Oracle E-Business Suite Maintenance Guide Release 12.2(Patching Utilities)

标签：

更多内容参考: http://docs.oracle.com/cd/E51111_01/current/acrobat/122ebsmt.zip

Oracle Patch Application Assistant

For patches that have manual steps, the patch readme file instructs you to use Oracle Patch Application Assistant (PAA) by running the admsi.pl script. For merged patches, PAA automatically merges the contents of the individual patch readme files.

The Oracle Patch Application Assistant Interface

The Patch Application Assistant is started from the command line, and collects your input in a graphical user interface.

Running Oracle Patch Application Assistant

The following is a summary of the steps you use to run Patch Application Assistant. For a complete description of all the steps, see Creating Customized Instructions for Patching Using PAA.

Step 1: Set the environment

You must set the environment to apply the configuration parameters that define your system. This task is common to many AD utilities.

Step 2: Unzip the patch

Download the patch into the patch top directory ($PATCH_TOP) and unzip it.

Step 3: Review the information in the readme file

In the directory where you unzipped the patch, you will find a README.txt file and a README.html file. Review either of these files for information about the patch and for instructions on using Oracle Patch Application Assistant to generate customized instructions for your system.

Step 4: Run Oracle Patch Application Assistant

Run PAA (admsi.pl) to generate customized instructions for your system. Follow the steps in the customized instructions to complete the patching process.

The adop Utility

You use the adop (AD Online Patching) utility to apply patches to the Oracle E-Business Suite file system or database. You can either allow adop to prompt for the information required to apply a patch, or enter the information without being prompted. Whichever method you choose, adop will then perform the tasks required to apply the patch:

• Reads patch metadata to determine patch dependencies and requirements

• Uploads patch information from a prior patch session to the database (if applicable)

• Reads and validate the patch driver file and reads the product driver files

• Compares version numbers of object modules from the product libraries and version numbers of the existing files against the patch files

• Backs up all existing files that will be changed by the patch

• Copies files

• Archive files in libraries

• Relinks executables

• Generates forms, reports, messages, graphics, and Java archive (JAR) files

• Compiles JSP files and invalid database objects

• Updates database objects

• Runs AutoConfig to update configuration files if any template files are introduced or updated by the patch

• Saves patch information to the database

Be aware of the following important points about adop:

• The adop utility always runs from the run edition file system. It automatically sets its environment correctly, regardless of the edition it is run from. Editions are described in more detail later.

• If a patch contains no new updates to files or database objects in your system, adop takes no action.

• If adop detects a previously failed patching session, it will attempt to recover that session.

adop Parameters

Run from the command line, adop accepts many parameters. Some are required, while others are optional. Some parameters override other parameters, and some have a higher order of precedence over others. All the parameters must be entered in name=value pairs.

Online Help

To obtain help about the basics of adop operation, enter the command:

adop -help

The help usage statement will also appear if you supply an invalid parameter on the adop command line.

Optionally, you can also display examples of the various adop parameters by entering the command:

adop -examples

The Input File

adop also accepts parameters in an input file. From the command line, you specify an input file by using the parameter input_file=<myinput.txt>, where myinput.txt is the name of your input file.

Input File Parameters

Note: You should always provide the full path to the input file.

Major parameters that can be specified in the input file include:

patches
phase
patchtop
merge
defaultsfile
abandon
restart
workers

Input file parameters must appear in name=value format, with one parameter per line. For example:

phase=apply
patches=123456
workers=8
autoskip=yes

Note: If you supply a parameter to the input file twice (for example, workers is defined on both lines 2 and lines 5 of your input file), the last definition (in this example, on line 5) will be used.

The Defaults File

Parameters can also be passed to adop into adop through a defaults file. From the command line, you can specify a defaults file by using the parameter defaultsfile=<mydefaults.txt>, where mydefaults.txt is the name of your file.

Your own defaults file will be checked the validity of its contents, and if issues are found an error will be raised. If you do not specify a custom defaults file, adop will use the one that is automatically generated by the system (using AutoConfig).

If adop is being run in hotpatch mode, your own defaults file should be located on the run file system, under $APPL_TOP/admin/$TWO_TASK. Otherwise, the defaults file should be in the same location, but on the patch file system.

