Oracle® Database Installation Guide
10g Release 1 (10.1) for hp OpenVMS Alpha
Part No. B13681-01
This chapter provides an overview of the procedures for configuring Oracle Database 10g Real Application Clusters (RAC) on OpenVMS.
You may also need to refer to the following RAC documents for more complete information about installing and administering RAC on OpenVMS:
Oracle Real Application Clusters Installation and Configuration Guide (B10766-08)
Oracle Real Application Clusters Administration and Deployment Guide (B14197-01)
Oracle Real Application Clusters Deployment and Performance Guide (B10768-02)
This chapter includes the following topics:
Oracle Universal Installer does not suggest a default location for the Oracle Cluster Registry (OCR) or the Oracle CRS voting disk. If you choose to create these files on a file system, use the following guidelines when deciding where to place them:
You must choose a disk mounted and available on all the nodes in the RAC cluster.
It must have at least 100 MB of free disk space for the OCR and 20 MB of free disk space for the CRS voting disk.
For improved reliability, you should choose a file system on a highly available storage device, for example, a RAID device that implements mirroring.
oracle user must have write permissions to create the files in the path that you specify.
If you choose to place the Oracle Database files on a file system, use the following guidelines when deciding where to place them:
You must select a disk that is accessible across all the nodes in the RAC cluster.
The default path suggested by Oracle Universal Installer for the database file directory is a subdirectory of the Oracle home directory. This default location is not recommended for production databases.
Choose a set of disks dedicated to the database.
For best performance and reliability, choose a RAID device or a logical volume on more than one physical device and implement the stripe-and-mirror-everything (SAME) methodology.
This method enables you to distribute physical I/O and create separate control files on different devices for increased reliability. You must choose either the Advanced database creation option or the Custom installation type during the installation to implement this method.
If you intend to create a preconfigured database during the installation, the file system (or file systems) that you choose must have at least 1.2 GB of free disk space.
For optimum performance, the file systems that you choose should be on physical devices that are used only by the database.
oracle user must have write permissions to create the files in the path that you specify.
Note:You must choose a location for recovery files only if you intend to enable automated backups during the installation.
If you choose to place the Oracle Database recovery files on a file system, use the following guidelines for placement of Oracle Database recovery files on the file system:
To prevent disk failure from making both the database files and the recovery files unavailable, place the recovery files in a file system on a different physical disk from the database files.
You should choose a device that is accessible on all the nodes of the RAC cluster.
The device that you choose should have at least 2 GB of free disk space.
The disk space requirement is the default disk quota configured for the flash recovery area (specified by the
DB_RECOVERY_FILE_DEST_SIZE initialization parameter).
If you choose the Custom installation type, then you can specify a different disk quota value. After you create the database, you can also use Oracle Enterprise Manager Grid Control to specify a different value.
See Also:For more information about sizing the flash recovery area, refer to Oracle Database Backup and Recovery Basics (B10735-01)
The default path suggested by Oracle Universal Installer for the flash recovery area is a subdirectory of the Oracle home directory. You can choose this path only if you are using an Oracle Database base directory that is on a shared file system. This default location is not recommended for production databases.
oracle user must have write permissions to create the files in the path that you specify.
Note:You must perform this procedure only if you want to place the Oracle CRS, database, or recovery files on a file system other than where the corresponding installations were created.
To create directories for the Oracle CRS, database, or recovery files on separate file systems from the installations:
If necessary, configure the disks that you want to use and mount them on each node.
show device command to determine the free disk space on each mounted file system.
From the display, identify the file systems that you want to use. Table 6-1 lists the disk space requirements for each file type.
Table 6-1 File Types and Their Disk Space Requirements
|File Type||File System Requirements|
|CRS files||Choose a file system with at least 120 MB of free disk space|
|Database files||Choose either:
|Recovery files||Choose a file system with at least 2 GB of free disk space.|
If you are using the same disk for more than one type of file, then add the disk space requirements for each type to determine the total disk space requirement.
Note the names of the root device directories for the disks that you identified.
Enter commands similar to the following to create the recommended subdirectories in each of the root device directories:
CRS file directory
$ create/dir/own=oinstall root_device:[oracrs]
Database file directory
$ create/dir/own=oinstall root_device:[oradata]
Recovery file directory (flash recovery area)
$ create/dir/own=oinstall root_device:[flash_recovery_area]
Creating the database and its instances
Setting up network configuration for the database, its instances, and database services
Oracle recommends that you use DBCA to create a RAC database, because DBCA preconfigured databases optimize the environment for features such as the server parameter file and automatic undo management.
