Oracle8 SQL Reference
Release 8.0

A58225-01

Library

Product

Contents

Index

Prev Next

4
Commands

This chapter describes, in alphabetical order, Oracle SQL commands and clauses.


Note:

Descriptions of commands and clauses preceded by are available only if the Oracle objects option is installed on your database server.

 

The description of each command or clause contains the following sections:

Purpose  

describes the basic uses of the command.  

Prerequisites  

lists privileges you must have and steps that you must take before using the command. In addition to the prerequisites listed, most commands also require that the database be opened by your instance, unless otherwise noted.  

Syntax  

shows the keywords and parameters that make up the command.  

Keywords and Parameters  

describes the purpose of each keyword and parameter. The conventions for keywords and parameters used in this chapter are explained in the Preface of this reference.  

 

Usage notes: Optional sections following "Keywords and Parameters" provide examples and discuss how and when to use the command.  

Related Topics  

lists related commands, clauses, and sections of this and other manuals.  


Summary of SQL Commands

The tables in the following sections provide a functional summary of SQL commands and are divided into these categories:

Data Definition Language (DDL) Commands

Data definition language (DDL) commands enable you to perform these tasks:

The CREATE, ALTER, and DROP commands require exclusive access to the object being acted upon. For example, an ALTER TABLE command fails if another user has an open transaction on the specified table.

The GRANT, REVOKE, ANALYZE, AUDIT, and COMMENT commands do not required exclusive access to the object being acted upon. For example, you can analyze a table while other users are updating the table.

Oracle implicitly commits the current transaction before and after every DDL statement.

Many DDL statements may cause Oracle to recompile or reauthorize schema objects. For information on how Oracle recompiles and reauthorizes schema objects and the circumstances under which a DDL statement would cause this, see Oracle8 Concepts.

DDL commands are not directly supported by PL/SQL, but may be available using packaged procedures supplied by Oracle corporation. For more information, see PL/SQL User's Guide and Reference.

Table 4-1 lists the DDL commands.

Table 4-1 Data Definition Language Commands
Command   Purpose  

ALTER CLUSTER  

Change the storage characteristics of a cluster.

Allocate an extent for a cluster.  

ALTER DATABASE  

Open/mount the database.

Convert an Oracle7 data dictionary when migrating to Oracle8.

Prepare to downgrade to an earlier release of Oracle.

Choose ARCHIVELOG/NOARCHIVELOG mode.

Perform media recovery.

Add/drop/clear redo log file groups members.

Rename a datafile/redo log file member.

Back up the current control file.

Back up SQL commands (that can be used to re-create the database) to the trace file.

Create a new datafile.

Resize one or more datafiles.

Create a new datafile in place of an old one for recovery purposes.

Enable/disable autoextending the size of datafiles.

Take a datafile online/offline.

Enable/disable a thread of redo log file groups.

Change the database's global name.  

ALTER FUNCTION  

Recompile a stored function.  

ALTER INDEX  

Redefine an index's future storage allocation.  

ALTER PACKAGE  

Recompile a stored package.  

ALTER PROCEDURE  

Recompile a stored procedure.  

ALTER PROFILE  

Add or remove a resource limit to or from a profile.  

ALTER RESOURCE COST  

Specify a formula to calculate the total cost of resources used by a session.  

ALTER ROLE  

Change the authorization needed to access a role.  

ALTER ROLLBACK SEGMENT  

Change a rollback segment's storage characteristics.

Bring a rollback segment online/offline.

Shrink a rollback segment to an optimal or given size.  

ALTER SEQUENCE  

Redefine value generation for a sequence.  

ALTER SNAPSHOT  

Change a snapshot's storage characteristics, automatic refresh time, or automatic refresh mode.  

ALTER SHAPSHOT LOG  

Change a snapshot log's storage characteristics.  

ALTER TABLE  

Add a column/integrity constraint to a table.

Redefine a column, to change a table's storage characteristics.

Enable/disable/drop an integrity constraint.

Enable/disable table locks on a table.

Enable/disable all triggers on a table.

Allocate an extent for the table.

Allow/disallow writing to a table.

Modify the degree of parallelism for a table.  

ALTER TABLESPACE  

Add/rename datafiles.

Change storage characteristics.

Take a tablespace online/offline.

Begin/end a backup.

Allow/disallow writing to a tablespace.  

ALTER TRIGGER  

Enable/disable a database trigger.  

ALTER TYPE  

Change a user-defined type.  

ALTER USER  

Change a user's password, default tablespace, temporary tablespace, tablespace quotas, profile, or default roles.  

ALTER VIEW  

Recompile a view.  

ANALYZE  

Collect performance statistics, validate structure, or identify chained rows for a table, cluster, or index.  

AUDIT  

Choose auditing for specified SQL commands or operations on schema objects.  

COMMENT  

Add a comment about a table, view, snapshot, or column to the data dictionary.  

CREATE CLUSTER  

Create a cluster that can contain one or more tables.  

CREATE CONTROLFILE  

Re-create a control file.  

CREATE DATABASE  

Create a database.  

CREATE DATABASE LINK  

Create a link to a remote database.  

CREATE DIRECTORY  

Create a directory database object for administering access to and use of BFILEs stored outside the database.  

CREATE FUNCTION  

Create a stored function.  

CREATE INDEX  

Create an index for a table or cluster.  

CREATE LIBRARY  

Create a library from which SQL and PL/SQL can call external third-generation language (3GL) functions and procedures.  

CREATE PACKAGE  

Create the specification of a stored package.  

CREATE PACKAGE BODY  

Create the body of a stored package  

CREATE PROCEDURE  

Create a stored procedure.  

CREATE PROFILE  

Create a profile and specify its resource limits.  

CREATE ROLE  

Create a role.  

CREATE ROLLBACK SEGMENT  

Create a rollback segment.  

CREATE SCHEMA  

Issue multiple CREATE TABLE, CREATE VIEW, and GRANT statements in a single transaction.  

CREATE SEQUENCE  

Create a sequence for generating sequential values.  

CREATE SHAPSHOT  

Create a snapshot of data from one or more remote master tables.  

CREATE SNAPSHOT LOG  

Create a snapshot log containing changes made to the master table of a snapshot.  

CREATE SYNONYM  

Create a synonym for a schema object.  

CREATE TABLE  

Create a table, defining its columns, integrity constraints, and storage allocation.  

CREATE TABLESPACE  

Create a place in the database for storage of schema objects, rollback segments, and temporary segments, naming the datafiles that make up the tablespace.  

CREATE TRIGGER  

Create a database trigger.  

CREATE TYPE  

Create an object type, named varying array (VARRAY), nested table type, or an incomplete object type.  

CREATE USER  

Create a database user.  

CREATE VIEW  

Define a view of one or more tables or views.  

DROP CLUSTER  

Remove a cluster from the database.  

DROP DATABASE LINK  

Remove a database link.  

DROP DIRECTORY  

Remove a directory from the database.  

DROP FUNCTION  

Remove a stored function from the database.  

DROP INDEX  

Remove an index from the database.  

DROP LIBRARY  

Remove a library object from the database.  

DROP PACKAGE  

Remove a stored package from the database.  

DROP PROCEDURE  

Remove a stored procedure from the database.  

DROP PROFILE  

Remove a profile from the database.  

DROP ROLE  

Remove a role from the database.  

DROP ROLLBACK SEGMENT  

Remove a rollback segment from the database.  

DROP SEQUENCE  

Remove a sequence from the database.  

DROP SNAPSHOT  

Remove a snapshot from the database.  

DROP SNAPSHOT LOG  

Remove a snapshot log from the database.  

DROP SYNONYM  

Remove a synonym from the database.  

DROP TABLE  

Remove a table from the database.  

DROP TABLESPACE  

Remove a tablespace from the database.  

DROP TRIGGER  

Remove a trigger from the database.  

DROP TYPE  

Remove a user-defined type from the database.  

DROP USER  

Remove a user and the objects in the user's schema from the database.  

DROP VIEW  

Remove a view from the database.  

GRANT  

Grant system privileges, roles, and object privileges to users and roles.  

NOAUDIT  

Disable auditing by reversing, partially or completely, the effect of a prior AUDIT statement.  

RENAME  

Change the name of a schema object.  

REVOKE  

Revoke system privileges, roles, and object privileges from users and roles.  

TRUNCATE  

Remove all rows from a table or cluster and free the space that the rows used.  

Data Manipulation Language (DML) Commands

Data manipulation language (DML) commands query and manipulate data in existing schema objects. These commands do not implicitly commit the current transaction.

Table 4-2 Data Manipulation Language Commands
Command   Purpose  

DELETE  

Remove rows from a table.  

EXPLAIN PLAN  

Return the execution plan for a SQL statement.  

INSERT  

Add new rows to a table.  

LOCK TABLE  

Lock a table or view, limiting access to it by other users.  

SELECT  

Select data in rows and columns from one or more tables.  

UPDATE  

Change data in a table.  

All DML commands except the EXPLAIN PLAN command are supported in PL/SQL.

Transaction Control Commands

Transaction control commands manage changes made by DML commands.

Table 4-3 Transaction Control Commands
Command   Purpose  

COMMIT  

Make permanent the changes made by statements issued since the beginning of the current transaction.  

ROLLBACK  

Undo all changes since the beginning of a transaction or since a savepoint.  

SAVEPOINT  

Establish a point back to which you may roll.  

SET TRANSACTION  

Establish properties for the current transaction.  

All transaction control commands except certain forms of the COMMIT and ROLLBACK commands are supported in PL/SQL. For information on the restrictions, see COMMIT and ROLLBACK.

Session Control Commands

Session control commands dynamically manage the properties of a user session. These commands do not implicitly commit the current transaction.

PL/SQL does not support session control commands.

Table 4-4 Session Control Commands
Command   Purpose  

ALTER SESSION  

Enable/disable the SQL trace facility.

Enable/disable global name resolution.

Change the values of the session's NLS parameters.

In a parallel server, indicate that the session must access database files as if the session were connected to another instance.

Close a database link.

Send advice to remote databases for forcing an in-doubt distributed transaction.

Permit or prohibit procedures and stored procedures from issuing COMMIT and ROLLBACK statements.

Change the goal of the cost-based optimization approach.  

SET ROLE  

Enable/disable roles for the current session.  

System Control Command

The single system control command dynamically manages the properties of an Oracle instance. This command does not implicitly commit the current transaction.

ALTER SYSTEM is not supported in PL/SQL.

Table 4-5 System Control Command
Command   Purpose  

ALTER SYSTEM  

Alter the Oracle instance by performing a specialized function.  

Embedded SQL Commands

Embedded SQL commands place DDL, DML, and transaction control statements within a procedural language program. Embedded SQL is supported by the Oracle precompilers and is documented in the following books:


ALTER CLUSTER

Purpose

Redefines storage and parallelism characteristics of a cluster. See also "Altering Clusters".

Prerequisites

The cluster must be in your own schema or you must have ALTER ANY CLUSTER system privilege.

Syntax

deallocate_unused_clause: See the DEALLOCATE UNUSED clause.
parallel_clause: See the PARALLEL clause.

Keywords and Parameters

schema  

is the schema containing the cluster. If you omit schema, Oracle assumes the cluster is in your own schema.  

cluster  

is the name of the cluster to be altered.  

physical_attributes_clause  

changes the values of the PCTUSED, PCTFREE, INITRANS, and MAXTRANS parameters of the cluster. See CREATE CLUSTER.  

 

storage_clause  

changes the storage characteristics for the cluster. See the STORAGE clause.  

SIZE  

determines how many cluster keys will be stored in data blocks allocated to the cluster. You can change the SIZE parameter only for an indexed cluster, not for a hash cluster. For a description of the SIZE parameter, see CREATE CLUSTER.  

allocate_extent_clause  

explicitly allocates a new extent for the cluster.  

 

SIZE  

specifies the size of the extent in bytes. Use K or M to specify the extent size in kilobytes or megabytes. If you omit this parameter, Oracle determines the size based on the values of the cluster's STORAGE parameters.  

 

DATAFILE  

specifies one of the datafiles in the cluster's tablespace to contain the new extent. If you omit this parameter, Oracle chooses the datafile.  

 

INSTANCE  

makes the new extent available to the specified instance. An instance is identified by the value of its initialization parameter INSTANCE_NUMBER. If you omit this parameter, the extent is available to all instances. Use this parameter only if you are using Oracle with the Parallel Server option in parallel mode.

Explicitly allocating an extent with this clause does not cause Oracle to evaluate the cluster's storage parameters and determine a new size for the next extent to be allocated. You can allocate a new extent only for an indexed cluster, not a hash cluster.  

deallocate_unused_clause  

explicitly deallocates unused space at the end of the cluster and makes the freed space available for other segments. Only unused space above the high-water mark can be freed. If KEEP is omitted, all unused space is freed. For syntax and complete information, see the DEALLOCATE UNUSED clause.  

 

KEEP  

specifies the number of bytes above the high-water mark that the cluster will have after deallocation. If the number of remaining extents are less than MINEXTENTS, then MINEXTENTS is set to the current number of extents. If the initial extent becomes smaller than INITIAL, then INITIAL is set to the value of the current initial extent.  

parallel_clause  

specifies the degree of parallelism for creating the cluster and the default degree of parallelism for queries on the cluster once created. For syntax and complete information, see the PARALLEL clause.  

Altering Clusters

You can perform these tasks with the ALTER CLUSTER command:

You cannot perform these tasks with the ALTER CLUSTER command:

Example I

The following statement alters the CUSTOMER cluster in the schema SCOTT:

ALTER CLUSTER scott.customer 
    SIZE 512 
    STORAGE (MAXEXTENTS 25); 

Oracle allocates 512 bytes for each cluster key value. Assuming a data block size of 2 kilobytes, future data blocks within this cluster contain 4 cluster keys per data block, or 2 kilobytes divided by 512 bytes.

The cluster can have a maximum of 25 extents.

Example II

The following statement deallocates unused space from CUSTOMER cluster, keeping 30 kilobytes of unused space for future use:

ALTER CLUSTER scott.customer 
    DEALLOCATE UNUSED KEEP 30 K;

Related Topics

CREATE CLUSTER
CREATE TABLE
DEALLOCATE UNUSED clause
DROP CLUSTER
DROP TABLE
STORAGE clause
PARALLEL clause

ALTER DATABASE

Purpose

To alter an existing database in one of these ways:

For illustrations of some of these purposes, see "Examples".

Prerequisites

You must have ALTER DATABASE system privilege.

Syntax


logfile_descriptor::=

autoextend_clause::=

recover_clause: See the RECOVER clause.

Keywords and Parameters

database  

identifies the database to be altered. The database name can contain only ASCII characters. If you omit database, Oracle alters the database identified by the value of the initialization parameter DB_NAME. You can alter only the database whose control files are specified by the initialization parameter CONTROL_FILES. Note that the database identifier is not related to the Net8 database specification.  

You can use the following options only when the database is not mounted by your instance:  

MOUNT  

mounts the database.  

 

STANDBY DATABASE  

mounts the standby database. For more information, see the Oracle8 Administrator's Guide.  

 

CLONE DATABASE  

mounts the clone database. For more information, see the Oracle8 Backup and Recovery Guide.  

CONVERT  

completes the conversion of the Oracle7 data dictionary. After you use this option, the Oracle7 data dictionary no longer exists in the Oracle database. Use this option only when you are migrating to Oracle8. For more information on using this option, see Oracle8 Migration.  

OPEN  

opens the database, making it available for normal use. You must mount the database before you can open it. You cannot open a standby database that has not been activated.  

 

RESETLOGS  

resets the current log sequence number to 1 and discards any redo information that was not applied during recovery, ensuring that it will never be applied. This effectively discards all changes that are in the redo log, but not in the database. You must use this option to open the database after performing media recovery with an incomplete recovery using the RECOVER UNTIL clause (see RECOVER clause) or with a backup control file. After opening the database with this option, you should perform a complete database backup.  

 

NORESETLOGS  

leaves the log sequence number and redo log files in their current state.  

 

You can specify the above options only after performing incomplete media recovery or complete media recovery with a backup control file. In any other case, Oracle uses the NORESETLOGS automatically.  

ACTIVATE STANDBY DATABASE  

changes the state of a standby database to an active database. For more information, see Oracle8 Administrator's Guide..  

Use the following options only if your instance has the database mounted in parallel server disabled mode, but not open:  

ARCHIVELOG  

establishes ARCHIVELOG mode for redo log file groups. In this mode, the contents of a redo log file group must be archived before the group can be reused. This option prepares for the possibility of media recovery. You can use this option only after shutting down your instance normally or immediately with no errors and then restarting it, mounting the database in parallel server disabled mode.  

NOARCHIVELOG  

establishes NOARCHIVELOG mode for redo log files. In this mode, the contents of a redo log file group need not be archived so that the group can be reused. This mode does not prepare for recovery after media failure.  

You can use any of the following options when your instance has the database mounted, open or closed, and the files involved are not in use:  

recover_clause  

performs media recovery. For syntax and more information, see the RECOVER clause. You recover the entire database only when the database is closed. You can recover tablespaces or datafiles when the database is open or closed, provided the tablespaces or datafiles to be recovered are offline. You cannot perform media recovery if you are connected to Oracle through the multithreaded server architecture. You can also perform media recovery with the Server Manager recovery dialog box.  

ADD LOGFILE  

adds one or more redo log file groups to the specified thread, making them available to the instance assigned the thread.  

 

THREAD  

is required only if you are using Oracle with the Parallel Server option in parallel mode. If you omit the THREAD parameter, the redo log file group is added to the thread assigned to your instance.  

 

GROUP  

uniquely identifies the redo log file group among all groups in all threads and can range from 1 to the MAXLOGFILES value. You cannot add multiple redo log file groups having the same GROUP value. If you omit this parameter, Oracle generates its value automatically. You can examine the GROUP value for a redo log file group through the dynamic performance view V$LOG.  

 

filespec  

Each filespec specifies a redo log file group containing one or more members, or copies. See the syntax description of filespec in "Filespec".  

ADD LOGFILE MEMBER  

adds new members to existing redo log file groups. Each new member is specified by 'filename'. If the file already exists, it must be the same size as the other group members, and you must specify the REUSE option. If the file does not exist, Oracle creates a file of the correct size. You cannot add a member to a group if all of the group's members have been lost through media failure.  

 

You can specify an existing redo log file group in one of these ways:  

 

GROUP  

Specify the value of the GROUP parameter that identifies the redo log file group.  

 

list of filenames  

List all members of the redo log file group. You must fully specify each filename according to the conventions of your operating system.  

DROP LOGFILE  

drops all members of a redo log file group. You can specify a redo log file group in the same manner as the ADD LOGFILE MEMBER clause. You cannot drop a redo log file group if it needs archiving or is the currently active group; nor can you drop a redo log file group if doing so would cause the redo thread to contain less than two redo log file groups.  

DROP LOGFILE MEMBER  

drops one or more redo log file members. Each 'filename' must fully specify a member using the conventions for filenames on your operating system.  

 

You cannot use this clause to drop all members of a redo log file group that contains valid data. To perform this operation, use the DROP LOGFILE clause.  

CLEAR LOGFILE  

reinitializes an online redo log, optionally without archiving the redo log. CLEAR LOGFILE is similar to adding and dropping a redo log, except that the command may be issued even if there are only two logs for the thread and also may be issued for the current redo log of a closed thread.  

 

UNARCHIVED  

You must specify UNARCHIVED if you want to reuse a redo log that was not archived.  

 

 

WARNING: Specifying UNARCHIVED makes backups unusable if the redo log is needed for recovery.  

 

You cannot use CLEAR LOGFILE to clear a log needed for media recovery. If it is necessary to clear a log containing redo after the database checkpoint, you must first perform incomplete media recovery. The current redo log of an open thread can be cleared. The current log of a closed thread can be cleared by switching logs in the closed thread.  

 

If the CLEAR LOGFILE command is interrupted by a system or instance failure, then the database may hang. If so, this command must be reissued once the database is restarted. If the failure occurred because of I/O errors accessing one member of a log group, then that member can be dropped and other members added.  

 

UNRECOVERABLE DATAFILE  

You must specify UNRECOVERABLE DATAFILE if the database has a datafile that is offline (not for drop) and if the unarchived log to be cleared is needed to recover the datafile before bringing it back online. In this case, you must drop the datafile and the entire tablespace once the CLEAR LOGFILE command completes.  

RENAME FILE  

renames datafiles or redo log file members. This clause renames only files in the control file; it does not actually rename them on your operating system. You must specify each filename using the conventions for filenames on your operating system.  

CREATE STANDBY CONTROLFILE  

creates a control file to be used to maintain a standby database. For more information, see Oracle8 Administrator's Guide.  

BACKUP CONTROLFILE  

backs up the current control file.  

 

TO 'filename'  

specifies the file to which the control file is backed up. You must fully specify the filename using the conventions for your operating system. If the specified file already exists, you must specify the REUSE option.  

 

TO TRACE  

writes SQL statements to the database's trace file rather than making a physical backup of the control file. The SQL commands can be used to start up the database, re-create the control file, and recover and open the database appropriately, based on the created control file.  

 

 

You can copy the commands from the trace file into a script file, edit the commands as necessary, and use the database if all copies of the control file are lost (or to change the size of the control file).  

 

RESETLOGS  

specifies that the SQL statement written to the trace file for starting the database is ALTER DATABASE OPEN RESETLOGS.  

 

NORESETLOGS  

specifies that the SQL statement written to the trace file for starting the database is ALTER DATABASE OPEN NORESETLOGS.  

RENAME GLOBAL_NAME  

changes the global name of the database. The database is the new database name and can be as long as eight bytes. The optional domain specifies where the database is effectively located in the network hierarchy.

Note: Renaming your database does not change global references to your database from existing database links, synonyms, and stored procedures and functions on remote databases. Changing such references is the responsibility of the administrator of the remote databases.  

 

For more information on global names, see Oracle8 Distributed Database Systems.  

RESET COMPATIBILITY  

marks the database to be reset to an earlier version of Oracle when the database is next restarted.  

 

Note: RESET COMPATIBILITY works only if you have successfully disabled Oracle features that affect backward compatibility. For more information on downgrading to an earlier version of Oracle, see Oracle8 Migration.  

You can use the following options only when your instance has the database open:  

ENABLE THREAD  

in a parallel server, enables the specified thread of redo log file groups. The thread must have at least two redo log file groups before you can enable it.  

 

PUBLIC  

makes the enabled thread available to any instance that does not explicitly request a specific thread with the initialization parameter THREAD. If you omit the PUBLIC option, the thread is available only to the instance that explicitly requests it with the initialization parameter THREAD.  

DISABLE THREAD  

disables the specified thread, making it unavailable to all instances. You cannot disable a thread if an instance using it has the database mounted.  

You can use any of the following options when your instance has the database mounted, open or closed, and the files involved are not in use:  

CREATE DATAFILE  

creates a new empty datafile in place of an old one. You can use this option to re-create a datafile that was lost with no backup. The 'filename' must identify a file that is or was once part of the database. The filespec specifies the name and size of the new datafile. If you omit the AS clause, Oracle creates the new file with the name and size as the file specified by 'filename'.  

 

During recovery, all archived redo logs written to since the original datafile was created must be applied to the new, empty version of the lost datafile.  

 

Oracle creates the new file in the same state as the old file when it was created. You must perform media recovery on the new file to return it to the state of the old file at the time it was lost.  

 

You cannot create a new file based on the first datafile of the SYSTEM tablespace.  

DATAFILE  

affects your database files as follows:  

 

ONLINE  

brings the datafile online.  

 

OFFLINE  

takes the datafile offline. If the database is open, you must perform media recovery on the datafile before bringing it back online, because a checkpoint is not performed on the datafile before it is taken offline.  

 

 

DROP takes a datafile offline when the database is in NOARCHIVELOG mode.  

 

RESIZE  

attempts to change the size of the datafile to the specified absolute size in bytes. You can also use K or M to specify this size in kilobytes or megabytes. There is no default, so you must specify a size.  

 

autoextend_clause  

enables or disables the automatic extension of a datafile. If you do not specify this clause, datafiles are not automatically extended.  

 

 

OFF disables autoextend if it is turned on. NEXT and MAXSIZE are set to zero. Values for NEXT and MAXSIZE must be respecified in further ALTER DATABASE AUTOEXTEND commands.  

 

 

ON enables autoextend.  

 

 

NEXT specifies the size in bytes of the next increment of disk space to be automatically allocated to the datafile when more extents are required. You can also use K or M to specify this size in kilobytes or megabytes. The default is one data block.  

 

 

MAXSIZE specifies the maximum disk space allowed for automatic extension of the datafile.  

 

 

UNLIMITED sets no limit on allocating disk space to the datafile.  

 

END BACKUP  

avoids media recovery on database startup after an online tablespace backup was interrupted by a system failure or instance failure or SHUTDOWN ABORT.  

 

WARNING: Do not use ALTER TABLESPACE ... END BACKUP if you have restored any of the files affected from a backup. Media recovery is fully described in the Oracle8 Backup and Recovery Guide andOracle8 Administrator's Guide..  

Examples

For more information on using the ALTER DATABASE command for database maintenance, see the Oracle8 Administrator's Guide.

Example I

The following statement adds a redo log file group with two members and identifies it with a GROUP parameter value of 3:

ALTER DATABASE stocks
  ADD LOGFILE GROUP 3 
    ('diska:log3.log' ,  
     'diskb:log3.log') SIZE 50K; 
Example II

The following statement adds a member to the redo log file group added in the previous example:

ALTER DATABASE stocks  
   ADD LOGFILE MEMBER 'diskc:log3.log'  
   TO GROUP 3; 
Example III

The following statement drops the redo log file member added in the previous example:

ALTER DATABASE stocks  
    DROP LOGFILE MEMBER 'diskc:log3.log'; 
Example IV

The following statement renames a redo log file member:

ALTER DATABASE stocks  
    RENAME FILE 'diskb:log3.log' TO 'diskd:log3.log'; 

The above statement only changes the member of the redo log group from one file to another. The statement does not actually change the name of the file 'DISKB:LOG3.LOG' to 'DISKD:LOG3.LOG'. You must perform this operation through your operating system.

Example V

The following statement drops all members of the redo log file group 3:

ALTER DATABASE stocks DROP LOGFILE GROUP 3; 
Example VI

The following statement adds a redo log file group containing three members to thread 5 and assigns it a GROUP parameter value of 4:

ALTER DATABASE stocks 
    ADD LOGFILE THREAD 5 GROUP 4  
        ('diska:log4.log', 
        'diskb:log4:log', 
        'diskc:log4.log' ); 
Example VII

The following statement disables thread 5 in a parallel server:

ALTER DATABASE stocks  
    DISABLE THREAD 5; 
Example VIII

The following statement enables thread 5 in a parallel server, making it available to any Oracle instance that does not explicitly request a specific thread:

ALTER DATABASE stocks  
    ENABLE PUBLIC THREAD 5; 
Example IX

The following statement creates a new datafile 'DISK1:DB1.DAT' based on the file 'DISK2:DB1.DAT':

ALTER DATABASE 
    CREATE DATAFILE 'disk1:db1.dat' AS 'disk2:db1.dat'; 
Example XI

The following statement changes the global name of the database and includes both the database name and domain:

ALTER DATABASE  
    RENAME GLOBAL_NAME TO sales.australia.acme.com; 
Example XII

The following statement attempts to change the size of datafile 'DISK1:DB1.DAT':

ALTER DATABASE  
    DATAFILE 'disk1:db1.dat' RESIZE 10 M;

For examples of performing media recovery, see Oracle8 Administrator's Guide and Oracle8 Backup and Recovery Guide.

Example XIII

The following statement clears a log file:

ALTER DATABASE  
    CLEAR LOGFILE 'disk3:log.dbf';

Related Topics

CREATE DATABASE
RECOVER clause
"Filespec"

ALTER FUNCTION

Purpose

To recompile a standalone stored function. See also "Recompiling Standalone Functions".

Prerequisites

The function must be in your own schema or you must have ALTER ANY PROCEDURE system privilege.

Syntax

Keywords and Parameters

schema  

is the schema containing the function. If you omit schema, Oracle assumes the function is in your own schema.  

function  

is the name of the function to be recompiled.  

COMPILE  

causes Oracle to recompile the function. The COMPILE keyword is required.  

Recompiling Standalone Functions

You can use the ALTER FUNCTION command to explicitly recompile a function that is invalid. Explicit recompilation eliminates the need for implicit run-time recompilation and prevents associated run-time compilation errors and performance overhead.

The ALTER FUNCTION command is similar to ALTER PROCEDURE. For information on how Oracle recompiles functions and procedures, see Oracle8 Concepts.


Note:

This command does not change the declaration or definition of an existing function. To redeclare or redefine a function, use the CREATE FUNCTION command with the OR REPLACE option; see CREATE FUNCTION.

 

Example

To explicitly recompile the function GET_BAL owned by the user MERRIWEATHER, issue the following statement:

ALTER FUNCTION merriweather.get_bal
COMPILE; 

If Oracle encounters no compilation errors while recompiling GET_BAL, GET_BAL becomes valid. Oracle can subsequently execute it without recompiling it at run time. If recompiling GET_BAL results in compilation errors, Oracle returns an error message and GET_BAL remains invalid.

Oracle also invalidates all objects that depend upon GET_BAL. If you subsequently reference one of these objects without explicitly recompiling it first, Oracle recompiles it implicitly at run time.

Related Topics

ALTER PROCEDURE
CREATE FUNCTION

ALTER INDEX

Purpose

Use ALTER INDEX to:

For illustrations of some of these purposes, see "Examples".

Prerequisites

The index must be in your own schema or you must have ALTER ANY INDEX system privilege.

Schema object privileges are granted on the parent index, not on individual index partitions. The following index partition operations require tablespace quota:

Syntax



parallel_clause: See PARALLEL clause.

storage_clause: See STORAGE clause.

deallocate_unused_clause: See DEALLOCATE UNUSED clause.


Keywords and Parameters

schema  

is the schema containing the index. If you omit schema, Oracle assumes the index is in your own schema.  

index  

is the name of the index to be altered.  

 

The following operations can be performed only on partitioned indexes:

  • drop partition
  • split partition
  • rename partition
  • rebuild partition
  • modify partition

Of these, drop partition and split partition can be performed only on global indexes.  

REBUILD  

re-creates an existing index.  

 

parallel_clause  

specifies that rebuilding the index, or some queries against the index or the index partition, are performed either in serial or parallel execution. For information about the syntax of this option and this clause, see the PARALLEL clause. For more information about parallelized operations see Oracle8 Parallel Server Concepts and Administration.  

 

LOGGING/NOLOGGING  

specifies whether ALTER INDEX...REBUILD (and ALTER INDEX...SPLIT) operations will be logged.  

 

REVERSE  

stores the bytes of the index block in reverse order, excluding the ROWID when the index is rebuilt.  

 

NOREVERSE  

stores the bytes of the index block without reversing the order when the index is rebuilt. Rebuilding a REVERSE index without the NOREVERSE keyword produces a rebuilt, reverse keyed index.  

 

index_physical_attributes_clause  

changes the values of the PCTFREE, INITRANS, and MAXTRANS parameters for a nonpartitioned index, index partition, or all partitions of a partitioned index, or default values of these parameters for a partitioned index. See these parameters in CREATE TABLE.

Note: You cannot change the value of the PCTFREE parameter for the index as a whole (ALTER INDEX) or to modify a partition (ALTER INDEX ... MODIFY PARTITION). You can change it in all other forms of the ALTER INDEX command.  

 

storage_clause  

changes the storage parameters for a nonpartitioned index, index partition, or all partitions of a partitioned index, or default values of these parameters for a partitioned index. See the STORAGE clause.  

 

TABLESPACE  

specifies the tablespace where the rebuilt index or index partition will be stored. The default is the default tablespace of the user issuing the command.  

deallocate_unused_clause  

explicitly deallocates unused space at the end of the index and make the freed space available for other segments. Only unused space above the high-water mark can be freed. If KEEP is omitted, all unused space is freed. See the DEALLOCATE UNUSED clause.  

 

KEEP  

specifies the number of bytes above the high-water mark that the index will have after deallocation. If the number of remaining extents are less than MINEXTENTS, then MINEXTENTS is set to the current number of extents. If the initial extent becomes smaller than INITIAL, then INITIAL is set to the value of the current initial extent.  

allocate_extent_clause  

explicitly allocates a new extent for the index.  

 

SIZE  

specifies the size of the extent in bytes. Use K or M to specify the extent size in kilobytes or megabytes. If you omit this parameter, Oracle determines the size based on the values of the index's STORAGE parameters.  

 

DATAFILE  

specifies one of the data files in the index's tablespace to contain the new extent. If you omit this parameter, Oracle chooses the data file.  

 

INSTANCE  

makes the new extent available to the specified instance. An instance is identified by the value of its initialization parameter INSTANCE_NUMBER. If you omit this parameter, the extent is available to all instances. Use this parameter only if you are using Oracle with the Parallel Server option in parallel mode.  

 

Explicitly allocating an extent with this clause does affect the size for the next extent to be allocated as specified by the NEXT and PCTINCREASE storage parameters.  

LOGGING/NOLOGGING  

LOGGING/NOLOGGING specifies that subsequent Direct Loader (SQL*Loader) and direct-load INSERT operations against a nonpartitioned index, index partition, or all partitions of a partitioned index will be logged (LOGGING) or not logged (NOLOGGING) in the redo log file.  