Note: Instead of using your own defaults file, it is generally preferable to supply your own parameters via the command line or in an input file. Parameters supplied in either of these ways take precedence over parameters in the the defaults file.

Parameters In Defaults File

Only one parameter, patchtop, can currently be defined in the defaultsfile. This parameter is used to specify the location where patches are unloaded. The default patchtop directory is on the non-editioned file system, at $APPL_TOP_NE/EBSapps/patch.

If you wish to use the patchtop supplied in the defaults file, you must specify the defaults file as a parameter either on the command line or in the input file. If you do not specify the defaults file in one of these two locations, the file will not be read and the defaults file patchtop will not be used.

Order of Parameters

As described above, most parameters can be defined in at least two locations, with patchtop able to be defined in three different locations. If multiple different definitions are specified, the following order is used.

1. Command Line: An adop parameter specified on the command line will take precedence over all others.

2. Input File: An adop parameter given here will only be lower in precedence to a parameter specified on the command line.

3. Defaults File: Parameters defined here have the lowest level of precedence.

Important: Because you can supply higher-priority parameters on the command line and in the input file, you should never need to edit the defaults file.

Preparing your System for Patching

Before you begin a patching cycle, there are some important tasks you need to complete.

In Release 12.2, it is more appropriate to think in terms of a patching cycle than a single patching operation. The online patching cycle consists of a number of phases:

1. Prepare

2. Apply

3. Finalize

4. Cutover

5. Cleanup

You specify the desired phase or phases as arguments to the adop utility. The actions taken in these phases are described in Oracle E-Business Suite Concepts. This and the next chapter of this book provide details of the available options.

Note: The adop utility sets its own environment. There is therefore no need to source the environment before running it.

Summary of Fundamental Patching Operations

Patching is performed in several phases, which are specified on the adop command line as follows:

adop phase=<phase_name>

Prepare phase - Used to start a new online patching cycle:

$ adop phase=prepare

Apply phase - Used to apply a patch to the patch edition of a system:

$ adop phase=apply patches=123456 workers=8

Finalize phase - Used to perform the final patching operations that can be executed while the application is still online:

$ adop phase=finalize

Cutover phase - Used to perform the transition to the patched environment:

$ adop phase=cutover

Cleanup phase - Used to remove old objects that are no longer needed:

$ adop phase=cleanup

Abort phase - Used only if the online patching cycle needs to be terminated before completion:

$ adop phase=abort

Important: The abort phase can only be run after the prepare phase and before completion of the cutover phase. After the cutover phase completes, it is no longer possible to abort the online patching cycle.

The adop phases are described in more detail in The Online Patching Cycle section of the Patching Procedures chapter.

Patch Log Files

It is advisable to review the relevant log files after any patching operation. The adop log files are located on the non-editioned file system (fs_ne), under:

$NE_BASE/EBSapps/log/adop/<adop_session_id>/<phase>_<date>_<time>/<context_name>/log

For example, if s_ne_base was /u01/R122_EBS/fs_ne, the session ID was 15, and the <CONTEXT_NAME> was patch01_testsys, the path to the adop log files from 9th July 2014 would resemble this:

/u01/R122_EBS/fs_ne/EBSapps/log/adop/15/apply_20140709_112226/patch01_testsys/log

Patch Log File Directory Structure

Note: The environment variable for s_ne_base is called $NE_BASE.

Other log files are created for specific purposes, for example, to record all the actions associated with parallel workers.These worker log files are written to the non-editioned file system, under EBSapps/log/adop/<adop_session_id>/<phase_timestamp>. For example, /u01/R122_EBS/fs_ne/EBSapps/log/adop. Review these files when the patching session is complete.

Other AD log files include those shown in the following table:

If adop does not perform an action, it does not generate the log file associated with that type of action.

Note: You can also review log files using the View Log Files feature of OAM Timing Reports. See: View Log Files.

JAR File List

In the apply phase of an online patching cycle, adop creates a file called jarlist.txt. This file is provided in case you wish to perform your own JAR file signing using a very secure certificate. In such cases, you will need to specify the adop command line parameter option=nojarsigning in order to bypass the standard JAR file signing activity that will otherwise performed by AD.

The jarlist.txt file is placed in the same directory as the patch log file. The following example is for patch 13358502, which was applied during a patching session that had ID=14: $NE_BASE/EBSapps/log/adop/14/apply_20130515_125116/testenv_sys3220410/13358502/log/jarlist.txt.

