Skip Headers
Oracle® Database Release Notes
10g Release 1 (10.1.0.2.0) for AIX 5L Based Systems (64-Bit)
B13611-08
 

 

Oracle® Database

Release Notes

10g Release 1 (10.1.0.2.0) for AIX 5L Based Systems (64-Bit)

B13611-08

January 2006

This document contains important information that was not included in the platform-specific or product-specific documentation for this release.

To check for updates to this document and to view other Oracle documentation, see the Documentation section on the Oracle Technology Network (OTN) Web site:

http://www.oracle.com/technology/documentation/

For additional information about this release, refer to the readme files located in the $ORACLE_HOME/relnotes directory.

This document contains the following topics:

1 Product Issues

The following sections contain information about issues related to Oracle Database 10g and associated products:

1.1 Oracle Real Application Clusters Installation and Configuration Guide

The Oracle Real Application Clusters Installation and Configuration Guide has been revised. You can download a copy of the revised book, part number B10766-08, from the Documentation section of the OTN Web site:

http://oracle.com/technology/documentation

1.2 Creating Raw Logical Volumes

If you are using raw logical volumes to store Oracle datafiles, Oracle recommends that you use a command similar to the following to create logical volumes with a zero offset:

/usr/sbin/mklv -y LVname -T O -w n -s n -r n VGname NumPPs

Using a zero offset improves database performance and fixes the issues described in Oracle bug 2620053.


Note:

On raw logical volumes, if you create tablespaces in datafiles that are not created in this way, a message is recorded in the alert.log file.


See Also:

For more information about this issue, contact Oracle Support.

1.3 Installing Oracle CRS on a System with More Than Three Network Interfaces

When you install Oracle Database on a system with more than 3 network interfaces, not all of the interfaces are displayed in the menu used to select the public and private interconnects. To work around this problem:

  1. Select public and private interconnects from the three interfaces which are listed in the menu.

  2. Complete the installation.

  3. After the installation, you can change the interconnects by entering the following commands as root, if necessary:

    # oifcfg delif -global
    # oifcfg setif -global interface/subnet:public
    # oifcfg setif -global interface/subnet:cluster_interconnect
    
    

    For example:

    # oifcfg delif -global
    # oifcfg setif -global en0/144.25.68.0:public
    # oifcfg setif -global en5/192.168.100.0:cluster_interconnect
    
    
  4. Enter the following command to verify the new settings:

    $ oifcfg getif
    
    

1.4 Installing Oracle CRS

When you install Oracle CRS on your system, the private interconnects may not be defined correctly. After installation, enter the following command to verify the public and private interconnects:

$ oifcfg getif

This command should return values for global public and global cluster_interconnect, for example:

en0 144.25.68.0 global public
en5 192.168.100.0 global cluster_interconnect

If the command does not return a value for global cluster_interconnect, enter the following commands:

# oifcfg delif -global
# oifcfg setif -global interface/subnet:public
# oifcfg setif -global interface/subnet:cluster_interconnect

For example:

# oifcfg delif -global
# oifcfg setif -global en0/144.25.68.0:public
# oifcfg setif -global en5/192.168.100.0:cluster_interconnect

Enter the following command to verify the new settings:

$ oifcfg getif

1.5 Oracle CRS Installation on an Oracle9i RAC Cluster

If you install Oracle CRS on a system that also contains an Oracle9i Real Application Clusters (RAC) cluster, you must restart all of the Oracle CRS nodes. You can restart the nodes one at a time to avoid affecting any Oracle9i databases that may be in use.

Oracle assumes that when you install Oracle CRS 10g software, you will also install Oracle Real Application Clusters 10g. If there will be a significant delay before you perform phase two of the RAC installation to install Oracle Real Application Clusters 10g, then run the /crs_home/bin/gsdctl start command to start the GSD manually to service the 9.2 SRVCTL tool and assistants. Then before installing Oracle Real Application Clusters 10g, run the /crs_home/bin/gsdctl stop command to stop the GSD.

1.6 Oracle CRS Silent Installation

If you perform a silent installation of Oracle CRS on multiple nodes, on a system that does not have other Oracle installations, the Installer does not set up the Oracle Inventory correctly. In this case, after the installation is complete, follow these steps:

  1. Run the orainstRoot.sh script on a local node.

  2. Copy the oraInventory directory from the local node to each of the remote nodes.

  3. Log in as the root user and run the following script on each remote node:

    oraInventory/orainstRoot.sh
    

1.7 Oracle CRS Installation Errors Caused by stty Commands

During an Oracle CRS installation, the Installer uses SSH (if available) to run commands and copy files to the other nodes. During the installation, you might see errors similar to the following if a "dot" file on the system (for example, .bashrc or .cshrc) contains stty commands:

 stty: standard input: Invalid argument 
 stty: standard input: Invalid argument 

To avoid this problem, Oracle recommends that you modify these files to suppress all output on STDERR, as follows:

  • Bourne, Bash, or Korn shell:

    if [ -t 0 ]; then
       stty intr ^C
    fi
    
    
  • C shell:

    test -t 0
    if ($status == 0) then
       stty intr ^C
    endif
    

Note:

When SSH is not available, the Installer uses the rsh and rcp commands instead of ssh and scp. If there are "dot" files that contain stty commands that are loaded by the remote shell, this error can also occur.

This issue is tracked through Oracle bug 3414362.

1.8 Using ASM on Your System

You must apply Program Technical Fix (PTF) U496549 or higher to your system before you use ASM.

