.
      • Perform a normal shutdown of the source system database
      • Copy the database (DBF) files from the source to the target system
      • Copy the source database ORACLE_HOME to the target system
      • Start up the source Applications system database and application tier processes

   Note: Unix/Linux users, make sure that the softlinks are preserved when copying.

3. Configure the Target SystemITPUB个人空间)eKT8L4W"M#f
   Execute the following commands to configure the target system. You will be prompted for the target system specific values (SID, Paths, Ports, etc)
   1. Configure the target system database serverITPUB个人空间,_.xF6X{v!VH&{C
      Log on to the target system as the ORACLE user and type the following commands to configure and start the database:ITPUB个人空间(X
      G@)N"x B2x)U
      cd <RDBMS ORACLE_HOME>/appsutil/clone/bin perl adcfgclone.pl dbTier Note:   To configure the target system database server with database files divided across more than 4 mount points, follow the instructions in Appendix B. -->

   2. Configure the target system application tier server nodesITPUB个人空间s*HMQu ]huVX
      Log on to the target system as the APPLMGR user and type the following commands:ITPUB个人空间Fu!L0u BaI3P8m
      cd <COMMON_TOP>/clone/bin perl adcfgclone.pl appsTier

      Attention: Windows user only: add <806 ORACLE_HOME>bin to the system path before running this step.

Section 3: Finishing Tasks

This section lists tasks that may be necessary depending on your implementation and the intended use of the cloned system.

1. Update profile optionsITPUB个人空间.}~ y0gRF@k5k
   Rapid Clone updates only site level profile options. If any other profile options are set to instance specific values, you must update them manually.

2. Update printer settingsITPUB个人空间:C4nau`%s
   If the new cloned system needs to utilize different printers, update the target system with the new printer settings now.

3. Update workflow configuration settingsITPUB个人空间.xCv8@M6R.Zz"Cn
   Cloning an Oracle Applications instance will not update the host and instance specific information used by Oracle Workflow. Review the following tables and columns to verify there is no instance specific data in the Workflow configuration on the target system.
   -->

   Table Name	Column Name	Column Value Details
   WF_NOTIFICATION_ATTRIBUTES	TEXT_VALUE	Value starts with http://<old web host> : Update to new web host
   WF_ITEM_ATTRIBUTE_VALUES	TEXT_VALUE	Value starts with "http://<old web host> : Update to new web host
   WF_SYSTEMS	GUID	Create a new system defined as the new global database name using the Workflow Administrator Web Applications responsibility.
   WF_SYSTEMS	NAME	Value needs to be replaced with the database global name
   WF_AGENTS	ADDRESS	Update database link with the new database global name.
   FND_FORM_FUNCTIONS	WEB_HOST_NAME	Update with the new web host name
   FND_FORM_FUNCTIONS	WEB_AGENT_NAME	Update to point at the new PLSQL listener name
   FND_FORM_FUNCTIONS	WEB_HTML_CALL
   FND_PROFILE_OPTION_VALUES	PROFILE_OPTION_VALUE
   FND_CONCURRENT_REQUESTS	LOGFILE_NAME	Update with the correct path to the logfile directory
   FND_CONCURRENT_REQUESTS	OUTFILE_NAME	Update with the new directory path on the target system

4. Verify the APPLCSF variable setting
   ?5O.W Xhd*cEx0Source the APPS environment and review that the variable APPLCSF (identifying the top-level directory for concurrent manager log and output files) points to an acceptable directory. To modify it, change the value of s_applcsf in the contextfile and run AutoConfig.

5. Update the SESSION_COOKIE_DOMAIN value in ICX_PARAMETERS
   .C)j[h G^n#E"\0If the target system is in a different domain name than the source system and SESSION_COOKIE_DOMAIN was not null in the source system, update that value to reflect the new domain name.

6. Windows users only reboot the target machineITPUB个人空间.?MO"| ^
   Once cloning is completed reboot to enable the services.

Section 4: Advanced Cloning Options

This section describes advanced cloning procedures.

1. Refreshing a target system
   biTS)\1a_+[eWW9b0You may need to refresh the target system periodically to synchronize it with changes from the source.

   To refresh the target system, perform the following steps as described in previous sections:

   1. Prepare the Source System

   2. Copy the Source System to the Target System
      • Copy the application tier file system if the APPL_TOP, 806 ORACLE_HOME, or iAS ORACLE_HOME needs to be refreshed. Copy the portion of the application tier file system which has been updated.
      • Copy the database tier file system if the RDBMS ORACLE_HOME or the database needs to be refreshed. If refreshing the database, the ORACLE_HOME should be refreshed at the same time.