Sessions

The adop utility maintains a session for each online patching cycle. A new session is created when you run the prepare phase. Each session is given a numeric ID number. The session is used to maintain the state of the online patching cycle across the various adop phases and commands. You can only run one adop session at a time on a particular Oracle E-Business Suite system.

Patch Directory (Patchtop)

The directory where the patch files have been unzipped is referred to as the patchtop. The default patch top directory is $NE_BASE/EBSapps/patch, which is pointed to by the $PATCH_TOP environment variable.

If specifying your own choice of directory, you must supply the full path, and the operating system user that is running adop must have write permissions to that directory. Oracle recommends using the default $PATCH_TOP direcotory.

Note: If you have a multi-node environment, you must download and unzip the patches (under $APPL_TOP_NE/EBSapps/patch) on all the nodes.

If you want to merge patches before applying them, you need to download and unzip all the individual patches in the same location as that of the merged driver file. For example, if you merge patches 111, 222, and 333 (using AD Merge Patch), and the merged patch driver file location is $PATCH_TOP/mergetest/mergetest.drv, you should then download and unzip the individual patches as $PATCH_TOP/patch/mergetest/111, $PATCH_TOP/mergetest/222, and $PATCH_TOP/mergetest/333.

Patch Driver File

The unified driver, named u<patchnum>.drv, contains the commands necessary to change files and database objects, and to generate new objects. It contains copy, database, and generate portions and performs the copy, database, and generate actions in the stated order. The adop utility runs the unified driver on all APPL_TOPs, but only performs the specific actions that are actually required for each APPL_TOP.

Copy Portion of a Unified Driver

When the copy portion of a unified driver runs, adop performs the following actions:

• Extracts the appropriate files from the C library of each product.

• Compares the extracted object modules with their corresponding files in the patch directory. It also makes this type of comparison with files such as forms, reports, and SQL scripts.

• Backs up any product file with a more recent version in the patch directory to a subdirectory in the patch directory. For example, if <patch_dir> is the patch directory, <system_name> is the applications system name, <appl_top_name> is the APPL_TOP name, and <prod> is the name of the product being patched, it backs up:

  <PROD>_TOP/<subdir(s)>/<old_file_name>
  

  to

  <patch_dir>/backup/<system_name>/<appl_top_name>/<prod>/<subdir(s)>/<old_file_name>
  

  Note: The Applications system name and the APPL_TOP name are determined during the Rapid Install process.

• Replaces the outdated files of each product with newer files from the patch directory.

• Loads the new object modules into the C libraries.

• Relinks the Oracle E-Business Suite products with the operating system, Oracle server, and other Oracle products libraries.

• Applies changed Java class files and regenerates JAR files as needed.

• Copies any specified HTML or media files to their respective destinations.

• Compiles out-of-date Java Server Page (JSP) files (if any JSP files are included in the patch).

Database Portion of a Unified Driver

When the database portion of a driver runs, adop performs these actions:

• Gets a list of current invalid objects in the APPS schema.

• Determines whether the action was performed in a previous patch.

• Runs SQL scripts and EXEC commands, which change Oracle E-Business Suite database objects. By default, adop runs scripts and commands in parallel.

• Compiles invalid objects in the database.

• Assembles a list of current invalid objects in the APPS schema.

Generate Portion of a Unified Driver

When the generate portion of a driver runs, adop performs these actions:

• Generates Oracle Forms PL/SQL library files

• Generates Oracle Forms menu files

• Generates Oracle Forms executable files

• Generates Oracle Reports PL/SQL library files

• Generates Oracle Reports files

• Generates message files

• Generates Oracle Workflow resource files

Number of Parallel Workers

By default, adop runs database updates and file generation commands in parallel using multiple workers. The default number of workers is computed based on the system hardware configuration, but the number can be specified explicitly using the ‘workers‘ parameter. Tasks are assigned to workers, the workers run the tasks to completion, and adop assigns new tasks.

adop runs (adop) all database actions based on phase order, a grouping of actions in the database portion of the patch that minimizes dependencies. This order is not necessarily the order in which the commands are listed in the database portion of the patch driver.

Note: For more information, see Using Parallel Processing in the Maintenance section of this book.