1.9 Silent Installations that Use Automatic Storage Management

You cannot use the silent installation method to install Oracle Database 10g and create a database that uses Automatic Storage Management (ASM) for database storage during the same installation. This is because the root.sh script must run before the Database Configuration Assistant (DBCA) can start an ASM instance.

If you want to use the silent installation method to install Oracle Database 10g and create a database that uses ASM, follow these steps:

  1. Use the enterprise.rsp response file to complete a software-only installation.

  2. Run $ORACLE_HOME/root.sh after the installation completes.

  3. Use the dbca.rsp response file to run DBCA in silent mode, using a command similar to the following:

    $ $ORACLE_HOME/bin/dbca -silent -responseFile
    /full_path/dbca.rsp
    

1.10 ASM Instance Clean Up Procedures for Node Deletion

To remove the ASM instances, the delete node procedure requires the following additional steps on UNIX-based systems:

  1. If this is the Oracle home from which listener for each node named LISTENER_nodename runs, then use NetCA to remove this listener and its CRS resources. If necessary, re-create this listener in another Oracle home.

  2. If this is the Oracle home from which the ASM instance runs, then enter the following commands to remove the ASM configuration:

    $ srvctl stop asm -n node
    $ remove asm -n node
    
    
  3. If you are using a cluster file system for your ASM Oracle home, then run the following commands on the local node:

    $ rm -r $ORACLE_BASE/admin/+ASM
    $ rm -f $ORACLE_HOME/dbs/*ASM*
    
    
  4. If you are not using a cluster file system for your ASM Oracle home, then run the rm commands listed in the previous step on each node on which the Oracle home exists.

  5. Remove oratab entries beginning with +ASM.

1.11 Backing Up the Voting Disk after Installing RAC

After installing Oracle RAC 10g and after ensuring that the system is functioning properly, make a backup of the voting disk. In addition, make a backup of the voting disk contents after you complete any node additions or node deletions and after running any de-installation procedures.

1.12 Remote Undo Tablespaces Do Not Autoexetend in RAC Seed Databases

If you create a RAC database with two or more instances and you choose to create General Purpose, Transaction Processing, or Data Warehouse databases, and if you use a shared cluster file system or Automatic Storage Management (ASM) for database files, then the Database Configuration Assistant (DBCA) creates undo tablespace datafiles with an initial size of 25 MB and AUTOEXTEND ON for the local instance but AUTOEXTEND OFF for remote instances.

You can set AUTOEXTEND ON for undo tablespace datafiles for remote instances after creating a RAC database using DBCA as follows:

  1. Connect to the database instance on the node from which you ran DBCA:

    $ sqlplus "/ AS SYSDBA"
    
    
  2. Enter the following command to find the datafile names for UNDOTBS tablespaces for remote instances:

    SQL> SELECT file_name FROM SYS.DBA_DATA_FILES
         WHERE tablespace_name LIKE 'UNDOTBS%' AND AUTOEXTENSIBLE='NO';
    
    
  3. Set AUTOEXTEND ON for the datafiles that you found in the previous step:

    SQL> ALTER DATABASE DATAFILE datafile_name AUTOEXTEND ON;
    

1.13 Upgrading a Database

If you choose the Custom installation path and have previous versions of Oracle databases installed, then the Oracle Universal Installer gives you the option of upgrading the existing databases. If you choose to upgrade an existing database, the Oracle Universal Installer displays another screen asking whether you want to create a starter database.If you choose to create a starter database, the DBCA upgrades the older version of the database to Oracle Database 10g release 1 (10.1), but does not create a starter database.

1.14 Upgrading an Oracle9i Database to Oracle Database 10g

If you upgrade an Oracle9i database to Oracle Database 10g release 1, Oracle Flashback features using a timestamp may fail. To work around this problem, enter the following SQL script from the Oracle Database 10g database:

SQL> DELETE FROM smon_scn_time WHERE orig_thread <> 0;
SQL> COMMIT;

This issue is tracked with Oracle bug 3994270.

1.15 Net Configuration Assistant Help

In the Net Configuration Assistant (NetCA) help, the link to the Select Oracle Context help topic is broken. The text for this topic is as follows:

Directory Usage Configuration, Select Oracle Context

Oracle administrative content has been found in more than one location in the directory. Oracle administrative content is stored in an Oracle Context, a subtree in the directory that stores Oracle entries.

From the list, select or enter the location you want to use as the default Oracle Context location from which this computer will access Oracle entries, such as connect identifiers.

1.16 Flashback Table or Flashback Analysis

If a user invokes the Flashback Table or Flashback Analysis operation, and that user has FLASHBACK ANY TABLE privileges but does not have specific flashback privileges on the objects that flashback is invoked on and does not have DBA privileges, then the following errors may occur:

ORA-02002: error while writing to audit trail 
ORA-00600: internal error code, arguments: [kzasps1], [4], [47], [],[],  

To fix this problem, as SYSDBA grant the user FLASHBACK privilege on the objects that are referred to in the FLASHBACK TABLE statement and then invoke the flashback operation.

For example:

SQL> GRANT FLASHBACK ON SCOTT.EMP_1 TO user1;

This issue is tracked through Oracle bug 3403666.

1.17 Oracle Workflow

Although Oracle Workflow is listed on the Companion CD installation screens, it is not included on the Companion CD for this release.

1.18 Enabling Automated Backups

While installing Oracle Database, the Specify Backup and Recovery Options screen may appear truncated if your system does not have the required fonts installed. If your system has only fixed-width fonts, you may not be able to fully specify the required information in the Backup Job Credentials area of the screen. To work around this issue, do not select Enable Automated Backups on this screen. After the installation is complete, use the Oracle Enterprise Manager 10g Database Control to enable automated backups.