   3. Configure the Target SystemITPUB个人空间Q:E&b5K;Ux)_ W5I8hI
      Specify the existing target system context file when running adcfgclone.pl commands:
      • perl adcfgclone.pl dbTier <Database target context file>where database context file is:ITPUB个人空间+x EA:](b/b@8c
        <RDBMS ORACLE_HOME>/appsutil/<Target CONTEXT_NAME>.xml
      • perl adcfgclone.pl appsTier <Appltop target context file>where appltop context file is:
        0J:M%ec3dm%t0 <APPL_TOP>/admin/<Target CONTEXT_NAME>.xml

   4. Finishing Tasks

2. Cloning a single-node system to a multi-node systemITPUB个人空间S/J v&`x
   This procedure allows the original single-node system to be cloned into a multi-node system. An Applications system comprises five server types:
   • Database server (database tier)
   • Forms server (application tier)
   • Web server (application tier)
   • Concurrent Processing server (application tier)
   • Administration server (application tier)

   During the single-node to multi-node cloning process, each of these servers can be placed on its own node, resulting in a multi-node target system.

   1. Perform prerequisitesITPUB个人空间N-O)o"URC^
      Perform these steps on all source and target nodes.

   2. Clone Oracle Applications 11iITPUB个人空间0p&e&yE(K}\
      Prepare, copy and configure the cloned Applications System. When creating more than one application tier server node from a single node system, the copy and configure steps must be performed on each target node. You can specify the server type for each target node while answering the prompts during the configuration step.
      &Y*aw-Kya.u?0The database ORACLE_HOME and database only need to be copied to the node on which the database will be run.

   3. Finishing Tasks

   Note: Another way of changing a single-node system to a multi-node system is to share the APPL_TOP. See document 233428.1 on OracleMetaLink document 233428.1 on OracleMetaLink for details.

3. Cloning a multi-node system to a multi-node system with same number of nodes
   T1G w6q z}~$W0A multi-node system can be cloned provided the number of nodes and distribution of servers in the target system matches that of the source system.

   To clone a multi-node system to a multi-node system, perform the cloning process on each node. For example, if the source system contains three nodes, perform the cloning process three times.

   • Clone source system node 1 to target system node 1.
   • Clone source system node 2 to target system node 2.
   • Clone source system node 3 to target system node 3.

   Attention: The database server node must be cloned first.

4. Adding a new node to an existing system
   6W+v6k)YdG2cC0You can use Rapid Clone to clone a node and add it to the existing Application System. The new node will run the same or a subset of the server types running on the source node. Follow the instructions in the Application Tier part of Clone Oracle Applications 11i:

   • Prepare the source system, copy it to the new node and configure it.
   • After adcfgclone.pl completes, source the Applications environment and run the following commands on the target system:
     • cd <COMMON_TOP>/clone/bin
     • perl adaddnode.pl
   Clone Oracle Applications 11i instructions: prepare the source node, and copy it to the new node. Configure the new node, providing the existing Application system database information during the prompts. This will add the cloned node to the existing system. -->

   Note: If the SQL*Net Access security is enabled in the existing system (enabled by default from 11i10), you first need to authorize the new node to access the database through SQL*Net. See Managed SQL*Net Access from HostsManaged SQL*Net Access from Hosts in document 281758.1 on OracleMetalink for instructions on how to achieve this from OAM.

5. Reducing the number of nodes of a multi-node system (merge APPL_TOP)ITPUB个人空间nM3U8o+N@
   You can use Rapid Clone to clone a multi-node system to a single-node system (or a multi-node system with fewer nodes) by merging the APPL_TOPs filesystem together. Follow the procedure for Merging existing APPL_TOPs in document 233428.1 Merging existing APPL_TOPs in document 233428.1 on OracleMetaLink.

6. Cloning a RAC system
   jn9u(J~N0You can use Rapid Clone to clone a RAC system. While cloning, you have the possibility to remove or add nodes to the cluster in the target RAC system. in document 165195.1 on OracleMetaLink in document 165195.1 on OracleMetaLink -->To clone a RAC system, perform the following tasks:
   1. Verify the prerequisites
      • Apply TXK patch 3571388 (AutoConfig support for Oracle RAC instances)
      • Migrate the source system cluster ORACLE_HOMES to AutoConfig as documented in document 165195.1 on OracleMetaLinkdocument 165195.1 on OracleMetaLink

   2. Prepare the Source SystemITPUB个人空间K1X2M.rtW@
      Choose one of the ORACLE_HOMEs on the source cluster. This ORACLE_HOME will be used as a master to create all the target cluster system ORACLE_HOME. Run "perl adpreclone.pl dbTier" on it.

      Note: You also have the option to clone each ORACLE_HOME from the source RAC cluster to the target cluster. In that case, run adpreclone.pl on every ORACLE_HOME.

   3. Copy the Source System to the Target System
      • Copy the master ORACLE_HOME from the source system to each node that will compose the target cluster. For example, if your source RAC system had 2 nodes, and you want to clone it to a 3 node target RAC system, you should copy the source master ORACLE_HOME to the 3 nodes on the target system.
      • Copy the database files from the source to target system.

