|Oracle® Real Application Clusters Administrator's Guide
10g Release 1 (10.1)
Part Number B10765-01
This chapter describes how to administer Oracle Real Application Clusters (RAC) database instances and RAC databases. This chapter explains the startup and shutdown tasks for database components and well as how to administer parameters and parameter files in RAC. The topics in this chapter are:
See Also:Chapter 3, " Administering Storage " for information about managing Automatic Storage Management (ASM) instances
RAC databases comprise a control file, redo logs, datafiles, and one server parameter file (SPFILE) or one or more PFILEs, or client-side parameter files. The instances comprise the System Global Area (SGA) and the instance background processes.
The following section introduces the three tools you will most likely to use to manage an existing RAC database and its instances, Oracle Enterprise Manager, SQL*Plus, and the SRVCTL utility. In many cases, you use these tools the same way to manage a single-instance database but the following sections identify the important differences when managing a RAC database:
Use the Web-based Enterprise Manager Database Control to manage a single RAC database. The Enterprise Manager Console provides a central point of control for the Oracle environment through a graphical user interface (GUI). You can use the Enterprise Manager Console to initiate a variety of cluster database management tasks. Use Enterprise Manager Grid Control to administer multiple RAC databases.
Enterprise Manager enables you to start, stop, and monitor databases, cluster database instances, and their listeners, as well as to schedule jobs or register events. You can perform these tasks simultaneously on multiple cluster databases. You can also use the Console to manage schemas, security, and cluster database storage features.
SQL*Plus commands operate on the current instance. The current instance can be either the local default instance on which you initiated your SQL*Plus session, or it can be a remote instance to which you connect with Net Services. Because the SQL*Plus prompt does not display the current instance, you should direct your commands to the correct instance. Starting a SQL*Plus session and connecting to the database without specifying an instance directs all SQL*Plus commands to the local instance. In this case, the default instance is also the current instance.
To connect to a different instance in SQL*Plus, issue a new CONNECT command specify a remote instance net service name as in the following example:
CONNECT <user name>/<password>@net_service_name
SYSDBA enables you to perform privileged operations, such as instance startup and shutdown. Multiple SQL*Plus sessions can connect to the same instance at the same time. SQL*Plus automatically disconnects you from the first instance whenever you connect to another one.
The SRVCTL tool manages configuration information that is used by several other Oracle tools. For example, Enterprise Manager uses the configuration information that SRVCTL generates to discover and monitor nodes in your cluster.
When you use SRVCTL to perform configuration operations on your cluster, SRVCTL stores configuration data in the Server Management (SRVM) configuration repository. SRVCTL performs other operations, such as starting and stopping instances, by calling SQL*Plus on each node. SRVCTL uses the same Oracle Cluster Registry (OCR) that is used with other Oracle administrative interfaces.
You can start up and shut down instances with Enterprise Manager, SQL*Plus or SRVCTL as described in the following sections. Both Enterprise Manager and SRVCTL provide options to startup and shutdown all of the instances in a RAC database with a single step.
You can only perform certain operations when the database is in a NOMOUNT or MOUNT state. Performing other operations requires that the database be OPEN. In addition, some operations require that only one instance be in the required state, while other operations require that all of the instances be in an identical state.
The procedures in this section assume that you are using a server parameter file (SPFILE) and are described in the following topics:
Before you can start a RAC instance your clusterware and any required operating system-specific processes. For more information about these processes, see your operating system documentation.
The procedure for shutting down RAC instances is identical to shutting down instances in single-instance Oracle, with the exceptions described here. Refer to the Oracle Database Administrator's Guide for more information about shutting down Oracle databases.
To shut down a RAC database mounted in shared mode, shut down every instance in the RAC environment.
IMMEDIATE shutdown, instance recovery is not required. Recovery is required, however, after you issue the
ABORT command or after an instance terminates abnormally. The instance that is still running performs instance recovery for the instance that shut down. If no other instances are running, the next instance to open the database performs instance recovery for any instances needing it.
TRANSACTIONAL command with the
LOCAL option is useful to shutdown an instance after all active transactions on the instance have either committed or rolled back. This is in addition to what this command does for
IMMEDIATE. Transactions on other instances do not block this operation. If you omit the
LOCAL option, then this operation waits until transactions on all other instances that started before the shutdown was issued either commit or rollback.
To access a cluster database instance, from the Home page, click the Targets tab, then click the cluster database name. On the Cluster Database Home page, the cluster database instances display at the bottom of the page. Click an instance name to go to the Cluster Database Instance Home page where you can start or stop the cluster database instance as well as see an overview of the cluster database instance activity such as CPU and space usage, active sessions, and so on.
To start a cluster database instance click Startup, or click Shutdown to stop it. To start or shutdown a cluster database, that is, all of the instances known to Enterprise Manager, select the database and click Startup or Shutdown on the Cluster Database page.
If you want to start or stop just one instance and you are connected to your local node, you should first ensure that your current environment includes the SID for the local instance. Note that any subsequent commands in your session, whether inside or outside a SQL*Plus session, will be associated with that same SID.