Customized Files

adop reviews the AD_FILES table to determine if any customized files (Register Flagged Files) will be replaced by the patch. If so, it displays a message listing the customized files it will replace.

Note: For more information, see Customization Standards, Oracle E-Business Suite Developer‘s Guide, and Register Flagged Files.

NLS

If the patch you are applying has an NLS-related version, and if you are an NLS customer, adop prompts you about the NLS-related version of the patch before allowing you to continue.

Principles of Non-Interactive Patching

Non-interactive patching saves time by automating the patching process. It is used with all the major phases of adop, including the apply phase.

The adop utility runs non-interactively by default. You must specify all required parameters for each adop command, either on the command line or in an input file.

After the patching actions are complete, you perform any post-patching steps listed in the patch readme file. See Performing Non-Interactive Patching, Oracle E-Business Suite Maintenance Guide.

Messages

adop generates several types of messages. Each message is recorded in a log file. See Log Files for a list and descriptions.

Informational Messages

Informational messages are written to the informational message file. This log file uses the same base file name as the main adop log file, but substitutes a .lgi extension for the .log extension. For example, if the adop log file is named u1234567.log, the adop informational log file is named u1234567.lgi.

For example, adop writes information pertaining to the files not updated because they are up-to-date in the informational log file.

File will not be copied to destination.

Version check:
/slot03/appmgr/prodappl/ad/12.2/xml/oam/patch/history/SearchFiles.uix
version is equal to or lower than
/slot03/appmgr/prodcomn/html/oam/patch/history/SearchFiles.uix.
File will not be copied to destination.

Version check:
/slot03/appmgr/prodappl/ad/12.2/xml/oam/patch/history/SearchFilesCriteriaAdvanced.uix
version is equal to or lower than

/slot03/appmgr/prodcomn/html/oam/patch/history/SearchFilesCriteriaAdvanced.uix

Error Messages

When adop is using parallel processing and an error occurs, the job fails. Review the main adop log file and the adworkxxx.log file to determine the source of the error, resolve the issues and continue. Restart adop using the adctrl command.

Note: See Monitoring and Controlling Parallel Processing, , for details on using the adctrl command.

If you cannot resolve the issue, you must:

• Verify that all steps in the readme file were completed.

• Check My Oracle Support for additional information regarding the patch you are applying.

If the message indicates that a worker has failed its job, you can fix the problem and restart the worker while the manager is running. Some failed jobs are deferred (not immediately reassigned) by the manager. These jobs do not cause the manager or other workers to stop.

See: Managing Worker Processes in this book.

Successful Completion Message

adop displays a success message when processing is complete. If you do not see a such a completion message, you should investigate and identify the reason.

Backup Directory

When adop runs, a backup directory is created in the directory where you unzip the patch. The old version of each file updated by the patch is copied into the backup directory. When applying large patches (such as release update packs, product family RUPs, and pre-upgrade patches), ensure there is enough disk space on the system where you unzip the patch, or the patching process may fail. We recommend having at least twice the amount of disk space as the unzipped patch file uses.

Tip: When there is no patching cycle in progress, you can if desired delete the files in the backup directory to free the space.

adop Patching Modes

The adop utility is normally used execute an online patching cycle, but patches can also be applied in other modes for special situations. If these special options are available or required, they will be documented in the patch readme file or other installation instructions.

Downtime Mode

To optimize the process of upgrading to Oracle E-Business Suite Release 12.2, support is provided for the capability to apply Oracle E-Business Suite patches in downtime mode. When applying patches in this mode, adop will first confirm that the application tier services are down, and will then proceed to apply the patch to the run edition of the Oracle E-Business Suite database and file system. Downtime mode patching does not use an online patching cycle. The process of applying a patch in downtime mode completes more quickly than in online mode, but at the cost of increased system downtime.

To run adop in downtime mode, you use the following command line options. In this example, patch 123456 is applied in downtime mode:

$ adop phase=apply patches=123456 apply_mode=downtime

Important: Be aware that:

• Release 12.2 patches are not normally tested in downtime mode.

• Downtime mode is only supported for production use where explicitly documented, or when directed by Oracle Support or Development.

Pre-Install Mode

Pre-install mode is generally used during the upgrade process to update AD utilities, apply pre-upgrade patches, or work around other patching issues. adop asks all startup questions except those relating to the database.