By using DBCA, you can create site-specific tablespaces as a part of database creation. If you have data file requirements that differ from those offered by DBCA templates, then create the database with DBCA and modify the data files later. You can also run user-specified scripts athes part of the database creation process.
DBCA also configures the RAC environment for various Oracle Database high-availability features such as services and cluster administration tools. It also starts any database instances required to support the defined configuration.
When you configure high-availability services with DBCA Database Services screen, you can also configure service instance preferences and transparent application failover (TAF) policies. These are described in the following subsections:
Use the DBCA Database Services screen to configure TAF failover policies. The DBCA Database Services screen also has a TAF policy selector row under the instance preference display. Make a selection in this row for your failover and reconnection policy preference as described in the following list:
To create a database by using DBCA in standalone mode, you must run the Oracle Net Configuration Assistant to configure the Oracle Net
listener.ora file. To start DBCA, connect to one of the nodes with Oracle Real Application Clusters (RAC) installed and then enter the
dbca command from the command line.
If DBCA does not display the Welcome screen for RAC, then DBCA was not able to detect whether or not the Oracle home is cluster installed. In this case, verify that
oraInventory is not damaged and that Oracle Universal Installer inventory,
oraInst.loc, is present. By default, this file is placed in a directory that has the same name as the node from where the installation was performed, which, in most cases, is the primary node of the RAC cluster. This directory is in the login directory of the user who carried out the installation. Otherwise, perform the following steps to create a RAC database:
Select Real Application Clusters database, click Next, and DBCA displays the Operations screen. DBCA enables the Configure Database Options, Delete a database, Instance Management, and Services Management options only if there is at least one RAC database configured on the cluster that runs from the Oracle Database home.
DBCA highlights the local node by default. Select the other nodes that you want to configure as members of the cluster database, click Next, and DBCA displays the Database Templates screen. If nodes that are part of the cluster installation are not displayed on the Node Selection screen, then perform inventory diagnostics and CRS diagnostics by running the
The templates on the Database Templates screen are Custom Database, Transaction Processing, Data Warehouse, and General Purpose. The Custom Database template does not include data files or options specially configured for a particular type of application. Use one of the other templates, which include data files, if you want to create a database with specifically configured options. Select the template from which you wish to create the cluster database, click Next, and DBCA displays the Database Identification screen.
Note:The global database name can be up to 30 characters in length and must begin with an alphabetic character. The
On the Management Options screen, you can choose to manage the database with Enterprise Manager. Despite choosing the Grid Control option, this will only configure EMAgent on OpenVMS. The Console component of Grid Control is not supported on OpenVMS platform. Refer to Appendix E, "Installing, Configuring, and Running EMAgent" for more information about how to configure the EMagent for use with the Grid Control Enterprise Manager, and how to manage and maintain the EMAgent for the OpenVMS platform.
Click Next, and DBCA displays the Database Credentials screen.
Enter the passwords for the database on the Database Credentials screen. You can enter the same or different passwords for the users
SYSMAN if you selected Enterprise Manager on the Management Options screen. Select the Use the Same Password for All Accounts option to assign the same password to the listed users. Alternatively, provide a different password for each of these users by selecting the Use Different Passwords option. Enter the password information, click Next, and DBCA displays the Storage Options screen.
Use the Storage Options screen to select a storage type for database creation. The Cluster File System option is the default. Select a storage option and click Next to proceed to the next screen. If you select Cluster File System, the Database File Locations screen is displayed.
The Database File Locations screen enables you to select the file storage for the database files: locations provided in a template, a common location for all database files (the files will not be Oracle Database-managed files), or Oracle Database-managed files in a common location. If you do not select the template option, then you can enter an existing directory path name in the space provided, or click Browse to open a selection list.
If you wish to multiplex the database redo log files and control files, click Multiplex Redo Logs and Control Files and provide the location for each copy you want. Click Ok when you have defined the multiplex locations to return to the Database File Locations screen.
You can also define variables for the file locations if you plan to use the Database Storage screen, explained in Step 14, to define individual file locations.