 

In NOLOGGING mode, data is modified with minimal logging (to mark new extents invalid and to record dictionary changes). When applied during media recovery, the extent invalidation records mark a range of blocks as logically corrupt, because the redo data is not logged. Therefore, if you cannot afford to lose this index, you must take a backup after the operation in NOLOGGING mode.  

 

If the database is run in ARCHIVELOG mode, media recovery from a backup taken before the an operation in LOGGING mode will re-create the index. However, media recovery from a backup taken before an operation in NOLOGGING mode will not re-create the index.  

 

An index segment can have logging attributes different from those of the base table and different from those of other index segments for the same base table.  

 

For more information about the LOGGING option and parallel DML, see Oracle8 Concepts and the Oracle8 Parallel Server Concepts and Administration.  

 

Note: The LOGGING/NOLOGGING keywords replace the RECOVERABLE/UNRECOVERABLE option. That option is still available as a valid keyword in Oracle8 when altering or rebuilding nonpartitioned indexes, but its use is not recommended.  

RENAME TO  

renames index to new_index_name. The new_index_name is a single identifier and does not include the schema name.  

MODIFY DEFAULT ATTRIBUTES  

is a valid option only for a partitioned index. Use this option to specify new values for the default attributes of a partitioned index.  

 

TABLESPACE  

specifies the tablespace where the default tablespace of a partitioned index will be stored. The default is the default tablespace of the user issuing the command.  

 

LOGGING/NOLOGGING  

specifies the default logging attribute of a partitioned index.  

Note: You can combine several operations on the base index into one ALTER INDEX statement (except RENAME and REBUILD), but you cannot combine partition operations with other partition operations or with operations on the base index.  

MODIFY PARTITION  

modifies the real physical attributes, logging option, or storage characteristics of index partition partition_name; partition_name is the name of the index partition to be altered. It must be a partition in index.  

UNUSABLE  

marks the index or index partition(s) as unusable. An unusable index must be rebuilt, or dropped and re-created, before it can be used. While one partition is marked unusable, the other partitions of the index are still valid; you can execute statements that require the index if the statements do not access the unusable partition. You can also split or rename the unusable partition before rebuilding it.  

RENAME PARTITION  

renames index partition_name to new_partition_name.  

DROP PARTITION  

removes a partition and the data in it from a partitioned global index. Dropping a partition of a global index marks the index's next partition as unusable. You cannot drop the highest partition of a global index.  

split_partition_clause  

splits a global partitioned index into two partitions, adding a new partition to the index.

Splitting a partition marked as unusable results in two partitions, both marked as unusable. You must rebuild the partitions before you can use them.  

 

Splitting a usable partition results in two partitions populated with index data, both marked as usable.  

 

AT (value_list)  

specifies the new noninclusive upper bound for split_partition_1. The value_list must compare less than the presplit partition bound for partition_name_old and greater than the partition bound for the next lowest partition (if there is one).  

 

INTO  

describes the two partitions resulting from the split.  

 

partition_description, partition_description  

specifies the names and physical attributes of the two partitions resulting from the split.  

REBUILD PARTITION  

rebuilds one partition of an index. You can also use this option to move an index partition to another tablespace or to change a create-time physical attribute. For more information about partition maintenance operations, see the Oracle8 Administrator's Guide.  

Examples

Example I

This statement alters SCOTT'S CUSTOMER index so that future data blocks within this index use 5 initial transaction entries and an incremental extent of 100 kilobytes:

ALTER INDEX scott.customer  
    INITRANS 5  
    STORAGE (NEXT 100K); 
Example II

The following example drops index partition IX_ANTARTICA:

ALTER INDEX sales_area_ix
  DROP PARTITION ix_antarctica;
Example III

This statement alters the real attributes of every partition of local partitioned index SALES_IX3. New partitions added in the future will use 5 initial transaction entries and an incremental extent of 100 K:

ALTER INDEX sales_ix3 INITRANS 5 STORAGE ( NEXT 100K );
Example III(a)

This statement alters the default attributes of local partitioned index SALES_IX3. New partitions added in the future will use 5 initial transaction entries and an incremental extent of 100 K:

ALTER INDEX sales_ix3 
  MODIFY DEFAULT ATTRIBUTES INITRANS 5 STORAGE ( NEXT 100K );
Example IV

The following statement marks the IDX_ACCTNO index as UNUSABLE:

ALTER INDEX idx_acctno UNUSABLE;
Example V

The following statement changes the maximum number of extents for partition BRIX_NY:

ALTER INDEX branch_ix MODIFY PARTITION brix_ny  
  STORAGE( MAXEXTENTS 30 ) LOGGING;
Example VI

The following example marks partition IDX_FEB96 of index IDX_ACCTNO as UNUSABLE:

ALTER INDEX idx_acctno MODIFY PARTITION idx_feb96 UNUSABLE;
Example VII

The following statement sets the parallel attributes for index ARTIST_IX:

ALTER INDEX artist_ix PARALLEL (DEGREE 4, INSTANCES 3);
Example VIII

The following statement sets the parallel attributes for index ARTIST_IX so that scans on the index will not be parallelized:

ALTER INDEX artist_ix NOPARALLEL;
Example IX

The following statement rebuilds partition P063 in index ARTIST_IX. The rebuilding of the index partition will not be logged:

ALTER INDEX artist_ix 
  REBUILD PARTITION p063 NOLOGGING;
Example X

The following example renames an index:

ALTER INDEX emp_ix1 RENAME TO employee_ix1;
Example XI

The following example renames an index partition:

ALTER INDEX employee_ix2 RENAME PARTITION emp_ix2_p3 
  TO employee_ix2_p3;
Example XII

The following example splits partition PARTNUM_IX_P6 in partitioned index PARTNUM_IX into PARTNUM_IX_P5 and PARTNUM_IX_P6:

ALTER INDEX partnum_ix
  SPLIT PARTITION partnum_ix_p6 AT ( 5001 )
  INTO ( PARTITION partnum_ix_p5 TABLESPACE ts017 LOGGING, 
         PARTITION partnum_ix_p6 TABLESPACE ts004 );

Note that the second partition retains the name of the old partition.

Example XIII

The following statement rebuilds index EMP_IX so that the bytes of the index block are stored in REVERSE order:

ALTER INDEX emp_ix REBUILD REVERSE;

Related Topics

CREATE INDEX
CREATE TABLE
PARALLEL clause
STORAGE clause

DEALLOCATE UNUSED clause


ALTER PACKAGE

Purpose

To recompile a stored package. See also "Recompiling Stored Packages".

Prerequisites

The package must be in your own schema or you must have ALTER ANY PROCEDURE system privilege.

Syntax

Keywords and Parameters

schema  

is the schema containing the package. If you omit schema, Oracle assumes the package is in your own schema.  

package  

is the name of the package to be recompiled.  

COMPILE  

recompiles the package specification or body. The COMPILE keyword is required.  

PACKAGE  

recompiles the package body and specification.  

BODY  

recompiles only the package body.  

 

The default option is PACKAGE.  

Recompiling Stored Packages

You can use the ALTER PACKAGE command to explicitly recompile either a package specification and body or only a package body. Explicit recompilation eliminates the need for implicit run-time recompilation and prevents associated run-time compilation errors and performance overhead.

Because all objects in a package are stored as a unit, the ALTER PACKAGE command recompiles all package objects together. You cannot use the ALTER PROCEDURE command or ALTER FUNCTION command to individually recompile a procedure or function that is part of a package.


Note:

This command does not change the declaration or definition of an existing package. To redeclare or redefine a package, use the CREATE PACKAGE or the CREATE PACKAGE BODY command with the OR REPLACE option.

 

Recompiling Package Specifications

You might want to recompile a package specification to check for compilation errors after modifying the specification. When you issue an ALTER PACKAGE statement with the COMPILE PACKAGE option, Oracle recompiles the package specification and body regardless of whether it is invalid. When you recompile a package specification, Oracle invalidates any local objects that depend on the specification, such as procedures that call procedures or functions in the package. Note that the body of a package also depends on its specification. If you subsequently reference one of these dependent objects without first explicitly recompiling it, Oracle recompiles it implicitly at run time.

Recompiling Package Bodies

You might want to recompile a package body after modifying it. When you issue an ALTER PACKAGE statement with the COMPILE BODY option, Oracle recompiles the package body regardless of whether it is invalid. When you recompile a package body, Oracle first recompiles the objects on which the body depends, if any of those objects are invalid. If Oracle recompiles the body successfully, the body becomes valid. If recompiling the body results in compilation errors, Oracle returns an error and the body remains invalid. You can then debug the body using the predefined package DBMS_OUTPUT. Note that recompiling a package body does not invalidate objects that depend upon the package specification.

For more information on debugging packages, see Oracle8 Application Developer's Guide. For information on how Oracle maintains dependencies among schema objects, including remote objects, see Oracle8 Concepts.

Example I

This statement explicitly recompiles the specification and body of the ACCOUNTING package in the schema BLAIR:

ALTER PACKAGE blair.accounting
COMPILE PACKAGE; 

If Oracle encounters no compilation errors while recompiling the ACCOUNTING specification and body, ACCOUNTING becomes valid. BLAIR can subsequently call or reference all package objects declared in the specification of ACCOUNTING without run-time recompilation. If recompiling ACCOUNTING results in compilation errors, Oracle returns an error message and ACCOUNTING remains invalid.

Oracle also invalidates all objects that depend upon ACCOUNTING. If you subsequently reference one of these objects without explicitly recompiling it first, Oracle recompiles it implicitly at run time.

Example II

To recompile the body of the ACCOUNTING package in the schema BLAIR, issue the following statement:

ALTER PACKAGE blair.accounting 
COMPILE BODY; 

If Oracle encounters no compilation errors while recompiling the package body, the body becomes valid. BLAIR can subsequently call or reference all package objects declared in the specification of ACCOUNTING without run-time recompilation. If recompiling the body results in compilation errors, Oracle returns an error message and the body remains invalid.

Because this statement recompiles the body and not the specification of ACCOUNTING, Oracle does not invalidate dependent objects.

Related Topics

CREATE PACKAGE
CREATE PACKAGE BODY

ALTER PROCEDURE

Purpose

To recompile a stand-alone stored procedure. See also "Recompiling Stored Procedures".

Prerequisites

The procedure must be in your own schema or you must have ALTER ANY PROCEDURE system privilege.

Syntax

Keywords and Parameters

schema  

is the schema containing the procedure. If you omit schema, Oracle assumes the procedure is in your own schema.  

procedure  

is the name of the procedure to be recompiled.  

COMPILE  

causes Oracle to recompile the procedure. The COMPILE keyword is required.  

Recompiling Stored Procedures

The ALTER PROCEDURE command is quite similar to the ALTER FUNCTION command. The following discussion of explicitly recompiling procedures also applies to functions.

You can use the ALTER PROCEDURE command to explicitly recompile a procedure that is invalid. Explicit recompilation eliminates the need for implicit run-time recompilation and prevents associated run-time compilation errors and performance overhead.

When you issue an ALTER PROCEDURE statement, Oracle recompiles the procedure regardless of whether it is valid or invalid.

You can use the ALTER PROCEDURE command only to recompile a standalone procedure. To recompile a procedure that is part of a package, recompile the entire package using the ALTER PACKAGE command.

When you recompile a procedure, Oracle first recompiles objects upon which the procedure depends, if any of those objects are invalid. Oracle also invalidates any local objects that depend upon the procedure, such as procedures that call the recompiled procedure or package bodies that define procedures that call the recompiled procedure. If Oracle recompiles the procedure successfully, the procedure becomes valid. If recompiling the procedure results in compilation errors, then Oracle returns an error and the procedure remains invalid. You can then debug procedures using the predefined package DBMS_OUTPUT. For information on debugging procedures, see Oracle8 Application Developer's Guide. For information on how Oracle maintains dependencies among schema objects, including remote objects, see Oracle8 Concepts.


Note:

This command does not change the declaration or definition of an existing procedure. To redeclare or redefine a procedure, use the CREATE PROCEDURE command with the OR REPLACE option.

 

Example

To explicitly recompile the procedure CLOSE_ACCT owned by the user HENRY, issue the following statement:

ALTER PROCEDURE henry.close_acct
COMPILE; 

If Oracle encounters no compilation errors while recompiling CLOSE_ACCT, CLOSE_ACCT becomes valid. Oracle can subsequently execute it without recompiling it at run time. If recompiling CLOSE_ACCT results in compilation errors, Oracle returns an error and CLOSE_ACCT remains invalid.

Oracle also invalidates all dependent objects. These objects include any procedures, functions, and package bodies that call CLOSE_ACCT. If you subsequently reference one of these objects without first explicitly recompiling it, Oracle recompiles it implicitly at run time.

Related Topics

ALTER FUNCTION
ALTER PACKAGE
CREATE PROCEDURE

ALTER PROFILE

Purpose

To add, modify, or remove a resource limit or password management in a profile. See also "Examples".

Prerequisites

You must have ALTER PROFILE system privilege to change profile resource limits. To modify password limits and protection, you must have ALTER PROFILE and ALTER USER system privileges. See also "Using Password History".

Syntax

Keywords and Parameters

profile  

is the name of the profile to be altered.  

integer  

defines a new limit for a resource in this profile.  

 

For information on parameter resource limits for ALTER PROFILE, see CREATE PROFILE.  

Note:

  • You cannot remove a limit from the DEFAULT profile.
  • You can use fractions of days for all parameters with days as units. Fractions are expressed as x/y. For example, 1 hour is 1/24 and 1 minute is 1/1440.
 

Using Password History

Changes made to a profile with an ALTER PROFILE statement affect users only in their subsequent sessions, not in their current sessions.

The following restrictions apply when specifying password history parameters:

Examples

Example I

The following example makes a password unavailable for reuse for 90 days:

ALTER PROFILE prof 
LIMIT PASSWORD_REUSE_TIME 90 
PASSWORD_REUSE_MAX UNLIMITED;
Example II

The following statement defaults the PASSWORD_REUSE_TIME value to its defined value in the DEFAULT profile:

ALTER PROFILE prof 
LIMIT PASSWORD_REUSE_TIME DEFAULT
PASSWORD_REUSE_MAX UNLIMITED;
Example III

The following example alters profile PROF with FAILED_LOGIN_ATTEMPTS set to 5 and PASSWORD_LOCK_TIME set to 1:

ALTER PROFILE prof LIMIT
FAILED_LOGIN_ATTEMPTS 5
PASSWORD_LOCK_TIME 1;

This command causes PROF's account to become locked for 1 day after 5 unsuccessful login attempts.

Example IV

The following example modifies profile PROF's PASSWORD_LIFE_TIME to 60 days and PASSWORD_GRACE_TIME to 10 days:

ALTER PROFILE prof LIMIT
PASSWORD_LIFE_TIME 60
PASSWORD_GRACE_TIME 10;
Example V

This statement defines a new limit of 5 concurrent sessions for the ENGINEER profile:

ALTER PROFILE engineer LIMIT SESSIONS_PER_USER  5; 

If the ENGINEER profile does not currently define a limit for SESSIONS_PER_USER, the above statement adds the limit of 5 to the profile. If the profile already defines a limit, the above statement redefines it to 5. Any user assigned the ENGINEER profile is subsequently limited to 5 concurrent sessions.

Example VI

This statement defines unlimited idle time for the ENGINEER profile:

ALTER PROFILE engineer LIMIT IDLE_TIME UNLIMITED; 

Any user assigned the ENGINEER profile is subsequently permitted unlimited idle time.

Example VII

This statement removes the IDLE_TIME limit from the ENGINEER profile:

ALTER PROFILE engineer LIMIT IDLE_TIME DEFAULT;

Any user assigned the ENGINEER profile is subject in their subsequent sessions to the IDLE_TIME limit defined in the DEFAULT profile.

Example VIII

This statement defines a limit of 2 minutes of idle time for the DEFAULT profile:

ALTER PROFILE default LIMIT IDLE_TIME  2; 

This IDLE_TIME limit applies to these users:

Related Topics

CREATE PROFILE

ALTER RESOURCE COST

Purpose

To specify a formula to calculate the total resource cost used in a session. For any session, this cost is limited by the value of the COMPOSITE_LIMIT parameter in the user's profile. See also "Altering Resource Costs".

Prerequisites

You must have ALTER RESOURCE COST system privilege.

Syntax

Keywords and Parameters

CPU_PER_SESSION  

is the amount of CPU time used by a session measured in hundredth of seconds.  

CONNECT_TIME  

is the amount of CPU time used by a session measured in hundredths of seconds.  

CPU_PER_SESSION  

is the elapsed time of a session measured in minutes.  

LOGICAL_READS_PER_SESSION  

is the number of data blocks read during a session including blocks read from both memory and disk.  

PRIVATE_SGA  

The number of bytes of private space in the system global area (SGA) used by a session. This limit only applies if you are using the multithreaded server architecture and allocating private space in the SGA for your session.  

integer  

is the weight of each resource.  

Altering Resource Costs

The ALTER RESOURCE COST command specifies the formula by which Oracle calculates the total resource cost used in a session. Oracle calculates the total resource cost by multiplying the amount of each resource used in the session by the resource's weight and summing the products for all four resources. Both the products and the total cost are expressed in units called service units.

Although Oracle monitors the use of other resources, only these four can contribute to the total resource cost for a session. For information on all resources, see CREATE PROFILE.

The weight that you assign to each resource determines how much the use of that resource contributes to the total resource cost. Using a resource with a lower weight contributes less to the cost than using a resource with a higher weight. If you do not assign a weight to a resource, the weight defaults to 0 and use of the resource subsequently does not contribute to the cost. The weights you assign apply to all subsequent sessions in the database.

Once you have specified a formula for the total resource cost, you can limit this cost for a session with the COMPOSITE_LIMIT parameter of the CREATE PROFILE command. If a session's cost exceeds the limit, Oracle aborts the session and returns an error. For information on establishing resource limits, see CREATE PROFILE. If you use the ALTER RESOURCE COST command to change the weight assigned to each resource, Oracle uses these new weights to calculate the total resource cost for all current and subsequent sessions.

Example I

The following statement assigns weights to the resources CPU_PER_SESSION and CONNECT_TIME:

ALTER RESOURCE COST 
CPU_PER_SESSION 100
CONNECT_TIME      1; 

The weights establish this cost formula for a session:

T = (100 * CPU) + CON

where:

T  

is the total resource cost for the session expressed in service units.  

CPU  

is the CPU time used by the session measured in hundredths of seconds.  

CON  

is the elapsed time of a session measured in minutes.  

Because the above statement assigns no weight to the resources LOGICAL_READS_PER_SESSION and PRIVATE_SGA, these resources do not appear in the formula.

If a user is assigned a profile with a COMPOSITE_LIMIT value of 500, a session exceeds this limit whenever T exceeds 500. For example, a session using 0.04 seconds of CPU time and 101 minutes of elapsed time exceeds the limit. A session 0.0301 seconds of CPU time and 200 minutes of elapsed time also exceeds the limit.

You can subsequently change the weights with another ALTER RESOURCE statement:

ALTER RESOURCE COST 
LOGICAL_READS_PER_SESSION 2
CONNECT_TIME 0; 

These new weights establish a new cost formula:

T = (100 * CPU) + (2 * LOG)

where:

T CPU  

are the same as in the previous formula.  

LOG  

is the number of data blocks read during the session.  

This ALTER RESOURCE COST statement changes the formula in these ways:

Related Topics

CREATE PROFILE


ALTER ROLE

Purpose

To change the authorization needed to enable a role. See also "Changing Authorizations".

Prerequisites

You must either have been granted the role with the ADMIN OPTION or have ALTER ANY ROLE system privilege.

Syntax

Keywords and Parameters

The keywords and parameters in the ALTER ROLE command all have the same meaning as in the CREATE ROLE command; see CREATE ROLE.

Changing Authorizations

Before you alter a role to IDENTIFIED GLOBALLY, you must:

The one exception to this rule is that you should not revoke the role from the user who is currently altering the role.

If a user with ALTER ANY ROLE changes a role that is IDENTIFIED GLOBALLY to any of the following, then Oracle grants the role with the ADMIN OPTION:

Example I

The following example changes the role ANALYST to IDENTIFIED GLOBALLY:

ALTER ROLE analyst IDENTIFIED GLOBALLY;
Example II

This statement changes the password on the TELLER role to LETTER:

ALTER ROLE teller 
IDENTIFIED BY letter; 

Users granted the TELLER role must subsequently enter the new password "letter" to enable the role.

Related Topics

CREATE ROLE
SET ROLE

ALTER ROLLBACK SEGMENT

Purpose

To alter a rollback segment by

For more information, see "Altering Rollback Segments".

Prerequisites

You must have ALTER ROLLBACK SEGMENT system privilege.

Syntax

storage_clause: See STORAGE clause.

Keywords and Parameters

rollback_segment  

specifies the name of an existing rollback segment.  

ONLINE  

brings the rollback segment online.  

OFFLINE  

takes the rollback segment offline.  

storage_clause  

changes the rollback segment's storage characteristics. See the STORAGE clause for syntax and additional information.  

SHRINK  

attempts to shrink the rollback segment to an optimal or given size.  

Altering Rollback Segments

When you create a rollback segment, it is initially offline. An offline rollback segment is not available for transactions.

The ONLINE option brings the rollback segment online, making it available for transactions by your instance. You can also bring a rollback segment online when you start your instance with the initialization parameter ROLLBACK_SEGMENTS.

The OFFLINE option takes the rollback segment offline. If the rollback segment does not contain information necessary to roll back any active transactions, Oracle takes it offline immediately. If the rollback segment does contain information for active transactions, Oracle makes the rollback segment unavailable for future transactions and takes it offline after all the active transactions are committed or rolled back. Once the rollback segment is offline, it can be brought online by any instance.

You cannot take the SYSTEM rollback segment offline.

You can tell whether a rollback segment is online or offline by querying the data dictionary view DBA_ROLLBACK_SEGS. Online rollback segments are indicated by a STATUS value of IN_USE. Offline rollback segments are indicated by a STATUS value of AVAILABLE.

For more information on making rollback segments available and unavailable, see Oracle8 Administrator's Guide.

The STORAGE clause of the ALTER ROLLBACK SEGMENT command affects future space allocation in the rollback segment. You cannot change the values of the INITIAL and MINEXTENTS for an existing rollback segment.

The SHRINK clause of the ALTER ROLLBACK SEGMENT command initiates an attempt to reduce the specified rollback segment to an optimum size. If size is not specified, then the size defaults to the OPTIMAL value of the STORAGE clause of the CREATE ROLLBACK SEGMENT command that created the rollback segment. If the OPTIMAL value was not specified, then the size defaults to the MINEXTENTS value of the STORAGE clause of the CREATE ROLLBACK SEGMENT command. The specified size in a SHRINK clause is valid for the execution of the command; thereafter, OPTIMAL reverts to the OPTIMAL value of the CREATE ROLLBACK SEGMENT command. Regardless of whether a size is specified or not, the rollback segment cannot shrink to less than two extents.

You can query the DBA_ROLLBACK_SEGS view to determine the actual size of a rollback segment after attempting to shrink a rollback segment.

For a parallel server, you can shrink only rollback segments that are online to your instance.

The SHRINK option is an attempt to shrink the size of the rollback segment; the success and amount of shrinkage depends on the following:

Example I

This statement brings the rollback segment RSONE online:

ALTER ROLLBACK SEGMENT rsone ONLINE; 
Example II

This statement changes the STORAGE parameters for RSONE:

ALTER ROLLBACK SEGMENT rsone 
STORAGE (NEXT 1000 MAXEXTENTS 20); 
Example III

This statement attempts to resize a rollback segment to an optimum size of 100 megabytes:

ALTER ROLLBACK SEGMENT rsone 
SHRINK TO 100 M;

Related Topics

CREATE ROLLBACK SEGMENT
CREATE TABLESPACE
STORAGE clause

ALTER SEQUENCE

Purpose

To change the sequence by

For illustrations of some of these purposes, see "Examples".

Prerequisites

The sequence must be in your own schema or you must have ALTER privilege on the sequence or you must have ALTER ANY SEQUENCE system privilege.

Syntax


Keywords and Parameters

The keywords and parameters in this command serve the same purpose that they do in CREATE SEQUENCE.

Note:

Examples

Example I

This statement sets a new maximum value for the ESEQ sequence:

ALTER SEQUENCE eseq 
MAXVALUE 1500; 
Example II

This statement turns on CYCLE and CACHE for the ESEQ sequence:

ALTER SEQUENCE eseq 
CYCLE
CACHE 5; 

Related Topics

CREATE SEQUENCE
DROP SEQUENCE

ALTER SESSION

Purpose

To alter your current session in one of the following ways:

Prerequisites

To enable and disable the SQL trace facility or to change the default label format, you must have ALTER SESSION system privilege.

To perform the other operations of this command, you do not need any privileges.

Syntax





Keywords and Parameters

ADVISE  

sends advice to a remote database to force a distributed transaction. This advice appears on the remote database in the ADVICE column of the DBA_2PC_PENDING data dictionary view in the event of an in-doubt distributed transaction. (See also "Forcing In-Doubt Distributed Transactions".)The following are advice options:  

 

COMMIT  

places the value 'C' in DBA_2PC_PENDING.ADVICE.  

 

ROLLBACK  

places the value 'R' in DBA_2PC_PENDING.ADVICE.  

 

NOTHING  

places the value ' ' in DBA_2PC_PENDING.ADVICE.  

CLOSE DATABASE LINK  

closes the database link dblink, eliminating your session's connection to the remote database. The database link cannot be currently in use by an active transaction or an open cursor. For more information, see "Closing Database Links".  

COMMIT IN PROCEDURE  

ENABLE  

permits procedures and stored functions to issue these statements.  

 

DISABLE  

prohibits procedures and stored functions from issuing these statements.  

 

See also "Transaction Control in Procedures and Stored Functions".  

PARALLEL DML  

specifies whether all subsequent DML transactions in the session will be considered for parallel execution. (See also "Parallel DML".)  

 

You can execute this option only between committed transactions. Uncommitted transactions must either be committed or rolled back prior to executing this command.  

 

ENABLE  

executes the session's DML statements in parallel mode if a parallel hint or a parallel clause is specified.  

 

DISABLE  

executes the session's DML statements serially. This is the default mode.  

 

FORCE  

forces parallel execution of subsequent DML statements in the session if none of the parallel DML restrictions are violated. If no parallel clause or hint is specified, then a default level of parallelism (for both degree and instances) is used.

Note: Using FORCE automatically causes all tables created in this session to be created with a default level of parallelism. The effect is the same as if you had specified the parallel clause (with default degree and default instances) with the CREATE TABLE statement.  

SET  

sets the session parameters that follow.  

CLOSE_OPEN_CACHED_CURSORS  

controls whether cursors opened and cached in memory by PL/SQL are automatically closed at each COMMIT or ROLLBACK.  

 

TRUE  

causes open cursors to be closed at each COMMIT or ROLLBACK.  

 

FALSE  

signifies that cursors opened by PL/SQL are held open so that subsequent executions need not open a new cursor.  

CONSTRAINT[S]  

determines when conditions specified by a deferrable constraint are enforced.  

 

IMMEDIATE  

indicates that the conditions specified by the deferrable constraint are checked immediately after each DML statement; equivalent to issuing the SET CONSTRAINTS ALL IMMEDIATE command at the beginning of each transaction in your session. See the IMMEDIATE parameter of SET CONSTRAINT(S).  

 

DEFERRED  

indicates that the conditions specified by the deferrable constraint are checked when the transaction is committed; equivalent to issuing the SET CONSTRAINTS ALL DEFERRED command at the beginning of each transaction in your session. See the DEFERRED parameter of SET CONSTRAINT(S).  

 

DEFAULT  

restores all constraints at the beginning of each transaction to their initial state of DEFERRED or IMMEDIATE.  

FLAGGER  

specifies FIPS flagging. See also "FIPS Flagging".  

 

ENTRY  

flags for SQL92 Entry level.  

 

INTERMEDIATE  

flags for SQL92 Intermediate level.  

 

FULL  

flags for SQL92 Full level.  

 

OFF  

turns off flagging  

GLOBAL_NAMES  

controls the enforcement of global name resolution for your session. For information on enabling and disabling global name resolution with this parameter, see ALTER SYSTEM.  

 

TRUE  

enables global name resolution.  

 

FALSE  

disables global name resolution.  

HASH_JOIN_ENABLED  

enables or disables the use of the hash join operation in queries. The default is TRUE, which enables hash joins.  

HASH_AREA_SIZE  

specifies in bytes the amount of memory to use for hash join operations. The default is twice the value of the SORT_AREA_SIZE initialization parameter.  

HASH_MULTIBLOCK_IO_COUNT  

specifies the number of data blocks to read and write during a hash join operation. The value multiplied by the DB_BLOCK_SIZE initialization parameter should not exceed 64 K. The default value for this parameter is 1. If the multithreaded server is used, the value is always 1, and any value given here is ignored.  

INSTANCE  

in a parallel server, accesses database files as if the session were connected to the instance specified by integer. For more information, see "Accessing the Database as if Connected to Another Instance in a Parallel Server".  

ISOLATION_LEVEL  

specifies how transactions containing database modifications are handled.  

 

SERIALIZABLE  

Transactions in the session use the serializable transaction isolation mode as specified in SQL92. That is, if a serializable transaction attempts to execute a DML statement that updates rows that are updated by another uncommitted transaction at the start of the serializable transaction, then the DML statement fails. A serializable transaction can see its own updates. The COMPATIBLE initialization parameter must be set to 7.3.0 or higher for SERIALIZABLE mode to work.  

 

READ COMMITTED  

Transactions in the session will use the default Oracle transaction behavior. Thus, if the transaction contains DML that requires row locks held by another transaction, then the DML statement will wait until the row locks are released.  

MAX_DUMP_FILE_SIZE  

specifies the upper limit of trace dump file size. Specify the maximum size as either a nonnegative integer that represents the number of blocks, or as 'UNLIMITED'. If 'UNLIMITED' is specified, no upper limit is imposed.  

For more information on the following NLS parameters, see "Using NLS Parameters".  

NLS_LANGUAGE  

changes the language in which Oracle returns errors and other messages. This parameter also implicitly specifies new values for these items:  

 

  • language for day and month names and abbreviations and spelled values of other elements
  • sort sequences
  • B.C. and A.D. indicators
  • A.M. and P.M. meridian indicators
 

NLS_TERRITORY  

implicitly specifies new values for these items:  

 

  • default date format
  • decimal character and group separators
  • local currency symbol
  • ISO currency symbol
  • first day of the week for D date format element
 

NLS_DATE_FORMAT  

explicitly specifies a new default date format. The 'fmt' value must be a date format model as specified in the section "Date Format Models".  

NLS_DATE_LANGUAGE  

explicitly changes the language for day and month names and abbreviations and spelled values of other date format elements.  

NLS_NUMERIC_CHARACTERS  

explicitly specifies a new decimal character and group separator. The 'text' value must have this form:

dg'

where: d is the new decimal character, and g is the new group separator.  

 

The decimal character and the group separator must be two different single-byte characters, and cannot be a numeric value or any of the following characters: "+" plus, "-" minus (or hyphen), "<" less-than, or ">" greater-than.  

NLS_ISO_CURRENCY  

explicitly specifies the territory whose ISO currency symbol should be used.  

NLS_CURRENCY  

explicitly specifies a new local currency symbol. The symbol cannot exceed 10 characters.  

NLS_SORT  

changes the sequence into which Oracle sorts character values.  

 

sort  

specifies the name of a linguistic sort sequence.  

 

BINARY  

specifies a binary sort.  

 

The default sort for all character sets is binary.  

NLS_CALENDAR  

explicitly specifies a new calendar type.  

OPTIMIZER_MODE  

specifies the approach and mode of the optimizer for your session. For more information on optimizer mode, see "Changing the Optimization Approach and Mode".  

 

ALL_ROWS  

specifies the cost-based approach and optimizes for best throughput.  

 

FIRST_ROWS  

specifies the cost-based approach and optimizes for best response time.  

 

RULE  

specifies the rule-based approach.  

 

CHOOSE  

causes the optimizer to choose an optimization approach based on the presence of statistics in the data dictionary.  

PARTITION_VIEW_ENABLED  

When set to TRUE, this parameter causes the optimizer to skip unnecessary table accesses in a partition view. For more information, see Oracle8 Reference.  

PLSQL_V2_COMPATABILITY  

modifies the compile-time behavior of PL/SQL programs to allow language constructs that are illegal in Oracle8 (PL/SQL V3), but were legal in Oracle7 (PL/SQL V2). See the PL/SQL User's Guide and Reference and Oracle8 Reference for more information about this session parameter.  

 

TRUE  

enables Oracle8 PL/SQL V3 programs to execute Oracle7 PL/SQL V2 constructs.  

 

FALSE  

disallows illegal Oracle7 PL/SQL V2 constructs. This is the default.  

REMOTE_DEPENDENCIES_MODE  

specifies how dependencies of remote stored procedures are handled by the session. For more information, refer Oracle8 Application Developer's Guide.  

SESSION_CACHED_CURSORS  

specifies the size of the session cache for holding frequently used cursors. integer specifies how many cursors can be retained in the cache. For more information on this parameter, see "Caching Session Cursors".  

SKIP_UNUSABLE_INDEXES  

 

 

controls the use and reporting of tables with unusable indexes or index partitions.  

 

TRUE  

disables error reporting of indexes marked as unusable. Allows inserts, deletes, and updates to tables with unusable indexes or index partitions.  

 

FALSE  

enables error reporting of indexes marked as unusable. Does not allow inserts, deletes, and updates to tables with unusable indexes or index partitions. This is the default.  