To start or shutdown your local instance, initiate a SQL*Plus session and connect with the SYSDBA or SYSOPER privilege and then issue the required command. For example to start and mount an instance on your local node, execute the following commands within your SQL*Plus session:
CONNECT / AS SYSDBA STARTUP MOUNT
You can start multiple instances from a single SQL*Plus session on one node by way of Oracle Net Services. To achieve this, you must connect to each instance in turn by using a Net Services connection string, typically an instance-specific alias from your
Note:To ensure that you connect to the correct instance, you must use an alias in the connect string that is associated with just one instance. If you use an alias to a service or with multiple addresses, you may not be connected to your intended instance.
For example, you can use a SQL*Plus session on a local node to perform a transactional shutdown for two instances on remote nodes by connecting to each in turn using the instance's individual alias name. Assume the alias name for the first instance is
db1 and that the alias for the second instance is
db2. Connect to the first instance and shut it down as follows:
Then connect to and shutdown the second instance by entering the following from you SQL*Plus session:
CONNECT /@db2 AS SYSDBA SHUTDOWN TRANSACTIONAL
Other startup and shut down keywords, such as NOMOUNT, MOUNT, IMMEDIATE, and so on, are described in the SQL*Plus User's Guide and Reference
It is not possible to start up or shut down more than one instance at a time in SQL*Plus, so you cannot start or stop all of the instances for a cluster database with a single SQL*Plus command. You may wish to create a script that will connect to each instance in turn and start it up and shut it down. However, you will need to maintain this script manually if you add or drop instances.
Enter the following SRVCTL syntax from the command line, providing the required database name and instance name, or include multiple instance names to start more than one specific instance:
Note that this command will also start all enabled and non-running services that have the listed instances either as preferred or available instances.
To stop one or more instances, enter the following SRVCTL syntax from the command line:
This command will also stop the services related to the terminated instances on the nodes where the instances were running.
To start or stop your entire cluster database, that is, all of the instances and its enabled services, enter the following SRVCTL commands:
See Also:Appendix B, " Server Control (SRVCTL) Reference " for information about SRVCTL options and information about other administrative tasks that you can perform with SRVCTL
When you create the database, Oracle creates an SPFILE in the file location that you specify. This location can be an ASM disk group, cluster file system file, or a shared raw device. If you manually create your database, then Oracle recommends that you create an SPFILE from an initialization parameter file (PFILE).
All instances in the cluster database use the same SPFILE at startup. Because the SPFILE is a binary file, do not edit it. Instead, change SPFILE parameter settings using Enterprise Manager or
SYSTEM SQL statements.
RAC uses a traditional PFILE only if an SPFILE does not exist or if you specify
PFILE in your
STARTUP command. Oracle recommends that you use SPFILE file to simplify administration, maintain parameter setting consistency, and to guarantee parameter setting persistence across database shutdown and startup events. In addition, you can configure RMAN to back up your SPFILE.
You can alter SPFILE settings with Enterprise Manager or by using the
SET clause of the
SYSTEM statement. In addition, the
SYSTEM syntax enables you to override the effects of SPFILE settings that you make manually. However, if your SPFILE contains instance-specific settings, then these settings take precedence over settings made with
The examples in this section appear in ASCII text although the SPFILE is a binary file. Assume that you start an instance with an SPFILE containing the following entries:
For the instance with the Oracle system identifier (sid)
OPEN_CURSORS parameter remains set to
1000 even though it has a database-wide setting of
500. The instance-specific parameter setting in the parameter file for an instance prevents database-wide alterations of the setting. This gives you control over parameter settings for instance prod1. These two types of settings can appear in any order in the parameter file.
If another DBA runs the following statement, then Oracle updates the setting on all instances except the instance with sid prod1:
ALTER SYSTEM SET OPEN_CURSORS=1500 sid='*' SCOPE=MEMORY;
In the example instance with sid prod1, the parameter begins accepting
SYSTEM values set by other instances if you change the parameter setting by running the following statement:
ALTER SYSTEM RESET OPEN_CURSORS SCOPE=MEMORY sid='prod1';
Then if you execute the following statement on another instance, the instance with sid prod1 also assumes the new setting of
ALTER SYSTEM SET OPEN_CURSORS=2000 sid='*' SCOPE=MEMORY;
In the following example, the server parameter file contains these entries:
Running the following statement makes Oracle disregard the first entry from the server parameter file:
ALTER SYSTEM RESET OPEN_CURSORS SCOPE=SPFILE sid='prod1';
To reset a parameter to its default value throughout your cluster database, enter the statement:
ALTER SYSTEM RESET OPEN_CURSORS SCOPE=SPFILE sid='*';
Note:Modifying SPIFLEs with anything except Enterprise Manager or SQL*Plus can corrupt the file and prevent database startup. To repair the file, you might need to create the PFILE and regenerate the SPFILE.
# first comment OPEN_CURSORS = value # second comment
comment is associated with the setting for
OPEN_CURSORS. Oracle displays this comment in the
V$PARAMETER2 views. Oracle also displays comments such as the entry
comment in the example.
You can revert to previous releases of RAC and convert from using the server parameter file to the traditional client-side PFILE. The procedure for this is described in Oracle Database Administrator's Guide.