On the Recovery Configuration screen, you can select redo log archiving by selecting Enable Archiving. You can also select the flash recovery area and size on the Recovery Configuration screen. The flash recovery area defaults to
You can also define variables for the file locations if you plan to use the Database Storage screen, explained in Step 14, to define individual file locations. When you have completed the entries, click Next, and the Database Content screen is displayed.
On the Database Content screen, if you chose the Custom Database option, you can select or deselect the database components and their corresponding tablespace assignment. For a seed database, you can select whether to include the sample schemas in the database and to run custom scripts as part of the database creation processing. After completing the selections, click Next to display the Databases Services screen.
To create services on the Database Services screen, expand the Services tree. Oracle Database displays the global database name in the top left corner of the screen. Select the global database name and click Add to add services to the database. Enter a service name in the Add a Service dialog box, and click OK to add the service and return to the Database Services screen.
The service name is displayed under the global database name. Select the service name and DBCA displays the service preferences for the service on the right side of the DBCA Database Services screen. Change the instance preference (Not Used, Preferred, or Available) and transparent application failover (TAF) policies for the service as needed.
Repeat this procedure for each service and when you are done configuring services for the database, click Next. DBCA displays the Initialization Parameters screen.
By default, the Initialization Parameters screen shows only the basic parameters and only enables you to change the parameter file definition if you are using raw storage. Each tab on the Initialization Parameters screen provides different sets of information that you can add or modify as follows:
Click Typical for default values based on the database type you selected or Custom to set values for the memory parameters. You can also see values for the advanced parameters by clicking All Initialization Parameters.
Carefully review the parameter settings displayed in this dialog box because DBCA configures these settings in the server parameter file. Instance-specific parameter settings for the RAC database are displayed at the bottom of this dialog box. The
sid prefixes for these entries are displayed in the left column.
To review the instance-specific parameter settings, scroll downward using the scroll bar on the right side of the dialog box. Use the check box in the Override Default column to indicate that DBCA should place the parameter setting in the server parameter file. DBCA only places a parameter entry into the server parameter file if the entry displays a check mark in the Override Default column of the All Initialization Parameters dialog box.
Use this screen to select the database standard block size and process count.
Character Sets Tab
Use this screen to set the database character set value.
Connection Mode Tab
You can use this tab to select either dedicated or shared database connections to the database.
When you have specified all the information on the Initialization Parameters screen, click Next, and the Database Storage screen is displayed.
If you selected a preconfigured database template, such as the General Purpose template, then DBCA displays the control files, data files, and redo logs on the Database Storage screen. Select the folder and the file name underneath the folder to edit the file name. However, if you selected the Custom Database template, the template without data files, then DBCA displays the control files, tablespaces, data files, and redo logs. To change the tablespace properties, such as the data file or the tablespace size, click the tablespaces icon to expand the object tree on the left side of the screen and click the tablespace. The tablespace property dialog box is displayed on the right side. Make the changes, and click Ok.
After you specify all the information on the Database Storage screen, click Next, and DBCA displays the Creation Options screen.
Save as a Database Template: Creates a template that records the database structure, including user-supplied inputs, initialization parameters, and so on. You can later use this template to create a database.
After you complete Step 16, DBCA performs the following steps:
Creates an operative RAC database and its instances
Creates the RAC data dictionary views
Configures the network for the cluster database
Starts the listeners and database instances and then starts the high-availability services
This section explains how to delete a RAC database by using DBCA. This process deletes a database and removes the initialization parameter files, instances, and Oracle net configuration of the database.
Start DBCA on one of the nodes by running the
dbca command at the DCL command prompt.
Select Oracle Real Application Clusters and click Next.
If your user ID and password are not authenticated by the operating system, then the List of Cluster Databases screen displays the user name and password fields. If these fields are displayed, then enter a user ID and password that has SYSDBA privileges.
Select the database to delete and click Finish.
Click Ok to begin the deletion of the database and its associated files, services, and environment settings, or click Cancel to stop the operation.
When you click Ok, DBCA continues the operation and deletes all of the associated instances for this database. The DBCA also removes the parameter files, password files, and oratab entries.
At this point, you have accomplished the following:
The following sections describe how to use the server parameter file (
spfile) in Oracle Database 10g Real Application Clusters (RAC) as well as how to configure RAC once it has been installed.
