Download and Unzip the Patch

Oracle recommends that you download and use the latest released OPatch version. Download the latest version from My Oracle Support:

My Oracle Support patch 6880880

You unzip the patch as the owner of the Oracle software binaries. This is the Oracle user for Oracle Database and Oracle Real Application Clusters (Oracle RAC).

ORACLE_HOME/OPatch

For exact instructions to install OPatch, follow the readme included with the tool download.

Guidelines for OPatch or OPatchAuto Maintenance

Release Updates (RUs) from Automated Release Update (ARU) or My Oracle Support are cumulative. That is, the content of all previous RUs for a given software release are included in the latest RU. To increase performance by deleting inactive patches, Oracle recommends that you run OPatch with the listorderedinactivepatches and deleteinactivepatches options. For more information, see:

OPatch 12.2.0.1.37+ Introduces a New Feature to Delete Inactive Patches in the ORACLE_HOME/.patch_storage Directory (Doc ID 2942102.1)

To use OPatch or OPatchAuto to install the latest software update for your Oracle software, the software home must have the latest release software already installed. For example, to install the latest RU for Oracle Database 23ai, you must first have installed Oracle Database 23ai in the Oracle home where you want to apply the most recent RU. You can find more information about your patch requirements and options in the Download Reference for software updates:

Assistant: Download Reference for Oracle Database/GI Update, Revision, PSU, SPU(CPU), Bundle Patches, Patchsets and Base Releases (Doc ID 2118136.2)

Patches typically include Java Development Kit (JDK) fixes released since the previous RU, and will update the JDK in the Oracle home.

The UTL_URI.ESCAPE function is now compliant with RFC 3896 and will treat "#" as a reserved character. For more information, see:

UTL_URI.ESCAPE function is now compliant with RFC 3896 and will treat “#” as a reserved character (Doc ID 2981395.1)

Deployment Guidelines for Oracle RAC or Oracle Grid Infrastructure

To deploy RUs or patches manually, you can use OPatchAuto. These updates can be installed as rolling patches for Oracle Grid Infrastructure or Oracle RAC.

You no longer require a separate Oracle Database Embedded JVM (OJVM) RU bundle patch. OJVM fixes can now be applied as rolling updates to Oracle RAC, and are included in the Database RU bundles. For more information, See:

Primary Note for Database Proactive Patch Program (Doc ID 888.1)

Validating the Oracle Inventory on Oracle Grid Infrastructure and Oracle RAC Cluster Members

Before beginning patch application, check the consistency of inventory information for the Grid home and each Oracle home that you plan to patch. Run this command as the respective Oracle home owner to check the consistency, where <ORACLE_HOME> is the Oracle home or Grid home:

$ <ORACLE_HOME>/OPatch/opatch lsinventory -detail -oh <ORACLE_HOME>

If this command succeeds, then it lists the Oracle components that are installed in the home. Save the output so that you have the status of the inventory before you start the patch application.

If this command fails, then contact Oracle Support for assistance.

Deployment Guidelines for Oracle Data Guard

To deploy RUs and patches in Oracle homes deployed with Oracle Data Guard, you can use OPatchAuto. For information about how to apply updates in homes configured with Oracle Data Guard, and how to reduce downtime, see:

Oracle Patch Assurance - Data Guard Standby-First Patch Apply (Doc ID 1265700.1)

For the latest Update with Security Fixes that should be used on Client-Only installations, refer to the Critical Patch Update (CPU) Program Patch Availability Document (PAD) section on Oracle Database for the cycle in which you are interested.