1.19 Oracle Internet Directory

This release includes the Oracle Internet Directory (OID) client tools, but it does not include OID server components. OID server components are included with Oracle Application Server 10g. If you require the OID server tools for Oracle Database components, then run them from an Oracle Application Server 10g installation.

The OID client tools include:

  • LDAP command-line tools

  • Oracle Internet Directory SDK

  • Oracle Directory Manager

The OID server components include the following servers and tools for starting and stopping them:

  • Directory server

  • Directory replication server

  • Directory integration server

1.20 Error When Viewing Period SQL Execution Plan in Korean

Viewing the execution plan of a Period SQL in Korean causes an internal server error. This problem is unique to Korean; it does not reproduce in Japanese or Chinese. The only workaround currently available is to run the product in a language other than Korean when you need to view this page.

1.21 Quick Tour Not Available in Oracle Change Management Pack

Quick Tour is not available in Oracle Change Management Pack. If you try to run it, then an error results.

1.22 extjob Executable Required Directory Permissions

To enable the extjob executable to locate required libraries, the $ORACLE_HOME/lib directory and all of its parent directories must have execute permissions for group and other.

1.23 Setting the AIXTHREAD_SCOPE Environment Variable

Threads in AIX can run with process-wide contention scope (M:N) or with systemwide contention scope (1:1). The AIXTHREAD_SCOPE environment variable controls which contention scope is used.

The default value of the AIXTHREAD_SCOPE environment variable is P which specifies process-wide contention scope. When using process-wide contention scope, Oracle threads are mapped to a pool of kernel threads. When Oracle is waiting on an event and its thread is swapped out, it may return on a different kernel thread with a different thread ID. Oracle uses the thread ID to post waiting processes so it is important for the thread ID to remain the same. When using systemwide contention scope, Oracle threads are mapped to kernel threads statically, one to one. For this reason Oracle recommends using systemwide contention. The use of systemwide contention is especially critical for Oracle Real Application Clusters (RAC) instances.Additionally, on AIX 5L version 5.2 or higher, if you set systemwide contention scope, significantly less memory is allocated to each Oracle process.

Oracle recommends that you set the value of the AIXTHREAD_SCOPE environment variable to S in the environment script that you use to set the ORACLE_HOME or ORACLE_SID environment variables for an Oracle database instance or an Oracle Net listener process, as follows:

  • Bourne, Bash, or Korn shell:

    Add the following line to the ~/.profile or /usr/local/bin/oraenv script:

    AIXTHREAD_SCOPE=S; export AIXTHREAD_SCOPE
    
    
  • C shell:

    Add the following line to the ~/.login or /usr/local/bin/coraenv script:

    setenv AIXTHREAD_SCOPE S
    
    

Doing this enables systemwide thread scope for running all Oracle processes.

1.24 Required Patch for Concurrent I/O

Authorized Problem Analysis Report (APAR) IY56024 is required to prevent concurrent I/O writes returning an incorrect length on AIX 5L systems.


See Also:

For more information on this patch, see the following Web site:
http://www-912.ibm.com/eserver/support/fixes/

1.25 Oracle Transparent Gateway for Sybase

Oracle Transparent Gateways for Sybase does not currently work with Sybase Open Client V12.0. A patch to correct this problem is available from the Download area of the OracleMetalink Web site:

http://metalink.oracle.com/

This issue is tracked through Oracle bug 3398064.

1.26 Problems Starting Oracle Enterprise Manager Database Control

If you choose to configure Oracle Enterprise Manager Database Control during an Oracle Database installation on AIX 5L version 5.3, Database Control might not start. This problem occurs because the Installer cannot obtain a free port in the default range 1830 to 1849.

To work around this problem, before installation edit the /etc/services file to remove or comment out the port assignments for ports in the range 1830 to 1849.

If you do not edit the /etc/services file before installation, perform the following steps after installation:

  1. Open the following file in a text editor:

    $ORACLE_HOME/hostname_sid/sysman/config/emd.properties
    
    
  2. Add an available TCP port to the value of the EMD_URL parameter as follows:

    EMD_URL=http://hostname:available_tcp_port/emd/main
    
    
  3. Open the following file in a text editor:

    $ORACLE_HOME/hostname_sid/sysman/emd/targets.xml
    
    
  4. Edit the port number in the oracle_emd target type to match the port number that you specified in step 2.

  5. Restart Database Control as follows:

    $ emctl start dbconsole
    
    

1.27 Starting Oracle Enterprise Manager Database Control

Before installing Oracle Database or before starting Oracle Enterprise Manager Database Control, you must set the DISPLAY environment variable to specify a valid X server as follows, where workstation_name is the name of a valid X server:

  • Bourne, Bash, or Korn shell:

    $ DISPLAY=workstation_name:0.0 ; export DISPLAY
    
    
  • C shell:

    % setenv DISPLAY workstation_name:0.0
    
    

If the DISPLAY environment variable specifies a server that is not an X server, the Installer fails or Oracle Enterprise Manager Database Control produces Java core files and users will not be able to connect to it from the browser.

1.28 Grid Features

Oracle Database New Features for Oracle Database 10g release 1 (10.1) lists two Grid features that are not available in the first release of Oracle Database 10g; Resonance and Transparent Session Migration. These features will be available in a future release.

1.29 Removing Cloned Oracle Homes


Note:

This issue occurs only if the cloned and source Oracle homes are on the same system.

