4.1.1 Authorized Problem Analysis Reports in Installation Guides

The Authorized Problem Analysis Reports in the installation guides are the minimum required Technical Levels for AIX 5L V5.3 Technical Level 09 and AIX 6.1 Technical Level 02 Service Pack 1. If you are using a later Technical Level than the minimum level listed here, then contact IBM to determine what Authorized Problem Analysis Reports are included in the Technical Level that you have on your system. If they are included, then you do not need to install them. If they are not included, then you must install the equivalent Authorized Problem Analysis Reports for the appropriate Technical Level.

4.2.1 Oracle Notification Service Issue After Shared Oracle RAC Database Upgrade

After you complete an upgrade from shared Oracle Clusterware Release 11.1 to shared Oracle Grid Infrastructure Release 11.2, you may encounter an error with Oracle Notification Service. The first node shows the Oracle Notification Service file, ora.ons, is offline.

Workaround:

Perform the following steps:

1. Log in as the Oracle Grid Infrastructure installation owner.

2. Run the following commands to find the Oracle Notification Service processes on the nodes and stop them:

   $ ps -ef | grep "ons -d" 
   $ kill -9 pid1 pid2 

3. Change directory to bin in the Oracle Grid Infrastructure home, and enter the command crsctl start res ora.ons. For example, if the grid home is /u01/grid, then enter the following commands:

   $ chdir /u01/grid/bin
   $ ./crsctl start res ora.ons
   

This issue is tracked with Oracle bug 9133235.

4.2.2 File System Configuration Data Not Found on Oracle Enterprise Manager Database Control Host Page

Oracle Enterprise Manager Database Control may not show the File System configuration data, and the Operating System Collection Warnings may appear on the Configuration page:

sh: mount: not found 

Workaround:

Perform the following steps:

1. Add /usr/sbin to PATH environment variable.

   For Bash shell:

   export PATH=$PATH:/usr/sbin
   

2. Restart Oracle Enterprise Manager Database Control.

This issue is tracked with Oracle bug 9206261.

4.2.3 ORA-00600 Error During Database Upgrade

When you upgrade from Oracle9i Database Release 2 to Oracle Database 11g Release 2, the database upgrade fails with the following error:

ORA-00600: internal error code, arguments: [ktsxaffs1] 

Workaround:

Set _affinity_on=false in the SPFILE of Oracle Database Release 9.2 before you upgrade.

This issue is tracked with Oracle bug 6615695.

4.2.4 Error During Deinstallation in JP locale

If you use the Deinstallation Tool (deinstall) in Japanese locale to remove Oracle RAC 11g Release 2, then the following error message appears:

IBM: 0403-009 The specified number is not valid for this command

Workaround:

Set LANG=C, before running the Deinstallation Tool.

This issue is tracked with Oracle bug 9227535.

4.2.5 Oracle Enterprise Manager Fails During Install or Upgrade

Oracle Database Configuration Assistant or Oracle Database Upgrade Assistant intermittently fails while running Oracle Enterprise Manager Configuration and gives a NULL error message. The configuration assistant prompts you to run the Oracle Enterprise Manager Configuration Assistant command line tool to configure or upgrade Oracle Enterprise Manager Database Control.

Workaround:

For a new Oracle Enterprise Manager Database Control setup, during database install or database upgrade, you must run the following command after the creating or upgrading the database:

emca -config dbcontrol db -repos recreate -cluster

For an Oracle Enterprise Manager Database Control upgrade process, you must run the following command after the upgrading the existing database:

emca -upgrade db -cluster

This issue is tracked with Oracle bug 9169706.

4.3.1 Help Files Translation Issue for Spanish and French

Oracle Universal Installer does not copy the translated help files to Oracle home directory when you select the Spanish or French language.

Workaround:

On the Select Product Languages page, select both the Latin American Spanish and Spanish to get the help files translated in Spanish; and select both the Canadian French and French for the translated help files in French.

This issue is tracked with Oracle bug 9147106.