Ensure that the $PATH definition has the following executables:

• make
• ar
• ld
• nm

The location of these executables depends on your operating system. On many operating systems, they are located in /usr/ccs/bin. Ensure that your path environment includes the location of the executables. For example:

export PATH=$PATH:/usr/ccs/bin

Oracle recommends that you use the "Patch Recommendations and Patch Plans" features on the Patches & Updates tab in the My Oracle Support search results page for a patch search. These features are the fastest and easiest way to determine whether you have interim patches in the Oracle home that conflict with a Release Update (RU), and to obtain the conflict resolution patches that you will require. The "Patch Recommendations and Patch Plans" features work in conjunction with the My Oracle Support Configuration Manager. You can find recorded training sessions for these features on My Oracle Support:

My Oracle Support How-to Video Training Series (Doc ID 603505.2)

If you choose not to use My Oracle Support Patch Plans, then My Oracle Support Conflict Checker tool enables you to upload an OPatch inventory and check the patches that you want to apply to your environment for conflicts.

For details about using this tool, see:

How to Use the My Oracle Support Conflict Checker Tool for Patches Installed with OPatch. (Doc ID 1091294.1)

The procedure to use the tool is as follows:

1. Download the patch from My Oracle Support. Patches are zip archive file format files in the form <patchnumber>_<version>_<platform>.zip, where <patchnumber> is the eight-digit number of the patch, <version> is the release of the Oracle software for which the patch is provided, and <platform> is the operating system platform for which the patch can be applied. For example: p12345678_23.0.0.0.0_Linux-x86-64.

2. Unzip the patch, and use the OPatch command opatch prereq CheckConflictAgainstOHWithDetail to find if any currently installed interim patches conflict with the patch being installed. For example, if the patch is p12345678_23.0.0.0.0_Linux-x86-64:

   unzip p12345678_23.0.0.0.0_Linux-x86-64.zip
   cd p12345678_23.0.0.0.0_Linux-x86-64
   opatch prereq CheckConflictAgainstOHWithDetail -ph ./
   

3. Review the report. If there are conflicts, then the report indicates the patches that conflict and the patches that are a superset.

4. Resolve the patch conflicts for each conflicting patch. Determine if a conflict resolution patch is already available, and if you need to request a new conflict resolution patch or if the conflict may be ignored. If you require additional conflict resolution patches, then you must wait until they become available before proceeding further.

   To resolve patch conflicts, use the following My Oracle Support document:

   Database Patch Conflict Resolution (Doc ID 1321267.1)

After you resolve patch conflicts, or receive any interim conflict resolution patches, you can proceed with updating your Oracle software.

The UPGRADE CATALOG command must be entered twice to confirm the upgrade.

Enter the following commands:

• For the Oracle Grid Infrastructure home (Grid home), as the Grid user:

  1. Create file /tmp/patch_list_gihome.txt with the following content (in this example, the patch location is /usr/home/oracle/patch/OPatch):

     % cat /tmp/patch_list_gihome.txt
     
     /usr/home/oracle/patch/OPatch/p12345678_23.0.0.0.0_Linux-x86-64/% ARU_Database_SubPatchTrackingBug %
     /usr/home/oracle/patch/OPatch/p12345678_23.0.0.0.0_Linux-x86-64/% ARU_OCW_SubPatchTrackingBug %
     /usr/home/oracle/patch/OPatch/p12345678_23.0.0.0.0_Linux-x86-64/% ARU_ACFS_SubPatchTrackingBug %
     /usr/home/oracle/patch/OPatch/p12345678_23.0.0.0.0_Linux-x86-64/% ARU_RHP_SubPatchTrackingBug %
     /usr/home/oracle/patch/OPatch/p12345678_23.0.0.0.0_Linux-x86-64/% ARU_TOMCAT_SubPatchTrackingBug %
     /usr/home/oracle/patch/OPatch/p12345678_23.0.0.0.0_Linux-x86-64/% ARU_DBWLM_SubPatchTrackingBug %
     

     Note:

     For HP-UX Itanium and Linux on IBM System z platforms, the last two rows in the previous example should not be added to the patch_list_gihome.txt file.