See Also:The Oracle Real Application Clusters Administration and Deployment Guide (B14197-01) for more information about parameters and the Oracle Real Application Clusters Deployment and Performance Guide (B10768-02) for a discussion of parallel execution-related parameters in RAC data warehouse environments
Oracle Database uses parameter settings in parameter files to determine how to control various database resources. You can use two types of files for parameter administration: the server parameter file (
spfile) or one or more traditional client-side parameter files.
Oracle recommends that you administer parameters using
spfile. If you use client-side parameter files, then Oracle Database does not preserve parameter changes made for self-tuning after shutdown.
See Also:The Oracle Database 10g Real Application Clusters documentation for more information about using client-side parameter files
By default, Oracle Database creates the server parameter file based on one
spfile. You can change parameter settings in the server parameter file only by using Oracle Enterprise Manager or
ALTER SYSTEM SET SQL statements; the server parameter file is a binary file that you should not edit.
Note:Oracle recommends that you avoid modifying the values for self-tuning parameters. Overriding these settings can adversely affect performance.
If you are upgrading from a previous Oracle Database release, then create and configure the server parameter file for RAC using the procedures described in the following section.
Oracle recommends that you use a
PFILE in this directory:
To use DBCA to create the database and to use the server parameter file, on the Initialization Parameters screen select Create server parameter file (
spfile) under the File Locations tab, which is visible only if you are using raw storage. Then enter a shared file system file name in the
Server Parameters Filename field.
Note:When you use DBCA to create the server parameter file, the default
For single-node cluster-enabled configurations, or if you are using a cluster file system, place the server parameter file on a file system.
Upgrade to the server parameter file by completing the following procedures:
Combine the initialization parameter files for all instances into one
.ora file by copying all shared
IFILE contents as is. All parameters defined in the
IFILE parameter files are global. Therefore, create them as
Copy all instance-specific parameter definitions from
.ora files using the following syntax where
sid is the
sid of the instance:
Create the server parameter file using the
CREATE SPFILE statement as in the following example:
CREATE SPFILE='ORA_DB:spfile_dbname.ora' FROM PFILE='ora_root:[dbs]initdbname.ora'
Oracle recommends that you use the server parameter file by running the
STARTUP command as in this example:
If you use this
STARTUP command syntax, then Oracle Database uses the server parameter file entry specified in
Oracle Database reports errors that occur during server parameter file creation or while reading the file during startup. If an error occurs during a parameter update, then Oracle Database records the error in the
ALERT.LOG file and ignores subsequent parameter updates to the file. If this happens, then do either of the following:
Shut down the instance, recover the server parameter file, and restart the instance.
Enable the instance to continue running without regard for subsequent parameter updates.
Oracle Database displays errors for parameter changes that you attempt when you incorrectly use the
ALTER SYSTEM SET statement. Oracle Database does this when an error occurs while reading from or writing to the server parameter file.
See Also:Oracle Real Application Clusters Administration and Deployment Guide (B14197-01) for more information about backing up the server parameter file
The following sections describe the RAC installed configuration:
The Oracle Net Configuration Assistant (NetCA) and DBCA configure the environment to meet the requirements for database creation and Enterprise Manager discovery of Real Application Cluster databases.
Note:Configuration files are created on each node in the cluster database.
The DBCA uses the Oracle Cluster Registry (OCR) for storing the configurations for the cluster databases that it creates. Oracle Universal Installer automatically initializes the OCR during the CRS installation.
Oracle Database creates an entry for each RAC database in the
oratab configuration file. Oracle Enterprise Manager uses this file during service discovery to determine the name of the RAC database as well as whether or not the database should be automatically started on restart. The database entry has the following syntax:
db_unique_name is the database name for the RAC database,
ORACLE_HOME is the UNIX-style directory path to the database, and
N indicates that the database should not be started at startup time. A sample entry for a database named
For both single-instance and cluster database environments, Oracle Database is divided into smaller logical areas of space known as tablespaces. Each tablespace corresponds to one or more data files stored on a disk. Table 6-2 shows the tablespace names used by a RAC database and the types of data they contain:
Table 6-2 Tablespace Names Used by RAC Databases
||Consists of the data dictionary, including definitions of tables, views, and stored procedures needed by the database. Oracle Database automatically maintains information in this tablespace.|
||An auxiliary system tablespace that contains the
||Consists of application data. As you create and enter data into tables, Oracle Database fills this space with your data.|
||Contains temporary tables and indexes created during SQL statement processing. You may need to expand this tablespace if you are running a SQL statement that involves significant sorting, such as
||These are the undo tablespaces for each instance that DBCA creates for automatic undo management.|
||If you do not use automatic undo management, then Oracle Database uses the
You cannot alter these tablespace names when using the preconfigured database configuration options from Oracle Universal Installer. However, you can change the names of the tablespaces if you use the advanced database creation method.