4.3.2 Incorrect Reverse Ping Command from Oracle Management Service

Oracle Management Service provides incorrect reverse ping command with Oracle Enterprise Manager Database Control. The emoms.propertiesfile shows:

emdrep.ping.pingCommand=/usr/sbin/ping 

Workaround:

Change the property in the emoms.properties file to:

emdrep.ping.pingCommand=/usr/sbin/ping -c 3 -w 30 

This issue is tracked with Oracle bug 9211800.

4.3.3 Patch Oracle Database Deployment Procedure Issue

When you apply a patch from Oracle Enterprise Manager Database Control using the Patch Oracle Database deployment procedure, the applied patch may not appear on the Oracle home details page.

Workaround:

Use the command, opatch lsinventory, at the operating system prompt to list the patches applied to the Oracle home.

This issue is tracked with Oracle bug 8791320.

4.3.4 Oracle Management Service Restart Fails Intermittently

Oracle Management Service restart fails intermittently as Oracle Enterprise Manager Database Control fails to start with the following error message:

The socket name is already in use.

Workaround:

If Oracle Enterprise Manager Database Control does not shut down successfully, then manually stop the process.

This issue is tracked with Oracle bug 9118229.

4.3.5 SQL Developer Release 1.5.5 Does Not Support JDK Version 1.5.x

The dialogs to create and edit objects in SQL Developer release 1.5.5 fail to open as JDK version 1.5.x is not supported.

Workaround:

Use any one of the following workarounds:

• Upgrade the JDK to version 1.6.

• Enter the SQL statement on the SQL Worksheet.

• Use SQL Developer from a client desktop (non-AIX).

• Use the sqlplus command to start SQL*Plus.

This issue is tracked with Oracle bug 9195473.

4.3.6 Deployment Procedure Issue with Oracle RAC

The Deployment Procedure to update the software using Oracle Enterprise Manager Database Control fails on Oracle RAC and Oracle Grid Infrastructure, with the following error:

ORA-00001: unique constraint (SYSMAN.PK_MGMT_JOB_EXECUTION) violated

Workaround:

Run the following commands as SYSMAN for the database:

SELECT MAX(step_id) FROM MGMT_JOB_HISTORY; (--&gt v1)
SELECT MGMT_JOB_SEQUENCE.CURRVAL FROM DUAL; (--&gt v2)
ALTER SEQUENCE MGMT_JOB_SEQUENCE INCREMENT BY (v2-v1+1);

Add the calculated value in the third command.

This issue is tracked with Oracle bug 9205553.

4.3.7 The addNode Command Fails on Oracle RAC

When you run the addNode command on an Oracle RAC environment, you may encounter the following updatenodelist error, where new_nodeis the name of the node you are adding:

SEVERE: 12/21/09 11:10:03 AM UTC: Remote 'UpdateNodeList' failed on nodes: 'new_ node'.

Workaround:

Run the updatenodelist command manually on the new node. Refer to the addnode output or logs for complete information aboutupdatenodelist command.

This issue is tracked with Oracle bug 9229777.

4.3.8 Wallet Upload Issue with Oracle Wallet Manager

Oracle Wallet Manager fails to upload wallet to Oracle Internet Directory server, giving the following error message:

Exception while initialization:java.security.NoSuchAlgorithmException:
Algorithm SunX509 not available

Workaround:

None

This issue is tracked with Oracle bug 10144879.

4.3.9 Help Pages Fail to Load for Oracle Database Client (32-Bit)

Oracle Universal Installer does not find the clientinstaller_help.map file and fails to load the help pages for locales other than English.

Workaround:

Set LANG=C before running Oracle Universal Installer. This changes the installer locale to English and the help content is available.

This issue is tracked with Oracle bug 9474214.

5.2.1 Role Permissions Issue with Oracle ASM Configuration Assistant, SQL*Plus, X Terminal

Ensure that the oracle_devmgmt role is set before running administrative tools, such as, Oracle ASM Configuration Assistant and SQL*Plus, for management of Oracle Automatic Storage Management Cluster File System and Oracle Automatic Storage Management Dynamic Volume Manager. Also, when entering a new X terminal (or using the associated commands, such as uxterm and aixterm), the oracle_devmgmt role necessary for the administration of Oracle Automatic Storage Management Cluster File System and Oracle Automatic Storage Management Dynamic Volume Manager is not properly set.