  2. Run the OPatch command to check if enough free space is available in the Grid home:

     % $ORACLE_HOME/OPatch/opatch prereq CheckSystemSpace -phBaseFile /tmp/patch_list_gihome.txt
     

• In the Oracle software home, as the software home user:

  1. Create file /tmp/patch_list_dbhome.txt with the patch content in the location where you have unzipped the patch. For example, if the patch is p12345678_23.0.0.0.0_Linux-x86-64::

     % cat /tmp/patch_list_dbhome.txt
     <patch-location>/p12345678_23.0.0.0.0_Linux-x86-64/% ARU_Database_SubPatchTrackingBug %
     <Patch-location>/p12345678_23.0.0.0.0_Linux-x86-64/% ARU_OCW_SubPatchTrackingBug %
     

  2. Run the OPatch command to check if enough free space is available in the Oracle home:

     % $ORACLE_HOME/OPatch/opatch prereq CheckSystemSpace -phBaseFile /tmp/patch_list_dbhome.txt
     

The command output reports pass and fail messages in response to the system space availability:

• If OPatch reports Prereq "checkSystemSpace" failed, then the required amount of space is not available, so you must free up space on the system.

• If OPatch reports Prereq "checkSystemSpace" passed, then no action is needed. Proceed with patch installation.

Oracle recommends that you use the latest OPatch and OPatchAuto version for your Oracle software release.

Follow these steps for a an Oracle Database environment deployed with Oracle Data Guard:

1. Disable REDO transport on the Primary database.

   1. If Data Guard Broker is in place, then you must use DGMGRL. For example:

      DGMGRL> connect /
      Connected.
      DGMGRL> show database verbose plb_prm
      Database
      Name: plb_prm
      Role: PRIMARY
      Enabled: YES
      Intended State: ONLINE
      Instance(s):
      plb
      Properties:
      InitialConnectIdentifier = 'plb_prm_dgmgrl'
      ..
      .
      Current status for "plb_prm":
      SUCCESS
       
      DGMGRL> edit database plb_prm set state='LOG-TRANSPORT-OFF';
      Succeeded.

   2. If Data Guard Broker is not used, then use SQL alter system set. For example:

      SQL> alter system set log_archive_dest_state_X=defer scope=both sid='*'

2. Shut down the standby database, all listeners, and all processes associated with the Oracle home standby site. For example:

   $ lsnrctl stop lsnrplb
   
   LSNRCTL for Linux: Version 10.2.0.4.0 - Production on 04-FEB-2010 08:41:29
   Copyright (c) 1991, 2007, Oracle. All rights reserved.
   Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=<host>)(PORT=1666)))
   The command completed successfully
   
   $ sqlplus / as sysdba
   
   SQL*Plus: Release 10.2.0.4.0 - Production on Thu Feb 4 08:42:02 2010
   Copyright (c) 1982, 2007, Oracle. All Rights Reserved.
   Connected to:
   Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Production
   With the Partitioning, OLAP, Data Mining and Real Application Testing options
   
   SQL> shutdown immediate
   
   ORA-01109: database not open
   Database dismounted.
   ORACLE instance shut down.

3. Set your current directory to the directory where the patch is located and then run the OPatchAuto utility by entering the following commands,where <patchnumber> is the eight-digit number of the patch,<version> is the release of the Oracle software for which the patch is provided, and <platform> is the operating system platform for which the patch can be applied.

   unzip p<patchnumber>_<version>_<platform>.zip
   cd p<patchnumber>_<version>_<platform>
   opatch apply
   

   For example:

   unzip p12345678_23.0.0.0.0_Linux-x86-64.zip
   cd p12345678_23.0.0.0.0_Linux-x86-64
   opatch apply
   

   Wait for the patch update to complete successfully. After the patch update is complete, start the standby site to mount and restart the listeners only. Do not restart managed recovery.

   Note:

   If the standby is an Oracle RAC environment, then you repeat the patch application across all nodes.

   If there are errors, refer to "Known Issues"

   Note:

   Do not apply any postupgrade changes to the standby Oracle home. Those changes must be performed using the REDO transport (catchpatch.sql, and so on) run against the standby RDBMS itself later.
4. Start the primary site, and re-enable log shipping to the standby.
5. At the standby site, start the MRP (managed recovery). RDBMS changes implemented in the Primary Site through catpatch, catbundle and catcpu are now applied to the standby.