Warning: Run adop in pre-install mode only if the patch readme instructs you to do so.

To run adop in pre-install mode, include preinstall=y on the adop command line. It performs the following actions:

• Compares version numbers

• Copies files

• Relinks FND and AD executables

• Saves patch information to the file system

Because adop does not read driver files in pre-install mode, it copies all product files in the patch to the APPL_TOP directory. Additionally, even if a file in the patch should be both in the APPL_TOP and in another directory (such as in $OA_HTML), adop copies the file only to the APPL_TOP.

In preinstall mode, adop validates codelevels against the files Preinstall_Codelevel_AD.txt and Preinstall_Codelevel_MP.txt. These files are located in the $APPL_TOP/admin directory, and contain codelevel information about AD and other products registered in the database tables.

Since no database connection is available in pre-install mode, adop tries to validate whether the current patch should be applied based on the codelevel information in these two files, as follows:

• If Preinstall_Codelevel_AD.txt is missing from the APPL_TOP, adop will apply the patch in pre-install mode without validating the patch for codelevel compatibility.

• If Preinstall_Codelevel_MP.txt is missing from the APPL_TOP, adop will proceed with patch application without validating the patch for codelevel compatibility of the entities.

• If both files are missing, adop will not validate codelevels in pre-install mode.

Note the following restrictions when applying a patch in pre-install mode:

• NLS patches cannot be applied on the instance.

• Baseline or codelevel-introducing patches cannot be applied on the instance.

• adop will not check to see if the patch is already applied on the system.

Test Mode

In test mode, adop does not apply the patch. Instead, it lists each file it would have copied, relinked, executed, or generated and shows exactly what actions it would have performed had it applied the patch. It also runs AutoConfig in test mode to determine any impending changes to the configuration files. This allows you to see the effects of the patch on your production system before you apply it.

To run adop in test mode, specify the apply=no parameter. In test mode, adop will go through the process of applying the patch but will not perform any of the following actions:

• Copy files from the patch directory to the Oracle E-Business Suite file system

• Archive object modules into the product libraries

• Relink executables

• Generate forms, reports, PL/SQL libraries, or menu files

• Run SQL or EXEC commands (commands that change the database)

• Instantiate new configuration files

• Update the patch information files

• Update patch information and release version in the database

See: Testing a Patch Before Applying It, Oracle E-Business Suite Maintenance Guide.

adop Command Line Arguments

You can adjust the way adop operates by supplying arguments to the various parameters that adop recognizes. Arguments can be passed either directly on the command line or through an input file. adop is non-interactive (except for passwords), so all required arguments must be specified when entering an adop command.

An input file is specified as follows:

$ adop phase=apply input_file=<input_file.txt> 

Arguments are specified on the command line or input file in "name=value" format, where "name" is the adop parameter name and "value" is your specified value. Parameter names are specified in lower case, and parameter values should be assumed to be case-sensitive.

You can enter more than one token=value argument on a single command line by separating them with a single space, as in the following example.

printdebug=y flags=hidepw

In some cases, you can include more than one value for a token. When doing so, you separate the values with commas and no spaces. For example:

flags=nohidepw,trace

is valid, but

flags=nohidepw, trace

is not valid.

adop Options

The "options" argument is used to pass options that control how the patch is applied. It takes the form of a comma-separated list. Enter a single option, or a comma-separated list of options such as adop options=nocopyportion,nogenerateportion.

Note: As with adop arguments, there must be no space after the comma.

 [WARNING] Skipping the application of patch
14125999_AR since it has been already applied
[WARNING]  Hint: Patches can be applied again by
specifying options=forceapply when invoking adop 

The adop Interface

adop is run from the command line. It prompts for required passwords, but expects all other input parameters to be specified on the command line or in an input file.

Running adop

The following is a summary of the steps you use to run adop. For a complete procedural description of all the steps, see Creating Customized Instructions for Patching Using PAA.

Step 1: Set the environment

You must set the environment to apply the configuration parameters that define your system. This task is common to many AD utilities. Your Oracle E-Business Suite installation includes a shell script in the root installation directory that helps you easily set the run edition or patch edition environment. For example:

$ . <EBS_ROOT>/EBSapps.env run

Step 2: Unzip the patch

Download and unzip the patch into the patch top directory, which is identified by the $PATCH_TOP environment variable.