Workaround:

Run the following command before running administrative tools or after opening a new terminal:

$ swrole oracle_devmgmt 

Check that the swrole is set to oracle_devmgmtby running the following command:

$ rolelist -e

This issue is tracked with Oracle 9965007.

5.2.2 Performance Statistics Issue with Oracle Enterprise Manager Database Control

The summary chart on the Oracle Enterprise Manager Cluster Performance page does not show any data.

Workaround:

Perform the following steps:

1. Stop Oracle Enterprise Manager Database Control.

2. Add the following to the properties file, $ORACLE_HOME/host_dbuniquename/sysman/config/emoms.properties:

   oracle.sysman.db.rac.useRemoteCollection=false
   

3. Restart the Oracle Enterprise Manager Database Control.

This issue is tracked with Oracle bug 10129539.

5.2.3 Error When Installing Oracle Grid Infrastructure for a Cluster with Grid Naming Service

When installing Oracle Grid Infrastructure for a Cluster with Grid Naming Service enabled, the root.sh script gives the following error:

PRCR-1079 : Failed to start resource ora.gns

Workaround:

None. This issue does not affect any functionality.

This issue is tracked with Oracle bug 10040115.

5.2.4 Oracle Notification Service Issue After Shared Oracle RAC Database Upgrade

After you complete an upgrade from shared Oracle Clusterware Release 11.1 to shared Oracle Grid Infrastructure Release 11.2, you may encounter an error with Oracle Notification Service. The first node shows the Oracle Notification Service file, ora.ons, is offline.

Workaround:

Perform the following steps:

1. Log in as the Oracle Grid Infrastructure installation owner.

2. Run the following commands to find the Oracle Notification Service processes on the nodes and stop them:

   $ ps -ef | grep "ons -d" 
   $ kill -9 pid1 pid2 

3. Change directory to bin in the Oracle Grid Infrastructure home, and enter the command crsctl start res ora.ons. For example, if the grid home is /u01/grid, then enter the following commands:

   $ chdir /u01/grid/bin
   $ ./crsctl start res ora.ons
   

This issue is tracked with Oracle bug 9133235.

5.2.5 ORA-00600 Error During Database Upgrade

When you upgrade from Oracle9i Database Release 2 to Oracle Database 11g Release 2, the database upgrade fails with the following error:

ORA-00600: internal error code, arguments: [ktsxaff1]

Workaround:

Set _affinity_on=false in the SPFILE of Oracle Database Release 9.2 before you upgrade.

This issue is tracked with Oracle bug 6615695.

5.2.6 Monitoring Issue with Oracle Enterprise Manager Database Control

Monitoring in memory access mode is not supported for databases.

This issue is tracked with Oracle bug 6498439.

5.2.7 Oracle Database Vault Administrator Login Issues

While trying to log in to Oracle Database Vault Administrator, you may get the following error:

java.security.NoSuchProviderException: no such provider: SUN 

Workaround:

Use any one of the following workarounds to configure Database Vault policy:

• Connect to the 11.2.0.2 AIX Database Vault enabled database using the 11.2.0.1 Oracle Database Vault Administrator from any platform.

• Use Database Vault administrative packages.

This issue is tracked with Oracle bug 10188865.

5.2.8 Error While Installing Oracle Grid Infrastructure Using Non-English Locales

While installing Oracle Grid Infrastructure using non-English locales, the root.sh script gives the following error message and the ora.registry.acfsresource shows offline:

CRS-2674: Start of 'ora.registry.acfs' on host_name failed

The error may also occur when running srvctl, asmca, or dbca utilities.

Workaround:

Change the locale to English before running the root.sh script:

# export LANG=C

This issue is tracked with Oracle bugs 10193581 and 10229727.

5.3.1 Oracle Management Service Restart Fails Intermittently

Oracle Management Service restart fails intermittently as Oracle Enterprise Manager Database Control fails to start with the following error message:

The socket name is already in use.
 