As mentioned, each tablespace has one or more data files. The data file names created by the preconfigured database configuration options vary by operating system. Microsoft Windows-based platforms use the symbolic link names for the data file and other database files shown in Table 6-3.
Table 6-3 Microsoft Windows-Based Platforms Symbolic Link Names and Files
|Microsoft Windows-Based Platforms Symbolic Link Names||Tablespace or Other Database Files|
||Control File 1|
||Control File 2|
|Redo Log Files
Each instance must have at least two redo log files. If the database name is
The link names of the redo log files of the second instance are:
Each instance is configured with at least two redo log files that are stored in the shared storage. If you selected cluster file system, then these files are shared file system files. The file names of the redo log files that are created with the preconfigured database configuration options vary by storage type.
Oracle Database stores rollback or undo information in undo tablespaces. To manage undo tablespaces, Oracle recommends that you use automatic undo management. Automatic undo management is an automated undo tablespace management mode that is easier to administer than manual undo management.
See Also:Oracle Real Application Clusters Administration and Deployment Guide (B14197-01) for more information about managing undo tablespaces
Two key benefits of RAC are connection load balancing and failover. RAC extends the ability of single-instance Oracle Database load balancing, where connections are distributed among local dispatchers, to the balancing of connections among all instances in a cluster database. In addition, RAC provides failover by configuring multiple listeners on multiple nodes to manage client connection requests for the same database service. Connection load balancing and failover increase availability by taking advantage of the redundant resources within a cluster database. These features, however, require cross-instance registration.
Cross-instance registration in RAC occurs when the
PMON process of an instance registers with the local listener and with all other listeners. Therefore, all instances in the cluster database register with all listeners running on nodes that run instances of the cluster database. This enables all listeners to manage connections across all instances for both load balancing and failover.
Cross-instance registration requires configuring the
REMOTE_LISTENER initialization parameters. The
LOCAL_LISTENER parameter identifies the local listener, and the
REMOTE_LISTENER parameter identifies the global list of listeners. The
REMOTE_LISTENER parameter is dynamic. Oracle Database changes the setting for
REMOTE_LISTENER dynamically when you reconfigure the cluster database, for example, when you add or delete instances.
By default, DBCA configures the environment with dedicated servers. However, if you select the Shared server option on DBCA, then Oracle Database configures the shared server. In this case, Oracle Database uses both dedicated and shared server processing. When shared servers are configured, the
DISPATCHERS parameter is specified as in the following example:
DISPATCHERS initialization parameter does not specify the
LISTENER attribute as in the previous example, then the
PMON process registers information for all dispatchers with the listeners specified by the
However, when the
LISTENER attribute is specified, the
PMON process registers dispatcher information with the listeners specified by the
LISTENER attribute. In this case, setting the
LISTENER attribute overrides
REMOTE_LISTENER settings for the specified dispatchers as in the following example:
See Also:Oracle Database Net Services Administrator's Guide (B10775-01) for more information about cross-instance registration, shared and dedicated server configurations, and connection load balancing
If you configured dedicated server mode by using the DBCA Connection Mode tab on the Initialization Parameters screen, then DBCA automatically configures the
LOCAL_LISTENER parameter when the listener uses a nondefault address port.
If you configured the dedicated server mode by setting the
REMOTE_LISTENER initialization parameter, then you must also configure the instance-specific
LOCAL_LISTENER initialization parameter.
For example, to configure the
LOCAL_LISTENER parameter, add the following entry to the initialization parameter file, where
sid is resolved to a listener address through either a
tnsnames.ora file or through the Oracle Names Server:
The following entry should be in the
If DBCA detects more than one listener on the node, it displays a list of the listeners. You can select one or all of these listeners with which to register the database.
Services coordinate their sessions using listener file entries by running a process on the server that receives connection requests on behalf of a client application. Listeners are configured to respond to connection requests sent to protocol addresses for a database service or nondatabase service.