SQL_TRACE  

controls the SQL trace facility for your session. See also "Enabling and Disabling the SQL Trace Facility".  

 

TRUE  

enables the SQL trace facility.  

 

FALSE  

disables the SQL trace facility.  

Enabling and Disabling the SQL Trace Facility

The SQL trace facility generates performance statistics for the processing of SQL statements. You can enable and disable the SQL trace facility for all sessions on an Oracle instance with the initialization parameter SQL_TRACE. When you begin a session, Oracle enables or disables the SQL trace facility based on the value of this parameter. You can subsequently enable or disable the SQL trace facility for your own session with the SQL_TRACE option of the ALTER SESSION command.

For more information on the SQL trace facility, including how to format and interpret its output, see Oracle8 Tuning.

Example I

To enable the SQL trace facility for your session, issue the following statement:

ALTER SESSION 
SET SQL_TRACE = TRUE; 

Using NLS Parameters

Oracle contains support for use in different nations and with different languages. When you start an instance, Oracle establishes support based on the values of initialization parameters that begin with "NLS". For information on these parameters, see Oracle8 Reference. You use the NLS clauses of the ALTER SESSION command to change NLS characteristics dynamically for your session. You can query the dynamic performance table V$NLS_PARAMETERS to see the current NLS attributes for your session. The sections that follow describe the use of specific NLS parameters.

Language for Error Messages

You can specify a new language for error messages with the NLS_LANGUAGE parameter. Note that this parameter also implicitly changes other language-related items. Oracle provides error messages in a wide range of languages on many platforms.

Example II

The following statement changes the language for error messages to the French:

ALTER SESSION 
  SET NLS_LANGUAGE = French 

Oracle returns error messages in French:

SELECT * FROM emp
ORA-00942: Table ou vue n'existe pas 



Note:

The language you select must already have been installed. Refer to your operating system specific installation instructions.

 

Default Date Format

You can specify a new default date format either explicitly with the NLS_DATE_FORMAT parameter or implicitly with the NLS_TERRITORY parameter. For information on the default date format models, see the section "Date Format Models".

Example III

The following statement dynamically changes the default date format for your session to 'YYYY MM DD-HH24:MI:SS':

ALTER SESSION 
SET NLS_DATE_FORMAT = 'YYYY MM DD HH24:MI:SS'

Oracle uses the new default date format:

SELECT TO_CHAR(SYSDATE) Today
FROM DUAL 
TODAY 
------------------- 
1997 08 12 14:25:56 

Language for Months and Days

You can specify a new language for names and abbreviations of months and days either explicitly with the NLS_DATE_LANGUAGE parameter or implicitly with the NLS_LANGUAGE parameter.

Example IV

The following statement changes the language for date format elements to the French:

ALTER SESSION 
SET NLS_DATE_LANGUAGE = French 

SELECT TO_CHAR(SYSDATE, 'Day DD Month YYYY') Today
FROM DUAL 

TODAY 
--------------------------- 
Mardi    28 Février   1997 

Decimal Character and Group Separator

You can specify new values for these number format elements either explicitly with the NLS_NUMERIC_CHARACTERS parameter or implicitly with the NLS_TERRITORY parameter:

D (decimal character)  

is the character that separates the integer and decimal portions of a number.  

G (group separator)  

is the character that separates groups of digits in the integer portion of a number.  

For information on how to use number format models, see "Number Format Models".

The decimal character and the group separator must be single-byte character and cannot be the same character. If the decimal character is not a period (.), you must use single quotation marks to enclose to enclose all number values that appear in expressions in your SQL statements. When not using a period for the decimal point, you should always use the TO_NUMBER function to ensure that a valid number is retrieved.

Example V

The following statement dynamically changes the decimal character to comma (,) and the group separator to period (.):

ALTER SESSION SET NLS_NUMERIC_CHARACTERS = ',.' ;

Oracle returns these new characters when you use their number format elements:

SELECT TO_CHAR( SUM(sal), 'L999G999D99') Total FROM emp ;

TOTAL 
-------------
FF29.025,00 

ISO Currency Symbol

You can specify a new value for the C number format element (the ISO currency symbol) either explicitly with the NLS_ISO_CURRENCY parameter or implicitly with the NLS_TERRITORY parameter. The value that you specify for these parameters is a territory whose ISO currency symbol becomes the value of the C number format element.

Example VI

The following statement dynamically changes the ISO currency symbol to the ISO currency symbol for the territory America:

ALTER SESSION
SET NLS_ISO_CURRENCY = America; 

SELECT TO_CHAR( SUM(sal), 'L999G999D99') Total
FROM emp; 

TOTAL 
-------------
USD29,025.00 

Local Currency Symbol

You can specify a new value for the L number format element, (the local currency symbol) either explicitly with the NLS_CURRENCY parameter or implicitly with the NLS_TERRITORY parameter.

Example VII

The following statement dynamically changes the local currency symbol to 'DM':

ALTER SESSION
SET NLS_CURRENCY = 'DM'; 

SELECT TO_CHAR( SUM(sal), 'L999G999D99') Total
FROM emp; 

TOTAL 
-------------
DM29.025,00 

Linguistic Sort Sequence

You can specify a new linguistic sort sequence or a binary sort either explicitly with the NLS_SORT parameter or implicitly with the NLS_LANGUAGE parameter.

Example VIII

The following statement dynamically changes the linguistic sort sequence to Spanish:

ALTER SESSION
SET NLS_SORT = XSpanish; 

Oracle sorts character values based on their position in the Spanish linguistic sort sequence.

Changing the Optimization Approach and Mode

The Oracle optimizer can use either of these approaches to optimize a SQL statement:

cost-based  

The optimizer optimizes a SQL statement by considering statistics describing the tables, indexes, and clusters accessed by the statement as well as the information considered with the rule-based approach.  

rule-based  

The optimizer optimizes a SQL statement based on the indexes and clusters associated with the accessed tables, the syntactic constructs of the statement, and a heuristically ranked list of these constructs.  

With the cost-based approach, the optimizer can optimize a SQL statement with one of these goals:

best throughput  

is the minimal time necessary to return all rows accessed by the statement.  

best response time  

is the minimal time necessary to return the first row accessed by the statement.  

When you start your instance, the optimization approach is established by the initialization parameter OPTIMIZER_MODE. If this parameter establishes the cost-based approach, the default goal is best throughput.

For information on how to choose a goal for the cost-based approach based on the characteristics of your application, see the Oracle8 Tuning.

FIPS Flagging

FIPS flagging causes an error message to be generated when a SQL statement is issued that is an extension of ANSI SQL92. In Oracle, there is currently no difference between Entry, Intermediate, or Full level flagging. Once flagging is set in a session, a subsequent ALTER SESSION SET FLAGGER command will work, but generates the message, ORA-00097. This allows FIPS flagging to be altered without disconnecting the session.

Caching Session Cursors

If an application repeatedly issues parse calls on the same set of SQL statements, the reopening of the session cursors can affect performance. The ALTER SESSION SET SESSION_CACHED_CURSORS command allows frequently used session cursors to be stored in a session cache even if they are closed. This is particularly useful for some Oracle tools. For example, Oracle Forms applications close all session cursors associated with a form when switching to another form; in this case, frequently used cursors would not have to be reparsed.

Oracle uses the shared SQL area to determine whether more than three parse requests were issued on a given statement. If so, Oracle moves the cursor into the session cursor cache. Subsequent requests to parse that SQL statement by the same session will find the cursor in the session cursor cache.

Session cursors are cached automatically if the initialization parameter SESSION_CACHED_CURSORS is set to a positive value. This parameter specifies the maximum number of session cursors to be kept in the cache. A least recently used algorithm ages out entries in the cache to make room for new entries when needed. You use the ALTER SESSION SET SESSION_CACHED_CURSORS command to dynamically enable session cursor caching.

For more information on session cursor caching, see Oracle8 Tuning.

Accessing the Database as if Connected to Another Instance in a Parallel Server

For optimum performance, each instance of a parallel server uses its own private rollback segments, freelist groups, and so on. A database is usually designed for a parallel server so that users connect to a particular instance and access data that is partitioned primarily for their use. If the users for that instance must connect to another instance, the data partitioning can be lost. The ALTER SESSION SET INSTANCE command allows users to access an instance as if they were connected to their usual instance.

Closing Database Links

A database link allows you to access a remote database in DELETE, INSERT, LOCK TABLE, SELECT, and UPDATE statements. When you issue a statement that uses a database link, Oracle creates a session for you on the remote database using the database link. The connection remains open until you end your local session or until the number of database links for your session exceeds the value of the initialization parameter OPEN_LINKS.

You can use the CLOSE DATABASE LINK clause of the ALTER SESSION command to close a database link explicitly if you do not plan to use it again in your session. You may want to close a database link explicitly if the network overhead associated with leaving it open is costly. Before closing a database link, you must first close all cursors that use the link and then end your current transaction if it uses the link.

Example

This example updates the employee table on the SALES database using a database link, commits the transaction, and explicitly closes the database link:

UPDATE emp@sales 
SET sal = sal + 200
WHERE empno = 9001;
COMMIT; 
ALTER SESSION
CLOSE DATABASE LINK sales; 

Forcing In-Doubt Distributed Transactions

If a network or machine failure occurs during the commit process for a distributed transaction, the state of the transaction may be unknown or in doubt. The transaction can be manually committed or rolled back on each database involved in the transaction with the FORCE clause of the COMMIT or ROLLBACK commands.

Before committing a distributed transaction, you can use the ADVISE clause of the ALTER SESSION command to send advice to a remote database in the event a distributed transaction becomes in doubt. If the transaction becomes in doubt, the advice appears in the ADVICE column of the DBA_2PC_PENDING view on the remote database. The administrator of that database can then use this advice to decide whether to commit or roll back the transaction on the remote database. For more information on distributed transactions and how to decide whether to commit or roll back in-doubt distributed transactions, see Oracle8 Distributed Database Systems.

You issue multiple ALTER SESSION statements with the ADVISE clause in a single transaction. Each such statement sends advice to the databases referenced in the following statements in the transaction until another such statement is issued. This allows you to send different advice to different databases.

Example

This transaction inserts an employee record into the EMP table on the database identified by the database link SITE1 and deletes an employee record from the EMP table on the database identified by SITE2:

ALTER SESSION
ADVISE COMMIT 

INSERT INTO emp@site1
VALUES (8002, 'FERNANDEZ', 'ANALYST', 7566,
TO_DATE('04-OCT-1992', 'DD-MON-YYYY'), 3000, NULL, 20) 

ALTER SESSION
ADVISE ROLLBACK; 
DELETE FROM emp@site2
WHERE empno = 8002; 
COMMIT; 

This transaction has two ALTER SESSION statements with the ADVISE clause. If the transaction becomes in-doubt, SITE1 is sent the advice 'COMMIT' by virtue of the first ALTER SESSION statement and SITE2 is sent the advice 'ROLLBACK' by virtue of the second.

Transaction Control in Procedures and Stored Functions

Procedures and stored functions are written in PL/SQL, and they can issue COMMIT and ROLLBACK statements. If your application performs record management that would be disrupted by a COMMIT or ROLLBACK statement not issued directly by the application itself, you may want to prevent procedures and stored functions called during your session from issuing these statements. You can do this with the following statement:

ALTER SESSION DISABLE COMMIT IN PROCEDURE;

If you subsequently call a procedure or a stored function that issues a COMMIT or ROLLBACK statement, Oracle returns an error and does not commit or roll back the transaction.

You can subsequently allow procedures and stored functions to issue COMMIT and ROLLBACK statements in your session by issuing the following statement:

ALTER SESSION ENABLE COMMIT IN PROCEDURE; 

This command does not apply to database triggers. Triggers can never issue COMMIT or ROLLBACK statements.


Note:

Some applications (such as SQL*Forms) automatically prohibit COMMIT and ROLLBACK statements in procedures and stored functions. Refer to your application documentation.

 

Parallel DML

When parallel DML is enabled for your session, all DML portions of statements issued are considered for parallel execution. Even with parallel DML enabled, however, some DML operations are restricted from parallelization, while others may still execute serially unless parallel hints and clauses are specified. For a detailed description of parallel DML features and hints, see Oracle8 Tuning.

The following restrictions apply to parallel DML operations:

Parallel DML mode can be modified only between committed transactions. Issuing this command following an uncommitted transaction will generate an error. Uncommitted transactions must be either committed or rolled back prior to issuing the ALTER SESSION ENABLE|DISABLE|FORCE PARALLEL DML command.

Example I

Issue the following statement to enable parallel DML mode for the current session:

ALTER SESSION ENABLE PARALLEL DML;
Example II

The following example modifies the current session to check all deferrable constraints immediately following each DML statement:

ALTER SESSION SET CONSTRAINTS IMMEDIATE;
Example III

The following statement modifies the current session to allow inserts into local index partitions marked as unusable:

ALTER SESSION SET SKIP_UNUSABLE_INDEXES=TRUE;

Related Topics

ALTER SESSION
SET CONSTRAINT(S)
PL/SQL User's Guide and Reference
Oracle8 Reference

Oracle8 Tuning

ALTER SNAPSHOT

Purpose

To alter a snapshot in one of the following ways:

For illustrations of some of these purposes, see "Examples".

For more information on snapshots, including refreshing snapshots, see CREATE SNAPSHOT.

Prerequisites

To alter a snapshot's storage parameters, the snapshot must be contained in your own schema, or you must have the ALTER ANY SNAPSHOT system privilege.

For detailed information about the prerequisites for ALTER SNAPSHOT, see Oracle8 Replication.

Syntax


parallel_clause: See the PARALLEL clause
storage_clause: See the STORAGE clause

For the syntax of the following clauses, see ALTER TABLE:

Keywords and Parameters

schema  

is the schema containing the snapshot. If you omit schema, Oracle assumes the snapshot is in your own schema.  

snapshot  

is the name of the snapshot to be altered.  

modify_default_attributes  

specifies new values for the default attributes of a partitioned table. For information about specifying the parameters of this clause, see ALTER TABLE.  

physical_attributes_clause  

change the values of the PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters and the storage characteristics for the internal table that Oracle uses to maintain the snapshot's data. For more information, see CREATE TABLE and the STORAGE clause.  

LOGGING/NOLOGGING  

specifies the logging attribute. For information about specifying this option, see ALTER TABLE.  

CACHE/NOCACHE  

for data that will be accessed frequently, specifies whether the blocks retrieved for this table are placed at the most recently used end of the LRU list in the buffer cache when a full table scan is performed. This option is useful for small lookup tables. For information about specifying this option, see ALTER TABLE.  

LOB_storage_clause  

specifies the LOB storage characteristics. For information about specifying the parameters of this clause, see ALTER TABLE.  

modify_LOB_storage_clause  

modifies the physical attributes of the LOB attribute lob_item or LOB object attribute. For information about specifying the parameters of this clause, see ALTER TABLE.  

For more information on the following partitioning clauses, see "Partitioned Snapshots".  

modify_partition_clause  

modifies the real physical attributes of a table partition. For information about specifying the parameters of this clause, see ALTER TABLE.  

move_partition_clause  

moves table partition partition_name to another segment. For information about specifying the parameters of this clause, see ALTER TABLE.  

add_partition_clause  

adds a new partition new_partition_name to the "high" end of a partitioned table. For information about specifying the parameters of this clause, see ALTER TABLE.  

split_partition_clause  

creates two new partitions, each with a new segment and new physical attributes, and new initial extents. For information about specifying the parameters of this clause, see ALTER TABLE.  

rename_partition_clause  

renames table partition partition_name to new_partition_name. For information about specifying the parameters of this clause, see ALTER TABLE.  

parallel_clause  

specifies the degree of parallelism for the snapshot. See the PARALLEL clause on page 4-1022. When this clause is set for master tables, performance for snapshot creation and refresh may improve (depending on the snapshot definition query).  

MODIFY PARTITION UNUSABLE LOCAL INDEXES  

 

marks all the local index partitions associated with partition_name as unusable.  

MODIFY PARTITION REBUILD UNUSABLE LOCAL INDEXES  

 

rebuilds the unusable local index partitions associated with partition_name.  

USING INDEX  

changes the value of INITRANS, MAXTRANS, and STORAGE parameters for the index Oracle uses to maintain the snapshot's data. If USING INDEX is not specified then default values are used for the index.  

REFRESH  

changes the mode and times for automatic refreshes.  

 

FAST  

specifies a fast refresh, or a refresh using the snapshot log associated with the master table.  

 

COMPLETE  

specifies a complete refresh, or a refresh that re-creates the snapshot during each refresh.  

 

FORCE  

specifies a fast refresh if one is possible or complete refresh if a fast refresh is not possible. Oracle decides whether a fast refresh is possible at refresh time.  

 

If you omit the FAST, COMPLETE, and FORCE options, Oracle uses FORCE by default.  

 

START WITH  

specifies a date expression for the next automatic refresh time.  

 

NEXT  

specifies a new date expression for calculating the interval between automatic refreshes.  

 

START WITH and NEXT values must evaluate to times in the future.  

 

WITH PRIMARY KEY  

changes a ROWID snapshot to a primary key snapshot. Primary key snapshots allow snapshot master tables to be reorganized without impacting the snapshot's ability to continue to fast refresh. The master table must contain an enabled primary key constraint. See also "Primary Key Snapshots".  

USING MASTER ROLLBACK SEGMENT  

changes remote master rollback segment to be used during snapshot refresh; rollback_segment is the name of the rollback segment to be used. (To change the local snapshot rollback segment, use the DBMS_REFRESH package in Oracle8 Replication.) See also "Specifying Rollback Segments",  

 

DEFAULT  

specifies that Oracle will choose which rollback segment to use. If you specify DEFAULT, you cannot specify rollback_segment.  

 

 

MASTER  

specifies the remote rollback segment to be used at the remote master for the individual snapshot.  

 

LOCAL  

specifies the remote rollback segment to be used for the local refresh group that contains the snapshot.  

Examples

Example I

The following statement changes the automatic refresh mode for the HQ_EMP snapshot to FAST:

ALTER SNAPSHOT hq_emp
REFRESH FAST; 

The next automatic refresh of the snapshot will be a fast refresh provided it is a simple snapshot and its master table has a snapshot log that was created before the snapshot was created or last refreshed.

Because the REFRESH clause does not specify START WITH or NEXT values, the refresh intervals established by the REFRESH clause when the HQ_EMP snapshot was created or last altered are still used.

Example II

The following statement stores a new interval between automatic refreshes for the BRANCH_EMP snapshot:

ALTER SNAPSHOT branch_emp
REFRESH NEXT SYSDATE+7;

Because the REFRESH clause does not specify a START WITH value, the next automatic refresh occurs at the time established by the START WITH and NEXT values specified when the BRANCH_EMP snapshot was created or last altered.

At the time of the next automatic refresh, Oracle refreshes the snapshot, evaluates the NEXT expression SYSDATE+7 to determine the next automatic refresh time, and continues to refresh the snapshot automatically once a week.

Because the REFRESH clause does not explicitly specify a refresh mode, Oracle continues to use the refresh mode specified by the REFRESH clause of a previous CREATE SNAPSHOT or ALTER SNAPSHOT statement.

Example III

The following statement specifies a new refresh mode, next refresh time, and new interval between automatic refreshes of the SF_EMP snapshot:

ALTER SNAPSHOT sf_emp
REFRESH COMPLETE   
START WITH TRUNC(SYSDATE+1) + 9/24  
NEXT SYSDATE+7;

The START WITH value establishes the next automatic refresh for the snapshot to be 9:00 am tomorrow. At that point, Oracle performs a fast refresh of the snapshot, evaluates the NEXT expression, and subsequently refreshes the snapshot every week.

Specifying Rollback Segments

You can specify the rollback segments to be used during a refresh for both the master site and the local site. The master rollback segment is stored on a per-snapshot basis and is validated during snapshot creation and refresh. If the snapshot is complex, the master rollback segment, if specified, is ignored.

You can change local snapshot rollback segments using the DBMS_REFRESH package and is stored at the refresh group level. For information about the DBMS_REFRESH package, see Oracle8 Replication. If the auto-refresh parameters (START WITH and NEXT) are specified, a new refresh group is automatically created to refresh the snapshot with a background process. The local rollback segment, if specified, is associated with this new refresh group. An error is raised if the auto-refresh parameters are not specified, but a local rollback segment is.


Note:

To direct Oracle to select the rollback segment automatically after one has been specified using CREATE SNAPSHOT or ALTER SNAPSHOT, specify the DEFAULT option with ALTER SNAPSHOT.

 

Example I

The following example changes the remote master rollback segment used during snapshot refresh to MASTER_SEG:

ALTER SNAPSHOT inventory 
  REFRESH USING MASTER ROLLBACK SEGMENT master_seg;  
Example II

The following example changes the remote master rollback segment used during snapshot refresh to one chosen by Oracle:

ALTER SNAPSHOT sales REFRESH USING DEFAULT MASTER ROLLBACK SEGMENT; 

Primary Key Snapshots

To change a ROWID snapshot to a primary key snapshot you must:

To fast refresh primary key snapshots you must first create a snapshot master log specifying WITH PRIMARY KEY. The snapshot master log can also store ROWIDs. The snapshot master log must be created before the snapshot is created in order for the snapshots to use the log to fast refresh.

For detailed information about primary key snapshots, see Oracle8 Replication.


Note:

Primary key snapshots cannot be altered to ROWID snapshots. You must drop the primary key snapshot and re-create it as a ROWID snapshot.

 

Example I

The following example changes a ROWID to a primary key snapshot:

ALTER SNAPSHOT emp_rs REFRESH WITH PRIMARY KEY; 

Partitioned Snapshots

Partitioned snapshots are the same as partitioned tables because snapshots are basically tables. The options have the same syntax and semantics as the partitioned table options for CREATE TABLE and ALTER TABLE. The only difference is that the following operations are not allowed on snapshots and snapshot logs:

You cannot perform bulk deletions by dropping or truncating partitions on master tables. Thus, after dropping or truncating a partition, all snapshots must be refreshed manually. A fast refresh will probably produce incorrect results, but Oracle will not raise an error.

Related Topics

CREATE SNAPSHOT
DROP SNAPSHOT
STORAGE clause

ALTER SNAPSHOT LOG

Purpose

Changes the storage characteristics of a snapshot log. For more information on snapshot logs, see CREATE SNAPSHOT.

Prerequisites

Only the owner of the master table or a user with the SELECT privilege for the master table can alter a snapshot log. For detailed information about the prerequisites for ALTER SNAPSHOT LOG, see Oracle8 Replication.

Syntax

For the syntax of the following clauses, see ALTER TABLE:

Keywords and Parameters

schema  

is the schema containing the master table. If you omit schema, Oracle assumes the snapshot log is in your own schema.  

table  

is the name of the master table associated with the snapshot log to be altered.  

physical_attributes_clause  

changes the value of PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters for the table, partition, the overflow data segment, or the default characteristics of a partitioned table. See the PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters of CREATE TABLE. See the example under "Modifying Physical Attributes".  

rename_partition_clause  

renames table partition partition_name to new_partition_name. For information about specifying the parameters of this clause, see ALTER TABLE.  

modify_partition_clause  

modifies the real physical attributes of a table partition. For information about specifying the parameters of this clause, see ALTER TABLE.  

move_partition_clause  

moves table partition partition_name to another segment. For information about specifying the parameters of this clause, see ALTER TABLE.  

add_partition_clause  

adds a new partition new_partition_name to the "high" end of a partitioned table. For information about specifying the parameters of this clause, see ALTER TABLE.  

split_partition_clause  

creates two new partitions, each with a new segment and new physical attributes, and new initial extents. For information about specifying the parameters of this clause, see ALTER TABLE.

For more information see "Partitioned Snapshot Logs".  

modify_default_attributes_clause  

is a valid option only for a partitioned index. Use this option to specify new values for the default attributes of a partitioned index.  

parallel_clause  

specifies the degree of parallelism for the snapshot. See the PARALLEL clause. When this clause is set for master tables, performance during snapshot creation and refresh may improve (depending on the snapshot definition query).  

LOGGING/NOLOGGING  

specifies the logging attribute. For information about specifying this option, see ALTER TABLE.  

CACHE/NOCACHE  

for data that will be accessed frequently, specifies whether the blocks retrieved for this table are placed at the most recently used end of the LRU list in the buffer cache when a full table scan is performed. This option is useful for small lookup tables. For information about specifying this option, see ALTER TABLE.  

ADD  

changes the snapshot log so that it records the primary key values or ROWID values when rows in the snapshot master table are updated. This clause can also be used to record additional filter columns.  

 

PRIMARY KEY  

specifies that the primary-key values of all rows updated should be recorded in the snapshot log.  

 

ROWID  

specifies that the ROWID values of all rows updated should be recorded in the snapshot log.  

 

filter_column(s)  

are non-primary-key columns referenced by snapshots. For information about filter columns, see Oracle8 Replication.  

 

For more information, see "Adding Primary Key, ROWID, and Filter Columns".  

Modifying Physical Attributes

Example

The following statement changes the MAXEXTENTS value of a snapshot log:

ALTER SNAPSHOT LOG ON dept  
STORAGE MAXEXTENTS 50;

Adding Primary Key, ROWID, and Filter Columns

Snapshot logs can be altered to additionally record primary key, ROWID, or filter column information when snapshot master tables are updated. To stop recording any of this information, you must first drop the snapshot log and then re-create it.

Example

The following example alters an existing ROWID snapshot log to also record primary key information:

ALTER SNAPSHOT LOG ON sales ADD PRIMARY KEY;

Partitioned Snapshot Logs

Partitioned snapshot logs are the same as partitioned tables, because snapshot logs are basically tables. The options have the same syntax and semantics as the partitioned table options for CREATE TABLE and ALTER TABLE. The only difference is that the following operations are not allowed on snapshots and snapshot logs:

You cannot perform bulk deletions by dropping or truncating partitions on master tables. Therefore, after dropping or truncating a partition, all snapshots must be manually refreshed. A fast refresh will probably produce incorrect results, but Oracle will not raise an error.

Related Topics

ALTER TABLE
CREATE SNAPSHOT
CREATE SNAPSHOT LOG
DROP SNAPSHOT LOG

ALTER SYSTEM

Purpose

To dynamically alter your Oracle instance in one of the following ways:

Prerequisites

You must have ALTER SYSTEM system privilege.

Syntax

archive_log_clause: See the ARCHIVE LOG clause.

set_clause::=





dispatch_clause::=

options_clause::=

Keywords and Parameters

You can use the following options regardless of whether your instance has the database dismounted or mounted, open or closed:  

RESTRICTED SESSION  

specifies whether logon to Oracle is restricted  

 

ENABLE  

allows only users with RESTRICTED SESSION system privilege to logon to Oracle.  

 

DISABLE  

reverses the effect of the ENABLE RESTRICTED SESSION option, allowing all users with CREATE SESSION system privilege to log on to Oracle.  

 

For more information, see "Restricting Logons".  

FLUSH SHARED_POOL  

clears all data from the shared pool in the system global area (SGA). For more information, see "Clearing the Shared Pool".  

You can use the following options when your instance has the database mounted, open or closed:  

CHECKPOINT  

performs a checkpoint.  

 

GLOBAL  

performs a checkpoint for all instances that have opened the database.  

 

LOCAL  

performs a checkpoint only for the thread of redo log file groups for your instance. You can use this option only when your instance has the database open.  

 

If you omit both the GLOBAL and LOCAL options, Oracle performs a global checkpoint. For more information, see "Performing a Checkpoint".  

CHECK DATAFILES  

GLOBAL  

verifies that all instances that have opened the database can access all online datafiles.  

 

LOCAL  

verifies that your instance can access all online datafiles.  

 

If you omit both the GLOBAL and LOCAL options, Oracle uses GLOBAL by default. For more information, see "Checking Datafiles".  

You can use the following parameters and options only when your instance has the database open:  

RESOURCE_LIMIT  

controls resource limits. TRUE enables resource limits; FALSE disables resource limits. See also "Using Resource Limits".  

GLOBAL_NAMES  

controls the enforcement of global name resolution for your session. TRUE enables the enforcement of global names; FALSE disables the enforcement of global names. For more information, see "Global Name Resolution".  

SCAN_INSTANCES  

in a parallel server, specifies the number of instances to participate in parallelized operations. This syntax will be obsolete in the next major release.  

CACHE_INSTANCES  

in a parallel server, specifies the number of instances that will cache a table. This syntax will be obsolete in the next major release.  

For more information on parallel operations, see Oracle8 Tuning.

For more information on the following multithreaded server parameters, see "Managing Processes for the Multithreaded Server".  

MTS_SERVERS  

specifies a new minimum number of shared server processes.  

MTS_DISPATCHERS  

specifies a new number of dispatcher processes:  

 

protocol  

is the network protocol of the dispatcher processes.  

 

integer  

is the new number of dispatcher processes of the specified protocol.  

 

You can specify multiple MTS_DISPATCHERS parameters in a single command for multiple network protocols.  

For more information on the following licensing parameters, see "Using Licensing Limits".  

JOB_QUEUE_PROCESSES  

specifies the number of job queue processes per instance (SNPn, where n is 0 to 9 followed by A to Z). Set this parameter to 1 or higher if you wish to have your snapshots updated automatically. One job queue process is usually sufficient unless you have many snapshots that refresh simultaneously.

Oracle also uses job queue processes to process requests created by the DBMS_JOB package. For more information on managing table snapshots, see Oracle8 Replication.  

LICENSE_MAX_SESSIONS  

limits the number OS sessions on your instance. A value of 0 disables the limit.  

LICENSE_SESSIONS_WARNING  

establishes a threshold of sessions over which Oracle writes warning messages to the ALERT file for subsequent sessions. A value of 0 disables the warning threshold.  

LICENSE_MAX_USERS  

limits number of concurrent users on your database. A value of 0 disables the limit.  

REMOTE_DEPENDENCIES_MODE  

specifies how dependencies of remote stored procedures are handled by the server. For more information, refer to Oracle8 Application Developer's Guide.

 

SWITCH LOGFILE  

switches redo log file groups. For more information, see "Switching Redo Log File Groups".  

DISTRIBUTED RECOVERY  

specifies whether or not distributed recovery is enabled.  

 

ENABLE  

enables distributed recovery. In a single-process environment, you must use this option to initiate distributed recovery.  

 

DISABLE  

switches redo log files.  

 

For more information, see "Enabling and Disabling Distributed Recovery".  

ARCHIVE LOG  

manually archives redo log files or enables or disables automatic archiving. See the ARCHIVE LOG clause.  

KILL SESSION  

terminates a session and any ongoing transactions. You must identify the session with both of the following values from the V$SESSION view:  

 

integer1  

is the value of the SID column.  

 

integer2  

is the value of the SERIAL# column.  

 

For more information, see "Terminating a Session".  

DISCONNECT SESSION  

disconnects the current session by destroying the dedicated server process (or virtual circuit if the connection was made via MTS). If configured, application failover will take effect. For more information about application failover see Oracle8 Tuning and Oracle8 Parallel Server Concepts and Administration. You must identify the session with both of the following values from the V$SESSION view:  

 

integer1  

is the value of the SID column.  

 

integer2  

is the value of the SERIAL# column.  

 

POST_TRANSACTION  

allows ongoing transactions to complete before the session is disconnected. This keyword is required when DISCONNECT SESSION is specified. For more information, see "Disconnecting a Session".  

PLSQL_V2_COMPATIBILITY  

modifies the compile-time behavior of PL/SQL programs to allow language constructs that are illegal in Oracle8 (PL/SQL V3), but were legal in Oracle7 (PL/SQL V2). See the PL/SQL User's Guide and Reference and Oracle8 Reference for more information about this system parameter.  

 

TRUE  

enables Oracle8 PL/SQL V3 programs to execute Oracle7 PL/SQL V2 constructs.  

 

FALSE  

disallows illegal Oracle7 PL/SQL V2 constructs. This is the default.  

MAX_DUMP_FILE_SIZE  

specifies the trace dump file size upper limit for all user sessions. Specify the maximum size as either a nonnegative integer that represents the number of blocks, or as 'UNLIMITED'. If you specify 'UNLIMITED', no upper limit is imposed.  

 

DEFERRED  

modifies the trace dump file size upper limit for future user sessions only.  

Restricting Logons

By default, any user granted CREATE SESSION system privilege can log on to Oracle. The ENABLE RESTRICTED SESSION option of the ALTER SYSTEM command prevents logons by all users except those having RESTRICTED SESSION system privilege. Existing sessions are not terminated.

You may want to restrict logons if you are performing application maintenance and you want only application developers with RESTRICTED SESSION system privilege to log on. To restrict logons, issue the following statement:

ALTER SYSTEM
ENABLE RESTRICTED SESSION; 

You can then terminate any existing sessions using the KILL SESSION clause of the ALTER SYSTEM command.

After performing maintenance on your application, issue the following statement to allow any user with CREATE SESSION system privilege to log on:

ALTER SYSTEM
DISABLE RESTRICTED SESSION; 

Clearing the Shared Pool

The FLUSH SHARED_POOL option of the ALTER SYSTEM command clears all information from the shared pool in the system global area (SGA). The shared pool stores this information:

You might want to clear the shared pool before beginning performance analysis. To clear the shared pool, issue the following statement:

ALTER SYSTEM
FLUSH SHARED_POOL; 

The above statement does not clear shared SQL and PL/SQL areas for SQL statements, stored procedures, functions, packages, or triggers that are currently being executed, or for SQL SELECT statements for which all rows have not yet been fetched.