Workaround:

If Oracle Enterprise Manager Database Control does not shut down successfully, then manually stop the process.

This issue is tracked with Oracle bug 9118229.

5.3.2 The addnode Command Fails on Oracle RAC

When you run the addnode command on an Oracle RAC environment, you may encounter the following updatenodelist error, where new_nodeis the name of the node you are adding:

SEVERE: 09/29/10 11:10:03 AM UTC: Remote 'UpdateNodeList' failed on nodes: 'new_node'.

Workaround:

Run the updatenodelist command manually on the new node. Refer to the addnode output or logs for complete information aboutupdatenodelist command.

This issue is tracked with Oracle bug 9229777.

5.3.3 SQL Developer Release 1.5.5 Does Not Support JDK Version 1.5.x

The dialogs to create and edit objects in SQL Developer release 1.5.5 fail to open as JDK version 1.5.x is not supported.

Workaround:

Use any one of the following workarounds:

• Upgrade the JDK to version 1.6.

• Enter the SQL statement on the SQL Worksheet.

• Use SQL Developer from a client desktop (non-AIX).

• Use the sqlplus command to start SQL*Plus.

This issue is tracked with Oracle bug 9195473.

5.3.4 Wallet Upload Issue with Oracle Wallet Manager

Oracle Wallet Manager fails to upload wallet to Oracle Internet Directory server, giving the following error message:

Exception while initialization:java.security.NoSuchAlgorithmException:
Algorithm SunX509 not available

Workaround:

None

This issue is tracked with Oracle bug 10144879.

5.3.5 Oracle ACFS Goes Into an Offline State

Under heavy load, resulting in operating system paging, Oracle ACFS may go into an offline state.

Workaround:

Request one-off patches for release 11.2.0.2.x.

fsck -y -o a /dev/asm/volume_name 

This issue is tracked with Oracle bugs 13146653 and 13255295.

6.2.1 Monitoring Issue with Oracle Enterprise Manager Database Control

Monitoring in memory access mode is not supported for databases.

This issue is tracked with Oracle bug 6498439.

6.2.2 ORA-00600 Error During Database Upgrade

When you upgrade from Oracle9i Database Release 2 to Oracle Database 11g Release 2, the database upgrade fails with the following error:

ORA-00600: internal error code, arguments: [ktsxaff1]

Workaround:

Set _affinity_on=false in the SPFILE of Oracle Database Release 9.2 before you upgrade.

This issue is tracked with Oracle bug 6615695.

6.2.3 Warning Message While Downloading Software Updates for Oracle Database Client

After specifying the oraparam.ini file, when you run the Oracle Universal Installer to download software updates for Oracle Database Client, the following warning message may appear:

WARNING: User specified paramFile: /local_dir/oraparam.ini CLUSTER_NODES={} not found

Workaround:

Ignore the warning message.

This issue is tracked with Oracle bug 12690717.

6.2.4 Role Permissions Issue with Oracle ASM Configuration Assistant, SQL*Plus, X Terminal

Ensure that the oracle_devmgmt role is set before running administrative tools, such as, Oracle ASM Configuration Assistant and SQL*Plus, for management of Oracle Automatic Storage Management Cluster File System and Oracle Automatic Storage Management Dynamic Volume Manager. Also, when entering a new X terminal (or using the associated commands, such as uxterm and aixterm), the oracle_devmgmt role necessary for the administration of Oracle Automatic Storage Management Cluster File System and Oracle Automatic Storage Management Dynamic Volume Manager is not properly set.

Workaround:

Run the following command before running administrative tools or after opening a new terminal:

$ swrole oracle_devmgmt 

Check that the swrole is set to oracle_devmgmtby running the following command:

$ rolelist -e

This issue is tracked with Oracle 9965007.

6.2.5 Cluster Verification Utility Issue in Checking Network Parameters

During an Oracle RAC database installation, the Cluster Verification Utility may not check the network parameters (tcp_sendspace, tcp_recvspace,udp_sendspace, udp_recvspace, sb_max, rfc1323) correctly. This may happen as part of the prerequisite check or if you run the following command:

cluvfy stage -pre crsinst

Workaround:

Establish a symlink /etc/no to /usr/sbin/no.