If you use the Installer to remove a cloned Oracle home, and the cloned Oracle home is on the same system as the source Oracle home, the Installer removes essential files from the source Oracle home. To avoid this issue, remove the cloned Oracle home as follows:

  1. Remove the cloned Oracle home manually by deleting the cloned Oracle home directory.

  2. Start the Installer from the source Oracle home directory:

    $ source_oracle_home/oui/bin/runInstaller
    
    
  3. In the Welcome window, click Deinstall Products.

    The Inventory window appears, listing all of the Oracle homes on the system.

  4. In the Inventory window, select the cloned Oracle home, then click Remove.

This issue is tracked through Oracle bug 3417663.

1.30 JDBC Driver and the Just-In-Time Java Compiler

If you use the JDBC driver, and the JAVA_COMPILER environment variable is set to any value other than NONE, Jserver will not start. This problem is fixed by setting JAVA_COMPILER to NONE. For example:

  • Set the JAVA_COMPILER environment variable.

    • Bourne, Bash, or Korn shell:

      $ JAVA_COMPILER=NONE ; export JAVA_COMPILER
      
      
    • C shell:

      % setenv JAVA_COMPILER NONE
      
      
  • Specify the following option with the java command:

    -Djava.compiler=none
    

1.31 Oracle Scheduler

You cannot use Oracle Scheduler to run jobs external to the database on your system.

1.32 Installing Enterprise Security Manager

To install Enterprise Security Manager (ESM), install Oracle Client and choose the Administrator installation type.

1.33 Using org.w3c.dom.ls Package with IBM JDK 1.4.2

When using IBM JDK 1.4.2 to build an application that uses the org.w3c.dom.ls package, the following errors might be thrown:

cannot resolve symbol
symbol  : method getConfig ()
location: interface org.w3c.dom.ls.LSParser

or

cannot resolve symbol
symbol  : method getConfig ()
location: interface org.w3c.dom.ls.LSSerializer

This issue is being tracked with Oracle bug 4286338.

1.34 Adding Node to Database on Automatic Storage Management

Adding a node to database on automatic storage management fails if $ORACLE_HOME is not shared.

Workaround: Do Custom install in single node and select ifile instead of pfile for ASM parameter file.

2 Documentation Updates

The following sections contain updates to Oracle documentation:

2.1 Updates for Real Application Clusters Documentation

The following sections describe corrections and supplemental information for the Oracle Real Application Clusters 10g documentation for UNIX platforms:

2.1.1 Installing Oracle CRS

The Oracle Real Application Clusters Installation and Configuration Guide, Chapter 8, "Installing Cluster Ready Services", under the heading "Installing Cluster Ready Services with the OUI", explains how to run the following script before running Oracle Universal Installer from the Oracle CRS CD-ROM:

ORACLE_HOME/bin/localconfig delete

Instead, run this script after starting the Oracle CRS installation when Oracle displays the warning to stop all Oracle services.


Note:

The documentation is correct by instructing you to stop the existing ASM instances before installing Oracle CRS. Only the timing of when to run localconfig delete is changed from before to during the installation.

2.1.2 Configuring Raw Disk Devices for CRS or Database File Storage for RAC Installations Without HACMP or GPFS

If you are installing RAC on an AIX cluster without HACMP or GPFS, you must use shared raw disk devices to store the Cluster Ready Services (CRS) files. This section describes how to configure the shared raw disk devices for the CRS files (Oracle Cluster Registry and CRS voting disk) and the database files.


Note:

You can also use shared raw disk devices to store database files. However, Oracle recommends that you use ASM for database file storage if you are not using HACMP or GPFS.

To configure shared raw disk devices for database files, configure the required shared devices as described in steps 2 through 7. The number and size of the required raw devices are listed in Step 2 of the "Create Raw Logical Volumes in the New Volume Group" section in the Oracle Real Application Clusters Installation and Configuration Guide.


The Oracle Cluster Registry (OCR) requires a disk with 100 MB of storage space and the CRS voting disk requires a disk with 20 MB of storage space.


Note:

Because each file requires exclusive use of a complete disk device, Oracle recommends that you use this storage option for CRS files only if you do not have HACMP or GPFS. You cannot use the disks that you choose for these files for any other purpose.

To configure raw shared disk devices for the OCR and CRS voting disk, follow these steps:

  1. Select two unused disks that are shared on all of the cluster nodes.

  2. As the root user, enter the following command on any node to identify the device names for the physical disks that you want to use:

    # /usr/sbin/lspv | grep -i none 
    
    

    This command displays information similar to the following for each disk that is not configured in a volume group:

    hdisk17         0009005fb9c23648                    None  
    
    

    In this example, hdisk17 is the device name of the disk and 0009005fb9c23648 is the physical volume ID (PVID).

  3. If the disks that you want to use for the CRS files do not have a PVID, enter a command similar to the following for each disk that you want to use:

    # /usr/sbin/chdev -l hdiskn -a pv=yes
    
    
  4. On each of the other nodes, enter a command similar to the following to identify the device names associated with the PVID on that node:

    # /usr/sbin/lspv | grep -i "0009005fb9c23648"
    
    

    The output from this command should be similar to the following:

    hdisk18         0009005fb9c23648                    None  
    
    

    In this example, the device name associated with the drive (hdisk18) is different on this node.

  5. If the device names are the same on all nodes, enter commands similar to the following to change the owner, group, and permissions on the character raw device files for the disks:

    # chown oracle:dba /dev/rhdiskn
    # chmod 660 /dev/rhdiskn
    
    
  6. If the device name associated with the PVID for a disk that you want to use is different on any node, you must create a new device file for the disk on each of the nodes using a common unused name.

    For the new device file, choose a device file name that identifies the purpose of the disk device, for example ocr_disk or crs_vote_disk.


    Note:

    Alternatively, you could choose a name that contains a number that will never be used on any of the nodes, for example hdisk99.