Protocol addresses are configured in the listener configuration file,
listener.ora, for a database service or a nondatabase service. Clients configured with the same addresses can connect to a service through the listener.
During a preconfigured database configuration installation, the Oracle Net Configuration Assistant creates and starts a default listener called
NODENAME. The listener is configured with a default protocol listening addresses for the database and external procedures. The advanced installation process prompts you to create at least one listener with the Oracle Net Configuration Assistant. The listener is configured to respond to connection requests that are directed at one protocol address that you specify, as well as an address for external procedures.
Both installation modes configure service information about the RAC database and external procedures. A database service automatically registers its information with the listener, such as its service name, instance names, and load information. This feature, called service registration, does not require configuration in the
listener.ora file. After listener creation, the Oracle Net Configuration Assistant starts the listener. A sample
listener.ora file with an entry for an instance named
listener_node1= (description= (address=(protocol=ipc)(key=extproc)) (address=(protocol=tcp)(host=node1-vip)(port=1521)) (address=(protocol=tcp)(host=node1-ip)(port=1521))) sid_list_listener_node1= (sid_list= (sid_desc= (sid_name=plsextproc) (oracle_home=
When a listener starts after the Oracle Database instance starts, and the listener is listed for service registration, registration does not occur until the next time the
PMON discovery routine runs. By default,
PMON discovery occurs every 60 seconds.
To override the 60-second delay, use the SQL
ALTER SYSTEM REGISTER statement. This statement forces
PMON to register the service immediately.
Oracle recommends that you create a script to process this statement immediately after starting the listener. If you run this statement while the listener is up and the instance is already registered, or while the listener is down, then the statement has no effect.
See Also:Oracle Database Net Services Administrator's Guide (B10775-01) for more information about the listener and the
If you configure access to a Lightweight Directory Access Protocol (LDAP)-compliant directory server with the Oracle Net Configuration Assistant during a Custom installation, then an
ldap.ora file is created. The
ldap.ora file contains the following types of information:
Type of directory
Location of the directory
Administrative context from which this server can look up, create, and modify a net service name, and the database service entries
See Also:Oracle Database Net Services Administrator's Guide (B10775-01) for more information about directory naming configuration and directory server access configuration
The network route to the service, including the location of the listener through a protocol address
SERVICE_NAME for an Oracle Database release 8.1 or later, or
SID for pre-8.1 Oracle Database releases.
Clients that connect to any instance of the database use the net service name entry for the database. This entry also enables Oracle Enterprise Manager to discover a RAC database.
A listener address is configured for each node that runs an instance of the database. The
LOAD_BALANCE option causes Oracle Database to choose the address randomly. If the chosen address fails, then the
FAILOVER option causes the connection request to fail over to the next address. Therefore, if an instance fails, then clients can still connect using another instance.
In the following example,
db.us.oracle.com is used by the client to connect to the target database,
db.us.acme.com= (description= (load_balance=on)
(address=(protocol=tcp)(host=node2-vip)(port=1521) (connect_data= (service_name=db.us.acme.com)))
When you set
DB_UNIQUE_NAME by entering a global database name that is longer than eight characters, excluding
DB_DOMAIN, then a net service entry similar to the following is created:
mydatabase.us.acme.com= (description = (address = (protocol = tcp)(host = node1-vip)(port = 1521)) (address = (protocol = tcp)(host = node2-vip)(port = 1521)) (load_balance = yes) (connect_data = (server = dedicated) (service_name = mydatabase.us.acme.com) ) )
Clients that connect to a particular instance of the database use the net service name entry for the instance. This entry, for example, enables Oracle Enterprise Manager to discover the instances in the cluster. These entries are also used to start and stop instances.
In the following example,
db1.us.acme.com, is used by Oracle Enterprise Manager to connect to an instance named
db1.us.acme.com= (description= (address=(protocol=tcp)(host=node1-vip)(port=1521)) (connect_data= (service_name=db.us.acme.com) (instance_name=db1)))
As discussed in Section 6.17, "Configuring Service Registration-Related Parameters in RAC", the
REMOTE_LISTENER parameter identifies the global list of listeners and it is dynamic. Oracle Database changes the setting for
REMOTE_LISTENER when you reconfigure the cluster database.