This issue is tracked with Oracle bug 13077654.

6.2.6 Oracle Notification Service Issue After Shared Oracle RAC Database Upgrade

After you complete an upgrade from shared Oracle Clusterware Release 11.1 to shared Oracle Grid Infrastructure Release 11.2, you may encounter an error with Oracle Notification Service. The first node shows the Oracle Notification Service file, ora.ons, is offline.

Workaround:

Perform the following steps:

1. Log in as the Oracle Grid Infrastructure installation owner.

2. Run the following commands to find the Oracle Notification Service processes on the nodes and stop them:

   $ ps -ef | grep "ons -d" 
   $ kill -9 pid1 pid2 

3. Change directory to bin in the Oracle Grid Infrastructure home, and enter the command crsctl start res ora.ons. For example, if the grid home is /u01/grid, then enter the following commands:

   $ chdir /u01/grid/bin
   $ ./crsctl start res ora.ons
   

This issue is tracked with Oracle bug 9133235.

6.2.7 Automatic SSH Configuration Issue

Automatic SSH configuration is not available at the time of this release due to an issue with the installer.

Workaround:

Perform manual SSH configuration as described in section "D.1 Configuring SSH Manually on All Cluster Nodes" in Oracle Grid Infrastructure Installation Guide for IBM AIX on POWER Systems.

This issue is tracked with Oracle bug 9979216.

6.2.8 Cluster Verification Utility Check Fails on AIX 6.1 and Later

The prerequisite checker checks only for the AIX fix, IZ97457, regardless of the operating system Technical Level version. The following are the correct Authorized Problem Analysis Reports (APARs) for the corresponding Technical Level versions:

6100-03 - use AIX APAR IZ89304
6100-04 - use AIX APAR IZ89302
6100-05 - use AIX APAR IZ89300
6100-06 - use AIX APAR IZ89514
7100-00 - use AIX APAR IZ89165

5300-11 - use AIX APAR IZ98424
5300-12 - use AIX APAR IZ98126
6100-04 - use AIX APAR IZ97605
6100-05 - use AIX APAR IZ97457
6100-06 - use AIX APAR IZ96155
7100-00 - use AIX APAR IZ97035

Workaround:

Before starting the Oracle Universal Installer, set the environment variable, IGNORE_PREADDNODE_CHECKS, to Y:

export IGNORE_PREADDNODE_CHECKS=Y

This issue is tracked with Oracle bug 13083120.

6.3.1 Required Oracle Warehouse Builder Patches

If you plan to install Oracle Warehouse Builder, then after you install Oracle Database, but before you create the Oracle Warehouse Builder repository, you must complete the following procedure:

1. Download patch 12973504 from the My Oracle Support website.

2. Extract the contents to a directory, for example, /tmp/12973504.

3. Verify that the ORACLE_HOME environment variable is set to the 11.2.0.2 Database home.

4. Run the Opatch utility to proceed with the Oracle Warehouse Builder installation:

   cd /tmp/12973504
   opatch apply
   

5. Navigate to $ORACLE_HOME/owb/patches/bug12973504.

6. Run the script, fix_bug12973504.sh.

7. After you have edited the files, you can install Oracle Warehouse Builder.

This issue is tracked with Oracle bug 11772716.

6.3.2 Oracle Messaging Gateway Fails to Start on AIX 6.1 or Higher Systems

When attempting to start the Oracle Messaging Gateway using the DBMS_MGWADM.STARTUP procedure, the job started by this procedure fails with following error:

ORA-32830: Result code -8 returned by Messaging Gateway agent

In the Oracle Messaging Gateway logs (located at $ORACLE_HOME/mgw/log), you see messages similar to the following:

>>2010-06-18 10:16:01 MGW C-Bootstrap 0 LOG process-id=2908272
Bootstrap program starting
>>2010-06-18 10:16:02 MGW C-Bootstrap 0 LOG process-id=2908272
Failed to create Java VM: retCode = -4
>>2010-06-18 10:16:02 MGW JVM log
JVMJ9VM015W Initialization error for library j9gc23(2): Failed to instantiate
heap; 64M requested

Problem Summary: Java Instantiation fails on systems running AIX 6.1 or higher, which leads to this issue. IBM is aware of this issue and is tracking this through Authorized Program Analysis Report (APAR) IZ55237. Contact IBM Support for specific patches for the AIX version and level running in your environment.