    To create a new common device file name for a disk on all nodes, follow these steps on each node:

    1. Enter the following command to determine the device major and minor numbers that identify the disk, where n is the disk number for the disk on this node:

      # ls -alF /dev/*hdiskn
      
      

      The output from this command is similar to the following:

      brw------- 1 root system    24,8192 Dec 05 2001  /dev/hdiskn
      crw------- 1 root system    24,8192 Dec 05 2001  /dev/rhdiskn
      
      

      In this example, the device file /dev/rhdiskn represents the character raw device, 24 is the device major number, and 8192 is the device minor number.

    2. Enter a command similar to the following to create the new device file, specifying the new device file name and the device major and minor numbers that you identified in the previous step:


      Note:

      In the following example, you must specify the character c to create a character raw device file.

      # mknod /dev/ocr_disk c 24 8192
      
      
    3. Enter commands similar to the following to change the owner, group, and permissions on the character raw device file for the disk:

      # chown oracle:dba /dev/ocr_disk
      # chmod 660 /dev/ocr_disk
      
      
    4. Enter a command similar to the following to verify that you have created the new device file successfully:

      # ls -alF /dev |grep "24,8192"
      
      

      The output should be similar to the following:

      brw------- 1 root   system   24,8192 Dec 05 2001  /dev/hdiskn
      crw-rw---- 1 oracle dba      24,8192 Dec 05 2001  /dev/ocr_disk
      crw------- 1 root   system   24,8192 Dec 05 2001  /dev/rhdiskn
      
      
  7. To enable simultaneous access to a disk device from multiple nodes, you must set the appropriate Object Data Manager (ODM) attribute listed in the following table to the value shown, depending on the disk type:

    Disk Type Attribute Value
    EMC PowerPath devices, SSA, or FAStT disks reserve_lock no
    ESS, HDS, or MPIO-capable disks reserve_policy no_reserve

    To determine whether the attribute has the correct value, do one of the following depending on your disk type:

    • SSA, FAStT, ESS, HDS, or MPIO-capable disks:

      Enter a command similar to the following on all cluster nodes for each disk device that you want to use:

      # /usr/sbin/lsattr -E -l hdiskn
      
      
    • EMC PowerPath devices:

      Enter a command similar to the following on each of the CRS and database devices that you want to use:

      # /usr/sbin/lsattr -E -l hdiskpowern 
      
      

    If necessary, change the attribute value as follows, depending on the disk type:

    • SSA or FAStT disks:

      # /usr/sbin/chdev -l hdiskn  -a reserve_lock=no
      
      
    • EMC PowerPath devices:

      # /usr/sbin/chdev -l hdiskpowern  -a reserve_lock=no
      
      
    • ESS, HDS, or MPIO-capable disks:

      # /usr/sbin/chdev -l hdiskn  -a reserve_policy=no_reserve
      
      

    Note:

    If the appropriate attribute is not set to the value listed in the table, problems may occur during access from multiple nodes.

  8. If necessary, enter the following command to remove all physical volume IDs (PVIDs) from the raw devices:

    # /usr/sbin/chdev -l hdiskn -a pv=clear
    
    
  9. When you are installing Oracle CRS, enter the paths to the appropriate device files when prompted for the path of the OCR and CRS voting disks, for example:

    /dev/ocr_disk
    

2.1.3 Adding Nodes

The following points describe corrections for the node addition procedures in Chapter 5, "Adding and Deleting Nodes and Instances" in the Oracle Database Oracle Clusterware and Oracle Real Application Clusters Administration and Deployment Guide.

  • Under the heading "Step 2 Adding Nodes at the Oracle Clusterware Layer (UNIX and Windows)", point 9 on page 5-6, the format for the racgons command should be:

    racgons add_config new_node_name:4948
    
    
  • Under the heading "Step 4: Adding Nodes at the Oracle RAC Database Layer", point 8 on page 5-9, explains that you must run the Virtual Internet Protocol Assistant (VIPCA). You must run the VIPCA with root user privileges.

  • Under the heading "Step 4: Adding Nodes at the Oracle RAC Database Layer", point 9 on page 5-10 explains that you must run the oifcfg command. The correct syntax for this command requires that you specify either the -n nodename option or the -global option to complete the command. The correct syntax for the oifcfg command is as follows:

    oifcfg iflist
    oifcfg setif {-node nodename | -global} {if_name/subnet:if_type}...
    oifcfg getif [-node nodename | -global] [ -if if_name[/subnet]        [-type if_type] ]
    oifcfg delif [-node nodename | -global] [if_name[/subnet]]
    oifcfg [-help]
    
    

    A standard Oracle CRS installation issues the oifcfg command as in the following example:

    oifcfg setif -global  eth0/146.56.76.0:public \
    eth1/192.0.0.0:cluster_interconnect'
    
    

    This sets both networks to global. Therefore, you do not need to run the oifcfg command manually after you add a node.

  • When you add a node to a RAC database, you must also add a listener to the new node by running the Net Configuration Assistant (NetCA).

2.1.4 Deleting Nodes

The following corrections are for the node deletion procedures for UNIX-based systems as described in Chapter 5, "Adding and Deleting Nodes and Instances" in the Oracle Database Oracle Clusterware and Oracle Real Application Clusters Administration and Deployment Guide. These corrections assume that you have a three-node cluster and that you are deleting the third node.

  • Step 2 in the node deletion procedure for UNIX-based systems should read:

    After you have deleted the instances from the nodes that you want to delete, delete the node applications for each node by running the following command where <node1, node2> are the nodes that you are removing from your cluster:

    rootdeletenode.sh <node1>,<node2>
    
    
  • For Step 3 in the node deletion procedure for UNIX-based systems, the runInstaller command is located in the $ORACLE_HOME/oui/bin directory. In addition, executing this command does not launch the Installer.