Performing a Checkpoint

The CHECKPOINT clause of the ALTER SYSTEM command explicitly forces Oracle to perform a checkpoint. You can force a checkpoint if you want to ensure that all changes made by committed transactions are written to the data files on disk. For more information on checkpoints, see the "Recovery Structures" chapter of Oracle8 Concepts.

If you are using Oracle with the Parallel Server option in parallel mode, you can specify either the GLOBAL option to perform a checkpoint on all instances that have opened the database or the LOCAL option to perform a checkpoint on only your instance.

The following statement forces a checkpoint:

ALTER SYSTEM 
CHECKPOINT; 

Oracle does not return control to you until the checkpoint is complete.

Checking Datafiles

The CHECK DATAFILES clause of the ALTER SYSTEM command verifies access to all online datafiles. If any datafile is not accessible, Oracle writes a message to an ALERT file. You may want to perform this operation after fixing a hardware problem that prevented an instance from accessing a datafile. For more information on using this clause, see Oracle8 Parallel Server Concepts and Administration.

The following statement verifies that all instances that have opened the database can access all online datafiles:

ALTER SYSTEM
CHECK DATAFILES GLOBAL;

Using Resource Limits

When you start an instance, Oracle enables or disables resource limits based on the value of the initialization parameter RESOURCE_LIMIT. You can issue an ALTER SYSTEM statement with the RESOURCE_LIMIT option to enable or disable resource limits for subsequent sessions.

Enabling resource limits only causes Oracle to enforce the resource limits already assigned to users. To choose resource limit values for a user, you must create a profile, or a set of limits, and assign that profile to the user. For more information on this process, see CREATE PROFILE and CREATE USER.

This ALTER SYSTEM statement dynamically enables resource limits:

ALTER SYSTEM
SET RESOURCE_LIMIT = TRUE; 

Global Name Resolution

When you start an instance, Oracle determines whether to enforce global name resolution for remote objects accessed in SQL statements based on the value of the initialization parameter GLOBAL_NAMES. You can subsequently enable or disable global name resolution while your instance is running with the GLOBAL_NAMES parameter of the ALTER SYSTEM command. You can also enable or disable global name resolution for your session with the GLOBAL_NAMES parameter of the ALTER SESSION command discussed earlier in this chapter.

Oracle recommends that you enable global name resolution if you use or plan to use distributed processing. For more information on global name resolution and how Oracle enforces it, see "Referring to Objects in Remote Databases" and Oracle8 Distributed Database Systems.

Managing Processes for the Multithreaded Server

When you start your instance, Oracle creates shared server processes and dispatcher processes for the multithreaded server architecture based on the values of the following initialization parameters:

MTS_SERVERS  

specifies the initial and minimum number of shared server processes. Oracle may automatically change the number of shared server processes if the load on the existing processes changes. While your instance is running, the number of shared server processes can vary between the values of the initialization parameters MTS_SERVERS and MTS_MAX_SERVERS.  

MTS_DISPATCHERS  

specifies one or more network protocols and the number of dispatcher processes for each protocol.  

For more information on the multithreaded server architecture, see Oracle8 Concepts.

You can use the MTS_SERVERS and MTS_DISPATCHERS parameters of the ALTER SYSTEM command to perform one of the following operations while the instance is running:

Example I

The following statement changes the minimum number of shared server processes to 25:

ALTER SYSTEM
SET MTS_SERVERS = 25; 

If there are currently fewer than 25 shared server processes, Oracle creates more. If there are currently more than 25, Oracle terminates some of them when they are finished processing their current calls if the load could be managed by the remaining 25.

Example II

The following statement dynamically changes the number of dispatcher processes for the TCP/IP protocol to 5 and the number of dispatcher processes for the DECNET protocol to 10:

ALTER SYSTEM 
SET MTS_DISPATCHERS = 'TCP, 5'
MTS_DISPATCHERS = 'DECnet, 10'; 

If there are currently fewer than 5 dispatcher processes for TCP, Oracle creates new ones. If there are currently more than 5, Oracle terminates some of them after the connected users disconnect.

If there are currently fewer than 10 dispatcher processes for DECnet, Oracle creates new ones. If there are currently more than 10, Oracle terminates some of them after the connected users disconnect.

If there are currently existing dispatchers for another protocol, the above statement does not affect the number of dispatchers for that protocol.

Using Licensing Limits

Oracle enforces concurrent usage licensing and named user licensing limits specified by your Oracle license. When you start your instance, Oracle establishes the licensing limits based on the values of the following initialization parameters:

LICENSE_MAX_SESSIONS  

establishes the concurrent usage licensing limit, or the limit for concurrent sessions. Once this limit is reached, only users with RESTRICTED SESSION system privilege can connect.  

LICENSE_SESSIONS_WARNING  

establishes a warning threshold for concurrent usage. Once this threshold is reached, Oracle writes warning messages to the database ALERT file for each subsequent session. Also, users with RESTRICTED SESSION system privilege receive warning messages when they begin subsequent sessions.  

LICENSE_MAX_USERS  

establishes the limit for users connected to your database. Once this limit is reached, more users cannot connect.  

You can dynamically change or disable limits or thresholds while your instance is running using the LICENSE_MAX_SESSIONS, LICENSE_SESSIONS_WARNING, and LICENSE_MAX_USERS parameters of the ALTER SYSTEM command. Do not disable or raise session or user limits unless you have appropriately upgraded your Oracle license. For information on upgrading your license, contact your Oracle sales representative.

New limits apply only to future sessions and users:

Example I

The following statement dynamically changes the limit on sessions for your instance to 64 and the warning threshold for sessions on your instance to 54:

ALTER SYSTEM 
SET LICENSE_MAX_SESSIONS = 64 
LICENSE_SESSIONS_WARNING = 54; 

If the number of sessions reaches 54, Oracle writes a warning message to the ALERT file for each subsequent session. Also, users with RESTRICTED SESSION system privilege receive warning messages when they begin subsequent sessions.

If the number of sessions reaches 64, only users with RESTRICTED SESSION system privilege can begin new sessions until the number of sessions falls below 64 again.

Example II

The following statement dynamically disables the limit for sessions on your instance:

ALTER SYSTEM
SET LICENSE_MAX_SESSIONS = 0; 

After you issue the above statement, Oracle no longer limits the number of sessions on your instance.

Example III

The following statement dynamically changes the limit on the number of users in the database to 200:

ALTER SYSTEM
SET LICENSE_MAX_USERS = 200; 

After you issue the above statement, Oracle prevents the number of users in the database from exceeding 200.

Switching Redo Log File Groups

The SWITCH LOGFILE option of the ALTER SYSTEM command explicitly forces Oracle to begin writing to a new redo log file group, regardless of whether the files in the current redo log file group are full. You may want to force a log switch to drop or rename the current redo log file group or one of its members, because you cannot drop or rename a file while Oracle is writing to it. The forced log switch affects only your instance's redo log thread. Note that when you force a log switch, Oracle begins to perform a checkpoint. Oracle returns control to you immediately rather than when the associated checkpoint is complete.

The following statement forces a log switch:

ALTER SYSTEM
SWITCH LOGFILE; 

Enabling and Disabling Distributed Recovery

Oracle allows you to perform distributed transactions, or transactions that modify data on multiple databases. If a network or machine failure occurs during the commit process for a distributed transaction, the state of the transaction may be unknown, or in doubt. Once the failure has been corrected and the network and its nodes are back online, Oracle recovers the transaction.

If you are using Oracle in multiple-process mode, this distributed recovery is performed automatically. If you are using Oracle in single-process (single user) mode, such as on the MS-DOS operating system, you must explicitly initiate distributed recovery with the following statement.

ALTER SYSTEM ENABLE DISTRIBUTED RECOVERY;

You may need to issue the above statement more than once to recover an in-doubt transaction, especially if the remote node involved in the transaction is not accessible. In-doubt transactions appear in the data dictionary view DBA_2PC_PENDING. You can tell that the transaction is recovered when it no longer appears in DBA_2PC_PENDING. For more information about distributed transactions and distributed recovery, see Oracle8 Distributed Database Systems.

You can disable distributed recovery in both single-process and multiprocess mode with the following statement:

ALTER SYSTEM DISABLE DISTRIBUTED RECOVERY; 

You may want to disable distributed recovery for demonstration purposes. You can then enable distributed recovery again by issuing an ALTER SYSTEM statement with the ENABLE DISTRIBUTED RECOVERY clause.

Terminating a Session

The KILL SESSION clause of the ALTER SYSTEM command terminates a session, immediately performing the following tasks:

You may want to kill the session of a user that is holding resources needed by other users. The user receives an error message indicating that the session has been killed; that user can no longer make calls to the database without beginning a new session. You can kill a session only on the same instance as your current session.

If you try to kill a session that is performing some activity that must be completed, such as waiting for a reply from a remote database or rolling back a transaction, Oracle waits for this activity to complete, kills the session, and then returns control to you. If the waiting lasts as long as a minute, Oracle marks the session to be killed and returns control to you with a message indicating that the session is marked to be killed. Oracle then kills the session when the activity is complete.

Example

Consider this data from the V$SESSION dynamic performance table:

SELECT sid, serial, username
FROM v$session 

  SID    SERIAL USERNAME
----- --------- ----------------
    1         1
    2         1
    3         1
    4         1 
    5         1 
    7         1 
    8        28 OPS$BQUIGLEY 
   10       211 OPS$SWIFT 
   11        39 OPS$OBRIEN 
   12        13 SYSTEM  
   13         8 SCOTT 

The following statement kills the session of the user SCOTT using the SID and SERIAL# values from V$SESSION:

ALTER SYSTEM
KILL SESSION '13, 8';

Disconnecting a Session

The DISCONNECT SESSION clause is similar to the KILL SESSION clause, but with two distinct differences.

First, the ALTER SYSTEM DISCONNECT SESSION 'X, Y' POST_TRANSACTION command waits until any current transaction that the session is working on completes before taking effect.

Second, the session is disconnected rather than killed, which means that the dedicated server process (or virtual circuit if the connection was made through MTS) is destroyed by this command. Termination of a session's connection causes application failover to take effect if the appropriate system parameters are configured.

Disconnecting a session essentially allows you to perform a manual application failover. Using this command in a parallel server environment allows you to disconnect sessions on an overloaded instance and shift them to another instance.

The POST_TRANSACTION keyword is required.

Example

The following statement disconnects user SCOTT's session, using the SID and SERIAL# values from V$SESSION:

ALTER SYSTEM
DISCONNECT SESSION '13, 8' POST_TRANSACTION;

For more information about application failover, see Oracle8 Parallel Server Concepts and Administration and Oracle8 Tuning.

Related Topics

ALTER SESSION
ALTER SESSION
CREATE USER
ARCHIVE LOG clause

ALTER TABLE

Purpose

To alter the definition of a table in one of the following ways:

Prerequisites

The table must be in your own schema, or you must have ALTER privilege on the table, or you must have ALTER ANY TABLE system privilege. For some operations you may also need the CREATE ANY INDEX privilege.

To use an object type in a column definition when modifying a table, either that object must belong to the same schema as the table being altered, or you must have either the EXECUTE ANY TYPE system privilege or the EXECUTE schema object privilege for the object type.

Syntax

add_column_options::=

column_constraint, table_constraint: See the CONSTRAINT clause

column_ref_clause::=

table_ref_clause::=

modify_column_options::=

physical_attributes_clause::=

storage_clause: See STORAGE clause.

LOB_storage_clause::=

LOB_parameters::=

LOB_index_clause::=

LOB_index_parameters::=

modify_LOB_storage_clause::=

modify_LOB_index_clause::=

nested_table_storage_clause::=

drop_clause: See the DROP clause.

allocate_extent_clause::=

deallocate_unused_clause: See the DEALLOCATE UNUSED clause.

index_organized_table_clauses::=

partitioning_clauses::=

rename_partition_clause::=

parallel_clause: See the PARALLEL clause.

Keywords and Parameters

schema  

is the schema containing the table. If you omit schema, Oracle assumes the table is in your own schema.  

table  

is the name of the table to be altered. You can alter the definition of an index-organized table.  

ADD  

adds a column or integrity constraint. You cannot ADD columns to an index-organized table. See also "Adding Columns".  

MODIFY  

modifies the definition of an existing column. If you omit any of the optional parts of the column definition (datatype, default value, or column constraint), these parts remain unchanged.You cannot MODIFY column definitions of index-organized tables. See also "Modifying Column Definitions".  

 

column  

is the name of the column to be added or modified.  

 

datatype  

specifies a datatype for a new column or a new datatype for an existing column.

You can omit the datatype only if the statement also designates the column as part of the foreign key of a referential integrity constraint. Oracle automatically assigns the column the same datatype as the corresponding column of the referenced key of the referential integrity constraint.  

 

DEFAULT  

specifies a default value for a new column or a new default for an existing column. Oracle assigns this value to the column if a subsequent INSERT statement omits a value for the column. If you are adding a new column to the table and specify the default value, Oracle inserts the default column value into all rows of the table.

The datatype of the default value must match the datatype specified for the column. The column must also be long enough to hold the default value. A DEFAULT expression cannot contain references to other columns, the pseudocolumns CURRVAL, NEXTVAL, LEVEL, and ROWNUM, or date constants that are not fully specified.  

 

column_constraint  

adds or removes a NOT NULL constraint to or from an existing column. See the syntax of column_constraint in the CONSTRAINT clause.  

 

table_constraint  

adds an integrity constraint to the table. See the syntax of table_constraint in the CONSTRAINT clause.

See also "REFs".  

modify_default_attributes_clause  

is a valid option only for partitioned tables. Use this option to specify new values for the default attributes of a partitioned table.  

physical_attributes_clause  

changes the value of PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters for the table, partition, the overflow data segment, or the default characteristics of a partitioned table. See the PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters of CREATE TABLE.  

 

storage_clause  

changes the storage characteristics of the table, partition, overflow data segment, or the default characteristics of a partitioned table. See the STORAGE clause.  

PCTTHRESHOLD  

specifies the percentage of space reserved in the index block for an index-organized table row. Any portion of the row that exceeds the specified threshold is stored in the overflow area. If OVERFLOW is not specified, then rows exceeding the THRESHOLD limit are rejected. PCTTHRESHOLD must be a value from 0 to 50.  

 

INCLUDING column_name  

specifies a column at which to divide an index-organized table row into index and overflow portions. All columns that follow column_name are stored in the overflow data segment. A column_name is either the name of the last primary key column or any non-primary-key column.  

 

OVERFLOW  

specifies the overflow data segment physical storage attributes to be modified for the index-organized table. Parameters specified in this clause are only applicable to the overflow data segment. See CREATE TABLE.  

 

ADD OVERFLOW  

adds an overflow data segment to the specified index-organized table.  

 

See also "Index-Organized Tables".  

LOB  

specifies the LOB storage characteristics for the newly added LOB column. You cannot use this clause to modify an existing LOB column.  

lob_item  

is the LOB column name or LOB object attribute for which you are explicitly defining tablespace and storage characteristics that are different from those of the table.  

STORE AS  

 

 

lob_segname  

specifies the name of the LOB data segment. You cannot use lob_segname if more than one lob_item is specified.  

 

ENABLE STORAGE IN ROW  

specifies that the LOB value is stored in the row (inline) if its length is less than approximately 4000 bytes minus system control information. This is the default.  

 

DISABLE STORAGE IN ROW  

specifies that the LOB value is stored outside of the row regardless of the length of the LOB value.  

 

Note that the LOB locator is always stored in the row regardless of where the LOB value is stored. You cannot change the STORAGE IN ROW once it is set.  

 

CHUNK integer  

specifies the number of bytes to be allocated for LOB manipulation. If integer is not a multiple of the database block size, Oracle rounds up (in bytes) to the next multiple. For example, if the database block size is 2048 and integer is 2050, Oracle allocates 4096 bytes (2 blocks).The maximum value is 32768 (32 K), which is the largest Oracle block size allowed.

Note: The value of CHUNK must be less than or equal to the values of both INITIAL and NEXT (either the default values or those specified in the storage clause). If CHUNK exceeds the value of either INITIAL or NEXT, Oracle returns an error.  

 

PCTVERSION integer  

is the maximum percentage of overall LOB storage space used for creating new versions of the LOB. The default value is 10, meaning that older versions of the LOB data rae not overwritten until 10% of the overall LOB storage space is used.  

 

INDEX lob_index_name  

is the name of the LOB index segment. You cannot use lob_index_name if more than one lob_item is specified.  

MODIFY LOB (lob_item)  

modifies the physical attributes of the LOB attribute lob_item or LOB object attribute. You can only specify one LOB column for each MODIFY LOB clause.

See also "LOB Columns".  

NESTED TABLE nested_item STORE AS storage_table  

 

specifies storage_table as the name of the storage table in which the rows of all nested_item values reside. You must include this clause when modifying a table with columns or column attributes whose type is a nested table.  

 

The nested_item is the name of a column or a column-qualified attribute whose type is a nested table.  

 

The storage_table is the name of the storage table. The storage table is modified in the same schema and the same tablespace as the parent table.

See also "Nested Table Columns".  

drop_clause  

drops an integrity constraint. See the DROP clause.  

ALLOCATE EXTENT  

explicitly allocates a new extent for the table, the partition, the overflow data segment, the LOB data segment, or the LOB index.  

 

SIZE  

specifies the size of the extent in bytes. You can use K or M to specify the extent size in kilobytes or megabytes. If you omit this parameter, Oracle determines the size based on the values of the table's overflow data segment's, or LOB index's STORAGE parameters.  

 

DATAFILE  

specifies one of the datafiles in the tablespace of the table, overflow data segment, LOB data tablespace, or LOB index to contain the new extent. If you omit this parameter, Oracle chooses the datafile.  

 

INSTANCE  

makes the new extent available to the freelist group associated with the specified instance. If the instance number exceeds the maximum number of freelist groups, the former is divided by the latter, and the remainder is used to identify the freelist group to be used. An instance is identified by the value of its initialization parameter INSTANCE_NUMBER. If you omit this parameter, the space is allocated to the table, but is not drawn from any particular freelist group. Rather, the master freelist is used, and space is allocated as needed. For more information, see Oracle8 Concepts. Use this parameter only if you are using Oracle with the Parallel Server option in parallel mode.  

 

Explicitly allocating an extent with this clause does affect the size for the next extent to be allocated as specified by the NEXT and PCTINCREASE storage parameters.  

DEALLOCATE UNUSED  

explicitly deallocates unused space at the end of the table, partition, overflow data segment, LOB data segment, or LOB index and makes the space available for other segments. You can free only unused space above the high-water mark. If KEEP is omitted, all unused space is freed. For more information, see DEALLOCATE UNUSED clause.  

 

KEEP  

specifies the number of bytes above the high-water mark that the table, overflow data segment, LOB data segment, or LOB index will have after deallocation. If the number of remaining extents are less than MINEXTENTS, then MINEXTENTS is set to the current number of extents. If the initial extent becomes smaller than INITIAL, then INITIAL is set to the value of the current initial extent.  

enable_clause  

enables a single integrity constraint or all triggers associated with the table. See the ENABLE clause.  

CACHE  

for data that is accessed frequently, specifies that the blocks retrieved for this table are placed at the most recently used end of the LRU list in the buffer cache when a full table scan is performed. This option is useful for small lookup tables.

CACHE is not a valid option for index-organized tables.  

NOCACHE  

for data that is not accessed frequently, specifies that the blocks retrieved for this table are placed at the least recently used end of the LRU list in the buffer cache when a full table scan is performed.

For LOBs, the LOB value is either not brought into the buffer cache or brought into the buffer cache and placed at the least recently used end of the LRU list. (The latter is the default behavior.)  

 

NOCACHE is not a valid option for index-organized tables.  

LOGGING/NOLOGGING  

LOGGING/NOLOGGING specifies that subsequent Direct Loader (SQL*Loader) and direct-load INSERT operations against a nonpartitioned table, table partition, or all partitions of a partitioned table will be logged (LOGGING) or not logged (NOLOGGING) in the redo log file.

When used with the modify_default_attributes_clause, this option affects the logging attribute of a partitioned table.  

 

LOGGING/NOLOGGING also specifies whether ALTER TABLE...MOVE and ALTER TABLE...SPLIT operations will be logged or not logged.  

 

In NOLOGGING mode, data is modified with minimal logging (to mark new extents invalid and to record dictionary changes). When applied during media recovery, the extent invalidation records mark a range of blocks as logically corrupt, because the redo data is not logged. Therefore, if you cannot afford to lose this table, it is important to take a backup after the NOLOGGING operation.  

 

If the database is run in ARCHIVELOG mode, media recovery from a backup taken before the LOGGING operation will restore the table. However, media recovery from a backup taken before the NOLOGGING operation will not restore the table.  

 

The logging attribute of the base table is independent of that of its indexes.  

 

For more information about the LOGGING option and Parallel DML, see Oracle8 Parallel Server Concepts and Administration.  

 

NOLOGGING is not a valid keyword for altering index-organized tables.  

RENAME TO  

renames table to new_table_name.  

partitioning_clauses  

See also "Modifying Table Partitions".  

MODIFY PARTITION [table partitions]  

modifies the real physical attributes of a table partition partition_name. You can specify any of the following as new physical attributes for the partition: the logging attribute; PCTFREE, PCTUSED, INITRANS, or MAXTRANS parameter; or storage parameters.  

MODIFY PARTITION [index partitions]  

modifies the attributes of an index partition partition_name. Note that you cannot specify the following options with clauses of the MODIFY PARTITION [table partitions] option.  

 

UNUSABLE LOCAL INDEXES  

marks all the local index partitions associated with partition_name as unusable.  

 

REBUILD UNUSABLE LOCAL INDEXES  

rebuilds the unusable local index partitions associated with partition_name.  

RENAME PARTITION  

renames table partition current_name to new_name.  

MOVE PARTITION  

moves table partition partition_name to another segment. You can move partition data to another tablespace, recluster data to reduce fragmentation, or change a create-time physical attribute.  

ADD PARTITION  

adds a new partition new_partition_name to the "high" end of a partitioned table. You can specify any of the following as new physical attributes for the partition: the logging attribute; the PCTFREE, PCTUSED, INITRANS, or MAXTRANS parameter; or storage parameters.  

 

VALUES LESS THAN (value_list)  

specifies the upper bound for the new partition. The value_list is a comma-separated, ordered list of literal values corresponding to column_list. The value_list must collate greater than the partition bound for the highest existing partition in the table.  

DROP PARTITION  

removes partition partition_name, and the data in that partition, from a partitioned table.  

TRUNCATE PARTITION  

removes all rows from the partition partition_name in a table.  

 

DROP STORAGE  

specifies that space from the deleted rows be deallocated and made available for use by other schema objects in the tablespace.  

 

REUSE STORAGE  

specifies that space from the deleted rows remains allocated to the partition. The space is subsequently available only for inserts and updates to the same partition.  

SPLIT PARTITION  

from an original partition partition_name_old, creates two new partitions, each with a new segment and new physical attributes, and new initial extents. The segment associated with partition_name_old is discarded.  

 

AT (value_list)  

specifies the new noninclusive upper bound for split_partition_1. The value_list must compare less than the pre-split partition bound for partition_name_old and greater than the partition bound for the next lowest partition (if there is one).  

 

INTO  

describes the two partitions resulting from the split.  

 

partition_description, partition_description  

specifies optimal names and physical attributes of the two partitions resulting from the split.  

EXCHANGE PARTITION  

converts partition partition_name into a nonpartitioned table, and a nonpartitioned table into a partition of a partitioned table by exchanging their data (and index) segments. The default behavior is EXCLUDING INDEXES WITH VALIDATION.  

 

WITH TABLE table  

specifies the table with which the partition will be exchanged.  

 

INCLUDING INDEXES  

specifies that the local index partitions be exchanged with the corresponding regular indexes.  

 

EXCLUDING INDEXES  

specifies that all the local index partitions corresponding to the partition and all the regular indexes on the exchanged table are marked as unusable.  

 

WITH VALIDATION  

specifies that any rows in the exchanged table that do not collate properly return an error.  

 

WITHOUT VALIDATION  

specifies that the proper collation of rows in the exchanged table is not checked.  

parallel_clause  

specifies the degree of parallelism for the table. PARALLEL is not a valid option for index-organized tables. See the PARALLEL clause.  

 

ENABLE TABLE LOCK  

enables DML and DDL locks on a table in a parallel server environment. For more information, see Oracle8 Parallel Server Concepts and Administration.  

 

disable_clause  

disables a single integrity constraint or all triggers associated with the tables. See the DISABLE clause.  

 

 

Integrity constraints specified in DISABLED clauses must be defined in the ALTER TABLE statements or in a previously issued statement. You can also enable and disable integrity constraints with the ENABLE and DISABLE keywords of the CONSTRAINT clause. If you define an integrity constraint but do not explicitly enable or disable it, Oracle enables it by default.  

 

DISABLE TABLE LOCK  

disables DML and DDL locks on a table to improve performance in a parallel server environment. For more information, see Oracle8 Parallel Server Concepts and Administration.  

Adding Columns

If you use the ADD clause to add a new column to the table, then the initial value of each row for the new column is null. You can add a column with a NOT NULL constraint only to a table that contains no rows.

If you create a view with a query that uses the asterisk (*) in the select list to select all columns from the base table and you subsequently add columns to the base table, Oracle will not automatically add the new column to the view. To add the new column to the view, you can re-create the view using the CREATE VIEW command with the OR REPLACE option.

Operations performed by the ALTER TABLE command can cause Oracle to invalidate procedures and stored functions that access the table. For information on how and when Oracle invalidates such objects, see Oracle8 Concepts.

Modifying Column Definitions

You can use the MODIFY clause to change any of the following parts of a column definition: datatype, size, default value, or NOT NULL column constraint.

The MODIFY clause need only specify the column name and the modified part of the definition, rather than the entire column definition.

Datatypes and Sizes

You can change a CHAR column to VARCHAR2 (or VARCHAR) and a VARCHAR2 (or VARCHAR) to CHAR only if the column contains nulls in all rows or if you do not attempt to change the column size. You can change any column's datatype or decrease any column's size if all rows for the column contain nulls. However, you can always increase the size of a character or raw column or the precision of a numeric column.

You cannot change a column's datatype to a LOB or REF datatype.

Default Values

A change to a column's default value only affects rows subsequently inserted into the table. Such a change does not change default values previously inserted.

To discontinue previously specified default values, so that they are no longer automatically inserted into newly added rows, replace the values with nulls, as shown in this example:

ALTER TABLE accounts
   MODIFY (bal DEFAULT NULL);

This statement has no effect on any existing values in existing rows.

Integrity Constraints

The only type of integrity constraint that you can add to an existing column using the MODIFY clause with the column constraint syntax is a NOT NULL constraint. However, you can define other types of integrity constraints (UNIQUE, PRIMARY KEY, referential integrity, and CHECK constraints) on existing columns using the ADD clause and the table constraint syntax.

You can define a NOT NULL constraint on an existing column only if the column contains no nulls.

Example I

The following statement adds a column named THRIFTPLAN of datatype NUMBER with a maximum of seven digits and two decimal places and a column named LOANCODE of datatype CHAR with a size of one and a NOT NULL integrity constraint:

ALTER TABLE emp 
   ADD (thriftplan NUMBER(7,2),
        loancode CHAR(1) NOT NULL); 
Example II

The following statement increases the size of the THRIFTPLAN column to nine digits:

ALTER TABLE emp
   MODIFY (thriftplan NUMBER(9,2)); 

Because the MODIFY clause contains only one column definition, the parentheses around the definition are optional.

Example III

The following statement changes the values of the PCTFREE and PCTUSED parameters for the EMP table to 30 and 60, respectively:

ALTER TABLE emp 
    PCTFREE 30
    PCTUSED 60; 
Example IV

The following statement allocates an extent of 5 kilobytes for the EMP table and makes it available to instance 4:

ALTER TABLE emp
  ALLOCATE EXTENT (SIZE 5K INSTANCE 4); 

Because this command omits the DATAFILE parameter, Oracle allocates the extent in one of the datafiles belonging to the tablespace containing the table.

Example V

This example modifies the BAL column of the ACCOUNTS table so that it has a default value of 0:

ALTER TABLE accounts
  MODIFY (bal  DEFAULT 0); 

If you subsequently add a new row to the ACCOUNTS table and do not specify a value for the BAL column, the value of the BAL column is automatically 0:

INSERT INTO accounts(accno, accname)
  VALUES (accseq.nextval, 'LEWIS') 
  SELECT * 
    FROM accounts
    WHERE accname = 'LEWIS'; 

ACCNO  ACCNAME BAL 
------ ------- ---
815234  LEWIS     0 

Index-Organized Tables

Index-organized tables are special kinds of tables that keep data sorted on the primary key and are therefore best suited for primary-key-based access and manipulation.

You cannot ADD columns to an index-organized table, but you can alter the definition of an index-organized table.

Example I

This example modifies the INITRANS parameter for the index segment of index-organized table DOCINDEX:

ALTER TABLE docindex INITRANS 4;
Example II

The following statement adds an overflow data segment to index-organized table DOCINDEX:

ALTER TABLE docindex ADD OVERFLOW;
Example III

This example modifies the INITRANS parameter for the overflow data segment of index-organized table DOCINDEX:

ALTER TABLE docindex OVERFLOW INITRANS 4;

LOB Columns

You can add a LOB column to a table, or modify the LOB data segment or index storage characteristics.

Example I

The following statement adds CLOB column RESUME to the EMPLOYEE table:

ALTER TABLE employee ADD (resume CLOB)
  LOB (resume) STORE AS resume_seg (TABLESPACE resume_ts);                   
Example II

To modify the LOB column RESUME to use caching, enter the following statement:

ALTER TABLE employee MODIFY LOB (resume) (CACHE); 

Nested Table Columns

You can add a nested table type column to a table. Specify a nested table storage clause for each column added.

Example I

The following example adds the nested table column SKILLS to the EMPLOYEE table:

ALTER TABLE employee ADD (skills skill_table_type)
    NESTED TABLE skills STORE AS nested_skill_table;

You can also modify a nested table's storage characteristics. Use the name of the storage table specified in the nested table storage clause to make the modification. You cannot query or perform DML statements on the storage table; only use the storage table to modify the nested table column storage characteristics.

Example II

The following example creates table VETSERVICE with nested table column CLIENT and storage table CLIENT_TAB. Nested table VETSERVICE is modified to specify constraints and modify a column length by altering nested storage table CLIENT_TAB:

CREATE TABLE vetservice (vet_name VARCHAR2(30),
                         client   pet_table)
  NESTED TABLE client STORE AS client_tab;
ALTER TABLE client_tab ADD UNIQUE (ssn);
ALTER TABLE client_tab MODIFY (pet_name VARCHAR2(35));
Example IV

The following statement adds a UNIQUE constraint to nested table NESTED_SKILL_TABLE:

ALTER TABLE nested_skill_table ADD UNIQUE (a);

For more information about nested table storage see the CREATE TABLE. For more information about nested tables, see Oracle8 Application Developer's Guide.

Example V

The following example alters the storage table for a nested table of REF values to specify that the REF is scoped:

CREATE TYPE emp_t AS OBJECT ( eno number, ename char(31)); 
CREATE TYPE emps_t AS TABLE OF REF emp_t; 
CREATE TABLE emptab OF emp_t; 
CREATE TABLE dept (dno NUMBER, employees EMPS_T) 
   NESTED TABLE employees STORE AS deptemps; 
ALTER TABLE deptemps ADD(SCOPE FOR (column_value) IS emptab); 

Similarly, to specify storing the REF with ROWID:

ALTER TABLE deptemps ADD (REF(column_value) WITH ROWID); 

Note that in order to execute these ALTER TABLE statements successfully, the storage table DEPTEMPS must be empty. Also, note that because the nested table is defined as a table of scalars (REFs), Oracle implicitly provides the column name COLUMN_VALUE for the storage table.

REFs

A REF value is a reference to a row in an object table. A table can have top-level REF columns or it can have REF attributes embedded within an object column. In general, if a table has a REF column, each REF value in the column could reference a row in a different object table. A SCOPE clause restricts the scope of references to a single table.

Use the ALTER TABLE command to add new REF columns or to add REF clauses to existing REF columns. You can modify any table, including named inner nested tables (storage tables). If a REF column is created WITH ROWID or with a scope table, you cannot modify the column to drop these options. However, if a table is created without any REF clauses, you can add them later with an ALTER TABLE statement.

Note: You can add a scope clause to existing REF columns of a table only if the table is empty. The scope_table_name must be in your own schema or you must have SELECT privilege on the table, or the SELECT ANY TABLE system privilege. This privilege is needed only while altering the table with the REF column.

Example I

In the following example an object type DEPT_T has been previously defined. Now, create table EMP as follows:

CREATE TABLE emp 
   (name VARCHAR(100), 
    salary NUMBER,
    dept REF dept_t); 

An object table DEPARTMENTS is created as:

CREATE TABLE departments OF dept_t; 

If the DEPARTMENTS table contains all possible departments, the DEPT column in EMP can only refer to rows in the DEPARTMENTS table. This can be expressed as a scope clause on the DEPT column as follows:

ALTER TABLE emp 
    ADD (SCOPE FOR (dept) IS departments); 

Note that the above ALTER TABLE statement will succeed only if the EMP table is empty.

Example II

If you want the REF values in the DEPT column of EMP to also store the ROWIDs, issue the following statement:

   ALTER TABLE emp 
    ADD (REF(dept) WITH ROWID);

Modifying Table Partitions

You can modify a table or table partition in any of the following ways. You cannot combine partition operations with other partition operations or with operations on the base table in one ALTER TABLE statement.