Workaround:

Modify the listener.ora file for the Messaging Gateway listener to implement MPROTECT_TXT=OFF. For example:

SID_LIST_LISTENER =
 
 (SID_LIST =
  (SID_DESC=
  (SID_NAME= mgwextproc)
  
(ENVS="LIBPATH=[ORACLE_HOME]/jdk/jre/bin:[ORACLE_HOME]/jdk/jre/bin/classic:
[ORACLE_HOME]/lib,MPROTECT_TXT=OFF")
  (ORACLE_HOME=[ORACLE_HOME])
  (PROGRAM = extproc))
 )
)

This issue is tracked with Oracle bug 12842870.

6.3.3 Oracle ACFS Goes Into Offline State

Under heavy load, resulting in operating system paging, Oracle ACFS may go into an offline state.

Workaround:

Install the latest 11.2.0.3 GI PSU (Oracle Grid Infrastructure Patch Set Update).

fsck -y -o a /dev/asm/volume_name 

This issue is tracked with Oracle bug 13255295.

6.3.4 Wallet Upload Issue with Oracle Wallet Manager

Oracle Wallet Manager fails to upload wallet to Oracle Internet Directory server, giving the following error message:

Exception while initialization:java.security.NoSuchAlgorithmException:
Algorithm SunX509 not available

Workaround:

None

This issue is tracked with Oracle bug 10144879.

6.3.5 Oracle RAC and IBM Virtual I/O Servers on AIX

If you use Virtual I/O (VIO) Servers with Oracle RAC, then there are special concerns for ensuring highly available interconnect networks. Refer to My Oracle Support note 1356461.1, which is available at the following URL:

https://support.oracle.com

6.3.6 SQL Developer Release 1.5.5 Does Not Support JDK Version 1.5.x

The dialogs to create and edit objects in SQL Developer release 1.5.5 fail to open as JDK version 1.5.x is not supported.

Workaround:

Use any one of the following workarounds:

• Upgrade the JDK to version 1.6.

• Enter the SQL statement on the SQL Worksheet.

• Use SQL Developer from a client desktop (non-AIX).

• Use the sqlplus command to start SQL*Plus.

This issue is tracked with Oracle bug 9195473.

6.3.7 Oracle Management Service Restart Fails Intermittently

Oracle Management Service restart fails intermittently as Oracle Enterprise Manager Database Control fails to start with the following error message:

The socket name is already in use.
 

Workaround:

If Oracle Enterprise Manager Database Control does not shut down successfully, then manually stop the process.

This issue is tracked with Oracle bug 9118229.

7.2.1 Role Permissions Issue with Oracle ASM Configuration Assistant, SQL*Plus, X Terminal

Ensure that the oracle_devmgmt role is set before running administrative tools, such as, Oracle ASM Configuration Assistant and SQL*Plus, for management of Oracle Automatic Storage Management Cluster File System and Oracle Automatic Storage Management Dynamic Volume Manager. Also, when entering a new X terminal (or using the associated commands, such as uxterm and aixterm), the oracle_devmgmt role necessary for the administration of Oracle Automatic Storage Management Cluster File System and Oracle Automatic Storage Management Dynamic Volume Manager is not properly set.

Workaround:

Run the following command before running administrative tools or after opening a new terminal:

$ swrole oracle_devmgmt 

Check that the swrole is set to oracle_devmgmtby running the following command:

$ rolelist -e

This issue is tracked with Oracle bug 9965007.

7.2.2 Monitoring Issue with Oracle Enterprise Manager Database Control

Monitoring in memory access mode is not supported for databases.

This issue is tracked with Oracle bug 6498439.