  • For Step 5 in the node deletion procedure for UNIX-based systems, only run the <CRS Home>/install/rootdelete.sh command once for that step. Use the argument nosharedhome when using local file systems. The default is sharedhome which prevents updating the permissions of local files such that they can be removed by the oracle user.

  • For Step 7 in the node deletion procedures for UNIX-based systems, the command is located in the Oracle CRS Home directory. In this step, ORACLE_HOME should read CRS_Home and the text for Step 7 should read:

    On the same node and as the oracle user, run the command <CRS home>/oui/bin/runInstaller -updateNodeList ORACLE_HOME=<CRS home> CLUSTER_NODES=node1,node2,... <nodeN> where node1 through <nodeN> is a comma-delimited list of nodes that are remaining in the cluster.

2.1.5 Real Application Clusters Services Deployment Example

The following two sections that describe shadow events and high availability callouts with Oracle Notification Services (ONS) events are supplements for Appendix A, "Services Deployment Example", in the Oracle Real Application Clusters Deployment and Performance Guide, part number B10768-01.

2.1.5.1 Events for Shadow Preconnect Services in Real Application Clusters

When using Transparent Application Failover (TAF) PRECONNECT, Real Application Clusters (RAC) high availability maintains a preconnect service to support TAF Preconnect and applications that are configured to manage work on secondary RAC instances. Secondary instances are RAC instances that are not supporting the primary service.

In this type of configuration, Oracle maintains the shadow service on all instances that do not support the primary service. You can use events to stop and start secondary work. The events are posted to callouts and to the Oracle Notification Service (ONS).

To use events, configure the payload with the following format:

Up Event Format:

SRV_PRECONNECT VERSION=1.0 service=db_unique_name.db_domain 
database=database_name instance instance_name host=host_name 
status=preconn_up reason=timestamp=27-Jan-2004 16:53:58 
reported=Tue Jan 27 16:53:59 PST 2004 

Down Event Format:

SRV_PRECONNECT VERSION=1.0 service=db_unique_name.db_domain
database=database_name instance=instance_name host=host_name
status=preconn_down reason=timestamp=27-Jan-2004 16:58:01
reported=Tue Jan 27 16:58:02 PST 2004

Up Event Example:

@ SRV_PRECONNECT VERSION=1.0 service=MYSERV.us.oracle.com 
database=RACEY instance=RACEY1 host=sun880-1 
status=preconn_up reason=timestamp=27-Jan-2004 16:53:58
reported=Tue Jan 27 16:53:59 PST 2004 

Down Event Example:

@ SRV_PRECONNECT VERSION=1.0 service=MYSERV.us.oracle.com
database=RACEY instance=RACEY1 host=sun880-1 
status=preconn_down reason=timestamp=27-Jan-2004 16:58:01
reported=Tue Jan 27 16:58:02 PST 2004
2.1.5.2 RAC High Availability Callouts and Oracle Notification Event

The notification interface is available as a server-side callout and as an Oracle Notification Services (ONS) event. The server-side callout is a script with the same payload as the ONS event that is immediately executed on the server when the condition occurs. Use this method to start and stop server-side applications, to relocate low-priority services when high priority services arrive, and to post tickets for fault tracking. The following table describes the event payload:

Parameter Description
Event type The event type for the component such as service, service_member, database, instance, or node
Service name The service name; matches the configured service in SERVICE$
Database name The database supporting the service; matches the initialization parameter value for DB_UNIQUE_NAME, which in turn defaults to the value of the initialization parameter DB_NAME
Instance The name of the instance that supports the service; matches the instance name
Node name The name of the node that supports the service or the node that has failed; matches the CSS node name
Status The new status; values are UP, DOWN, and NOT_RESTARTING
Cardinality Cardinality for the service on UP events
Time stamp The local time zone to use when ordering notification events
Incarnation Cluster incarnation for node down

When a session connects, mid-tiers can record the following values that match the HA event payload:

sys_context('userenv', 'instance_name');
sys_context('userenv', 'server_host');
sys_context('userenv', 'service_name');
sys_context('userenv', 'db_unique_name');

2.1.6 Configuring Disks for ASM

If you intend to use ASM as the storage mechanism for a RAC database, configure the raw disk devices used in the ASM disk group to enable simultaneous access from multiple nodes. To enable simultaneous access to a disk device from multiple nodes, you must set the appropriate Object Data Manager (ODM) attribute listed in the following table to the value shown, depending on the disk type:

Disk Type Attribute Value
EMC PowerPath devices, SSA, or FAStT disks reserve_lock no
ESS, HDS, or MPIO-capable disks reserve_policy no_reserve

To determine whether the attribute has the correct value, do one of the following depending on your disk type:

  • SSA, FAStT, ESS, HDS, or MPIO-capable disks:

    Enter a command similar to the following on all cluster nodes for each disk device that you want to use:

    # /usr/sbin/lsattr -E -l hdiskn
    
    
  • EMC PowerPath devices:

    Enter a command similar to the following on each of the CRS and database devices that you want to use:

    # /usr/sbin/lsattr -E -l hdiskpowern 
    
    

If the required attribute is not set to the correct value on any node, enter a command similar to one of the following on that node:

  • SSA or FAStT disks

    # /usr/sbin/chdev -l hdiskn  -a reserve_lock=no
    
    
  • EMC PowerPath devices

    # /usr/sbin/chdev -l hdiskpowern  -a reserve_lock=no
    
    
  • ESS, HDS, or MPIO-capable disks

    # /usr/sbin/chdev -l hdiskn  -a reserve_policy=no_reserve
    
    

Before raw devices are used, remove all physical volume IDs (PVIDs) from the raw devices. PVIDs may have been assigned by the user to correlate the hdisk numbers for disks shared across RAC nodes. Enter the following command to remove the PVID.