ADD PARTITION

Use ALTER TABLE ADD PARTITION to add a partition to the high end of the table (after the last existing partition). If the first element of the partition bound of the high partition is MAXVALUE, you cannot add a partition to the table. You must split the high partition.

You can add a partition to a table even if one or more of the table indexes or index partitions are marked UNUSABLE.

You must use the SPLIT PARTITION clause to add a partition at the beginning or the middle of the table.

The following example adds partition JAN97 to tablespace TSX:

ALTER TABLE sales 
  ADD PARTITION jan97 VALUES LESS THAN( '970201' ) 
  TABLESPACE tsx; 

DROP PARTITION

ALTER TABLE DROP PARTITION drops a partition and its data. If you want to drop a partition but keep its data in the table, you must merge the partition into one of the adjacent partitions. For information about merging two tables partitions, see the Oracle8 Administrator's Guide.

If you drop a partition and later insert a row that would have belonged to the dropped partition, the row will be stored in the next higher partition. However, if you drop the highest partition, the insert will fail because the range of values represented by the dropped partition is no longer valid for the table.

This statement also drops the corresponding partition in each local index defined on table. The index partitions are dropped even if they are marked as unusable.

If there are global indexes defined on table, and the partition you want to drop is not empty, dropping the partition marks all the global, nonpartitioned indexes and all the partitions of global partitioned indexes as unusable.

When a table contains only one partition, you cannot drop the partition. You must drop the table.

The following example drops partition DEC95:

ALTER TABLE sales DROP PARTITION dec95;

EXCHANGE PARTITION

This form of ALTER TABLE converts a partition to a nonpartitioned table and a NONPARTITIONED table to a partition by exchanging their data segments. You must have ALTER TABLE privileges on both tables to perform this operation.

The statistics of the table and partition-including table, column, index statistics and histograms-are exchanged. The aggregate statistics of the partitioned table are recalculated.

The logging attribute of the table and partition is exchanged.

The following example converts partition FEB97 to table SALES_FEB97 without exchanging local index partitions with corresponding indexes on SALES_FEB97 and without verifying that data in SALES_FEB97 falls within the bounds of partition FEB97:

ALTER TABLE sales 
  EXCHANGE PARTITION feb97 WITH TABLE sales_feb97 
   WITHOUT VALIDATION;

MODIFY PARTITION

Use the MODIFY PARTITION options of ALTER TABLE to

The following example marks all the local index partitions corresponding to the NOV96 partition of the SALES table UNUSABLE:

ALTER TABLE sales MODIFY PARTITION nov96 
  UNUSABLE LOCAL INDEXES;

The following example rebuilds all the local index partitions that were marked UNUSABLE:

ALTER TABLE sales MODIFY PARTITION jan97
  REBUILD UNUSABLE LOCAL INDEXES;

The following example changes MAXEXTENTS and logging attribute for partition BRANCH_NY:

ALTER TABLE branch MODIFY PARTITION branch_ny  
  STORAGE(MAXEXTENTS 75) LOGGING;

MOVE PARTITION

This ALTER TABLE option moves a table partition to another segment. MOVE PARTITION always drops the partition's old segment and creates a new segment, even if you do not specify a new tablespace.

If partition partition_name is not empty, MOVE PARTITION marks all corresponding local index partitions and all global nonpartitioned indexes, and all the partitions of global partitioned indexes as unusable.

ALTER TABLE MOVE PARTITION obtains its parallel attribute from the PARALLEL clause, if specified. If not specified, the default PARALLEL attributes of the table, if any, are used. If neither is specified, it performs the move without using parallelism.

The PARALLEL clause on MOVE PARTITION does not change the default PARALLEL attributes of table.

The following example moves partition DEPOT2 to tablespace TS094:

ALTER TABLE parts 
  MOVE PARTITION depot2 TABLESPACE ts094 NOLOGGING;

RENAME

Use the RENAME option of ALTER TABLE to rename a table or to rename a partition.

The following example renames a table:

ALTER TABLE emp RENAME TO employee;

In the following example, partition EMP3 is renamed:

ALTER TABLE employee RENAME PARTITION emp3 TO employee3;

SPLIT PARTITION

The SPLIT PARTITION option divides a partition into two partitions. A new segment is allocated for each partition resulting from the split. The attributes of the new partitions are inherited from the partition that was split, except for attributes whose values you explicitly override in the SPLIT clause. The segment associated with the old partition is discarded.

This statement also performs a matching split on the corresponding partition in each local index defined on table. The index partitions are split even if they are marked unusable.

With the exception of the TABLESPACE attribute, the physical attributes of the LOCAL index partition being split are used for both new index partitions. If the parent LOCAL index lacks a default TABLESPACE attribute, new LOCAL index partitions will reside in the same tablespace as the corresponding newly created partitions of the underlying table.

If you do not specify physical attributes (PCTFREE, PCTUSED, INITRANS, MAXTRANS, STORAGE) for the new partitions, the current values of the partition being split are used as the default values for both partitions.

If partition_name is not empty, SPLIT PARTITION marks all affected index partitions as unusable. This includes all global index partitions as well as the local index partitions that result from the split.

The PARALLEL clause on SPLIT PARTITION does not change the default PARALLEL attributes of table.

The following example splits the old partition DEPOT4, creating two new partitions, naming one DEPOT9 and reusing the name of the old partition for the other:

ALTER TABLE parts
  SPLIT PARTITION depot4 AT ( '40-001' )
  INTO ( PARTITION depot4 TABLESPACE ts009 (MINEXTENTS 2),
         PARTITION depot9 TABLESPACE ts010 )
  PARALLEL ( DEGREE 10 );

TRUNCATE PARTITION

Use TRUNCATE PARTITION to remove all rows from a partition in a table. Freed space is deallocated or reused depending on whether DROP STORAGE or REUSE STORAGE is specified in the clause.

This statement truncates the corresponding partition in each local index defined on table. The local index partitions are truncated even if they are marked as unusable. The unusable local index partitions are marked valid, resetting the UNUSABLE indicator.

If any global indexes are defined on table, and the partition you want to truncate is not empty, truncating the partition marks all the global nonpartitioned indexes and all the partitions of global partitioned indexes as unusable.

If you want to truncate a partition that contains data, you must first disable any referential integrity constraints on the table. Alternatively, you can delete the rows and then truncate the partition.

The following example deletes all the data in the SYS_P017 partition and deallocates the freed space:

ALTER TABLE deliveries
  TRUNCATE PARTITION sys_p017 DROP STORAGE;

For examples of defining integrity constraints with the ALTER TABLE command, see the CONSTRAINT clause.

For examples of enabling, disabling, and dropping integrity constraints and triggers with the ALTER TABLE command, see the ENABLE clause, the DISABLE clause, and the DROP clause.

For examples of changing the value of a table's storage parameters, see the STORAGE clause.

Related Topics

CREATE TABLE
CONSTRAINT clause
DISABLE clause
DISABLE clause
ENABLE clause
STORAGE clause
CREATE VIEW

ALTER TABLESPACE

Purpose

To alter an existing tablespace in one of the following ways:

See also "Using ALTER TABLESPACE".

Prerequisites

If you have ALTER TABLESPACE system privilege, you can perform any of this command's operations. If you have MANAGE TABLESPACE system privilege, you can only perform the following operations:

Before you can make a tablespace read-only, the following conditions must be met. Performing this function in restricted mode may help you meet these restrictions, since only users with RESTRICTED SESSION system privilege can be logged on.

Syntax

filespec: See "Filespec".

storage_clause: See STORAGE clause.

Keywords and Parameters

tablespace  

is the name of the tablespace to be altered.  

LOGGING/ NOLOGGING  

specifies the default logging attribute of all tables, indexes, and partitions within the tablespace.

The tablespace-level logging attribute can be overridden by logging specifications at the table, index, and partition levels.  

 

When an existing tablespace logging attribute is changed by an ALTER TABLESPACE statement, all tables, indexes, and partitions created after the statement will have the new default logging attribute (which you can still subsequently override); the logging attributes of existing objects are not changed.  

 

Only the following operations support NOLOGGING mode:

  • DML: direct-load INSERT (serial or parallel); Direct Loader (SQL*Loader)
 

 

  • DDL: CREATE TABLE... AS SELECT, CREATE INDEX, ALTER INDEX... REBUILD, ALTER INDEX... REBUILD PARTITION, ALTER INDEX... SPLIT PARTITION, ALTER TABLE... SPLIT PARTITION, ALTER TABLE... MOVE PARTITION.
 

 

In NOLOGGING mode, data is modified with minimal logging (to mark new extents invalid and to record dictionary changes). When applied during media recovery, the extent invalidation records mark a range of blocks as logically corrupt, because the redo data is not logged. Therefore, if you cannot afford to lose the object, it is important to take a backup after the NOLOGGING operation.  

ADD DATAFILE  

adds the datafile specified by filespec to the tablespace. (See the syntax description of Filespec). You can add a datafile while the tablespace is online or offline. Be sure that the datafile is not already in use by another database.  

AUTOEXTEND  

enables or disables the autoextending of the size of the datafile in the tablespace.  

 

OFF  

disables autoextend if it is turned on. NEXT and MAXSIZE are set to zero. Values for NEXT and MAXSIZE must be respecified in further ALTER TABLESPACE AUTOEXTEND commands.  

 

ON  

enables autoextend.  

 

NEXT  

specifies the size in bytes of the next increment of disk space to be allocated automatically to the datafile when more extents are required. You can use K or M to specify this size in kilobytes or megabytes. The default is one data block.  

 

MAXSIZE  

specifies maximum disk space allowed for automatic extension of the datafile.  

 

UNLIMITED  

sets no limit on allocating disk space to the datafile.  

RENAME DATAFILE  

renames one or more of the tablespace's datafiles. Take the tablespace offline before renaming the datafile. Each 'filename' must fully specify a datafile using the conventions for filenames on your operating system.  

 

This clause only associates the tablespace with the new file rather than the old one. This clause does not actually change the name of the operating system file. You must change the name of the file through your operating system.  

COALESCE  

for each datafile in the tablespace, coalesces all contiguous free extents into larger contiguous extents.  

 

COALESCE cannot be specified with any other command option.  

DEFAULT storage_clause  

specifies the new default storage parameters for objects subsequently created in the tablespace. See the STORAGE clause.  

MINIMUM EXTENT integer  

controls free space fragmentation in the tablespace by ensuring that every used and/or free extent size in a tablespace is at least as large as, and is a multiple of, integer. For more information about using MINIMUM EXTENT to control space fragmentation, see Oracle8 Administrator's Guide.  

ONLINE  

brings the tablespace online.  

OFFLINE  

takes the tablespace offline and prevents further access to its segments.  

 

NORMAL  

performs a checkpoint for all datafiles in the tablespace. All of these datafiles must be online. This is the default. You need not perform media recovery on this tablespace before bringing it back online. You must use this option if the database is in NOARCHIVELOG mode.  

 

TEMPORARY  

performs a checkpoint for all online datafiles in the tablespace but does not ensure that all files can be written. Any offline files may require media recovery before you bring the tablespace back online.  

 

IMMEDIATE  

does not ensure that tablespace files are available and does not perform a checkpoint. You must perform media recovery on the tablespace before bringing it back online.  

 

FOR RECOVER  

takes the production database tablespaces in the recovery set offline. Use this option when one or more datafiles in the tablespace are unavailable.  

Suggestion: Before taking a tablespace offline for a long time, you may want to alter any users who have been assigned the tablespace as either a default or temporary tablespace. When the tablespace is offline, these users cannot allocate space for objects or sort areas in the tablespace. You can reassign to such users new default and temporary tablespaces with the ALTER USER command.  

BEGIN BACKUP  

signifies that an open backup is to be performed on the datafiles that make up this tablespace. This option does not prevent users from accessing the tablespace. You must use this option before beginning an open backup. You cannot use this option on a read-only tablespace.  

 

Note: While the backup is in progress, you cannot: take the tablespace offline normally, shutdown the instance, or begin another backup of the tablespace.  

END BACKUP  

signifies that an open backup of the tablespace is complete. Use this option as soon as possible after completing an open backup. You cannot use this option on a read-only tablespace.

If you forget to indicate the end of an online tablespace backup, and an instance failure or SHUTDOWN ABORT occurs, Oracle assumes that media recovery (possibly requiring archived redo log) is necessary at the next instance start up. To restart the database without media recovery, see Oracle8 Administrator's Guide.  

READ ONLY  

signifies that no further write operations are allowed on the tablespace. The tablespace becomes read only.

Once a tablespace is read-only, you can copy its files to read-only media. You must then rename the datafiles in the control file to point to the new location by using the SQL command ALTER DATABASE RENAME.  

READ WRITE  

signifies that write operations are allowed on a previously read-only tablespace.  

PERMANENT  

specifies that the tablespace is to be converted from a temporary to a permanent one. A permanent tablespace is one wherein permanent database objects can be stored. This is the default when a tablespace is created.  

TEMPORARY  

specifies that the tablespace is to be converted from a permanent to a temporary one. A temporary tablespace is one in which no permanent database objects can be stored.  

Using ALTER TABLESPACE

The following examples illustrate the use of the ALTER TABLESPACE COMMAND.

Example I

The following statement signals to the database that a backup is about to begin:

ALTER TABLESPACE accounting 
    BEGIN BACKUP; 
Example II

The following statement signals to the database that the backup is finished:

ALTER TABLESPACE accounting 
    END BACKUP; 
Example III

This example moves and renames a datafile associated with the ACCOUNTING tablespace from 'DISKA:PAY1.DAT' to 'DISKB:RECEIVE1.DAT':

  1. Take the tablespace offline using an ALTER TABLESPACE statement with the OFFLINE option:
    ALTER TABLESPACE accounting OFFLINE NORMAL; 
    
    
  2. Copy the file from 'DISKA:PAY1.DAT' to 'DISKB:RECEIVE1.DAT' using your operating system's commands.
  3. Rename the datafile using the ALTER TABLESPACE command with the RENAME DATAFILE clause:
    ALTER TABLESPACE accounting
      RENAME DATAFILE 'diska:pay1.dbf'
      TO     'diskb:receive1.dbf'; 
    
    
  4. Bring the tablespace back online using an ALTER TABLESPACE statement with the ONLINE option:
    ALTER TABLESPACE accounting ONLINE; 
    
    
Example IV

The following statement adds a datafile to the tablespace and changes the default logging attribute to NOLOGGING; when more space is needed new extents of size 10 kilobytes will be added up to a maximum of 100 kilobytes:

ALTER TABLESPACE accounting NOLOGGING
    ADD DATAFILE 'disk3:pay3.dbf'
    AUTOEXTEND ON
    NEXT 10 K
    MAXSIZE 100 K;

Altering a tablespace logging attribute has no affect on the logging attributes of the existing schema objects within the tablespace. The tablespace-level logging attribute can be overridden by logging specifications at the table, index, and partition levels.

Example V

The following statement changes the allocation of every extent of TABSPACE_ST to a multiple of 128K:

ALTER TABLESPACE tabspace_st MINIMUM EXTENT 128K;

Related Topics

CREATE TABLESPACE
CREATE DATABASE
DROP TABLESPACE
STORAGE clause
"Filespec"

ALTER TRIGGER

Purpose

To enable, disable, or compile a database trigger.

Prerequisites

The trigger must be in your own schema or you must have ALTER ANY TRIGGER system privilege.

Syntax

Keywords and Parameters

schema  

is the schema containing the trigger. If you omit schema, Oracle assumes the trigger is in your own schema.  

trigger  

is the name of the trigger to be altered. See also "Invalid Triggers".  

ENABLE  

enables the trigger. See also "Enabling and Disabling Triggers".  

DISABLE  

disables the trigger. See also "Enabling and Disabling Triggers".  

COMPILE  

compiles the trigger.  

 

DEBUG  

instructs the PL/SQL compiler to generate and store the code for use by the PL/SQL debugger. This option can be used for normal triggers and for instead-of triggers.  

Invalid Triggers

You can use the ALTER TRIGGER command to explicitly recompile a trigger that is invalid. Explicit recompilation eliminates the need for implicit run-time recompilation and prevents associated run-time compilation errors and performance overhead.

When you issue an ALTER TRIGGER statement, Oracle recompiles the trigger regardless of whether it is valid or invalid.

When you recompile a trigger, Oracle first recompiles objects upon which the trigger depends, if any of these objects are invalid. If Oracle recompiles the trigger successfully, the trigger becomes valid. If recompiling the trigger results in compilation errors, then Oracle returns an error and the trigger remains invalid. You can then debug triggers using the predefined package DBMS_OUTPUT. For information on debugging procedures, see Oracle8 Application Developer's Guide. For information on how Oracle maintains dependencies among schema objects, including remote objects, see Oracle8 Concepts.


Note:

This command does not change the declaration or definition of an existing trigger. To redeclare or redefine a trigger, you must use the CREATE TRIGGER command with the OR REPLACE option.

 

Enabling and Disabling Triggers

A database trigger is always either enabled or disabled. If a trigger is enabled, Oracle fires the trigger when a triggering statement is issued. If the trigger is disabled, Oracle does not fire the trigger when a triggering statement is issued.

When you create a trigger, Oracle enables it automatically. You can use the ENABLE and DISABLE options of the ALTER TRIGGER command to enable and disable a trigger.

You can also use the ENABLE and DISABLE clauses of the ALTER TABLE command to enable and disable all triggers associated with a table.


Note:

The ALTER TRIGGER command does not change the definition of an existing trigger. To redefine a trigger, you must use the CREATE TRIGGER command with the OR REPLACE option.

 

Example

Consider a trigger named REORDER created on the INVENTORY table. The trigger is fired whenever an UPDATE statement reduces the number of a particular part on hand below the part's reorder point. The trigger inserts into a table of pending orders a row that contains the part number, a reorder quantity, and the current date.

When this trigger is created, Oracle enables it automatically. You can subsequently disable the trigger with the following statement:

ALTER TRIGGER reorder
    DISABLE;
 

When the trigger is disabled, Oracle does not fire the trigger when an UPDATE statement causes the part's inventory to fall below its reorder point.

After disabling the trigger, you can subsequently enable it with the following statement:

ALTER TRIGGER reorder
    ENABLE; 

After you reenable the trigger, Oracle fires the trigger whenever a part's inventory falls below its reorder point as a result of an UPDATE statement. Note that a part's inventory may have fallen below its reorder point while the trigger was disabled. When you reenable the trigger, Oracle does not automatically fire the trigger for this part until another transaction further reduces the inventory.

Related Topics

CREATE TRIGGER
DROP TRIGGER
DISABLE clause
ENABLE clause
ALTER TABLE

ALTER TYPE

Purpose

To recompile the specification and/or body, or to change the specification of an object type by adding new object member subprogram specifications.


Note:

This command is available only if the Oracle objects option is installed on your database server.

 

Prerequisites

The object type must be in your own schema and you must have CREATE TYPE or CREATE ANY TYPE system privilege, or you must have ALTER ANY TYPE system privileges.

Syntax

Keywords and Parameters

schema  

is the schema that contains the type. If you omit schema, Oracle creates the type in your current schema.  

type_name  

is the name of an object type, a nested table type, or a VARRAY type.  

COMPILE  

compiles the object type specification and body. This is the default if no option is specified.  

 

SPECIFICATION  

compiles only the object type specification.  

 

BODY  

compiles only the object type body.  

REPLACE AS OBJECT  

adds new member subprogram specifications. This option is valid only for object types.  

attribute_name  

is an object attribute name. Attributes are data items with a name and a type specifier that form the structure of the object.  

MAP/ORDER MEMBER function_specification  

 

MAP  

specifies a member function (MAP method) that returns the relative position of a given instance in the ordering of all instances of the object. A map method is called implicitly and induces an ordering of object instances by mapping them to values of a predefined scalar type. PL/SQL uses the ordering to evaluate Boolean expressions and to perform comparisons.  

 

 

A scalar value is always manipulated as a single unit. Scalars are mapped directly to the underlying hardware. An integer, for example, occupies 4 or 8 contiguous bytes of storage, in memory or on disk.  

 

 

An object specification can contain only one map method, which must be a function. The result type must be a predefined SQL scalar type, and the map function can have no arguments other than the implicit SELF argument.  

 

ORDER  

specifies a member function (ORDER method) that takes an instance of an object as an explicit argument and the implicit SELF argument and returns either a negative, zero, or positive integer. The negative, zero, or positive indicates that the implicit SELF argument is less than, equal to, or greater than the explicit argument.  

 

 

When instances of the same object type definition are compared in an ORDER BY clause, the order method function_specification is invoked.  

 

 

An object specification can contain only one ORDER method, which must be a function having the return type INTEGER.  

 

You can declare either a MAP method or an ORDER method, but not both. If you declare either method, you can compare object instances in SQL.  

 

If you do not declare either method, you can compare object instances only for equality or inequality. Note that instances of the same type definition are equal only if each pair of their corresponding attributes is equal. No comparison method needs to be specified to determine the equality of two object types. For more information about object value comparisons, "Object Values".  

MEMBER  

specifies a function or procedure subprogram associated with the object type which is referenced as an attribute. For information about overloading subprogram names within a package, see the PL/SQL User's Guide and Reference. See also "Restriction".

You must specify a corresponding method body in the object type body for each procedure or function specification. See CREATE TYPE BODY.  

 

procedure_specification  

is the specification of a procedure subprogram.  

 

function_specification  

is the specification of a function subprogram.  

PRAGMA RESTRICT_REFERENCES  

is a complier directive that denies member functions read/write access to database tables, packaged variables, or both, and thereby helps to avoid side effects. For more information, see the PL/SQL User's Guide and Reference.  

 

method_name  

is the name of the MEMBER function or procedure to which the pragma is being applied.  

 

WNDS  

specifies constraint writes no database state (does not modify database tables).  

 

WNPS  

specifies constraint writes no package state (does not modify packaged variables).  

 

RNDS  

specifies constraint reads no database state (does not query database tables).  

 

RNPS  

specifies constraint reads no package state (does not reference packages variables).  

Restriction

You cannot change the existing properties (attributes, member subprograms, map or order functions) of an object type, but you can add new member subprogram specifications.

Example I

In the following example, member function QTR is added to the type definition of DATA_T:

CREATE TYPE data_t AS OBJECT 
   ( year NUMBER, 
     MEMBER FUNCTION prod(invent NUMBER) RETURN NUMBER 
   ); 
 
   CREATE TYPE BODY data_t IS   
      MEMBER FUNCTION prod (invent NUMBER) RETURN NUMBER IS 
        BEGIN 
           RETURN (year + invent); 
        END; 
      END; 
 
   ALTER TYPE data_t REPLACE AS OBJECT 
   ( year NUMBER, 
     MEMBER FUNCTION  prod(invent NUMBER) RETURN NUMBER, 
     MEMBER FUNCTION  qtr(der_qtr DATE) RETURN CHAR 
   ); 
 
   CREATE OR REPLACE TYPE BODY data_t IS   
      MEMBER FUNCTION prod (invent NUMBER) RETURN NUMBER IS 
        BEGIN 
           RETURN (year + invent); 
        END; 
      MEMBER FUNCTION qtr(der_qtr DATE) RETURN CHAR IS 
        BEGIN 
           RETURN 'FIRST'; 
        END; 
      END; 
Example II

The following example recompiles type LOAN_T:

CREATE TYPE loan_t AS OBJECT
  ( loan_num        INTEGER,
    interest_rate   FLOAT,
    amount          FLOAT,
    start_date      DATE,
    end_date        DATE );

ALTER TYPE loan_t COMPILE;
Example III

The following example compiles the type body of LINK2:

CREATE TYPE link1 AS OBJECT
  (a NUMBER); 

CREATE TYPE link2 AS OBJECT
  (a NUMBER, 
   b link1, 
   MEMBER FUNCTION p(c1 NUMBER) RETURN NUMBER); 

CREATE TYPE BODY link2 AS
   MEMBER FUNCTION p(c1 NUMBER) RETURN NUMBER IS t13 link1; 
     BEGIN t13 := link1(13); 
      dbms_output.put_line(t13.a);
        RETURN 5; 
     END;
END; 

CREATE TYPE link3 AS OBJECT (a link2); 
CREATE TYPE link4 AS OBJECT (a link3); 
CREATE TYPE link5 AS OBJECT (a link4); 
ALTER TYPE link2 COMPILE BODY; 
Example IV

The following example compiles the type specification of LINK2:

CREATE TYPE link1 AS OBJECT
  (a NUMBER); 

CREATE TYPE link2 AS OBJECT
  (a NUMBER, 
   b link1, 
   MEMBER FUNCTION p(c1 NUMBER) RETURN NUMBER); 

CREATE TYPE BODY link2 AS
    MEMBER FUNCTION p(c1 NUMBER) RETURN NUMBER IS t14 link1;
      BEGIN t14 := link1(14); 
      dbms_output.put_line(t14.a);
       RETURN 5; 
      END;
END; 

CREATE TYPE link3 AS OBJECT (a link2); 
CREATE TYPE link4 AS OBJECT (a link3); 
CREATE TYPE link5 AS OBJECT (a link4); 
ALTER TYPE link2 COMPILE SPECIFICATION;

Related Topics

CREATE TYPE
CREATE TYPE BODY
PL/SQL User's Guide and Reference
Oracle8 Application Developer's Guide

ALTER USER

Purpose

To change any of the following characteristics of a database user:

Prerequisites

You must have the ALTER USER system privilege. However, you can change your own password without this privilege.

Syntax

Keywords and Parameters

The keywords and parameters in the ALTER USER command all have the same meaning as in the CREATE USER command. For information on these keywords and parameters, see CREATE USER.

For more information on default roles, see "Establishing Default Roles". For more information on security domains, see "Changing Authentication Methods".

Establishing Default Roles

The DEFAULT ROLE clause can only contain roles that have been granted directly to the user with a GRANT statement. You cannot use the DEFAULT ROLE clause to enable:

Note that Oracle enables default roles at logon without requiring the user to specify their passwords.

Example I

The following statement changes the user SCOTT's password to LION and default tablespace to the tablespace TSTEST:

ALTER USER scott 
    IDENTIFIED BY lion
    DEFAULT TABLESPACE tstest; 
Example II

The following statement assigns the CLERK profile to SCOTT:

ALTER USER scott 
    PROFILE clerk; 

In subsequent sessions, SCOTT is restricted by limits in the CLERK profile.

Example III

The following statement makes all roles granted directly to SCOTT default roles, except the AGENT role:

ALTER USER scott 
    DEFAULT ROLE ALL EXCEPT agent; 

At the beginning of SCOTT's next session, Oracle enables all roles granted directly to SCOTT except the AGENT role.

Changing Authentication Methods

You can change a user's access verification method to IDENTIFIED GLOBALLY AS 'external_name' only if all external roles granted directly to the user are revoked.

You can change a user created as IDENTIFIED GLOBALLY AS 'external_name' to IDENTIFIED BY password or IDENTIFIED EXTERNALLY.

Example I

The following example changes user TOM's authentication mechanism:

ALTER USER tom IDENTIFIED GLOBALLY AS 'CN=tom';
Example II

The following example causes user FRED's password to expire:

ALTER USER fred PASSWORD EXPIRE;

If you cause a database user's password to expire with PASSWORD EXPIRE, the user must change the password before attempting to log in to the database following the expiration. However, tools such as SQL*Plus allow you to change the password on the first attempted login following the expiration.

Related Topics

CREATE PROFILE
CREATE ROLE
CREATE USER

CREATE TABLESPACE


ALTER VIEW

Purpose

To recompile a view or an object view. See also "Recompiling Views".


Note: :

Descriptions of commands and clauses preceded by are available only if the Oracle objects option is installed on your database server.  


Prerequisites

The view must be in your own schema or you must have ALTER ANY TABLE system privilege.

Syntax

Keywords and Parameters

schema

 

is the schema containing the view. If you omit schema, Oracle assumes the view is in your own schema.

 

view

 

is the name of the view to be recompiled.

 

COMPILE

 

causes Oracle to recompile the view. The COMPILE keyword is required.

 

Recompiling Views

You can use the ALTER VIEW command to explicitly recompile a view that is invalid. Explicit recompilation allows you to locate recompilation errors before run time. You may want to explicitly recompile a view after altering one of its base tables to ensure that the alteration does not affect the view or other objects that depend on it.

When you issue an ALTER VIEW statement, Oracle recompiles the view regardless of whether it is valid or invalid. Oracle also invalidates any local objects that depend on the view. For more about dependencies among schema objects, see Oracle8 Concepts.


Note:

This command does not change the definition of an existing view. To redefine a view, you must use the CREATE VIEW command with the OR REPLACE option.

 

Example

To recompile the view CUSTOMER_VIEW, issue the following statement:

ALTER VIEW customer_view
    COMPILE; 

If Oracle encounters no compilation errors while recompiling CUSTOMER_VIEW, CUSTOMER_VIEW becomes valid. If recompiling results in compilation errors, Oracle returns an error and CUSTOMER_VIEW remains invalid.

Oracle also invalidates all dependent objects. These objects include any procedures, functions, package bodies, and views that reference CUSTOMER_VIEW. If you subsequently reference one of these objects without first explicitly recompiling it, Oracle recompiles it implicitly at run time.

Related Topics

CREATE VIEW


ANALYZE

Purpose

To perform one of the following functions on an index or index partition, table or table partition, index-organized table, or cluster:

Prerequisites

The schema object to be analyzed must be in your own schema or you must have the ANALYZE ANY system privilege.

If you want to list chained rows of a table or cluster into a list table, the list table must be in your own schema, or you must have INSERT privilege on the list table, or you must have INSERT ANY TABLE system privilege. If you want to validate a partitioned table, you must have INSERT privilege on the table into which you list analyzed ROWIDS, or you must have INSERT ANY TABLE system privilege.

See also "Restrictions".

Syntax

Keywords and Parameters

schema  

is the schema containing the index, table, or cluster. If you omit schema, Oracle assumes the index, table, or cluster is in your own schema.  

index  

identifies an index to be analyzed (if no FOR clause is used).  

table  

identifies a table to be analyzed. When you collect statistics for a table, Oracle also automatically collects the statistics for each of the table's indexes, provided that no FOR clauses are used.  

PARTITION  

specifies that statistics will be gathered for (partition_name). You cannot use this option when analyzing clusters.  

cluster  

identifies a cluster to be analyzed. When you collect statistics for a cluster, Oracle also automatically collects the statistics for all the cluster's tables and all their indexes, including the cluster index. See also "Clusters".  

VALIDATE REF UPDATE  

validates the REFs in the specified table, checks the ROWID portion in each REF, compares it with the true ROWID, and corrects, if necessary. You can use this option only when analyzing a table.  

COMPUTE STATISTICS  

computes exact statistics about the analyzed object and stores them in a data dictionary. See also "Collecting Statistics".  

ESTIMATE STATISTICS  

estimates statistics about the analyzed object and stores them in the data dictionary.  

 

SAMPLE  

specifies the amount of data from the analyzed object Oracle samples to estimate statistics. If you omit this parameter, Oracle samples 1064 rows. If you specify more than half of the data, Oracle reads all the data and computes the statistics.  

 

ROWS  

causes Oracle to sample integer rows of the table or cluster or integer entries from the index. The integer must be at least 1.  

 

PERCENT  

causes Oracle to sample integer percent of the rows from the table or cluster or integer percent of the index entries. The integer can range from 1 to 99.  

for_clause  

specifies whether an entire table or index, or just particular columns, will be analyzed. The following clauses apply only to the ANALYZE TABLE version of this command:  

 

FOR TABLE  

collects table statistics for the table.  

 

FOR ALL COLUMNS  

collects column statistics for all columns and scalar object attributes.  

 

 

INDEX collects column statistics for all indexed columns in the table.  

 

FOR COLUMNS  

collects column statistics for the specified columns and scalar object attributes.  

 

attribute  

specifies the qualified column name of an item in an object.  

 

FOR ALL INDEXES  

all indexes associated with the table will be analyzed.  

 

FOR ALL LOCAL INDEXES  

specifies that all local index partitions are analyzed. You must specify the keyword LOCAL if the PARTITION (partition_name) clause and the index option are specified.  

 

SIZE  

specifies the maximum number of partitions in the histogram. The default value is 75, minimum value is 1, and maximum value is 254.  

 

Histogram statistics are described in Oracle8 Tuning. See also "Columns".  

DELETE STATISTICS  

deletes any statistics about the analyzed object that are currently stored in the data dictionary. See also "Deleting Statistics".  

VALIDATE STRUCTURE  

validates the structure of the analyzed object. If you use this option when analyzing a cluster, Oracle automatically validates the structure of the cluster's tables. If you use this option when analyzing a partitioned table, Oracle also verifies that the row belongs to the correct partition. See also "Validating Structures".  

 

INTO  

specifies a table into which Oracle lists the ROWIDs of the partitions whose rows do not collate correctly. If you omit schema, Oracle assumes the list is in your own schema. If you omit this clause all together, Oracle assumes that the table is named INVALID_ROWS. The SQL script used to create this table is UTLVALID.SQL.  

 

CASCADE  

validates the structure of the indexes associated with the table or cluster. If you use this option when validating a table, Oracle also validates the table's indexes. If you use this option when validating a cluster, Oracle also validates all the clustered tables' indexes, including the cluster index.  

LIST CHAINED ROWS  

identifies migrated and chained rows of the analyzed table or cluster. You cannot use this option when analyzing an index.  