   Caution:

   You must perform this step immediately after upgrading the Oracle Database binaries on the Standby Database. This step ensures that the Data Dictionary (CATPROC) version matches the version of the database binaries. If the data dictionary does not match (for example, if you upgrade the Standby database first and perform a role change on the standby before you upgrade the Primary), it is possible for you to run into severe problems. for more information , see Mixed Oracle Version support with Data Guard Redo Transport Services (Doc ID 785347.1).

6. Set your current directory to the directory where the patch storage is located and then run the OPatch utility by entering the following commands,where <patchnumber> is the eight-digit number of the patch,<version> is the release of the Oracle software for which the patch is provided, and <platform> is the operating system platform for which the patch can be applied.

   unzip p<patchnumber>_<version>_<platform>.zip
   cd p<patchnumber>_<version>_<platform>
   opatch apply
   

   For example:

   unzip p12345678_23.0.0.0.0_Linux-x86-64.zip
   cd p12345678_23.0.0.0.0_Linux-x86-64
   opatch apply
   

   Wait for the patch update to complete successfully.

   If there are errors, refer to "Known Issues"

7. If you are using Data Guard Broker with Oracle Data Guard, then to avoid Data Guard Broker from starting the MRP automatically, change the state to state=APPLY-OFF. For example:

   DGMGRL> EDIT DATABASE 'plb_prm' SET STATE='APPLY-OFF';

9. Shut down the primary database, all listeners, and all processes associated with the Oracle home standby site. For example:

   $ lsnrctl stop lsnrplb
   
   LSNRCTL for Linux: Version 10.2.0.4.0 - Production on 04-FEB-2010 08:41:29
   Copyright (c) 1991, 2007, Oracle. All rights reserved.
   Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=<host>)(PORT=1666)))
   The command completed successfully
   
   $ sqlplus / as sysdba
   
   SQL*Plus: Release 10.2.0.4.0 - Production on Thu Feb 4 08:42:02 2010
   Copyright (c) 1982, 2007, Oracle. All Rights Reserved.
   Connected to:
   Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Production
   With the Partitioning, OLAP, Data Mining and Real Application Testing options
   
   SQL> shutdown immediate
   
   ORA-01109: database not open
   Database dismounted.
   ORACLE instance shut down.

   For an Oracle RAC primary:

   $ srvctl stop listener -n <host> -l lsnrplb_<host>
   $ srvctl stop listener -n <host> -l lsnrplb_<host>
   $ srvctl stop database -d plb

10. Unzip the patch into the primary database patch storage location.

    Set your current directory to the directory where the patch is located and then run the OPatch utility by entering the following commands,where <patchnumber> is the eight-digit number of the patch,<version> is the release of the Oracle software for which the patch is provided, and <platform> is the operating system platform for which the patch can be applied.

    unzip p<patchnumber>_<version>_<platform>.zip
    cd p<patchnumber>_<version>_<platform>
    opatch apply
    

    For example:

    unzip p12345678_23.0.0.0.0_Linux-x86-64.zip
    cd p12345678_23.0.0.0.0_Linux-x86-64
    opatch apply
    

    Wait for the patch update to complete successfully.

    If there are errors, refer to "Known Issues"

11. Upgrade and patch the RDBMS and dictionary objects.

    In an Oracle RAC environment, you perform this task from one instance only, with the other instances shut down and not running.

12. After the patch is installed successfully, if you obtained patch conflict resolution interim patches after checking for patch conflicts, then apply them now.

13. To reestablish the Oracle Data Guard environment on the primary site. start the listeners.

    In an Oracle RAC environment, you reestablish Oracle Data Guard on the primary site one instance only, while the remaining cluster instances are down and not running.