# /usr/sbin/chdev -l hdiskn -a pv=clear

Enter commands similar to the following on every node to change the owner, group, and permissions on the character raw device file for each disk that you want to add to the disk group:

# chown oracle:dba /dev/rhdiskn 
# chmod 660 /dev/rhdiskn
 

Note:

If you are using a multi-pathing disk driver with ASM, make sure that you set the permissions only on the correct logical device name for the disk. The device name associated with a disk might be different on other nodes. Make sure that you specify the correct device name on each node.

2.1.7 Running DBCA on RAC

The following text appears on page 5-10 of the Oracle Real Application Clusters Administrator's Guide:

Execute the following procedures on each new node to add instances:

  1. Start the Database Configuration Assistant (DBCA) by entering dbca at the system prompt from the bin directory in the $ORACLE_HOME on UNIX.

This text should be changed as follows:

Execute the following procedures for each new node to add instances:

2.1.8 srvctl is Spelled Incorrectly

In Oracle Real Application Clusters Installation and Configuration Guide10g Release 1 (10.1) for AIX-Based Systems, Apple Mac OS X, hp HP-UX, hp Tru64 UNIX, Linux, Solaris Operating System, and Windows Platforms, srvclt is incorrectly spelled as svrctl in the following sections:

  • Chapter 11, "Installing Oracle Database 10g with Real Application Clusters"

  • Chapter 13, "Real Application Clusters Post-Installation Procedures"

  • Appendix B, "Using Scripts to Create Real Application"

2.2 Updates for Other Documentation

The following sections contain updates to the Oracle Database 10g documentation:

2.2.1 Building Pro*C/C++ Precompiler Demonstrations

When entering the commands shown in the "Pro*C/C++ Precompiler" section of the Oracle Database 10g Administrator's Reference, include the -r option, for example:

$ make -r -f demo_proc.mk sample1

If you do not use the -r option, the demonstrations may not compile correctly.

2.2.2 Installing Oracle HTML DB

Although the Oracle Database Companion CD Installation Guide states that you can install Oracle HTML DB into an existing Oracle home that contains Oracle Application Server 10g or Oracle Enterprise Manager 10g, this type of installation is not supported with this release. You must install Oracle HTML DB with the Apache Standalone component in a separate Oracle home directory.

2.2.3 Globalization Support

The instructions about the ORA_NLS10 environment variable in the "Using Oracle9i Database Language and Territory Definition Files with Oracle Database 10g" section of the Oracle Database 10g README file are incorrect. This is because the $ORACLE_HOME/nls/data/old directory contains definitions files which have changed in Oracle Database 10g release 1 only, and not the complete set of the database locale definition files. To include the Oracle9i definitions:

  1. Create a new directory:

    $ mkdir $ORACLE_HOME/nls/data/9i
    
    
  2. Copy the contents of the $ORACLE_HOME/nls/data directory to the new directory:

    $ cp $ORACLE_HOME/nls/data/*.nlb $ORACLE_HOME/nls/data/9i
    
    
  3. Copy the contents of the $ORACLE_HOME/nls/data/old directory to the directory that you created in step 1:

    $ cp $ORACLE_HOME/nls/data/old/*.* $ORACLE_HOME/nls/data/9i
    
    
  4. Set the ORA_NLS10 environment variable to specify the directory that you created, for example:

    • Bourne, Bash, or Korn shell:

      $ ORA_NLS10=$ORACLE_HOME/nls/data/9i ; export ORA_NLS10
      
      
    • C shell:

      % setenv ORA_NLS10 $ORACLE_HOME/nls/data/9i
      
      

    See Also:

    Refer to the Oracle Database 10g Globalization Support Guide and the Oracle Database 10g README for more information about the changes in the language and territory definition files between Oracle9i and Oracle Database 10g.

2.2.4 Full-Text Searching with Oracle Text

For full-text searching with Oracle Text, you must create XML tables manually.

If you will need to use Oracle Text indexes for text-based ora:contains searches over a collection of XML elements, then do not use XML schema annotation storeVarrayAsTable="true". This annotation causes element collections to be persisted as rows in an Index Organized Table (IOT). Oracle Text does not support IOTs.

To be able to use Oracle Text to search the contents of element collections, set parameter genTables="false" during schema registration. Then create the necessary tables manually, without using the clause ORGANIZATION INDEX OVERFLOW. The tables will then be heap-organized instead of index-organized (IOT), as shown in the following example:

CREATE TABLE PurchaseOrder of XMLTYPE
  XMLSCHEMA http://localhost:8080/home/SCOTT/poSource/xsd/purchaseOrder.xsd
  ELEMENT "PurchaseOrder"
  VARRAY "XMLDATA"."ACTIONS"."ACTION"
    STORE AS TABLE ACTION_TABLE ((PRIMARY KEY (NESTED_TABLE_ID, ARRAY_INDEX)))
  VARRAY "XMLDATA"."LINEITEMS"."LINEITEM"
    STORE AS TABLE LINEITEM_TABLE ((PRIMARY KEY (NESTED_TABLE_ID, ARRAY_INDEX)));
       

2.2.5 View Descriptions

Oracle Database Performance Tuning Guide, Chapter 10, "Instance Tuning Using Performance Views", has two incorrect view descriptions. The view names and correct descriptions are as follows:

  • V$EVENT_HISTOGRAM

    The V$EVENT_HISTOGRAM view displays a histogram of the number of waits, the maximum wait, and total wait time on an event basis.