 

INTO  

specifies a table into which Oracle lists the migrated and chained rows. If you omit schema, Oracle assumes the list table is in your own schema. If you omit this clause altogether, Oracle assumes that the table is named CHAINED_ROWS. The script used to create this table is UTLCHAIN.SQL. The list table must be on your local database.  

 

To analyze index-organized tables, you must create a separate chained-rows table for each index-organized table created to accommodate the primary key storage of index-organized tables. Use the SQL scripts DBMSIOTC.SQL and PRVTIOTC.PLB to define the BUILD_CHAIN_ROWS_TABLE package, and then execute this procedure to create an IOT_CHAINED_ROWS table for an index-organized table.

See also "Listing Chained Rows".  

Restrictions

Do not use ANALYZE to collect statistics on data dictionary tables.

You cannot compute or estimate statistics for the following column types:

Collecting Statistics

You can collect statistics about the physical storage characteristics and data distribution of an index, table, column, or cluster and store them in the data dictionary. For computing or estimating statistics:

Use estimation, rather than computation, unless you feel you need exact values. Some statistics are always computed exactly, regardless of whether you specify computation or estimation. If you choose estimation and the time saved by estimating a statistic is negligible, Oracle computes the statistic exactly.

If the data dictionary already contains statistics for the analyzed object, Oracle updates the existing statistics with the new ones.

Example I

The following statement calculates statistics for a scalar object attribute:

ANALYZE TABLE emp COMPUTE STATISTICS FOR COLUMNS addr.street;

The statistics are used by the Oracle optimizer to choose the execution plan for SQL statements that access analyzed objects. These statistics may also be useful to application developers who write such statements. For information on how these statistics are used, see Oracle8 Tuning.

The following sections list the statistics for that are collected for indexes, tables, columns, and clusters. The statistics marked with asterisks (*) are always computed exactly.

Indexes

For an index, Oracle collects the following statistics:

Index statistics appear in the data dictionary views USER_INDEXES, ALL_INDEXES, and DBA_INDEXES.

Tables

For a table, Oracle collects the following statistics:

Table statistics appear in the data dictionary views USER_TABLES, ALL_TABLES, and DBA_TABLES.

Columns

Column statistics can be based on the entire column or can use a histogram. A histogram partitions the values in the column into bands, so that all column values in a band fall within the same range. In some cases, it is useful to see how many values fall in various ranges. Oracle's histograms are height balanced as opposed to width balanced. This means that the column values are divided into bands so that each band contains approximately the same number of values. The useful information the histogram provides, then, is where in the range of values the endpoints fall. Width-balanced histograms, in contrast, divide the data into a number of ranges, all of which are the same size, and then count the number of values falling into each range.

Oracle collects the following column statistics:

When to Use Histograms

For uniformly distributed data, the cost-based approach makes fairly accurate guesses at the cost of executing a particular statement. For non-uniformly distributed data, Oracle allows you to store histograms describing the data distribution of a particular column. These histograms are stored in the dictionary and can be used by the cost-based optimizer.

Histograms are persistent objects, so there is a maintenance and space cost for using them. You should compute histograms only for columns that you know have highly skewed data distribution. Also, be aware that histograms, as well as all optimizer statistics, are static. If the data distribution of a column changes frequently, you must reissue the ANALYZE command to recompute the histogram for that column.

Histograms are not useful for columns with the following characteristics:

Create histograms on columns that are frequently used in WHERE clauses of queries and have a highly skewed data distribution. You create a histogram by using the ANALYZE TABLE command. For example, if you want to create a 10-band histogram on the SAL column of the EMP table, issue the following statement:

ANALYZE TABLE emp 
   COMPUTE STATISTICS FOR COLUMNS sal SIZE 10;

You can also collect histograms for a single partition of a table. The following statement analyzes the EMP table partition P1:

ANALYZE TABLE emp PARTITION (p1) COMPUTE STATISTICS;

Column statistics appear in the data dictionary views: USER_TAB_COLUMNS, ALL_TAB_COLUMNS, and DBA_TAB_COLUMNS.

Histograms appear in the data dictionary views USER_HISTOGRAMS, DBA_HISTOGRAMS, and ALL_HISTOGRAMS.

Clusters

For an indexed cluster, Oracle collects the average number of data blocks taken up by a single cluster key value and all of its rows. For a hash clusters, Oracle collects the average number of data blocks taken up by a single hash key value and all of its rows. These statistics appear in the data dictionary views USER_CLUSTERS and DBA_CLUSTERS.

Example II

The following statement estimates statistics for the CUST_HISTORY table and all of its indexes:

ANALYZE TABLE cust_history
   ESTIMATE STATISTICS; 

Deleting Statistics

With the DELETE STATISTICS option of the ANALYZE command, you can remove existing statistics about an object from the data dictionary. You may want to remove statistics if you no longer want the Oracle optimizer to use them.

When you use the DELETE STATISTICS option on a table, Oracle also automatically removes statistics for all the table's indexes. When you use the DELETE STATISTICS option on a cluster, Oracle also automatically removes statistics for all the cluster's tables and all their indexes, including the cluster index.

Example

The following statement deletes statistics about the CUST_HISTORY table and all its indexes from the data dictionary:

ANALYZE TABLE cust_history
   DELETE STATISTICS; 

Validating Structures

With the VALIDATE STRUCTURE option of the ANALYZE command, you can verify the integrity of the structure of an index, table, or cluster. If Oracle successfully validates the structure, a message confirming its validation is returned to you. If Oracle encounters corruption in the structure of the object, an error message is returned to you. In this case, drop and re-create the object.

Validating the structure of a object prevents SELECT, INSERT, UPDATE, and DELETE statements from concurrently accessing the object. Therefore, do not use this option on the tables, clusters, and indexes of your production applications during periods of high database activity.

Indexes

For an index, the VALIDATE STRUCTURE option verifies the integrity of each data block in the index and checks for block corruption. Note that this option does not confirm that each row in the table has an index entry or that each index entry points to a row in the table. You can perform these operations by validating the structure of the table with the CASCADE option.

When you use the VALIDATE STRUCTURE option on an index, Oracle also collects statistics about the index and stores them in the data dictionary view INDEX_STATS. Oracle overwrites any existing statistics about previously validated indexes. At any time, INDEX_STATS can contain only one row describing only one index. The INDEX_STATS view is described in the Oracle8 Reference.

The statistics collected by this option are not used by the Oracle optimizer. Do not confuse these statistics with the statistics collected by the COMPUTE STATISTICS and ESTIMATE STATISTICS options.

Example I

The following statement validates the structure of the index PARTS_INDEX:

ANALYZE INDEX parts_index
  VALIDATE STRUCTURE; 

Tables

For a table, the VALIDATE STRUCTURE option verifies the integrity of each of the table's data blocks and rows. You can use the CASCADE option to also validate the structure of all indexes on the table as well and to perform cross-referencing between the table and each of its indexes. For each index, the cross-referencing involves the following validations:

Example II

The following statement analyzes the EMP table and all of its indexes:

ANALYZE TABLE emp
   VALIDATE STRUCTURE CASCADE; 

For a table, the VALIDATE REF UPDATE option verifies the REFs in the specified table, checks the ROWID portion of each REF, and then compares it with the true ROWID. If the result is an incorrect ROWID, the REF is updated so that the ROWID portion is correct.

Example III

The following statement validates the REFs in the EMP table:

ANALYZE TABLE emp VALIDATE REF UPDATE;

Clusters

For a cluster, the VALIDATE STRUCTURE option verifies the integrity of each row in the cluster and automatically validates the structure of each of the cluster's tables. You can use the CASCADE option to also validate the structure of all indexes on the cluster's tables as well, including the cluster index.

Example IV

The following statement analyzes the ORDER_CUSTS cluster, all of its tables, and all of their indexes, including the cluster index:

ANALYZE CLUSTER order_custs
    VALIDATE STRUCTURE CASCADE; 

Partitioned Tables

There is no rule-based optimizer for partitioned tables, so it is important to analyze partitioned tables and indexes regularly.

For a partitioned table, the VALIDATE STRUCTURE option verifies each row in the partition to verify whether the column values of the partitioning columns collate less than the partition bound of that partition and greater than the partition bound of the previous partition (except the first partition). If the row does not collate correctly, the ROWID is inserted into the INVALID_ROWS table.

Listing Chained Rows

With the LIST option of the ANALYZE command, you can collect information about the migrated and chained rows in a table or cluster. A migrated row is one that has been moved from one data block to another. For example, Oracle migrates a row in a cluster if its cluster key value is updated. A chained row is one that is contained in more than one data block. For example, Oracle chains a row of a table or cluster if the row is too long to fit in a single data block. Migrated and chained rows may cause excessive I/O. You may want to identify such rows to eliminate them. For information on eliminating migrated and chained rows, see Oracle8 Tuning.

You can use the INTO clause to specify an output table into which Oracle places this information. The definition of a sample output table CHAINED_ROWS is provided in a SQL script available on your distribution media. Your list table must have the same column names, types, and sizes as the CHAINED_ROWS table. On many operating systems, the name of this script is UTLCHAIN.SQL. The actual name and location of this script depends on your operating system.

Example

The following statement collects information about all the chained rows of the table ORDER_HIST:

ANALYZE TABLE order_hist
    LIST CHAINED ROWS INTO cr; 

The preceding statement places the information into the table CR. You can then examine the rows with this query:

SELECT * 
    FROM cr 
OWNER_NAME  TABLE_NAME  CLUSTER_NAME  HEAD_ROWID         TIMESTAMP
----------  ----------  ------------  ------------------ ---------
SCOTT       ORDER_HIST                AAAAZzAABAAABrXAAA 15-MAR-96 

Related Topics

Oracle8 Tuning


ARCHIVE LOG clause

Purpose

To manually archive redo log file groups or to enable or disable automatic archiving. See also "Restrictions".

Prerequisites

The ARCHIVE LOG clause must appear in an ALTER SYSTEM command. You must have the privileges necessary to issue this statement. For information on these privileges, see ALTER SYSTEM.

You must also have the OSDBA or OSOPER role enabled.

You can use most of the options of this clause when your instance has the database mounted, open or closed. Options that require your instance to have the database open are noted.

Syntax

Keywords and Parameters

THREAD  

specifies the thread containing the redo log file group to be archived. You need to specify this parameter only if you are using Oracle with the Parallel Server option in parallel mode.  

SEQUENCE  

manually archives the online redo log file group identified by the log sequence number integer in the specified thread. If you omit the THREAD parameter, Oracle archives the specified group from the thread assigned to your instance.  

CHANGE  

manually archives the online redo log file group containing the redo log entry with the system change number (SCN) specified by integer in the specified thread. If the SCN is in the current redo log file group, Oracle performs a log switch. If you omit the THREAD parameter, Oracle archives the groups containing this SCN from all enabled threads. You can use this option only when your instance has the database open.  

CURRENT  

manually archives the current redo log file group of the specified thread, forcing a log switch. If you omit the THREAD parameter, Oracle archives all redo log file groups from all enabled threads, including logs previous to current logs. You can use this option only when your instance has the database open.  

GROUP  

manually archives the online redo log file group with the GROUP value specified by integer. You can determine the GROUP value for a redo log file group by examining the data dictionary view DBA_LOG_FILES. If you specify both the THREAD and GROUP parameters, the specified redo log file group must be in the specified thread.  

LOGFILE  

manually archives the online redo log file group containing the redo log file member identified by 'filename'. If you specify both the THREAD and LOGFILE parameters, the specified redo log file group must be in the specified thread.  

NEXT  

manually archives the next online redo log file group from the specified thread that is full but has not yet been archived. If you omit the THREAD parameter, Oracle archives the earliest unarchived redo log file group from any enabled thread.  

ALL  

manually archives all online redo log file groups from the specified thread that are full but have not been archived. If you omit the THREAD parameter, Oracle archives all full unarchived redo log file groups from all enabled threads.  

START  

enables automatic archiving of redo log file groups. You can enable automatic archiving only for the thread assigned to your instance.  

TO  

specifies the location to which the redo log file group is archived. The value of this parameter must be a fully specified file location following the conventions of your operating system. If you omit this parameter, Oracle archives the redo log file group to the location specified by the initialization parameter LOG_ARCHIVE_DEST.  

STOP  

disables automatic archiving of redo log file groups. You can disable automatic archiving only for the thread assigned to your instance.  

Restrictions

You must archive redo log file groups in the order in which they are filled. If you specify a redo log file group for archiving with the LOGFILE parameter, and earlier redo log file groups are not yet archived, Oracle returns an error. If you specify a redo log file group for archiving with the CHANGE parameter or CURRENT option, and earlier redo log file groups are not yet archived, Oracle archives all unarchived groups up to and including the specified group.

You can also manually archive redo log file groups with the ARCHIVE LOG Server Manager command. For information on this command, see the Oracle Server Manager User's Guide.

You can also choose to have Oracle archive redo log files groups automatically. For information on automatic archiving, see Oracle8 Administrator's Guide. Note that you can always manually archive redo log file groups regardless of whether automatic archiving is enabled.

Example I

The following statement manually archives the redo log file group with the log sequence number 4 in thread number 3:

ALTER SYSTEM ARCHIVE LOG THREAD 3 SEQUENCE 4; 
Example II

The following statement manually archives the redo log file group containing the redo log entry with the SCN 9356083:

ALTER SYSTEM ARCHIVE LOG CHANGE 9356083; 
Example III

The following statement manually archives the redo log file group containing a member named 'DISKL:LOG6.LOG' to an archived redo log file in the location 'DISKA:[ARCH$]':

ALTER SYSTEM ARCHIVE LOG 
    LOGFILE 'diskl:log6.log' 
    TO 'diska:[arch$]'; 

Related Topics

ALTER SYSTEM


AUDIT (SQL Statements)

Purpose

To choose specific SQL statements for auditing in subsequent user sessions. To choose particular schema objects for auditing, see AUDIT (Schema Objects). See also "Auditing".


Note:

Descriptions of commands and clauses preceded by are available only if the Oracle objects option is installed on your database server.

 

Prerequisites

You must have AUDIT SYSTEM system privilege.

Syntax

Keywords and Parameters

statement_opt  

chooses specific SQL statements for auditing. For a list of these statement options and the SQL statements they audit, see Table 4-6 and Table 4-7. See also "Statement Options for Database Objects" and "Statement Options for Commands".  

system_priv  

chooses SQL statements that are authorized by the specified system privilege for auditing. For a list of all system privileges and the SQL statements that they authorize, see Table 4-19. See also "Shortcuts for System Privileges and Statement Options".  

BY user  

chooses only SQL statements issued by specified users for auditing. If you omit this clause, Oracle audits all users' statements.  

BY SESSION  

causes Oracle to write a single record for all SQL statements of the same type issued in the same session.  

BY ACCESS  

causes Oracle to write one record for each audited statement.  

 

If you specify statement options or system privileges that audit data definition language (DDL) statements, Oracle automatically audits by access regardless of whether you specify the BY SESSION or BY ACCESS option.  

 

For statement options and system privileges that audit other types of SQL statements other than DDL, you can specify either the BY SESSION or BY ACCESS option. BY SESSION is the default.  

WHENEVER SUCCESSFUL  

chooses auditing only for statements that succeed.  

 

NOT chooses auditing only for statements that fail or result in errors.  

 

If you omit the WHENEVER clause, Oracle audits SQL statements regardless of success or failure.  

Auditing

Auditing keeps track of operations performed by database users. For each audited operation, Oracle produces an audit record containing this information:

Oracle writes audit records to the audit trail. The audit trail is a database table that contains audit records. You can review database activity by examining the audit trail through data dictionary views. For information on these views, see the Oracle8 Reference.

To generate audit records, you must perform the following steps:

Enable auditing

You must enable auditing by setting the initialization parameter AUDIT_TRAIL = DB.

Specify auditing options

To specify auditing options, you must use the AUDIT command. Auditing options choose which SQL commands, operations, database objects, and users Oracle audits. After you specify auditing options, they appear in the data dictionary. For more information on data dictionary views containing auditing options see the Oracle8 Reference.

You can specify auditing options regardless of whether auditing is enabled. However, Oracle does not generate audit records until you enable auditing.

Auditing options specified by the AUDIT command (SQL Statements) apply only to subsequent sessions, rather than to current sessions.

Statement Options for Database Objects

Table 4-6 lists the statement options relating to database objects and the statements that they audit.

Table 4-6 Statement Auditing Options for Database Objects
Statement Option   SQL Statements and Operations  
CLUSTER
 
CREATE CLUSTER
AUDIT CLUSTER
DROP CLUSTER
TRUNCATE CLUSTER
 
DATABASE 
LINK
 
CREATE DATABASE LINK
DROP DATABASE LINK
 
DIRECTORY
 
CREATE DIRECTORY
DROP DIRECTORY
 
INDEX
 
CREATE INDEX
ALTER INDEX
DROP INDEX
 
NOT EXISTS
 
All SQL statements that fail because a 
specified object does not exist.
 
PROCEDURE
 
CREATE FUNCTION
CREATE LIBRARY
CREATE PACKAGE
CREATE PACKAGE BODY
CREATE PROCEDURE
DROP FUNCTION
DROP LIBRARY
DROP PACKAGE
DROP PROCEDURE
 
PROFILE
 
CREATE PROFILE
ALTER PROFILE
DROP PROFILE
 
PUBLIC 
DATABASE 
LINK
 
CREATE PUBLIC DATABASE LINK
DROP PUBLIC DATABASE LINK
 
PUBLIC 
SYNONYM
 
CREATE PUBLIC SYNONYM
DROP PUBLIC SYNONYM
 
ROLE
 
CREATE ROLE
ALTER ROLE
DROP ROLE
SET ROLE
 
ROLLBACK 
STATEMENT
 
CREATE ROLLBACK SEGMENT
ALTER ROLLBACK SEGMENT
DROP ROLLBACK SEGMENT
 
SEQUENCE
 
CREATE SEQUENCE
DROP SEQUENCE
 
SESSION
 
Logons
 
SYNONYM
 
CREATE SYNONYM
DROP SYNONYM
 
SYSTEM 
AUDIT
 
AUDIT (SQL Statements)
NOAUDIT (SQL Statements)
 
SYSTEM 
GRANT
 
GRANT (System Privileges and Roles)
REVOKE (System Privileges and Roles)
 
TABLE
 
CREATE TABLE
DROP TABLE
TRUNCATE TABLE
 
TABLESPACE
 
CREATE TABLESPACE
ALTER TABLESPACE
DROP TABLESPACE
 
TRIGGER
 
CREATE TRIGGER
ALTER TRIGGER 
   with ENABLE and DISABLE options
DROP TRIGGER
ALTER TABLE
   with ENABLE ALL TRIGGERS
   and DISABLE ALL TRIGGERS clauses
 
  TYPE
 
CREATE TYPE
CREATE TYPE BODY
ALTER TYPE
DROP TYPE
DROP TYPE BODY
 
USER
 
CREATE USER
ALTER USER
DROP USER
 
VIEW
 
CREATE VIEW
DROP VIEW
 

Statement Options for Commands

Table 4-7 lists additional statement options related to commands and the SQL statements and operations that they audit.

Table 4-7 Statement Auditing Options for Commands
Statement Option   SQL Statements and Operations  
ALTER SEQUENCE
 
ALTER SEQUENCE
 
ALTER TABLE
 
ALTER TABLE
 
COMMENT TABLE
 
COMMENT ON TABLE table, view, snapshot
COMMENT ON COLUMN table.column, 
view.column, snapshot.column
 
DELETE TABLE
 
DELETE FROM table, view
 
EXECUTE PROCEDURE
 

Execution of any procedure or function or access to any variable, library, or cursor inside a package.

 
GRANT DIRECTORY
 
GRANT privilege ON directory
REVOKE privilege ON directory
 
GRANT PROCEDURE
 
GRANT privilege ON procedure, function, 
package 
REVOKE privilege ON procedure, 
function, package
 
GRANT SEQUENCE
 
GRANT privilege ON sequence
REVOKE privilege ON sequence
 
GRANT TABLE
 
GRANT privilege ON table, view, 
snapshot.
REVOKE privilege ON table, view, 
snapshot
 
 GRANT TYPE
 
GRANT privilege ON TYPE
REVOKE privilege ON TYPE
 
INSERT TABLE
 
INSERT INTO table, view
 
LOCK TABLE
 
LOCK TABLE table, view
 
SELECT SEQUENCE
 

Any statement containing sequence.CURRVAL or sequence.NEXTVAL

 
SELECT TABLE
 
SELECT FROM table, view, snapshot
 
UPDATE TABLE
 
UPDATE table, view
 

Example I

To choose auditing for every SQL statement that creates, alters, drops, or sets a role, regardless of whether the statement completes successfully, issue the following statement:

AUDIT ROLE; 

To choose auditing for every statement that successfully creates, alters, drops, or sets a role, issue the following statement:

AUDIT ROLE
    WHENEVER SUCCESSFUL; 

To choose auditing for every CREATE ROLE, ALTER ROLE, DROP ROLE, or SET ROLE statement that results in an Oracle error, issue the following statement:

AUDIT ROLE
    WHENEVER NOT SUCCESSFUL; 
Example II

To choose auditing for any statement that queries or updates any table, issue the following statement:

AUDIT SELECT TABLE, UPDATE TABLE; 

To choose auditing for statements issued by the users SCOTT and BLAKE that query or update a table or view, issue the following statement:

AUDIT SELECT TABLE, UPDATE TABLE
    BY scott, blake; 
Example III

To choose auditing for statements issued using the DELETE ANY TABLE system privilege, issue the following statement:

AUDIT DELETE ANY TABLE; 
Example IV

To choose auditing for statements issued using the CREATE ANY DIRECTORY system privilege, issue the following statement:

AUDIT CREATE ANY DIRECTORY;
Example V

To choose auditing for CREATE DIRECTORY (and DROP DIRECTORY) statements that do NOT use the CREATE ANY DIRECTORY system privilege, issue the following statement:

AUDIT DIRECTORY;

Shortcuts for System Privileges and Statement Options

Oracle provides shortcuts for specifying groups of system privileges and statement options at once. However, Oracle encourages you to choose individual system privileges and statement options for auditing, because these shortcuts may not be supported in future versions of Oracle. The shortcuts are follows:

CONNECT  

is equivalent to specifying the CREATE SESSION system privilege.  

RESOURCE  

is equivalent to specifying the following system privileges:  

 

  • ALTER SESSION
  • CREATE CLUSTER
  • CREATE DATABASE LINK
  • CREATE PROCEDURE
  • CREATE ROLLBACK SEGMENT
  • CREATE SEQUENCE
  • CREATE SYNONYM
  • CREATE TABLE
  • CREATE TABLESPACE
  • CREATE VIEW
 

DBA  

is equivalent to the SYSTEM GRANT statement option and the following system privileges:  

 

  • AUDIT SYSTEM
  • CREATE PUBLIC DATABASE LINK
  • CREATE PUBLIC SYNONYM
  • CREATE ROLE
  • CREATE USER
 

ALL  

equivalent to specifying all statement options shown in Table 4-6, but not the additional statement options shown in Table 4-7.  

ALL PRIVILEGES  

is equivalent to specifying all system privileges.  

Related Topics

AUDIT (Schema Objects)
NOAUDIT (Schema Objects)

AUDIT (Schema Objects)

Purpose

To choose a specific schema object for auditing. To choose particular SQL commands for auditing, see AUDIT (SQL Statements).

Auditing keeps track of operations performed by database users. For a brief conceptual overview of auditing, including how to enable auditing, see the AUDIT (SQL Statements). Note that auditing options established by the AUDIT command (Schema Objects) apply to current sessions as well as to subsequent sessions.

Prerequisites

The object you choose for auditing must be in your own schema or you must have AUDIT ANY system privilege. In addition, if the object you choose for auditing is a directory object, even if you created it, you must have AUDIT ANY system privilege.

Syntax

Keywords and Parameters

object_opt  

specifies a particular operation for auditing. Table 4-8 shows each object option and the types of objects to which it applies. See also "Object Options".  

schema  

is the schema containing the object chosen for auditing. If you omit schema, Oracle assumes the object is in your own schema.  

object  

identifies the object chosen for auditing. The object must be a table; view; sequence; stored procedure, function, or package; snapshot; or library.  

 

You can also specify a synonym for a table, view, sequence, procedure, stored function, package, or snapshot.  

ON DEFAULT  

establishes the specified object options as default object options for subsequently created objects. See also "Default Auditing".  

DIRECTORY directory_name  

identifies the name of the directory chosen for auditing.  

BY SESSION  

means that Oracle writes a single record for all operations of the same type on the same object issued in the same session.  

BY ACCESS  

means that Oracle writes one record for each audited operation.  

If you omit both of the preceding options, Oracle audits by session.  

WHENEVER SUCCESSFUL  

chooses auditing only for SQL statements that complete successfully.  

 

NOT chooses auditing only for statements that fail, or result in errors.  

 

If you omit the WHENEVER clause entirely, Oracle audits all SQL statements, regardless of success or failure.  

Object Options

Table 4-8 shows the object options you can choose for each type of object.

Table 4-8 Object Auditing Options
Object Option   Table   View   Sequence   Procedure Function Package   Snapshot   Library   Directory  

ALTER  

X  

 

X  

 

X  

 

 

AUDIT  

X  

X  

X  

X  

X  

 

X  

COMMENT  

X  

X  

 

 

X  

 

 

DELETE  

X  

X  

 

 

X  

 

 

EXECUTE  

 

 

 

X  

 

X  

 

GRANT  

X  

X  

X  

X  

X  

X  

X  

INDEX  

X  

 

 

 

X  

 

 

INSERT  

X  

X  

 

 

X  

 

 

LOCK  

X  

X  

 

 

X  

 

 

READ  

 

 

 

 

 

 

X  

RENAME  

X  

X  

 

X  

X  

 

 

SELECT  

X  

X  

X  

 

X  

 

 

UPDATE  

X  

X  

 

 

X  

 

 

The name of each object option specifies a command to be audited. For example, if you choose to audit a table with the ALTER option, Oracle audits all ALTER TABLE statements issued against the table. If you choose to audit a sequence with the SELECT option, Oracle audits all statements that use any of the sequence's values.

Shortcuts for Object Options

Oracle provides a shortcut for specifying object auditing options:

ALL  

is equivalent to specifying all object options applicable for the type of object. You can use this shortcut rather than explicitly specifying all options for an object.  

Default Auditing

You can use the DEFAULT option of the AUDIT command to specify auditing options for objects that have not yet been created. Once you have established these default auditing options, any subsequently created object is automatically audited with those options. Note that the default auditing options for a view are always the union of the auditing options for the view's base tables.

If you change the default auditing options, the auditing options for previously created objects remain the same. You can change the auditing options for an existing object only by specifying the object in the ON clause of the AUDIT command.

Example I

To choose auditing for every SQL statement that queries the EMP table in the schema SCOTT, issue the following statement:

AUDIT SELECT
    ON scott.emp; 

To choose auditing for every statement that successfully queries the EMP table in the schema SCOTT, issue the following statement:

AUDIT SELECT 
    ON scott.emp
    WHENEVER SUCCESSFUL; 

To choose auditing for every statement that queries the EMP table in the schema SCOTT and results in an Oracle error, issue the following statement:

AUDIT SELECT 
    ON scott.emp
    WHENEVER NOT SUCCESSFUL; 
Example II

To choose auditing for every statement that inserts or updates a row in the DEPT table in the schema BLAKE, issue the following statement:

AUDIT INSERT, UPDATE
    ON blake.dept; 
Example III

To choose auditing for every statement that performs any operation on the ORDER sequence in the schema ADAMS, issue the following statement:

AUDIT ALL
    ON adams.order; 

The above statement uses the ALL short cut to choose auditing for the following statements that operate on the sequence:

Example IV

To choose auditing for every statement that reads files from the BFILE_DIR1 directory, issue the following statement:

AUDIT READ ON DIRECTORY bfile_dir1;
Example V

The following statement specifies default auditing options for objects created in the future:

AUDIT ALTER, GRANT, INSERT, UPDATE, DELETE
    ON DEFAULT; 

Any objects created later are automatically audited with the specified options that apply to them, provided that auditing has been enabled:

Related Topics

AUDIT (SQL Statements)
NOAUDIT (Schema Objects)

COMMENT

Purpose

To add a comment about a table, view, snapshot, or column into the data dictionary. See also "Using Comments".

Prerequisites

The table, view, or snapshot must be in your own schema or you must have COMMENT ANY TABLE system privilege.

Syntax

Keywords and Parameters

TABLE  

specifies the schema and name of the table, view, or snapshot to be commented.  

COLUMN  

specifies the name of the column of a table, view, or snapshot to be commented. If you omit schema, Oracle assumes the table, view, or snapshot is in your own schema.  

IS 'text'  

is the text of the comment. See the syntax description of 'text' in "Text".  

Using Comments

You can effectively drop a comment from the database by setting it to the empty string ' '. For information on the data dictionary views that contain comments, see Oracle8 Reference.

Example

To insert an explanatory remark on the NOTES column of the SHIPPING table, you might issue the following statement:

COMMENT ON COLUMN shipping.notes
    IS 'Special packing or shipping instructions'; 

To drop this comment from the database, issue the following statement:

COMMENT ON COLUMN shipping.notes IS ' '; 

Related Topics

"Comments"


COMMIT

Purpose

To end your current transaction and make permanent all changes performed in the transaction. This command also erases all savepoints in the transaction and releases the transaction's locks. See also "About Transactions".

You can also use this command to commit an in-doubt distributed transaction manually. See "Ending Transactions" for more information on transactions.

Prerequisites

You need no privileges to commit your current transaction.

To manually commit a distributed in-doubt transaction that you originally committed, you must have FORCE TRANSACTION system privilege. To manually commit a distributed in-doubt transaction that was originally committed by another user, you must have FORCE ANY TRANSACTION system privilege.

Syntax

Keywords and Parameters

WORK  

is supported only for compliance with standard SQL. The statements COMMIT and COMMIT WORK are equivalent.  

COMMENT  

specifies a comment to be associated with the current transaction. The 'text' is a quoted literal of up to 50 characters that Oracle stores in the data dictionary view DBA_2PC_PENDING along with the transaction ID if the transaction becomes in-doubt.  

FORCE  

manually commits an in-doubt distributed transaction. The transaction is identified by the 'text' containing its local or global transaction ID. To find the IDs of such transactions, query the data dictionary view DBA_2PC_PENDING. You can also use the integer to specifically assign the transaction a system change number (SCN). If you omit the integer, the transaction is committed using the current SCN.

COMMIT statements using the FORCE clause are not supported in PL/SQL.  

About Transactions

A transaction (or a logical unit of work) is a sequence of SQL statements that Oracle treats as a single unit. A transaction begins with the first executable SQL statement after a COMMIT, ROLLBACK, or connection to the database. A transaction ends with a COMMIT, ROLLBACK, or disconnection (intentional or unintentional) from the database. Note that Oracle issues an implicit COMMIT before and after any data definition language (DDL) statement.

You can also use a COMMIT or ROLLBACK statement to terminate a read-only transaction begun by a SET TRANSACTION statement.

Example I

This example inserts a row into the DEPT table and commits this change:

INSERT INTO dept   VALUES (50, 'MARKETING', 'TAMPA'); 
COMMIT WORK; 
Example II

The following statement commits the current transaction and associates a comment with it:

COMMIT WORK 
    COMMENT 'In-doubt transaction Code 36, Call (415) 555-2637'; 

If a network or machine failure prevents this distributed transaction from committing properly, Oracle stores the comment in the data dictionary along with the transaction ID. The comment indicates the part of the application in which the failure occurred and provides information for contacting the administrator of the database where the transaction was committed.

Distributed Transactions

Oracle with the distributed option allows you to perform distributed transactions, or transactions that modify data on multiple databases. To commit a distributed transaction, you need only issue a COMMIT statement as you would to commit any other transaction. Each component of the distributed transaction is then committed on each database.

If a network or machine failure occurs during the commit process for a distributed transaction, the state of the transaction may be unknown, or in-doubt. After consultation with the administrators of the other databases involved in the transaction, you may decide to manually commit or roll back the transaction on your local database. You can manually commit the transaction on your local database by using the FORCE clause of the COMMIT command. For more information on these topics, see Oracle8 Distributed Database Systems.

Note that a COMMIT statement with a FORCE clause only commits the specified transaction. Such a statement does not affect your current transaction.

Example

The following statement manually commits an in-doubt distributed transaction:

COMMIT FORCE '22.57.53'; 

Ending Transactions

Oracle recommends that you explicitly end every transaction in your application programs with a COMMIT or ROLLBACK statement, including the last transaction, before disconnecting from Oracle. If you do not explicitly commit the transaction and the program terminates abnormally, the last uncommitted transaction is automatically rolled back.

A normal exit from most Oracle utilities and tools causes the current transaction to be committed. A normal exit from an Oracle precompiler program does not commit the transaction and relies on Oracle to roll back the current transaction.

Related Topics

COMMENT
COMMENT
SET TRANSACTION

CONSTRAINT clause

Purpose

To define an integrity constraint. An integrity constraint is a rule that restricts the values for one or more columns in a table or an index-organized table.

Prerequisites

CONSTRAINT clauses can appear in either CREATE TABLE or ALTER TABLE commands. To define an integrity constraint, you must have the privileges necessary to issue one of these commands. See CREATE TABLE and ALTER TABLE.

Defining a constraint may also require additional privileges or preconditions, depending on the type of constraint. For information on these privileges, see the descriptions of each type of integrity constraint in "Defining Integrity Constraints".

Syntax

table_constraint::=

column_constraint::=


storage_clause: See the STORAGE clause.