14. Force the primary site to register its services to the listener:

    $ sqlplus / as sysdba
    SQL> alter system register;
    

15. Re-enable Oracle Data Guard on the Standby.

    1. In a single instance Oracle Data Guard deployment, disable restricted session to allow end user connectivity.

       SQL> alter system disable restricted session;

    2. In an Oracle RAC Oracle Data Guard deployment, restart the primary site and all of its instances:

       $ srvctl stop database -d plb
       $ srvctl start database -d plb

       If there are additional Oracle RAC services used to connect the database then restart those services. For example:

       $ srvctl start service -d plb

16. Re-enable log shipping to the Standby Site.

    When you re-enable log-shipping to the standby, the RDBMS changes made through running catupgrade. catbundle and catcpu can be applied to the standby.

    1. If you are using Data Guard Broker, then connect and update the state to state=ONLINE. For example:

       
       DGMGRL> edit database plb_prm set state='ONLINE';

After installing the patches needed, complete these steps:

1. Load modified SQL files into the database, as explained in "Load Modified SQL Files into the Database" .
2. Upgrade Oracle Recovery Manager Catalog, as explained in "Upgrade Oracle Recovery Manager Catalog" .
3. Check bug fixes that may change an existing optimizer execution plan, as explained in Managing "installed but disabled" bug fixes in Database Release Updates using DBMS_OPTIM_BUNDLE (Doc ID 2147007.1).

If there are errors, refer to "Known Issues"

Datapatch is the patching tool that completes the post-patch SQL actions for RDBMS patches, such as running scripts to load modified SQL files into the database registry. Datapatch identifies the postinstallation steps required after the database is updated, and applies and removes or rolls back SQL changes in the database as required. Unlike a maintenance operation with OPatchAuto, which calls Datapatch automatically to complete post-patch actions after the new binary patch is installed and the database is restarted, you must run Datapatch manually after running OPatch.

For more information about Datapatch, see:

• Datapatch: Database 12c or later Post Patch SQL Automation (Doc ID 1585822.1)
• Datapatch User Guide (Doc ID 2680521.1)

1. For your databases running on the same Oracle home in the multitenant environment being patched, run the datapatch tool. In this example, we run datapatch with the -sanity_checks option:

   % sqlplus /nolog
   SQL> Connect / as sysdba
   SQL> startup
   SQL> alter pluggable database all open; 
   SQL> quit
   % cd $ORACLE_HOME/OPatch
   % ./datapatch -sanity_checks
   % ./datapatch -verbose

   Oracle highly recommends that you run datapatch with the -sanity_checks option. When started with this option, datapatch runs a series of environment and database checks to validate if conditions are optimal for patching. Results are shown on screen with severity and recommendations for potential actions you can take. For more information about this feature, see:

   Datapatch User Guide (Doc ID 2680521.1)

   Oracle also recommends that you run datapatch on all pluggable databases (PDBs) in the multitenant environment at the same time. However, you can choose to run datapatch on individual pluggable databases. If you choose to do this, then datapatch is run only on the multitenant container database (CDB) and opened pluggable databases (PDBs). You then are required to update the PDBs that are not updated at this time by using the alter pluggable database statement, and running the datapatch tool again at a later time to complete post-patch SQL actions on those PDBs. See:

   Multitenant Unplug/Plug Best Practices (Doc ID 1935365.1)

   The datapatch utility runs apply scripts to load the modified SQL files into the database. An entry is added to the dba_registry_sqlpatch view, recording the patch application.

2. In the dba_registry_sqlpatch view, verify that the Status for the APPLY is SUCCESS. For any other status, see:

   Troubleshooting Assistant:12c Datapatch Issues (Doc ID 2335899.2)

3. After datapatch completes, check its output to see if it has reported any errors. The output of datapatch contains log file locations for the patches installed. You can find all the relevant log files of that particular datapatch process in the same directory as sqlpatch_invocation.log. Check the initial lines of the output for "Log file for this invocation:"

The UPGRADE CATALOG command must be entered twice to confirm the upgrade.

Enter the following commands:

Access to Oracle Support

Oracle customer access to and use of Oracle support services will be pursuant to the terms and conditions specified in their Oracle order for the applicable services.