Step 3: Review the information in the readme file

In the directory where you unzipped the patch, you will find a README.txt file and a README.html file. Review either readme file for information about the patch and for instructions on using Oracle Patch Application Assistant (PAA) to generate customized instructions for your system.

Step 4: Run Oracle Patch Application Assistant

Run PAA (admsi.pl) to generate customized instructions for your system. Follow the steps in the customized instructions to start the patching process.

Step 5: Run adop

The customized instructions generated by PAA describe how to run adop using the adop command.

Note: You can add arguments on the command line to refine the way adop runs. See adop Modes and Command Line Arguments.

A simple example of the commands to execute a complete online patching cycle for patch 123456 is as follows:

$ . <EBS_ROOT>/EBSapps.env run
$ adop phase=prepare
$ adop phase=apply patches=123456
$ adop phase=finalize
$ adop phase=cutover
$ . <EBS_ROOT>/EBSapps.env run
$ adop phase=cleanup

Monitoring Status

You can obtain a brief report for the current patching session by running the command:

$ adop -status

This will display information that includes phases completed and the time taken. In the example below, the current patching session ID is 5:

Current Patching Session ID: 5
Node Name       Node Type       Phase       Status          Started          
              Finished                       Elapsed
 --------------- --------------- ----------- ---------------
 ------------------------------ ------------------------------ ------------
 patchtest1      master          PREPARE     COMPLETED       06-MAY-13
 11:31:38 -07:00      07-MAY-13 12:27:51 -07:00      0:56:13
                                 APPLY       COMPLETED       07-MAY-13
 04:19:17 -07:00      07-MAY-13 04:43:12 -07:00      0:23:55
                                 CUTOVER     COMPLETED       07-MAY-13
 05:54:03 -07:00      07-MAY-13 05:57:57 -07:00      0:03:54
                                 CLEANUP     COMPLETED       07-MAY-13
 09:14:33 -07:00      07-MAY-13 09:22:46 -07:00      0:08:13

The output is also recorded in a log file, called /u01/R122_EBS/fs_ne/EBSapps/log/status_<YYYYMMDD>_<HHMMSS>/adzdshowstatus.out. For example, /u01/R122_EBS/fs_ne/EBSapps/log/status_20130507_111647/adzdshowstatus.out.

Two additional options with this command are as follows.

• If you want information about a particular session, specify the relevant session ID:

  $ adop -status <session ID>

• If you want additional details of operations performed:

  $ adop -status -detail

  This option will give a summary of last ten adop session IDs, and full details of the file system and database changes introduced by a patch. It also shows the log file location of the current patching cycle.

Aborting an Online Patching Cycle.

If a patching cycle is failing and the issue cannot be resolved quickly, it is possible to abort the patching cycle and return to normal runtime operation. The patch edition will be dropped.

You can abandon a patching cycle (without applying any patches) by running the command:

$ adop phase=abort

Important: The abort command is only available until successful completion of the cutover phase. After cutover completes, the system is running on the new edition and abort is no longer available for that patching cycle.

Warning: Aborting a patching cycle will drop the patch edition, but you must then run the cleanup and fs_clone phases before starting a new patching cycle. The cleanup must be a full cleanup ($ adop phase=cleanup cleanup_mode=full).

For example:

$ adop phase=prepare
$ adop phase=apply patches=123456
[Patch application encounters problems and you want to abort]
$ adop phase=abort
$ adop phase=cleanup cleanup_mode=full
$ adop phase=fs_clone

Restarting adop

If you have shut down the workers, or if adop quits while performing processing actions, it saves all the actions completed up to that point in restart files. Investigate and resolve the problem that caused the failure, then restart adop. After you restart adop, it will ask if you want to continue with the previous session (at the point where the processing stopped), or start a new session.

Note: A difference from adpatch is that adop restart behavior is controlled by the abandon=yes/no and restart=yes/no options in the input_file that can be passed to the adop command in the apply phase.

See: Restarting a Utility in this book.

AD Merge Patch

Important: The functionality of AD Merge Patch is now included in the adop tool. If you want adop to merge patches, you must explicitly specify merge=yes when invoking adop. AD Merge Patch is still supported, however, and its usage is described in this section.