7.2.3 Oracle Notification Service Issue After Shared Oracle RAC Database Upgrade

After you complete an upgrade from shared Oracle Clusterware Release 11.1 to shared Oracle Grid Infrastructure Release 11.2, you may encounter an error with Oracle Notification Service. The first node shows the Oracle Notification Service file, ora.ons, is offline.

Workaround:

Perform the following steps:

1. Log in as the Oracle Grid Infrastructure installation owner.

2. Run the following commands to find the Oracle Notification Service processes on the nodes and stop them:

   $ ps -ef | grep "ons -d" 
   $ kill -9 pid1 pid2 

3. Change directory to bin in the Oracle Grid Infrastructure home, and enter the command crsctl start res ora.ons. For example, if the grid home is /u01/grid, then enter the following commands:

   $ chdir /u01/grid/bin
   $ ./crsctl start res ora.ons
   

This issue is tracked with Oracle bug 9133235.

7.2.4 SQL Developer Release 1.5.5 Does Not Support JDK Version 1.5.x

The dialogs to create and edit objects in SQL Developer release 1.5.5 fail to open as JDK version 1.5.x is not supported.

Workaround:

Use any one of the following workarounds:

• Upgrade the JDK to version 1.6.

• Enter the SQL statement on the SQL Worksheet.

• Use SQL Developer from a client desktop (non-AIX).

• Use the sqlplus command to start SQL*Plus.

This issue is tracked with Oracle bug 9195473.

7.2.5 Warning Message While Downloading Software Updates for Oracle Database Client

After specifying the oraparam.ini file, when you run the Oracle Universal Installer to download software updates for Oracle Database Client, the following warning message may appear:

WARNING: User specified paramFile: /local_dir/oraparam.ini CLUSTER_NODES={} not found

Workaround:

Ignore the warning message.

This issue is tracked with Oracle bug 12690717.

7.2.6 Automatic SSH Configuration Issue

Automatic SSH configuration is not available at the time of this release due to an issue with the installer.

Workaround:

Perform manual SSH configuration as described in section "D.1 Configuring SSH Manually on All Cluster Nodes" in Oracle Grid Infrastructure Installation Guide for IBM AIX on POWER Systems.

This issue is tracked with Oracle bug 9979216.

7.2.7 Oracle Database Downgrade Issue

After an initial upgrade of Oracle Database release 11.1.0.7 to release 11.2.0.4, if you try to downgrade the database to release 11.1.0.7 using the Database Upgrade Assistant, then the srvctl command may fail to start the database from the earlier Oracle home location for release 11.1.0.7.

Workaround:

Manually run dbsid_restore.sh generated by the Database Upgrade Assistant after making the following change in the ORACLE_HOME/bin/srvctlscript:

< ${JRE} ${JRE_OPTIONS} -DORACLE_HOME=${ORACLE_HOME} -classpath ${CLASSPATH} ${SRVM_PROPERTY_DEFS} oracle.ops.opsctl.OPSCTLDriver "$@"
> ${JRE} ${JRE_OPTIONS} -Xms64M -DORACLE_HOME=${ORACLE_HOME} -classpath ${CLASSPATH} ${SRVM_PROPERTY_DEFS} oracle.ops.opsctl.OPSCTLDriver "$@"

This issue is tracked with Oracle bug 16966994.

7.2.8 ORA-00600 Error During Database Upgrade

When you upgrade from Oracle9i Database Release 2 to Oracle Database 11g Release 2, the database upgrade fails with the following error:

ORA-00600: internal error code, arguments: [ktsxaffs1] 

Workaround:

Set _affinity_on=false in the SPFILE of Oracle Database Release 9.2 before you upgrade.

This issue is tracked with Oracle bug 6615695.

7.2.9 Cluster Verification Utility Check Fails

During an Oracle Cluster Ready Services (CRS) installation, the Cluster Verification Utility (CVU) prerequisite check fails for the OS patch IZ89514, although the patch is correctly installed.

Workaround:

Ignore the prerequisite check for the OS patch and proceed with the installation.

This issue is tracked with Oracle bug 17459487.