   4. Configure the Target System
      • Log on to any ORACLE_HOME in the target cluster and run
        • adcfgclone.pl dbTier
        This step configures the first node of the target system RAC cluster and recreates the database control files. The prompts, however, will gather information about every node in the target RAC cluster and the data will be re-used when configuring each subsequent node in the next step.
      • For each remaning RDBMS ORACLE_HOME in the target system, run the following command as the ORACLE file system owner:
        • adcfgclone.pl dbTier
      • If your database is version 10g or higher, run the following commands to register the database and the instance to the cluster manager .
        • srvctl add database -d <database_name> -o <oracle_home>
        • srvctl add instance -d <database_name> ITPUB个人空间)f&h%k#rtB
-i <instance_name> -n <host_name>

      Note: If the cluster manager is not on Linux (ORACM) or the database is version 10g or higher, manually start up the cluster manager on the target system before running adcfgclone.pl

   5. Clone the application tierITPUB个人空间 bV*WT7tw;UNUR
      Follow the steps in section 2 to prepare, copy and configure the application tier. When prompted for the database SID, specify any one of the RAC service names.

   Note: To clone from RAC to non-RAC, follow the same above steps but copy the master ORACLE_HOME to one target node only, and answer "No" to the question "Target instance is a Real Application Cluster (RAC) instance (y/n)", when prompted by adcfgclone.pl.

7. Adding a node to an existing RAC Cluster
   @t2u}n0You can use Rapid Clone to add one or several nodes to an existing RAC Cluster. Perform the following tasks:
   1. Choose any one of the ORACLE_HOMEs in the existing cluster and run "perl adpreclone.pl dbTier" on it. This ORACLE_HOME will be used as a master to create the additional node(s).
   2. Copy the master ORACLE_HOME filesystem to the new node(s).
   3. Reconfigure the RAC Cluster to include the new node(s):
      • Log on to the new node, or any one of the new nodes if adding more than one, and run the following command:
        • cd /appsutil/clone/bin
        • perl adcfgclone.pl dbTier

      Note: You will be prompted for information about all the nodes forming the new cluster and asked to perform the next step before completing adcfgclone.pl on this node.

      • When instructed so by the previous step, log on to every node from the original cluster and run the following command on each of them:
        • cd /appsutil/clone/bin
        • perl adcfgclone.pl addracnode <context file>
      • Go back to the first node and complete adcfgclone.pl
      • If you are adding more than one node to the original cluster, log on to each of the remaining new nodes and run
        • cd /appsutil/clone/bin
        • perl adcfgclone.pl dbTier

   Note: If the cluster manager is not on Linux (ORACM) or the database is version 10g or higher, manually start up the cluster manager on the target systems before running adcfgclone.pl

8. Cloning a Shared File System
   7WBQ!fqK5Xm9o0You can use Rapid Clone to clone a shared APPL_TOP system. Since all the nodes share a unique APPL_TOP you only need to fully clone one of the source system nodes to its target machine and add more nodes directly sharing the APPL_TOP on the target system. Choose one of the shared APPL_TOP nodes on source system (the rest of this section will refer to it as Node A) and perform the following tasks:
   1. Perform prerequisites
      0G3K,} {;AlB0Perform these steps on Node A and and every target nodes.

   2. Clone Oracle Applications 11i
      3zZP1T4WWn0Perform a full clone (Prepare, copy and configure steps) of the Database Tier and Node A (Application tier).

   3. Add the shared file system nodes to the target system
      K {8J(^H#AX b3JzH6h0On the target system, follow the instructions for adding a node to a shared APPL_TOP system in document 233428.1adding a node to a shared APPL_TOP system in document 233428.1 on OracleMetalink and add more nodes sharing the target system APPL_TOP to match the source system topology.

9. Future cloning optionsITPUB个人空间H;|5RjLWM
   Oracle Development is working on additional cloning configuration options. Additional options currently under development include:
   • Using Rapid Clone from OAM.

   -->

Appendix A: Install JRE 1.3.1 into RDBMS ORACLE_HOME

Windows customers will need to perform the following steps:

• Download JRE 1.3.1 from Sun Microsystems.
• Run the install executable
• When prompted for the location to install jre, click browse and enter the location:
  f
  jT}Nk/D0<RDBMS ORACLE_HOME>jre1.3.1
• Accept the default installation options.

JRE 1.3.1 will be installed into the <RDBMS ORACLE_HOME>jre1.3.1 directory.

Appendix B: Recreating database control files manually in Rapid Clone.

This Appendix documents the steps to allow manual creation of the target database control files within the Rapid Clone process. Examples of when to use this method are for databases on raw partitions or hot backup cloning. Replace section 2.3a (Configure the target system database server) with the following steps:

• Log on to the target system as the ORACLE user
• Configure the <RDBMS ORACLE_HOME>
  • cd <RDBMS ORACLE_HOME>/appsutil/clone/bin
  • perl adcfgclone.pl dbTechStack
• Create the target database control files manually
• Start the target system database in open mode
• Run the library update script against the database
  • cd <RDBMS ORACLE_HOME>/appsutil/install/<CONTEXT NAME>
  • sqlplus "/ as sysdba" @adupdlib.sql <libext>where <libext> is "sl" for HP-UX, "so" for any other UNIX platform and not required for Windows.
• Configure the target database (the database must be open)
  • cd <RDBMS ORACLE_HOME>/appsutil/clone/bin
  • perl adcfgclone.pl dbconfig <target context file>where target context file is:
    <RDBMS ORACLE_HOME>/appsutil/<Target CONTEXT_NAME>.xml