When a series of patches are applied individually, some patching actions (such as linking executables) may need to be performed repeatedly, which can take a lot of time. Also, in some special cases a corrective patch must be merged with a base patch in order to have the desired effect. In both these scenarios it is beneficial to merge multiple patches into a single merged patch, and then apply the merged patch.

An alternative is to use AD Merge Patch. This utility merges multiple patches into a single patch, allowing you to reduce patch application time by eliminating the tasks you would otherwise have to have performed for each individual patch.

When merging compatible patches, AD Merge Patch bases its actions on metadata. It removes duplicate driver lines from the database portions of the driver. When merging two or more patches that have manual steps, the steps and readme files of both patches are also merged.

Source and Destination Directories

You extract the patches to be merged from the source directory. The destination directory is where the merged patch is created. AD Merge Patch reads the patch driver files for each patch in the source directory and merges them to create patch driver files in the destination directory. If a file exists in more than one source patch, only the highest revision of the file is copied to the destination directory.

The source and destination directories should be created under the same parent directory. For example, if the parent directory is named <top>, both the source and destination directories should be subdirectories of <top>. The source and the destination directories cannot be child or parent directories of each other.

Directory Structure for Source and Destination Directories - Basic Example

The source directory must have all patches to be merged as immediate child directories. The patch directories cannot be in a lower directory. For example, a directory structure for merging four patches would look like this:

Directory Structure for Source and Destination Directories - Merging Four Patches

Naming the Merged Patch

You should indicate the name of the merged patch on the command line, using the -merge_name option to provide a meaningful name. If you do not use this option, the patch will be given the default name of merged.

Merging Zipped Patches

The manifest file is a text file in which you document the location and names of the patch zip files. The contents of a manifest file resemble the following:

/d01/prodappl/patches/p3903945_12_GENERIC.zip
/d01/prodappl/patches/p3892799_12_GENERIC.zip
/d01/prodappl/patches/p3874740_12_LINUX.zip

You can use the -manifest option to create a manifest file. AD Merge Patch references this file, and unzips the patches listed. It copies the unzipped files into the source directory and includes them, along with any other files in the source directory, in the merged patch.

The AD Merge Patch Interface

You run AD Merge Patch and supply the information it needs from the command line. There are no menus or input screens.

Running AD Merge Patch

AD Merge Patch is located in the AD_TOP/bin directory. However, you run it from the parent directory (<top>) of the source directory. The following is a summary of the steps you use to run AD Merge Patch.

Step 1: Set the environment

You must set the environment to indicate the location of the configuration parameters that define your system. This task is common to many AD utilities.

Step 2: Run AD Merge Patch

From the parent directory (<top>), run AD Merge patch using the admrgpch command.

Using AD Merge Patch With adop

You can use AD Merge Patch to merge patches, and then apply them with adop.

In this example, patches 111 and 222 are to be applied after merging. The two patches are staged in <FS_NE>/patch.

$ admrgpch -s <source_directory> -d <destination_directory>
$ pwd
/u01/R122_EBS/fs_ne/EBSapps/patch
$ admrgpch -s /u01/R122_EBS/fs_ne/EBSapps/patch -d /u01/R122_EBS/fs_ne/EBSapps/patch/test                   

Now patches 111 and 222 have been merged, and the merged patch‘s patchtop is /u01/R122_EBS/fs_ne/EBSapps/patch/test:

$ pwd
/u01/R122_EBS/fs_ne/EBSapps/patch/test
$ ls
fnd  u_merged.drv

You can then apply this patch:

$ adop phase=apply

During the next prepare phase, the synchronization steps expect the individual patches (111, 222) to be present under the merged patch‘s patchtop; in this example, /u01/R122_EBS/fs_ne/EBSapps/patch/test. You therefore need to move 111 and 222 to under /u01/R122_EBS/fs_ne/EBSapps/patch/test before running the next prepare:

$ pwd
$ /u01/R122_EBS/fs_ne/EBSapps/patch
$ mv 111 /u01/R122_EBS/fs_ne/EBSapps/patch/test
$ mv 222 /u01/R122_EBS/fs_ne/EBSapps/patch/test
$ cd /u01/R122_EBS/fs_ne/EBSapps/patch/test
$ ls
111  222  fnd  u_merged.drv

Oracle E-Business Suite Maintenance Guide Release 12.2(Patching Utilities)

标签：

原文地址：http://blog.csdn.net/cai_xingyun/article/details/42044097