7.3.1 Oracle ACFS Offline After Downgrade

When you downgrade from release 11.2.0.x to a previous 11.2.0.x release, you must run acfsroot install from the old home after a successful downgrade. Until you run this command and it completes, the Oracle ACFS may appear offline.

Workaround:

Complete the following procedures during a downgrade for each cluster member node to ensure that Oracle ACFS is available:

1. Downgrade the node.

2. As the root user, run the following command to stop the Oracle Grid Infrastructure stack:

   $ crsctl stop crs
   

3. Run acfsroot install from the Oracle Grid Infrastructure binary home as the root user. For example:

   $ /u01/app/11.2.0/grid/bin/acfsroot install
   

4. Use the following command to start the Oracle Grid Infrastructure stack:

   $ crsctl start crs 
   

5. Repeat this procedure on each node until you have run the command on all cluster member nodes.

This issue is tracked with Oracle bug 16804348.

7.3.2 Input/Output Error Displayed During Oracle Grid Infrastructure Shutdown If Any Oracle ACFS is Offline

Before shutting down the Oracle Grid stack, if any Oracle ACFS is reported offline by the command, acfsutil info fs, or is reported stale by the command, srvctl status filesystem, then bring it online. If you are not able to bring the file system online, then restart the node. Doing this ensures that no processes are active on an offline Oracle ACFS resulting in the input/output error.

This issue is tracked with Oracle bug 17357831.

7.3.3 Issues When Oracle Database Home Is Using Oracle ACFS

With Oracle Database home on Oracle ACFS, shutting down the database may mark the file system busy and may lead to unmount failures when there are no processes using files on that file system.

Workaround:

Restart the node.

Fix:

Contact IBM AIX Support for the appropriate IBM APAR titled, "UMOUNT FAILS WITH DEVICE BUSY ERROR EVEN WITHOUT ACTIVE PROCESS", or a Technical Level or a Service Pack that includes the fix.

Here is the link:

http://www-01.ibm.com/support/docview.wss?uid=isg1IV37940

This issue is tracked with Oracle bug 17244827.

7.3.4 Wallet Upload Issue with Oracle Wallet Manager

Oracle Wallet Manager fails to upload wallet to Oracle Internet Directory server, giving the following error message:

Exception while initialization:java.security.NoSuchAlgorithmException:
Algorithm SunX509 not available

Workaround:

None

This issue is tracked with Oracle bug 10144879.

7.3.5 Oracle Messaging Gateway Fails to Start on AIX 6.1 or Higher Systems

When attempting to start the Oracle Messaging Gateway using the DBMS_MGWADM.STARTUP procedure, the job started by this procedure fails with following error:

ORA-32830: Result code -8 returned by Messaging Gateway agent

In the Oracle Messaging Gateway logs (located at $ORACLE_HOME/mgw/log), you see messages similar to the following:

>>2010-06-18 10:16:01 MGW C-Bootstrap 0 LOG process-id=2908272
Bootstrap program starting
>>2010-06-18 10:16:02 MGW C-Bootstrap 0 LOG process-id=2908272
Failed to create Java VM: retCode = -4
>>2010-06-18 10:16:02 MGW JVM log
JVMJ9VM015W Initialization error for library j9gc23(2): Failed to instantiate
heap; 64M requested

Problem Summary: Java Instantiation fails on systems running AIX 6.1 or higher, which leads to this issue. IBM is aware of this issue and is tracking this through Authorized Program Analysis Report (APAR) IZ55237. Contact IBM Support for specific patches for the AIX version and level running in your environment.

Workaround:

Modify the listener.ora file for the Messaging Gateway listener to implement MPROTECT_TXT=OFF. For example:

SID_LIST_LISTENER =
 
 (SID_LIST =
  (SID_DESC=
  (SID_NAME= mgwextproc)
  
(ENVS="LIBPATH=[ORACLE_HOME]/jdk/jre/bin:[ORACLE_HOME]/jdk/jre/bin/classic:
[ORACLE_HOME]/lib,MPROTECT_TXT=OFF")
  (ORACLE_HOME=[ORACLE_HOME])
  (PROGRAM = extproc))
 )
)

This issue is tracked with Oracle bug 12842870.