  • V$SYSTEM_WAIT_CLASS

    The V$SYSTEM_WAIT_CLASS view provides the instance-wide time totals for the number of waits and the time spent in each class of wait events. (This description removes the reference to object number.)

2.2.6 Configuring Management Service Processes

In Oracle High Availability Architecture and Best Practices, Chapter 8, "Using Enterprise Manager for Monitoring and Detection", the section "Configure At Least Two Service Processes and Load Balance Them" should read as follows: For the middle tier, the baseline recommendation is to have a minimum of two Management Service processes, using a hardware server load balancer to mask the location of an individual Management Service process and a failure of any individual component. This provides immediate coverage for a single failure in the most critical components in the Enterprise Manager architecture with little interruption of service for all systems monitored using Enterprise Manager. Management Service processes connect to the repository instances using Oracle Net.

2.2.7 Monolingual Linguistic Sorting

In Oracle Database Globalization Support Guide, Chapter 5, "Linguistic Sorting and String Searching", the section called "Monolingual Linguistic Sorts" should end with the following note:

Monolingual linguistic sorting is not available for nonUnicode multibyte database character sets. If a monolingual linguistic sort is specified when the database character set is nonUnicode multibyte, then the default sort order is the binary sort order of the database character set. One exception is UNICODE_BINARY. This sort is available for all character sets.

2.2.8 XDK Error Messages

XDK error messages are available at the XML Technology Center on the OTN Web site:

http://www.oracle.com/technology/tech/xml/doc/production10g/Javaerrormsgs.html

2.2.9 Support for Hearing and Speech Impaired Customers

The Text Telephone (TTY) phone number for nontechnical support is listed incorrectly in several documents. The correct phone number is 1.800.464.2330.

3 Documentation Accessibility

Our goal is to make Oracle products, services, and supporting documentation accessible, with good usability, to the disabled community. To that end, our documentation includes features that make information available to users of assistive technology. This documentation is available in HTML format, and contains markup to facilitate access by the disabled community. Accessibility standards will continue to evolve over time, and Oracle is actively engaged with other market-leading technology vendors to address technical obstacles so that our documentation can be accessible to all of our customers. For more information, visit the Oracle Accessibility Program Web site at http://www.oracle.com/accessibility/.

Accessibility of Code Examples in Documentation

Screen readers may not always correctly read the code examples in this document. The conventions for writing code require that closing braces should appear on an otherwise empty line; however, some screen readers may not always read a line of text that consists solely of a bracket or brace.

Accessibility of Links to External Web Sites in Documentation

This documentation may contain links to Web sites of other companies or organizations that Oracle does not own or control. Oracle neither evaluates nor makes any representations regarding the accessibility of these Web sites.

TTY Access to Oracle Support Services

Oracle provides dedicated Text Telephone (TTY) access to Oracle Support Services within the United States of America 24 hours a day, seven days a week. For TTY support, call 800.446.2398.


Oracle Database Release Notes, 10g Release 1 (10.1.0.2.0) for AIX 5L Based Systems (64-Bit)

B13611-08

Copyright © 1996, 2006, Oracle. All rights reserved.

The Programs (which include both the software and documentation) contain proprietary information; they are provided under a license agreement containing restrictions on use and disclosure and are also protected by copyright, patent, and other intellectual and industrial property laws. Reverse engineering, disassembly, or decompilation of the Programs, except to the extent required to obtain interoperability with other independently created software or as specified by law, is prohibited.

The information contained in this document is subject to change without notice. If you find any problems in the documentation, please report them to us in writing. This document is not warranted to be error-free. Except as may be expressly permitted in your license agreement for these Programs, no part of these Programs may be reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose.

If the Programs are delivered to the United States Government or anyone licensing or using the Programs on behalf of the United States Government, the following notice is applicable:

U.S. GOVERNMENT RIGHTS Programs, software, databases, and related documentation and technical data delivered to U.S. Government customers are "commercial computer software" or "commercial technical data" pursuant to the applicable Federal Acquisition Regulation and agency-specific supplemental regulations. As such, use, duplication, disclosure, modification, and adaptation of the Programs, including documentation and technical data, shall be subject to the licensing restrictions set forth in the applicable Oracle license agreement, and, to the extent applicable, the additional rights set forth in FAR 52.227-19, Commercial Computer Software—Restricted Rights (June 1987). Oracle Corporation, 500 Oracle Parkway, Redwood City, CA 94065

The Programs are not intended for use in any nuclear, aviation, mass transit, medical, or other inherently dangerous applications. It shall be the licensee's responsibility to take all appropriate fail-safe, backup, redundancy and other measures to ensure the safe use of such applications if the Programs are used for such purposes, and we disclaim liability for any damages caused by such use of the Programs.

Oracle, JD Edwards, PeopleSoft, and Retek are registered trademarks of Oracle Corporation and/or its affiliates. Other names may be trademarks of their respective owners.

The Programs may provide links to Web sites and access to content, products, and services from third parties. Oracle is not responsible for the availability of, or any content provided on, third-party Web sites. You bear all risks associated with the use of such content. If you choose to purchase any products or services from a third party, the relationship is directly between you and the third party. Oracle is not responsible for: (a) the quality of third-party products or services; or (b) fulfilling any of the terms of the agreement with the third party, including delivery of products or services and warranty obligations related to purchased products or services. Oracle is not responsible for any loss or damage of any sort that you may incur from dealing with any third party.