Keywords and Parameters

CONSTRAINT  

identifies the integrity constraint by the name constraint. Oracle stores this name in the data dictionary along with the definition of the integrity constraint. If you omit this identifier, Oracle generates a name with this form: SYS_Cn. See also "Defining Integrity Constraints".  

 

If you do not specify NULL or NOT NULL in a column definition, NULL is the default.  

UNIQUE  

designates a column or combination of columns as a unique key. You cannot define UNIQUE constraints on index-organized tables. See also "UNIQUE Constraints".  

PRIMARY KEY  

designates a column or combination of columns as the table's primary key. See also "PRIMARY KEY Constraints".  

FOREIGN KEY  

designates a column or combination of columns as the foreign key in a referential integrity constraint.  

REFERENCES  

identifies the primary or unique key that is referenced by a foreign key in a referential integrity constraint. See also "Referential Integrity Constraints".  

ON DELETE CASCADE  

specifies that Oracle maintains referential integrity by automatically removing dependent foreign key values if you remove a referenced primary or unique key value.  

NULL  

specifies that a column can contain null values.  

NOT NULL  

specifies that a column cannot contain null values. See also "NOT NULL Constraints".  

CHECK  

specifies a condition that each row in the table must satisfy. See also "CHECK Constraints".  

DEFERRABLE  

indicates that constraint checking can be deferred until the end of the transaction by using the SET CONSTRAINT(S) command.  

NOT DEFERRABLE  

indicates that this constraint is checked at the end of each DML statement. You cannot defer a NOT DEFERRABLE constraint with the SET CONSTRAINT(S) command. If you do not specify DEFERRABLE or NOT DEFERRABLE, then NOT DEFERRABLE is the default. See also "DEFERRABLE Constraints".  

 

INITIALLY IMMEDIATE  

indicates that at the start of every transaction, the default is to check this constraint at the end of every DML statement. If no INITIALLY clause is specified, INITIALLY IMMEDIATE is the default.  

 

INITIALLY DEFERRED  

implies that this constraint is DEFERRABLE and specifies that, by default, the constraint is checked only at the end of each transaction.  

USING INDEX  

specifies parameters for the index Oracle uses to enable a UNIQUE or PRIMARY KEY constraint. The name of the index is the same as the name of the constraint. You can choose the values of the INITRANS, MAXTRANS, TABLESPACE, STORAGE, PCTFREE, LOGGING, and NOLOGGING parameters for the index. For information on these parameters, see CREATE TABLE.  

 

Use this clause only when enabling UNIQUE and PRIMARY KEY constraints.  

NOSORT  

indicates that the rows are stored in the database in ascending order and therefore Oracle does not have to sort the rows when creating the index.  

EXCEPTIONS INTO  

specifies a table into which Oracle places the ROWIDs of all rows violating the constraint.

Note: You must create an appropriate exceptions report table to accept information from the EXCEPTIONS option of the ENABLE clause before enabling the constraint. You can create an exception table by submitting the script UTLEXCPT.SQL, which creates a table named EXCEPTIONS. You can create additional exceptions tables with different names by modifying and resubmitting the script.  

 

The EXCEPTIONS INTO clause is a valid option only when validating a constraint (see the ENABLE clause) or when enabling a constraint with an ALTER TABLE command. See ALTER TABLE.  

ENABLE VALIDATE  

ensures that all new insert, delete, and update operations on the constrained data comply with the constraint. Checks that all old data also obeys the constraint. An enabled and validated constraint guarantees that all data is and will continue to be valid. This is the default.  

ENABLE NOVALIDATE  

ensures that all new insert, update, and delete operations on the constrained data comply with the constraint. Oracle does not verify that existing data in the table complies with the constraint.  

DISABLE  

disables the integrity constraint. If an integrity constraint is disabled, Oracle does not enable it. If you do not specify this option, Oracle automatically enables the integrity constraint.  

You can also enable and disable integrity constraints with the ENABLE and DISABLE clauses of the CREATE TABLE and ALTER TABLE commands. See the ENABLE clause and the DISABLE clause. See also "Enabling and Disabling Constraints".  

Disabled constraints can be made enabled with ALTER TABLE.  

Defining Integrity Constraints

To define an integrity constraint, include a CONSTRAINT clause in a CREATE TABLE or ALTER TABLE statement. The CONSTRAINT clause has two syntactic forms:

table_constraint  

The table_constraint syntax is part of the table definition. An integrity constraint defined with this syntax can impose rules on any columns in the table.  

 

The table_constraint syntax can appear in a CREATE TABLE or ALTER TABLE statement. This syntax can define any type of integrity constraint except a NOT NULL constraint.  

column_constraint  

The column_constraint syntax is part of a column definition. Usually, an integrity constraint defined with this syntax can impose rules only on the column in which it is defined.  

 

The column_constraint syntax that appears in a CREATE TABLE statement can define any type of integrity constraint. Column_constraint syntax that appears in an ALTER TABLE statement can only define or remove a NOT NULL constraint.  

The table_constraint syntax and the column_constraint syntax are simply different syntactic means of defining integrity constraints. A constraint that references more than one column must be defined as a table constraint. There is no other functional difference between an integrity constraint defined with table_constraint syntax and the same constraint defined with column_constraint syntax.


Note:

You cannot create a constraint on columns or attributes whose type is user-defined, LOB, or REF. The only exception is that Oracle supports creation of a NOT NULL constraint on columns or attributes of OBJECT type, VARRAY type, LOB, or REF.

 

NOT NULL Constraints

The NOT NULL constraint specifies that a column cannot contain nulls. To satisfy this constraint, every row in the table must contain a value for the column.

The NULL keyword indicates that a column can contain nulls. It does not actually define an integrity constraint. If you do not specify either NOT NULL or NULL, the column can contain nulls by default.

You can specify NOT NULL or NULL with column_constraint syntax only in a CREATE TABLE or ALTER TABLE statement, not with table_constraint syntax.

Example

The following statement alters the EMP table and defines and enables a NOT NULL constraint on the SAL column:

ALTER TABLE emp 
   MODIFY (sal  NUMBER  CONSTRAINT nn_sal NOT NULL); 

NN_SAL ensures that no employee in the table has a null salary.

UNIQUE Constraints

The UNIQUE constraint designates a column or combination of columns as a unique key. To satisfy a UNIQUE constraint, no two rows in the table can have the same value for the unique key. However, the unique key made up of a single column can contain nulls.

A unique key column cannot be of datatype LONG or LONG RAW. You cannot designate the same column or combination of columns as both a unique key and a primary key or as both a unique key and a cluster key. However, you can designate the same column or combination of columns as both a unique key and a foreign key.

Defining Unique Keys

You can define a unique key on a single column with column_constraint syntax.

Example

The following statement creates the DEPT table and defines and enables a unique key on the DNAME column:

CREATE TABLE dept 
    (deptno  NUMBER(2), 
     dname   VARCHAR2(9)  CONSTRAINT unq_dname UNIQUE,
      loc     VARCHAR2(10) ); 

The constraint UNQ_DNAME identifies the DNAME column as a unique key. This constraint ensures that no two departments in the table have the same name. However, the constraint does allow departments without names.

Alternatively, you can define and enable this constraint with the table_constraint syntax:

CREATE TABLE dept 
    (deptno  NUMBER(2), 
     dname   VARCHAR2(9), 
     loc     VARCHAR2(10), 
        CONSTRAINT unq_dname 
        UNIQUE (dname) 
    USING INDEX PCTFREE 20
        TABLESPACE user_x
        STORAGE (INITIAL 8K  NEXT 6K) ); 

The above statement also uses the USING INDEX option to specify storage characteristics for the index that Oracle creates to enable the constraint.

Defining Composite Unique Keys

A composite unique key is a unique key made up of a combination of columns. Oracle creates an index on the columns of a unique key, so a composite unique key can contain a maximum of 16 columns. To define a composite unique key, you must use table_constraint syntax rather than column_constraint syntax.

To satisfy a constraint that designates a composite unique key, no two rows in the table can have the same combination of values in the key columns. Any row that contains nulls in all key columns automatically satisfies the constraint. However, two rows that contain nulls for one or more key columns and the same combination of values for the other key columns violate the constraint.

Example

The following statement defines and enables a composite unique key on the combination of the CITY and STATE columns of the CENSUS table:

ALTER TABLE census 
    ADD CONSTRAINT unq_city_state 
    UNIQUE (city, state) 
    USING INDEX PCTFREE 5  
        TABLESPACE user_y
    EXCEPTIONS INTO bad_keys_in_ship_cont; 

The UNQ_CITY_STATE constraint ensures that the same combination of CITY and STATE values does not appear in the table more than once.

The CONSTRAINT clause also specifies other properties of the constraint:

PRIMARY KEY Constraints

A PRIMARY KEY constraint designates a column or combination of columns as the table's primary key. To satisfy a PRIMARY KEY constraint, both of the following conditions must be true:

A table can have only one primary key.

A primary key column cannot be of datatype LONG or LONG RAW. You cannot designate the same column or combination of columns as both a primary key and a unique key or as both a primary key and a cluster key. However, you can designate the same column or combination of columns as both a primary key and a foreign key.

Defining Primary Keys

You can use the column_constraint syntax to define a primary key on a single column.

Example

The following statement creates the DEPT table and defines and enables a primary key on the DEPTNO column:

CREATE TABLE dept 
    (deptno  NUMBER(2) CONSTRAINT pk_dept PRIMARY KEY, 
    dname   VARCHAR2(9), 
    loc     VARCHAR2(10) ); 

The PK_DEPT constraint identifies the DEPTNO column as the primary key of the DEPT table. This constraint ensures that no two departments in the table have the same department number and that no department number is NULL.

Alternatively, you can define and enable this constraint with table_constraint syntax:



CREATE TABLE dept 
    (deptno  NUMBER(2), 
    dname   VARCHAR2(9), 
    loc   VARCHAR2(10), 
    CONSTRAINT pk_dept PRIMARY KEY (deptno) ); 

Defining Composite Primary Keys

A composite primary key is a primary key made up of a combination of columns. Oracle creates an index on the columns of a primary key; therefore, a composite primary key can contain a maximum of 16 columns. To define a composite primary key, you must use the table_constraint syntax rather than the column_constraint syntax.

Example

The following statement defines a composite primary key on the combination of the SHIP_NO and CONTAINER_NO columns of the SHIP_CONT table:

ALTER TABLE ship_cont 
    ADD PRIMARY KEY (ship_no, container_no) DISABLE; 

This constraint identifies the combination of the SHIP_NO and CONTAINER_NO columns as the primary key of the SHIP_CONT table. The constraint ensures that no two rows in the table have the same values for both the SHIP_NO column and the CONTAINER_NO column.

The CONSTRAINT clause also specifies the following properties of the constraint:

Referential Integrity Constraints

A referential integrity constraint designates a column or combination of columns as a foreign key and establishes a relationship between that foreign key and a specified primary or unique key, called the referenced key. In this relationship, the table containing the foreign key is called the child table and the table containing the referenced key is called the parent table. Note the following restrictions:

To satisfy a referential integrity constraint, each row of the child table must meet one of the following conditions:

A referential integrity constraint is defined in the child table. A referential integrity constraint definition can include any of the following keywords:

FOREIGN KEY  

identifies the column or combination of columns in the child table that makes up the foreign key. Ise this keyword only when you define a foreign key with a table constraint clause.  

REFERENCES  

identifies the parent table and the column or combination of columns that make up the referenced key.  

 

If you identify only the parent table and omit the column names, the foreign key automatically references the primary key of the parent table.  

 

The corresponding columns of the referenced key and the foreign key must match in number and datatypes.  

ON DELETE CASCADE  

allows deletion of referenced key values in the parent table that have dependent rows in the child table and causes Oracle to automatically delete dependent rows from the child table to maintain referential integrity.  

 

If you omit this option, Oracle forbids deletions of referenced key values in the parent table that have dependent rows in the child table.  

Before you define a referential integrity constraint in the child table, the referenced UNIQUE or PRIMARY KEY constraint on the parent table must already be defined. Also, the parent table must be in your own schema or you must have REFERENCES privilege on the columns of the referenced key in the parent table. Before you enable a referential integrity constraint, its referenced constraint must be enabled.

You cannot define a referential integrity constraint in a CREATE TABLE statement that contains an AS clause. Instead, you can create the table without the constraint and then add it later with an ALTER TABLE statement.

A foreign key column cannot be of datatype LONG or LONG RAW. You can designate the same column or combination of columns as both a foreign key and a primary or unique key. You can also designate the same column or combination of columns as both a foreign key and a cluster key.

You can define multiple foreign keys in a table. Also, a single column can be part of more than one foreign key.

Defining Referential Integrity Constraints

You can use column_constraint syntax to define a referential integrity constraint in which the foreign key is made up of a single column.

Example

The following statement creates the EMP table and defines and enables a foreign key on the DEPTNO column that references the primary key on the DEPTNO column of the DEPT table:

CREATE TABLE emp 
   (empno      NUMBER(4), 
    ename      VARCHAR2(10), 
    job        VARCHAR2(9), 
    mgr        NUMBER(4), 
    hiredate   DATE, 
    sal        NUMBER(7,2), 
    comm       NUMBER(7,2), 
    deptno     CONSTRAINT fk_deptno REFERENCES dept(deptno) ); 

The constraint FK_DEPTNO ensures that all departments given for employees in the EMP table are present in the DEPT table. However, employees can have null department numbers, meaning they are not assigned to any department. If you wish to prevent the latter, you could create a NOT NULL constraint on the deptno column in the EMP table, in addition to the REFERENCES constraint.

Before you define and enable this constraint, you must define and enable a constraint that designates the DEPTNO column of the DEPT table as a primary or unique key. For the definition of such a constraint, see the example.

Note that the referential integrity constraint definition does not use the FOREIGN KEY keyword to identify the columns that make up the foreign key. Because the constraint is defined with a column constraint clause on the DEPTNO column, the foreign key is automatically on the DEPTNO column.

Note that the constraint definition identifies both the parent table and the columns of the referenced key. Because the referenced key is the parent table's primary key, the referenced key column names are optional.

Note that the above statement omits the DEPTNO column's datatype. Because this column is a foreign key, Oracle automatically assigns it the datatype of the DEPT.DEPTNO column to which the foreign key refers.

Alternatively, you can define a referential integrity constraint with table_constraint syntax:

CREATE TABLE emp 
  (empno     NUMBER(4), 
   ename     VARCHAR2(10), 
   job       VARCHAR2(9), 
   mgr       NUMBER(4), 
   hiredate  DATE, 
   sal       NUMBER(7,2), 
   comm      NUMBER(7,2), 
   deptno, 
   CONSTRAINT fk_deptno 
      FOREIGN  KEY (deptno) 
      REFERENCES  dept(deptno) ); 

Note that the foreign key definitions in both statements of this example omit the ON DELETE CASCADE option, causing Oracle to forbid the deletion of a department if any employee works in that department.

Maintaining Referential Integrity with ON DELETE CASCADE

If you use the ON DELETE CASCADE option, Oracle permits deletions of referenced key values in the parent table and automatically deletes dependent rows in the child table to maintain referential integrity.

Example

This example creates the EMP table, defines and enables the referential integrity constraint FK_DEPTNO, and uses the ON DELETE CASCADE option:





CREATE TABLE emp 
  (empno    NUMBER(4), 
   ename    VARCHAR2(10), 
   job      VARCHAR2(9), 
   mgr      NUMBER(4), 
   hiredate DATE, 
   sal      NUMBER(7,2), 
   comm     NUMBER(7,2), 
   deptno   NUMBER(2)   CONSTRAINT fk_deptno 
            REFERENCES dept(deptno) 
            ON DELETE CASCADE ); 

Because of the ON DELETE CASCADE option, Oracle cascades any deletion of a DEPTNO value in the DEPT table to the DEPTNO values of its dependent rows of the EMP table. For example, if Department 20 is deleted from the DEPT table, Oracle deletes the department's employees from the EMP table.

Referential Integrity Constraints with Composite Keys

A composite foreign key is a foreign key made up of a combination of columns. A composite foreign key can contain as many as 16 columns. To define a referential integrity constraint with a composite foreign key, you must use table_constraint syntax. You cannot use column_constraint syntax, because this syntax can impose rules only on a single column. A composite foreign key must refer to a composite unique key or a composite primary key.

To satisfy a referential integrity constraint involving composite keys, each row in the child table must satisfy one of the following conditions:

Example

The following statement defines and enables a foreign key on the combination of the AREACO and PHONENO columns of the PHONE_CALLS table:

ALTER TABLE phone_calls 
    ADD CONSTRAINT fk_areaco_phoneno 
        FOREIGN KEY (areaco, phoneno) 
        REFERENCES customers(areaco, phoneno)
        EXCEPTIONS INTO wrong_numbers; 

The constraint FK_AREACO_PHONENO ensures that all the calls in the PHONE_CALLS table are made from phone numbers that are listed in the CUSTOMERS table. Before you define and enable this constraint, you must define and enable a constraint that designates the combination of the AREACO and PHONENO columns of the CUSTOMERS table as a primary or unique key.

The EXCEPTIONS option causes Oracle to write information to the WRONG_NUMBERS table about any rows in the PHONE_CALLS table that violate the constraint.

CHECK Constraints

The CHECK constraint explicitly defines a condition. To satisfy the constraint, each row in the table must make the condition either TRUE or unknown (due to a null). For information on conditions, see the syntax description of condition in "Conditions". The condition of a CHECK constraint can refer to any column in the table, but it cannot refer to columns of other tables. CHECK constraint conditions cannot contain the following constructs:

Whenever Oracle evaluates a CHECK constraint condition for a particular row, any column names in the condition refer to the column values in that row.

If you create multiple CHECK constraints for a column, design them carefully so their purposes do not conflict. Oracle does not verify that CHECK conditions are not mutually exclusive.

Example I

The following statement creates the DEPT table and defines a CHECK constraint in each of the table's columns:

CREATE TABLE dept  (deptno NUMBER  CONSTRAINT check_deptno
          CHECK (deptno BETWEEN 10 AND 99) 
          DISABLE, 
dname VARCHAR2(9)  CONSTRAINT check_dname 
          CHECK (dname = UPPER(dname)) 
          DISABLE, 
loc VARCHAR2(10)  CONSTRAINT check_loc 
          CHECK (loc IN ('DALLAS','BOSTON',
          'NEW YORK','CHICAGO')) 
          DISABLE); 

Each constraint restricts the values of the column in which it is defined:

CHECK_DEPTNO  

ensures that no department numbers are less than 10 or greater than 99.  

CHECK_DNAME  

ensures that all department names are in uppercase.  

CHECK_LOC  

restricts department locations to Dallas, Boston, New York, or Chicago.  

Because each CONSTRAINT clause contains the DISABLE option, Oracle only defines the constraints and does not enable them.

Unlike other types of constraints, a CHECK constraint defined with column_constraint syntax can impose rules on any column in the table, rather than only on the column in which it is defined.

Example II

The following statement creates the EMP table and uses a table constraint clause to define and enable a CHECK constraint:

CREATE TABLE emp 
    (empno          NUMBER(4), 
     ename          VARCHAR2(10), 
     job            VARCHAR2(9),
     mgr            NUMBER(4), 
     hiredate       DATE, 
     sal            NUMBER(7,2), 
     comm           NUMBER(7,2),
     deptno         NUMBER(2),
     CHECK (sal + comm <= 5000) );

This constraint uses an inequality condition to limit an employee's total compensation, the sum of salary and commission, to $5000:

Because the CONSTRAINT clause in this example does not supply a constraint name, Oracle generates a name for the constraint.

Example III

The following statement defines and enables a PRIMARY KEY constraint, two referential integrity constraints, a NOT NULL constraint, and two CHECK constraints:

CREATE TABLE order_detail 
  (CONSTRAINT pk_od PRIMARY KEY (order_id, part_no), 
   order_id NUMBER 
      CONSTRAINT fk_oid REFERENCES scott.order (order_id), 
   part_no            NUMBER 
      CONSTRAINT fk_pno REFERENCES scott.part (part_no), 
   quantity            NUMBER 
      CONSTRAINT nn_qty NOT NULL 
      CONSTRAINT check_qty_low CHECK (quantity > 0), 
   cost            NUMBER 
      CONSTRAINT check_cost CHECK (cost > 0) ); 

The constraints enable the following rules on table data:

PK_OD  

identifies the combination of the ORDER_ID and PART_NO columns as the primary key of the table. To satisfy this constraint, the following conditions must be true:  

 

  • No two rows in the table can contain the same combination of values in the ORDER_ID and the PART_NO columns.
  • No row in the table can have a null in either the ORDER_ID column or the PART_NO column.
 

FK_OID  

identifies the ORDER_ID column as a foreign key that references the ORDER_ID column in the ORDER table in SCOTT's schema. All new values added to the column ORDER_DETAIL.ORDER_ID must already appear in the column SCOTT.ORDER.ORDER_ID.  

FK_PNO  

identifies the PART_NO column as a foreign key that references the PART_NO column in the PART table owned by SCOTT. All new values added to the column ORDER_DETAIL.PART_NO must already appear in the column SCOTT.PART.PART_NO.  

NN_QTY  

forbids nulls in the QUANTITY column.  

CHECK_QTY  

ensures that values in the QUANTITY column are always greater than zero.  

CHECK_COST  

ensures the values in the COST column are always greater than zero.  

This example also illustrates the following points about constraint clauses and column definitions:

DEFERRABLE Constraints

You can specify table and column constraints as DEFERRABLE or NOT DEFERRABLE. DEFERRABLE means that the constraint will not be checked until the transaction is committed. The default is NOT DEFERRABLE.

If you specify DEFERRABLE, you can also specify the constraint's initial state as INITIALLY DEFERRED and thereby start the transaction in DEFERRED mode. Or you can specify a DEFERRABLE constraint's initial state as INITIALLY IMMEDIATE and start the transaction in NOT DEFERRED mode.

Example I

The following statement creates table GAMES with a NOT DEFERRABLE INITIALLY IMMEDIATE constraint check on the SCORES column:

CREATE TABLE games (scores NUMBER CHECK (scores >= 0));
Example III

To define a unique constraint on a column as INITIALLY DEFERRED DEFERRABLE, issue the following statement:

CREATE TABLE orders
  (ord_num NUMBER CONSTRAINT unq_num UNIQUE (ord_num)
   INITIALLY DEFERRED DEFERRABLE);

A constraint cannot be defined as NOT DEFERRABLE INITIALLY DEFERRED.

Use SET CONSTRAINT(S) to set, for a single transaction, whether a deferrable constraint is checked following each DML statement or when the transaction is committed. You cannot alter a constraint's deferrability status; you must drop the constraint and re-create it.

See Oracle8 Administrator's Guide and Oracle8 Concepts for more information about deferred constraints.

Enabling and Disabling Constraints

Constraints can have one of three states: DISABLE, ENABLE NOVALIDATE, or ENABLE VALIDATE.

Taking a constraint from a disabled to enable validated state requires an exclusive lock on the table, because while all old data is being checked for validity, no new data can be entered into the table. Due to this behavior, only one constraint can be enabled at a time, and each new constraint must check all existing rows by serial scan.

To avoid locking the table, place the constraint in the ENABLE NOVALIDATE state, using the ENABLE clause. This state ensures that all new DML statements on the table are validated, therefore Oracle does not need to prevent concurrent access to the table.

ENABLE NOVALIDATE also allows you to place several of the table's constraints in the ENABLE VALIDATE state concurrently. Each scan that Oracle performs to validate existing data can also be performed in parallel when possible.

Placing constraints concurrently in the ENABLE VALIDATE state requires that you issue multiple ALTER TABLE commands from separate sessions.

Enabling Primary Key and Unique Key Constraints

Enabling a primary key or unique key constraint automatically creates a unique index to enforce the constraint. This index is dropped if the constraint is subsequently disabled, thus causing Oracle to rebuild the index every time the constraint is enabled.

To avoid this behavior, create new primary key and unique key constraints initially disabled; then create nonunique indexes or use existing nonunique indexes to enforce the constraint. Because Oracle does not drop the nonunique index when the constraint is disabled, any ENABLE operation on a primary key or unique key constraint occurs almost instantly, because the index already exists. Redundant indexes are also eliminated.

For more information about PRIMARY KEY and UNIQUE constraints, see the ENABLE clause.

Related Topics

CREATE TABLE
ALTER TABLE
ENABLE clause
DISABLE clause
SET CONSTRAINT(S)
ALTER SESSION

CREATE CLUSTER

Purpose

To create a cluster. A cluster is a schema object that contains one or more tables, all of which have one or more columns in common. See also "About Clusters" and "Adding Tables to a Cluster".

Prerequisites

To create a cluster in your own schema, you must have CREATE CLUSTER system privilege. To create a cluster in another user's schema, you must have CREATE ANY CLUSTER system privilege. Also, the owner of the schema to contain the cluster must have either space quota on the tablespace containing the cluster or UNLIMITED TABLESPACE system privilege.

Syntax

storage_clause: See the STORAGE clause.

parallel_clause: See the PARALLEL clause.

Keywords and Parameters

schema  

is the schema to contain the cluster. If you omit schema, Oracle creates the cluster in your current schema.  

cluster  

is the name of the cluster to be created.  

column  

is the name of a column in the cluster key. See also "Cluster Keys".  

datatype  

is the datatype of a cluster key column. A cluster key column can have any datatype except LONG or LONG RAW. You cannot use the HASH IS clause if any column datatype is not INTEGER or NUMBER with scale 0. For information on datatypes, see the section "Datatypes".  

physical_attributes_clause:  

PCTUSED  

specifies the limit that Oracle uses to determine when additional rows can be added to a cluster's data block. The value of this parameter is expressed as a whole number and interpreted as a percentage.  

PCTFREE  

specifies the space reserved in each of the cluster's data blocks for future expansion. The value of the parameter is expressed as a whole number and interpreted as a percentage.  

INITRANS  

specifies the initial number of concurrent update transactions allocated for data blocks of the cluster. The value of this parameter for a cluster cannot be less than 2 or more than the value of the MAXTRANS parameter. The default value is the greater of the INITRANS value for the cluster's tablespace and 2.  

MAXTRANS  

specifies the maximum number of concurrent update transactions for any given data block belonging to the cluster. The value of this parameter cannot be less than the value of the INITRANS parameter. The maximum value of this parameter is 255. The default value is the MAXTRANS value for the tablespace to contain the cluster.  

 

For a complete description of the PCTUSED, PCTFREE, INITRANS, and MAXTRANS parameters, see CREATE TABLE.  

SIZE  

specifies the amount of space in bytes to store all rows with the same cluster key value or the same hash value. You can use K or M to specify this space in kilobytes or megabytes. If you omit this parameter, Oracle reserves one data block for each cluster key value or hash value. See also "Cluster Size".  

TABLESPACE  

specifies the tablespace in which the cluster is created.  

storage_clause  

specifies how data blocks are allocated to the cluster. See the STORAGE clause.  

INDEX  

creates an indexed cluster. In an indexed cluster, rows are stored together based on their cluster key values. See also "Types of Clusters".  

HASHKEYS  

creates a hash cluster and specifies the number of hash values for a hash cluster. Oracle rounds the HASHKEYS value up to the nearest prime number to obtain the actual number of hash values. The minimum value for this parameter is 2. If you omit both the INDEX option and the HASHKEYS parameter, Oracle creates an indexed cluster by default. See also "Types of Clusters".  

HASH IS  

specifies a expression to be used as the hash function for the hash cluster. The expression:  

 

  • must evaluate to a positive value
  • must contain at least one column with referenced columns of any datatype as long as the entire expression evaluates to a number of scale 0-for example, NUM_COLUMN * length(VARCHAR2_COLUMN)
  • cannot reference user-defined PL/SQL functions
  • cannot reference SYSDATE, USERENV, TO_DATE, UID, USER, LEVEL, ROWNUM
  • cannot evaluate to a constant
  • cannot contain a subquery
  • cannot contain columns qualified with a schema or object name (other than the cluster name)
 

 

If you omit the HASH IS clause, Oracle uses an internal hash function for the hash cluster.  

 

The cluster key of a hash column can have one or more columns of any datatype. Hash clusters with composite cluster keys or cluster keys made up of noninteger columns must use the internal hash function.  

parallel_clause  

specifies the degree of parallelism to use when creating the cluster and the default degree of parallelism to use when querying the cluster after creation. See the PARALLEL clause.  

CACHE  

specifies that the blocks retrieved for this table are placed at the most recently used end of the LRU list in the buffer cache when a full table scan is performed. This option is useful for small lookup tables.  

NOCACHE  

specifies that the blocks retrieved for this table are placed at the least recently used end of the LRU list in the buffer cache when a full table scan is performed. This is the default behavior.  

About Clusters

A cluster is a schema object that contains one or more tables that all have one or more columns in common. Rows of one or more tables that share the same value in these common columns are physically stored together within the database.

Clustering provides more control over the physical storage of rows within the database. Clustering can reduce both the time it takes to access clustered tables and the space needed to store the table. After you create a cluster and add tables to it, the cluster is transparent. You can access clustered tables with SQL statements just as you can nonclustered tables.

If you cannot fit all rows for one hash value into a data block, do not use hash clusters. Performance is very poor in this circumstance because an insert or update of a row in a hash cluster with a size exceeding the data block size fills the block and performs row chaining to contain the rest of the row.

Generally, you should only cluster tables that are frequently joined on the cluster key columns in SQL statements. Clustering multiple tables improves the performance of joins, but it is likely to reduce the performance of full table scans, INSERT statements, and UPDATE statements that modify cluster key values. Before clustering, consider its benefits and trade-offs in light of the operations you plan to perform on your data. For more information on the performance implications of clustering, see Oracle8 Tuning.

Cluster Keys

The columns defined by the CREATE CLUSTER command make up the cluster key. These cluster columns must correspond in both datatype and size to columns in each of the clustered tables, although they need not correspond in name.

You cannot specify integrity constraints as part of the definition of a cluster key column. Instead, you can associate integrity constraints with the tables that belong to the cluster.

Types of Clusters

A cluster can be either an indexed cluster or a hash cluster.

Indexed Clusters

In an indexed cluster, Oracle stores together rows having the same cluster key value. Each distinct cluster key value is stored only once in each data block, regardless of the number of tables and rows in which it occurs. This saves disk space and improves performance for many operations.

You may want to use indexed clusters in the following cases:

After you create an indexed cluster, you must create an index on the cluster key before you can issue any data manipulation language (DML) statements against a table in the cluster. This index is called the cluster index. For information on creating a cluster index, see CREATE INDEX. As with the columns of any index, the order of the columns in the cluster key affects the structure of the cluster index.

A cluster index provides quick access to rows within a cluster based on the cluster key. If you issue a SQL statement that searches for a row in the cluster based on its cluster key value, Oracle searches the cluster index for the cluster key value and then locates the row in the cluster based on its ROWID.

Hash Clusters

In a hash cluster, Oracle stores together rows that have the same hash key value. The hash value for a row is the value returned by the cluster's hash function. When you create a hash cluster, you can either specify a hash function or use the Oracle internal hash function. Hash values are not actually stored in the cluster, although cluster key values are stored for every row in the cluster.

You may want to use hash clusters in the following cases:

The hash function provides access to rows in the table based on the cluster key value. If you issue a SQL statement that locates a row in the cluster based on its cluster key value, Oracle applies the hash function to the given cluster key value and uses the resulting hash value to locate the matching rows. Because multiple cluster key values can map to the same hash value, Oracle must also check the row's cluster key value. This process often results in less I/O than the process for the indexed cluster, because the index search is not required.

Oracle's internal hash function returns values ranging from 0 to the value of HASHKEYS - 1. If you specify a column with the HASH IS clause, the column values need not fall into this range. Oracle divides the column value by the HASHKEYS value and uses the remainder as the hash value. The hash value for null is HASHKEYS - 1. Oracle also rounds the HASHKEYS value up to the nearest prime number to obtain the actual number of hash values. This rounding reduces the likelihood of hash collisions, or multiple cluster key values having the same hash value.

You cannot create a cluster index for a hash cluster, and you need not create an index on a hash cluster key.

Cluster Size

Oracle uses the value of the SIZE parameter to determine the space reserved for rows corresponding to one cluster key value or one hash value. This space then determines the maximum number of cluster or hash values stored in a data block. If the SIZE value is not a divisor of the data block size, Oracle uses the next largest divisor. If the SIZE value is larger than the data block size, Oracle uses the operating system block size, reserving at least one data block per cluster or hash value.

Oracle also considers the length of the cluster key when determining how much space to reserve for the rows having a cluster key value. Larger cluster keys require larger sizes. To see the actual size, query the KEY_SIZE column of the USER_CLUSTERS data dictionary view. This does not apply to hash clusters because hash values are not actually stored in the cluster.

Although the maximum number of cluster and hash key values per data block is fixed on a per `-cluster basis, Oracle does not reserve an equal amount of space for each cluster or hash key value. Varying this space stores data more efficiently, because the data stored per cluster or hash key value is rarely fixed.

A SIZE value smaller than the space needed by the average cluster or hash key value may require the data for one cluster key or hash key value to occupy multiple data blocks. A SIZE value much larger results in wasted space.

When you create a hash cluster, Oracle immediately allocates space for the cluster based on the values of the SIZE and HASHKEYS parameters. For more information on how Oracle allocates space for clusters, see Oracle8 Concepts.

Adding Tables to a Cluster

You can add tables to an existing cluster by issuing a CREATE TABLE statement with the CLUSTER clause. A cluster can contain as many as 32 tables, although the performance gains of clustering are often lost in clusters of more than four or five tables.