Most initialization parameters can have different values on different instances as described in the Oracle Database Reference. A parameter that can optionally have a different value for each instance has a default value that is typically the same on all instances. You can change the value on one or more instances and your new values can be unique across your instances or the values can be the same on one or more of your instances. Other parameters must either be identical or unique as described in the following sections.
Certain initialization parameters that are critical at database creation or that affect certain database operations must have the same value for every instance in RAC. Specify these parameter values in the SPFILE, or within each
init_dbname.ora file on each instance. The following list contains the parameters must be identical on every instance:
The setting for
DML_LOCKS must be identical on every instance only if set to zero.
If you use the
ROLLBACK_SEGMENTS parameters, then Oracle recommends setting unique values for them by using the
sid identifier in the SPFILE. However, you must set a unique value for
INSTANCE_NUMBER for each instance and you cannot use a default value.
Oracle uses the
INSTANCE_NUMBER parameter to distinguish among instances at startup. Oracle uses the
THREAD number to assign redo log groups to specific instances. To simplify administration, use the same number for both the
Enables a database to be started in cluster mode. Set this parameter to
Sets the number of instances in your RAC environment. A proper setting for this parameter can improve memory use.
Specifies the cluster interconnect when there is more than one interconnect. Refer to your Oracle platform-specific documentation for the use of this parameter, its syntax, and its behavior.
You typically do not need to set the
CLUSTER_INTERCONNECTS parameter. For example, do not set this parameter for the following common configurations:
If you have only one cluster interconnect.
If the default cluster interconnect meets the bandwidth requirements of your RAC database, which is typically the case.
Oracle uses information from
CLUSTER_INTERCONNECTS to distribute interconnect traffic among the various network interfaces if you specify more than one interconnect with this parameter. Note that the specified configuration inherits any limitations of the listed interconnects and the associated operating system IPC services, such as availability. Consider setting
CLUSTER_INTERCONNECTS when a single cluster interconnect cannot meet your bandwidth requirements. You may need to set this parameter in data warehouse environments with high interconnect bandwidth demands from one or more databases as described here.
For example, if you have two databases with high interconnect bandwidth requirements, then you can override the default interconnect provided by your operating system and nominate a different interconnect for each database using the following syntax in each server parameter file where ip
n is an IP address in standard dot-decimal format, for example:
Database One: CLUSTER_INTERCONNECTS = ip1 Database Two: CLUSTER_INTERCONNECTS = ip2
If you have one database with high bandwidth demands, then you can nominate multiple interconnects using the following syntax:
CLUSTER_INTERCONNECTS = ip1:ip2:...:ipn
If you set multiple values for
CLUSTER_INTERCONNECTS as in the preceding example, then Oracle uses all of the interconnects that you specify. This provides load balancing as long as all of the listed interconnects remain operational.
If there is an operating system error writing to the interconnect that you specify with
CLUSTER_INTERCONNECTS, then Oracle returns an error even if some other interfaces are available. This is because the communication protocols between Oracle and the interconnect can vary greatly depending on your platform. Refer to your Oracle platform-specific documentation for more information.
If you set a value for
DB_NAME in instance-specific parameter files, the setting must be identical for all instances.
DISPATCHERS parameter to enable a shared server configuration, that is a server that is configured to allow many user processes to share very few server processes. With shared server configurations, many user processes connect to a dispatcher. The
DISPATCHERS parameter may contain many attributes.
Oracle recommends that you configure at least the
PROTOCOL specifies the network protocol for which the dispatcher process generates a listening end point.
LISTENER specifies an alias name for the Oracle Net Services listeners. Set the alias to a name that is resolved through a naming method such as a
tnsnames.ora file. The
tnsnames.ora file contains net service names. This file is needed on clients, nodes, the Enterprise Manager Central Control, and the Oracle Performance Manager node. Refer to Oracle Net Services Administrator's Guide for complete information about configuring the
DISPATCHER parameter and its attributes and for configuring the shared server.
This is a RAC-specific parameter. Do not alter the default setting for this parameter except under a limited set of circumstances. This parameter specifies the maximum amount of time allowed before the system change number (SCN) held in the SGA of an instance is refreshed by the log writer process (
LGWR). It determines whether the local
SCN should be refreshed from the
SGA when getting the snapshot
SCN for a query.
When you use an SPFILE, all RAC database instances must use the SPFILE and the file must be on shared storage.
Each instance maintains its own
SESSIONS_PER_USER count. If
SESSIONS_PER_USER is set to
1 for a user, the user can log on to the database more than once as long as each connection is from a different instance.
If specified, this parameter must have unique values on all instances. The
THREAD parameter specifies the number of the redo thread to be used by an instance. You can specify any available redo thread number as long as that thread number is enabled and is not used.
CREATE PFILE='?/dbs/initdbname.ora' FROM SPFILE='/dev/vx/rdsk/oracle_dg/dbspfile'
You can also recover by starting up an instance using a client-side initialization parameter file. Then re-create the server parameter file using the
SPFILE statement. You can also use RMAN (Recovery Manager) to create backups of the server parameter file.