Whether using shared servers or dedicated servers, the list of remote listeners is supplied using the
REMOTE_LISTENERS parameter, for example:
This enables the instance to register with remote listeners on the other nodes;
db_unique_name is resolved through a naming method such as a
In the following example,
listeners_db.us.acme.com is resolved to a list of listeners available on the nodes on which the cluster database has instances:
The instance uses this list to determine the addresses of the remote listeners with which to register its information.
As discussed in Section 6.18.1, "Local Listeners" and Section 6.18.2, "Multiple Listeners", the
LOCAL_LISTENER parameter is set in the
.ora file if a nondefault listener is configured. For example:
sid is resolved to a listener address through a naming method such as a
In the following example,
listener_db1.us.acme.com is resolved to the nondefault listener address:
When you configure high-availability services using the DBCA Services screen, then DBCA creates net service entries similar to the following. The three services in the following examples,
db_svc3, have TAF policies of
db_svc1.us.acme.com= (description = (address=(protocol=tcp)(host=node1-vip)(port=1521)) (address=(protocol=tcp)(host=node2-vip)(port=1521)) (load_balance=yes) (connect_data= (server = dedicated) (service_name = db_svc1.us.acme.com) ) ) db_svc2.us.acme.com= (description= (address=(protocol=tcp)(host=node1-vip)(port=1521)) (address=(protocol=tcp)(host=node2-vip)(port=1521)) (load_balance=yes) (connect_data = (server = dedicated) (service_name=db_svc2.us.acme.com) (failover_mode = (type=select) (method=basic) (retries=180) (delay=5) ) ) ) db_svc3.us.acme.com= (description= (address=(protocol=tcp)(host=node1-vip)(port=1521)) (address=(protocol=tcp)(host=node2-vip)(port=1521)) (load_balance=yes) (connect_data= (server=dedicated) (service_name=db_svc3.us.acme.com) (failover_mode= (backup=db_svc3_preconnect.us.acme.com) (type=select) (method=preconnect) (retries=180) (delay=5) ) ) )
When a service has a TAF policy of
PRECONNECT, then a
preconnect net service entry is also created as in the following example:
db_svc3_preconnect.us.acme.com = (description = (address = (protocol = tcp)(host = node1-vip)(port = 1521)) (address = (protocol = tcp)(host = node2-vip)(port = 1521)) (load_balance = yes) (connect_data = (server = dedicated) (service_name = db_svc3_preconnect.us.amce.com) (failover_mode = (backup = db_svc3.us.acme.com) (type = select) (method = basic) (retries = 180) (delay = 5) ) ) )
An entry for connections to external procedures. This enables an Oracle Database to connect to external procedures.
extproc_connection_data.us.acme.com= (description= (address_list= (address=(protocol=ipc)(key=extproc0)) (connect_data= (sid=plsextproc)))
Example 6-1 Example tnsnames.ora File
The following is a sample
tnsnames.ora file that is created during a preconfigured database configuration installation:
db.us.acme.com= (description= (load_balance=on) (address=(protocol=tcp)(host=node1-vip)(port=1521)) (address=(protocol=tcp)(host=node2-vip)(port=1521)) (connect_data= (service_name=db.us.acme.com))) db1.us.acme.com= (description= (address=(protocol=tcp)(host=node1-vip)(port=1521)) (connect_data= (service_name=db.us.acme.com) (instance_name=db1))) db2.us.acme.com= (description= (address=(protocol=tcp)(host=node2-vip)(port=1521)) (connect_data= (service_name=db.us.acme.com) (instance_name=db2))) listeners_db.us.acme.com= (address_list= (address=(protocol=tcp)(host=node1-vip)(port=1521)) (address=(protocol=tcp)(host=node2-vip)(port=1521))) extproc_connection_data.us.acme.com= (description= (address_list= (address=(protocol=ipc)(key=extproc))) (connect_data= (sid=plsextproc) (presentation=RO)))
See Also:Oracle Database Net Services Administrator's Guide (B10775-01) for more information about the
The computer domain
This domain is automatically appended to any unqualified net service name. For example, if the default domain is set to
us.acme.com, then Oracle Database resolves
db in the connect string
CONNECT scott/tiger@db as:
A naming method the server uses to resolve a name to a connect descriptor
The order of naming methods is as follows: directory naming (for Custom installation only),
tnsnames.ora file, Oracle Names server, and host naming.
names.default_domain=us.acme.com names.directory_path=(tnsnames, onames,hostname)
See Also:Oracle Database Net Services Administrator's Guide (B10775-01) for more information about the