All tables in the cluster have the cluster's storage characteristics as specified by the PCTUSED, PCTFREE, INITRANS, MAXTRANS, TABLESPACE, and STORAGE parameters.

Example I

The following statement creates an indexed cluster named PERSONNEL with the cluster key column DEPARTMENT_NUMBER, a cluster size of 512 bytes, and storage parameter values:

CREATE CLUSTER personnel 
    ( department_number  NUMBER(2) ) 
    SIZE 512 
    STORAGE (INITIAL 100K NEXT 50K PCTINCREASE 10); 

The following statements add the EMP and DEPT tables to the cluster:

CREATE TABLE emp 
    (empno     NUMBER        PRIMARY KEY, 
     ename     VARCHAR2(10)  NOT NULL 
                             CHECK (ename = UPPER(ename)), 
     job       VARCHAR2(9), 
     mgr       NUMBER        REFERENCES scott.emp(empno), 
     hiredate  DATE          CHECK (hiredate >= SYSDATE), 
     sal       NUMBER(10,2)  CHECK (sal > 500), 
     comm      NUMBER(9,0)   DEFAULT NULL, 
     deptno   NUMBER(2)      NOT NULL ) 
     CLUSTER personnel (deptno); 
 
CREATE TABLE dept 
    (deptno  NUMBER(2), 
     dname   VARCHAR2(9), 
     loc     VARCHAR2(9))
     CLUSTER personnel (deptno); 

The following statement creates the cluster index on the cluster key of PERSONNEL:

CREATE INDEX idx_personnel ON CLUSTER personnel;
 

After creating the cluster index, you can insert rows into either the EMP or DEPT tables.

Example II

The following statement creates a hash cluster named PERSONNEL with the cluster key column DEPARTMENT_NUMBER, a maximum of 503 hash key values, each of size 512 bytes, and storage parameter values:

CREATE CLUSTER personnel
( department_number  NUMBER )
SIZE 512  HASHKEYS 500 
    STORAGE (INITIAL 100K  NEXT 50K  PCTINCREASE 10); 

Because the above statement omits the HASH IS clause, Oracle uses the internal hash function for the cluster.

Example III

The following statement creates a hash cluster named PERSONNEL with the cluster key made up of the columns HOME_AREA_CODE and HOME_PREFIX, and uses a SQL expression containing these columns for the hash function:

CREATE CLUSTER personnel 
    ( home_area_code  NUMBER,
    home_prefix     NUMBER ) 
    HASHKEYS 20
    HASH IS MOD(home_area_code + home_prefix, 101); 

Related Topics

CREATE INDEX
CREATE TABLE
"Index-Organized Tables"
STORAGE clause

CREATE CONTROLFILE

Purpose

To re-create a control file in one of the following cases:

See also "Re-creating Control Files".


warning:

Oracle recommends that you perform a full backup of all files in the database before using this command.

 

Prerequisites

You must have the OSDBA role enabled. The database must not be mounted by any instance.

Syntax

filespec: See "Filespec".

Keywords and Parameters

REUSE  

specifies that existing control files identified by the initialization parameter CONTROL_FILES can be reused, thus ignoring and overwriting any information they may currently contain. If you omit this option and any of these control files already exists, Oracle returns an error.  

DATABASE  

specifies the name of the database. The value of this parameter must be the existing database name established by the previous CREATE DATABASE statement or CREATE CONTROLFILE statement.  

SET DATABASE  

changes the name of the database. The name of a database can be as long as eight bytes.  

LOGFILE  

specifies the redo log file groups for your database. You must list all members of all redo log file groups. See the syntax description of filespec in "Filespec".  

RESETLOGS  

ignores the contents of the files listed in the LOGFILE clause. These files do not have to exist. Each filespec in the LOGFILE clause must specify the SIZE parameter. Oracle assigns all redo log file groups to thread 1 and enables this thread for public use by any instance. After using this option, you must open the database using the RESETLOGS option of the ALTER DATABASE command.  

NORESETLOGS  

specifies that all files in the LOGFILE clause should be used as they were when the database was last open. These files must exit and must be the current redo log files rather than restored backups. Oracle reassigns the redo log file groups to the threads to which they were previously assigned and reenables the threads as they were previously enabled. If you specify GROUP values, Oracle verifies these values with the GROUP values when the database was last open.  

DATAFILE  

specifies the datafiles of the database. You must list all datafiles. These files must all exist, although they may be restored backups that require media recovery. See the syntax description of filespec in "Filespec".  

MAXLOGFILES  

specifies the maximum number of redo log file groups that can ever be created for the database. Oracle uses this value to determine how much space in the control file to allocate for the names of redo log files. The default and maximum values depend on your operating system. The value that you specify should not be less than the greatest GROUP value for any redo log file group.  

 

Note that the number of redo log file groups accessible to your instance is also limited by the initialization parameter LOG_FILES.  

MAXLOGMEMBERS  

specifies the maximum number of members, or copies, for a redo log file group. Oracle uses this value to determine how much space in the control file to allocate for the names of redo log file. The minimum value is 1. The maximum and default values depend on your operating system.  

MAXLOGHISTORY  

specifies the maximum number of archived redo log file groups for automatic media recovery of the Oracle8 Parallel Server. Oracle uses this value to determine how much space in the control file to allocate for the names of archived redo log files. The minimum value is 0. The default value is a multiple of the MAXINSTANCE value and depends on your operating system. The maximum value is limited only by the maximum size of the control file. This parameter is useful only if you are using Oracle with the Parallel Server option in both parallel mode and archivelog mode.  

MAXDATAFILES  

specifies the initial sizing of the datafiles section of the control file at CREATE DATABASE or CREATE CONTROLFILE time. An attempt to add a file whose number is greater than MAXDATAFILES, but less than or equal to DB_FILES, causes the Oracle8 control file to expand automatically so that the datafiles section can accommodate more files.  

 

Note that the number of datafiles accessible to your instance is also limited by the initialization parameter DB_FILES.  

MAXINSTANCES  

specifies the maximum number of instances that can simultaneously have the database mounted and open. This value takes precedence over the value of the initialization parameter INSTANCES. The minimum value is 1. The maximum and default values depend on your operating system.  

ARCHIVELOG  

establishes the mode of archiving the contents of redo log files before reusing them. This option prepares for the possibility of media recovery as well as instance recovery.  

NOARCHIVELOG  

If you omit both the ARCHIVELOG and NOARCHIVELOG options, Oracle chooses noarchivelog mode by default. After creating the control file, you can change between archivelog mode and noarchivelog mode with the ALTER DATABASE command.  

Re-creating Control Files

Oracle recommends that you take a full backup of all files in the database before issuing a CREATE CONTROLFILE statement.

When you issue a CREATE CONTROLFILE statement, Oracle creates a new control file based on the information you specify in the statement. If you omit any of the options from the statement, Oracle uses the default options, rather than the options for the previous control file. After successfully creating the control file, Oracle mounts the database in the mode specified by the initialization parameter PARALLEL_SERVER. You then must perform media recovery before opening the database. It is recommended that you then shutdown the instance and take a full backup of all files in the database.

For more information about using this command, see the Oracle8 Administrator's Guide.

Example

This example re-creates a control file:

CREATE CONTROLFILE REUSE 
DATABASE orders_2 
LOGFILE GROUP 1 ('diskb:log1.log', 'diskc:log1.log') SIZE 50K, 
GROUP 2 ('diskb:log2.log', 'diskc:log2.log') SIZE 50K 
NORESETLOGS 
DATAFILE 'diska:dbone.dat' SIZE 2M 
MAXLOGFILES 5 
MAXLOGHISTORY 100 
MAXDATAFILES 10 
MAXINSTANCES 2 
ARCHIVELOG; 

Related Topics

CREATE DATABASE command on page 4-220
"Filespec"

CREATE DATABASE

Purpose

To create a database, making it available for general use, with the following options:

For examples of some of these purposes, see "Examples".


warning:

This command prepares a database for initial use and erases any data currently in the specified files. Use this command only when you understand its ramifications.

 

This command erases all data in any specified datafiles that already exist to prepare them for initial database use. If you use the command on an existing database, all data in the datafiles is lost.

After creating the database, this command mounts it in the mode specified by the PARALLEL_SERVER initialization parameter and opens it, making it available for normal use.

Prerequisites

You must have the OSDBA role enabled.

Syntax

Keyword and Parameters

database  

is the name of the database to be created and can be up to eight bytes long. The database name can contain only ASCII characters. Oracle writes this name into the control file. If you subsequently issue an ALTER DATABASE statement and that explicitly specifies a database name, Oracle verifies that name with the name in the control file. Database names should also adhere to the rules described in "Schema Object Naming Rules".  

 

Note: You cannot use special characters from European or Asian character sets in a database name. For example, the umlaut is not allowed.  

 

If you omit the database name from a CREATE DATABASE statement, Oracle uses the name specified by the initialization parameter DB_NAME.  

CONTROLFILE REUSE  

reuses existing control files identified by the initialization parameter CONTROL_FILES, thus ignoring and overwriting any information they currently contain. Normally you use this option only when you are re-creating a database, rather than creating one for the first time. You cannot use this option if you also specify a parameter value that requires that the control file be larger than the existing files. These parameters are MAXLOGFILES, MAXLOGMEMBERS, MAXLOGHISTORY, MAXDATAFILES, and MAXINSTANCES.  

 

If you omit this option and any of the files specified by CONTROL_FILES already exist, Oracle returns an error message.  

LOGFILE  

specifies one or more files to be used as redo log files. Each filespec specifies a redo log file group containing one or more redo log file members, or copies. See the syntax description of filespec in "Filespec". All redo log files specified in a CREATE DATABASE statement are added to redo log thread number 1.  

 

GROUP  

uniquely identifies a redo log file group and can range from 1 to the value of the MAXLOGFILES parameter. You cannot specify multiple redo log file groups having the same GROUP value. If you omit this parameter, Oracle generates its value automatically. You can examine the GROUP value for a redo log file group through the dynamic performance table V$LOG.  

 

If you omit the LOGFILE clause, Oracle creates two redo log file groups by default. The names and sizes of the default files depends on your operating system.  

MAXLOGFILES  

specifies the maximum number of redo log file groups that can ever be created for the database. Oracle uses this value to determine how much space in the control file to allocate for the names of redo log files. The default, minimum, and maximum values depends on your operating system.  

 

The number of redo log file groups accessible to your instance is also limited by the initialization parameter LOG_FILES.  

MAXLOGMEMBERS  

specifies the maximum number of members, or copies, for a redo log file group. Oracle uses this value to determine how much space in the control file to allocate for the names of redo log files. The minimum value is 1. The maximum and default values depend on your operating system.  

MAXLOGHISTORY  

specifies the maximum number of archived redo log files for automatic media recovery of Oracle with the Parallel Server option. Oracle uses this value to determine how much space in the control file to allocate for the names of archived redo log files. The minimum value is 0. The default value is a multiple of the MAXINSTANCES value and depends on your operating system. The maximum value is limited only by the maximum size of the control file.

Note: This parameter is useful only if you are using Oracle with the Parallel Server option in parallel mode, and archivelog mode enabled.  

MAXDATAFILES  

specifies the initial sizing of the datafiles section of the control file at CREATE DATABASE or CREATE CONTROLFILE time. An attempt to add a file whose number is greater than MAXDATAFILES, but less than or equal to DB_FILES, causes the Oracle8 control file to expand automatically so that the datafiles section can accommodate more files.  

 

Note that the number of datafiles accessible to your instance is also limited by the initialization parameter DB_FILES.  

MAXINSTANCES  

specifies the maximum number of instances that can simultaneously have this database mounted and open. This value takes precedence over the value of initialization parameter INSTANCES. The minimum value is 1. The maximum and default values depend on your operating system.  

ARCHIVELOG  

establishes archivelog mode for redo log file groups. In this mode, the contents of a redo log file group must be archived before the group can be reused. This option prepares for the possibility of media recovery.  

NOARCHIVELOG  

establishes noarchivelog mode for redo log files groups. In this mode, the contents of a redo log file group need not be archived before the group can be reused. This option does not prepare for the possibility of media recovery.  

 

The default is noarchivelog mode. After creating the database, you can change between archivelog mode and noarchivelog mode with the ALTER DATABASE command.  

CHARACTER SET  

specifies the character set the database uses to store data. You cannot change the database character set after creating the database. The supported character sets and default value of this parameter depend on your operating system.  

 

You can specify any supported character set except the following fixed-width, multibyte character sets, which can be used only as the national character set:

JA16SJISFIXED

JA16EUCFIXED

JA16DBCSFIXED

For more information about valid character sets, see in the Oracle8 Reference.  

NATIONAL CHARACTER SET  

specifies the national character set used to store data in columns specifically defined as NCHAR, NCLOB, or NVARCHAR2. You cannot change the national character set after creating the database. If not specified, the national character set defaults to the database character set. See Oracle8 Reference for valid character set names.  

DATAFILE  

specifies one or more files to be used as datafiles. See the syntax description of filespec in "Filespec". All these files become part of the SYSTEM tablespace. If you omit this clause, Oracle creates one datafile by default. The name and size of this default file depend on your operating system.

Note: Oracle recommends that the total initial space allocated for the SYSTEM tablespace be a minimum of 5 megabytes.  

AUTOEXTEND  

enables or disables the automatic extension of a datafile. If you do not specify this clause, datafiles are not automatically extended.  

 

OFF  

disables autoextend if it is turned on. NEXT and MAXSIZE are set to zero. Values for NEXT and MAXSIZE must be respecified in ALTER DATABASE AUTOEXTEND or ALTER TABLESPACE AUTOEXTEND commands.  

 

ON  

enables autoextend.  

 

NEXT  

specifies the size in bytes of the next increment of disk space to be allocated to the datafile automatically when more extents are required. You can also use K or M to specify this size in kilobytes or megabytes. The default is one data block.  

 

MAXSIZE  

specifies the maximum disk space allowed for automatic extension of the datafile.  

 

UNLIMITED  

sets no limit on the allocation of disk space to the datafile.  

Examples

Example I

The following statement creates a small database using defaults for all arguments:

CREATE DATABASE; 
Example II

The following statement creates a database and fully specifies each argument:

CREATE DATABASE newtest 
CONTROLFILE REUSE 
LOGFILE
GROUP 1 ('diskb:log1.log', 'diskc:log1.log') SIZE 50K, 
GROUP 2 ('diskb:log2.log', 'diskc:log2.log') SIZE 50K 
MAXLOGFILES 5 
MAXLOGHISTORY 100 
DATAFILE 'diska:dbone.dat' SIZE 2M 
MAXDATAFILES 10 
MAXINSTANCES 2 
ARCHIVELOG 
CHARACTER SET US7ASCII
NATIONAL CHARACTER SET JA16SJISFIXED
DATAFILE  
'disk1:df1.dbf' AUTOEXTEND ON
'disk2:df2.dbf' AUTOEXTEND ON NEXT 10M MAXSIZE UNLIMITED;

Related Topics

ALTER DATABASE
CREATE ROLLBACK SEGMENT
CREATE TABLESPACE

CREATE DATABASE LINK

Purpose

To create a database link. A database link is a schema object in the local database that allows you to access objects on a remote database. The remote database can be either an Oracle or a non-Oracle system. See also "Creating Database Links".

Prerequisites

To create a private database link, you must have CREATE DATABASE LINK system privilege. To create a public database link, you must have CREATE PUBLIC DATABASE LINK system privilege. Also, you must have CREATE SESSION privilege on the remote Oracle database. Net8 must be installed on both the local and remote Oracle databases. To access non-Oracle systems you must use the Oracle8 Heterogeneous Services.

Syntax

Keyword and Parameters

SHARED  

uses a single network connection to create a public database link that can be shared between multiple users. This option is available only with the multithreaded server configuration. For more information about shared database links, see Oracle8 Distributed Database Systems.  

PUBLIC  

creates a public database link available to all users. If you omit this option, the database link is private and is available only to you.  

dblink  

is the complete or partial name of the database link. For guidelines for naming database links, see "Referring to Objects in Remote Databases" and "Current-User Database Links".  

CONNECT TO  

enables a connection to the remote database.  

 

CURRENT_USER  

creates a current user database link. To use a current database link, the current user must be a global user authenticated by the Oracle Security Server. See also "Current-User Database Links".  

 

user IDENTIFIED BY password  

is the username and password used to connect to the remote database (fixed user database link). If you omit this clause, the database link uses the username and password of each user who is connected to the database (connected user database link).  

authenticated_clause  

specifies the username and password on the target instance. This clause authenticates the user to the remote server and is required for security. The specified username and password must be a valid username and password on the remote instance. The username and password are used only for authentication; no other operations are performed on behalf of this user.  

 

You must specify this clause when using the SHARED option.  

USING 'connect string'  

specifies the service name of a remote database. For information on specifying remote databases, see Net8 Administrator's Guide.  

Creating Database Links

You cannot create a database link in another user's schema, and you cannot qualify dblink with the name of a schema. Periods are permitted in names of database links, so Oracle interprets the entire name, such as RALPH.LINKTOSALES, as the name of a database link in your schema rather than as a database link named LINKTOSALES in the schema RALPH.

Once you have created a database link, you can use it to refer to tables and views on the remote database. You can refer to a remote table or view in a SQL statement by appending @dblink to the table or view name. You can query a remote table or view with the SELECT command. If you are using Oracle with the distributed option, you can also access remote tables and views using any of the following commands:

See Oracle8 Application Developer's Guide for information about accessing remote tables or views with PL/SQL functions, procedures, packages, and datatypes.

The number of different database links that can appear in a single statement is limited to the value of the initialization parameter OPEN_LINKS.

Current-User Database Links

A current user database link is one that contains no user credentials and that enables a connection to a remote database as the current user. To use the link, the current user must be a global user with global accounts on both the local and remote databases. Both databases must be members of the same security domain.

To create a global user, see CREATE USER. For detailed information about current database links, see Oracle8 Distributed Database Systems.

CURRENT_USER

When executing a stored object (such as a procedure, view, or trigger) that initiates a database link, CURRENT_USER is the username that created the stored object, and not the username that called the object. For example if the database link appears inside procedure SCOTT.P (created by SCOTT), and user JANE calls procedure SCOTT.P, the current user is SCOTT.

If the database link is used directly, that is, NOT from within a stored object, then the current user is the same as the connected user.

Examples

Example I

The following example defines a current-user database link:

CREATE DATABASE LINK sales.hq.acme.com
CONNECT TO CURRENT_USER
USING 'sales';
Example II

The following statement defines a fixed-user database link named SALES.HQ.ACME.COM:

CREATE DATABASE LINK sales.hq.acme.com 
CONNECT TO scott IDENTIFIED BY tiger 
USING 'sales' 

Once this database link is created, you can query tables in the schema SCOTT on the remote database in this manner:

SELECT *
FROM emp@sales.hq.acme.com 

You can also use DML commands to modify data on the remote database:

INSERT INTO accounts@sales.hq.acme.com(acc_no, acc_name, balance)
VALUES (5001, 'BOWER', 2000) 

UPDATE accounts@sales.hq.acme.com 
SET balance = balance + 500 

DELETE FROM accounts@sales.hq.acme.com 
WHERE acc_name = 'BOWER' 

You can also access tables owned by other users on the same database. This example assumes SCOTT has access to ADAM's DEPT table:

SELECT *
FROM adams.dept@sales.hq.acme.com 

The previous statement connects to the user SCOTT on the remote database and then queries ADAM's DEPT table.

A synonym may be created to hide the fact that SCOTT's EMP table is on a remote database. The following statement causes all future references to EMP to access a remote EMP table owned by SCOTT:

CREATE SYNONYM emp 
FOR scott.emp@sales.hq.acme.com; 
Example III

The following statement defines a shared public fixed user database link named SALES.HQ.ACME.COM that refers to user SCOTT with password TIGER on the database specified by the string service name 'SALES':

CREATE SHARED PUBLIC DATABASE LINK sales.hq.acme.com 
CONNECT TO scott IDENTIFIED BY tiger 
AUTHENTICATED BY anupam IDENTIFIED BY bhide
USING 'sales'; 
Example IV

The following example creates a current user database link:

CREATE DATABASE LINK sales.hq.acme.com
CONNECT TO CURRENT_USER
USING 'sales';

Related Topics

CREATE SYNONYM
CREATE USER
DELETE
INSERT
LOCK TABLE
SELECT
UPDATE
PL/SQL User's Guide and Reference
Oracle8 Distributed Database Systems

CREATE DIRECTORY

Purpose

Use CREATE DIRECTORY to create a directory object, which represents an operating system directory for administering access to, and the use of, BFILEs stored outside the database. A directory is an alias for a full pathname on the server's file system where the files are actually located.

Prerequisites

You must have CREATE ANY DIRECTORY system privileges to create directories.

You must also create a corresponding operating system directory for file storage. Your system or database administrator must ensure that the operating system directory has the correct read permissions for Oracle processes.

Syntax

Keywords and Parameters

OR REPLACE  

re-creates the directory database object if it already exists. You can use this option to change the definition of an existing directory without dropping, re-creating, and regranting database object privileges previously granted on the directory.

Users who had previously been granted privileges on a redefined directory can still access the directory without being regranted the privileges.  

directory  

is the name of the directory object to be created. The maximum length of directory is 30 bytes. You cannot qualify a directory object with a schema name. See also "Directory Objects".

Note: Oracle does not verify that the directory you specify actually exists; therefore, take care that you specify a valid directory in your operating system. In addition, if your operating system uses case-sensitive pathnames, be sure you specify the directory in the correct format. (However, you need not include a trailing slash at the end of the pathname.)  

'path_name'  

is the full pathname of the operating system directory on the server where the files are located. Note that the single quotes are required, with the result that the path name is case sensitive.  

Directory Objects

A directory object specifies an alias name for a directory on the server's file system where external binary file LOBs (BFILEs) are located. You can use directory names when referring to BFILEs in your PL/SQL code and OCI calls, rather than hard-coding the operating system pathname, thereby allowing greater file management flexibility.

The Oracle BFILE datatype provides access to the external file system. A BFILE column or attribute contains a locator to an external file on the operating system, rather than the file itself. The locator maintains the directory alias and the filename.

All directories are created in a single namespace and are not owned by an individual's schema. You can secure access to the BFILEs stored within the directory structure by granting object privileges on the directories to specific users. When you create a directory, you are automatically granted the READ object privilege and can grant READ privileges to other users and roles. The DBA can also grant this privilege to other users and roles.

Privileges granted for the directory are created independently of the permissions defined for the operating system directory; therefore, the two may or may not correspond exactly. For example, an error occurs if user SCOTT is granted READ privilege on the directory schema object, but the corresponding operating system directory does not have READ permission defined for Oracle processes.

Example

The following statement redefines directory database object BFILE_DIR to enable access to BFILEs stored in the operating system directory /PRIVATE1/LOB/FILES:

CREATE OR REPLACE DIRECTORY bfile_dir AS '/private1/LOB/files';

Related Topics

DROP DIRECTORY
GRANT (System Privileges and Roles)
"Large Object (LOB) Datatypes"

CREATE FUNCTION

Purpose

To create a stored function or to register an external function.

A stored function (also called a user function) is a set of PL/SQL statements you can call by name. Stored functions are very similar to procedures, except that a function returns a value to the environment in which it is called. User functions can be used as part of a SQL expression. For a general discussion of procedures and functions, see CREATE PROCEDURE. For examples of creating functions, see "Examples"

An external function is a third-generation language (3GL) routine stored in a shared library that can be called from SQL or PL/SQL. To call an external function, you must provide information in your PL/SQL function about where to find the external function, how to call it, and what to pass to it.

The CREATE FUNCTION command creates a function as a standalone schema object. You can also create a function as part of a package. For information on creating packages, see CREATE PACKAGE.

For more information about registering external functions, see the PL/SQL User's Guide and Reference.

Prerequisites

Before a stored function can be created, the user SYS must run the SQL script DBMSSTDX.SQL. The exact name and location of this script depend on your operating system.

To create a function in your own schema, you must have CREATE PROCEDURE system privilege. To create a function in another user's schema, you must have CREATE ANY PROCEDURE system privilege.

To call an external function, you must have EXECUTE privileges on the callout library in which the function resides.

To create a stored function, you must be using Oracle with PL/SQL installed. For more information, see PL/SQL User's Guide and Reference.

To embed a CREATE FUNCTION statement inside an Oracle precompiler program, you must terminate the statement with the keyword END-EXEC followed by the embedded SQL statement terminator for the specific language.

Syntax

external_body::=

Keywords and Parameters

OR REPLACE  

re-creates the function if it already exists. Use this option to change the definition of an existing function without dropping, re-creating, and regranting object privileges previously granted on the function. If you redefine a function, Oracle recompiles it. For information on recompiling functions, see ALTER FUNCTION.  

 

Users who had previously been granted privileges on a redefined function can still access the function without being regranted the privileges.  

schema  

is the schema to contain the function. If you omit schema, Oracle creates the function in your current schema.  

function  

is the name of the function to be created.  

argument  

is the name of an argument to the function. If the function does not accept arguments, you can omit the parentheses following the function name.  

IN  

specifies that you must supply a value for the argument when calling the function. This is the default.  

OUT  

specifies the function will set the value of the argument.  

IN OUT  

specifies that a value for the argument can be supplied by you and may be set by the function.  

datatype  

is the datatype of an argument. An argument can have any datatype supported by PL/SQL.  

 

The datatype cannot specify a length, precision, or scale. Oracle derives the length, precision, or scale of an argument from the environment from which the function is called.  

RETURN datatype  

specifies the datatype of the function's return value. Because every function must return a value, this clause is required. The return value can have any datatype supported by PL/SQL.  

 

The datatype cannot specify a length, precision, or scale. Oracle derives the length, precision, or scale of the return value from the environment from which the function is called. For information on PL/SQL datatypes, see PL/SQL User's Guide and Reference.  

pl/sql_subprogram_body  

is the definition of the function. Function definitions are writing in PL/SQL. For information on PL/SQL, see PL/SQL User's Guide and Reference.

 

external_body_clause  

identifies the external function to be registered.  

AS EXTERNAL  

identifies an external 3GL function stored in a shareable library. The AS EXTERNAL clause is the interface between PL/SQL and the external function.  

LIBRARY  

specifies the shared library in which the external function is stored. You must have EXECUTE privileges on the library. See CREATE LIBRARY for the syntax.  

library_name  

is a PL/SQL identifier. Enclosing library_name in double quotes makes it case sensitive, but quotes are not required.  

NAME external_function_name  

specifies the external function to be called. Enclosing external_function_name in double quotes makes it case sensitive, but quotes are not required. If you omit the name, it defaults to the PL/SQL subprogram (uppercase) name.  

LANGUAGE  

specifies the 3GL in which the external function was written. Currently, the only language name supported is C. If you omit the name, it defaults to C.  

CALLING STANDARD  

specifies the calling standard (C or Pascal) under which the external function was compiled. If you omit the calling standard, it defaults to C.  

WITH CONTEXT  

specifies that a context pointer will be the first parameter passed to the external function. The context is opaque to the external function but is available to access functions called by the external function. For more information about the WITH CONTEXT clause, see PL/SQL User's Guide and Reference.  

PARAMETERS  

specifies the positions and datatypes of parameters passed to the external function. It can also specify parameter properties such as current length and maximum length, and the preferred parameter passing method (by value or by reference). For more information about parameter passing see PL/SQL User's Guide and Reference.  

Examples

Example I

The following statement creates the function GET_BAL:

CREATE FUNCTION get_bal(acc_no IN NUMBER) 
RETURN NUMBER 
IS 
acc_bal NUMBER(11,2); 
BEGIN 
SELECT balance 
INTO acc_bal 
FROM accounts 
WHERE account_id = acc_no; 
RETURN(acc_bal); 
END; 

The GET_BAL function returns the balance of a specified account.

When you call the function, you must specify the argument ACC_NO, the number of the account whose balance is sought. The datatype of ACC_NO is NUMBER.

The function returns the account balance. The RETURN clause of the CREATE FUNCTION statement specifies the datatype of the return value to be NUMBER.

The function uses a SELECT statement to select the BALANCE column from the row identified by the argument ACC_NO in the ACCOUNTS table. The function uses a RETURN statement to return this value to the environment in which the function is called.

The function created above can be used in a SQL statement. For example:

SELECT get_bal(100) FROM DUAL;
Example II

The following statement creates PL/SQL standalone function GET_VAL that registers the C routine C_GET_VAL as an external function:

CREATE FUNCTION get_val
( x_val IN BINARY_INTEGER,
y_val IN BINARY_INTEGER,
image IN LONG RAW )
RETURN BINARY_INTEGER AS EXTERNAL LIBRARY c_utils
NAME "c_get_val"
LANGUAGE C;

Related Topics

ALTER FUNCTION
CREATE LIBRARY
CREATE PACKAGE
CREATE PACKAGE BODY
CREATE PROCEDURE
DROP FUNCTION
PL/SQL User's Guide and Reference

CREATE INDEX

Purpose

To create an index on

An index is a schema object that contains an entry for each value that appears in the indexed column(s) of the table or cluster and provides direct, fast access to rows. A partitioned index consists of partitions containing an entry for each value that appears in the indexed column(s) of the table. See also "Creating Indexes".


Note:

Descriptions of commands and clauses preceded by are available only if the Oracle objects option is installed on your database server.

 

Prerequisites

To create an index in your own schema, one of the following conditions must be true:

To create an index in another schema, you must have CREATE ANY INDEX system privilege.

Also, the owner of the schema to contain the index must have either space quota on the tablespaces to contain the index or index partitions, or UNLIMITED TABLESPACE system privilege.

See also "Index Columns".

Syntax

parallel_clause: See PARALLEL clause.

storage_clause: See STORAGE clause

Keywords and Parameters

UNIQUE  

specifies that the value of the column (or combination of columns) in the table to be indexed must be unique.  

 

If the index is local nonprefixed (see LOCAL clause below), then the index key must contain the partitioning key.  

BITMAP  

specifies that index is to be created as a bitmap, rather than as a B-tree. See also "Creating Bitmap Indexes".

Note: You cannot use this keyword when creating a global partitioned index.  

You can specify either UNIQUE or BITMAP, but you cannot create a unique bitmap index.  

schema  

is the schema to contain the index. If you omit schema, Oracle creates the index in your own schema.  

index  

is the name of the index to be created. (See also "Multiple Indexes Per Table".) An index can contain several partitions.  

 

You cannot range partition a cluster index or an index defined on a clustered table.  

table  

is the name of the table for which the index is to be created. If you do not qualify table with schema, Oracle assumes the table is contained in your own schema.  

 

If the index is LOCAL, then table must be partitioned.  

 

You cannot create an index on an index-organized table.  

 

You can create an index on a nested table storage table.  

column  

is the name of a column in the table. An index can have as many as 32 columns. A column of an index cannot be of datatype LONG or LONG RAW. See also "Index Columns".  

 

You can create an index on a scalar object attribute column or on the system-defined NESTED_TABLE_ID column of the nested table storage table. If you specify an object attribute column, the column name must be qualified with the table name. If you specify a nested table column attribute, it must be qualified with the outermost table name, the containing column name, and all intermediate attribute names leading to the nested table column attribute. See also "Creating Indexes on Nested Table Columns".

See also "Nulls".  

ASC / DESC  

are allowed for DB2 syntax compatibility, although indexes are always created in ascending order. Indexes on character data are created in ascending order of the character values in the database character set.  

CLUSTER  

specifies the cluster for which a cluster index is to be created. If you do not qualify cluster with schema, Oracle assumes the cluster is contained in your current schema. You cannot create a cluster index for a hash cluster. See also "Creating Cluster Indexes".  

index_physical_attributes_clause  

establishes values for the INITRANS, MAXTRANS, and PCTFREE parameters and storage characteristics for the index. See CREATE TABLE.  

PCTFREE  

is the percentage of space to leave free for updates and insertions within each of the index's data blocks.  

storage_clause  

establishes the storage characteristics for the index. See the STORAGE clause.  

TABLESPACE  

is the name of the tablespace to hold the index or index partition. If you omit this option, Oracle creates the index in the default tablespace of the owner of the schema containing the index.  

 

For a partitioned index, this is the tablespace name.  

 

For a local index, you can specify the keyword DEFAULT in place of tablespace. New partitions added to the local index will be created in the same tablespace(s) as the corresponding partition(s) of the underlying table.  

NOSORT  

indicates to Oracle that the rows are stored in the database in ascending order; therefore Oracle does not have to sort the rows when creating the index. You cannot specify REVERSE with this option. See also "The NOSORT Option".  

REVERSE  

stores the bytes of the index block in reverse order, excluding the ROWID. You cannot specify NOSORT with this option.  

 

You cannot reverse a bitmap index.  

LOGGING /NOLOGGING  

specifies that the creation of the index will be logged (LOGGING) or not logged (NOLOGGING) in the redo log file. It also specifies that subsequent Direct Loader (SQL*Loader) and direct-load INSERT operations against the index are logged or not logged. LOGGING is the default.  

 

If index is nonpartitioned, this is the logging attribute of the index.  

 

For partitioned index, the logging attribute specified is the default physical attribute of the segments associated with the index partitions. The default logging value applies to all partitions specified in the CREATE statement (and on subsequent ALTER TABLE ADD PARTITION statements) unless you specify LOGGING/NOLOGGING in the PARTITION description clause.  

 

In NOLOGGING mode, data is modified with minimal logging (to mark new extents invalid and to record dictionary changes). When applied during media recovery, the extent invalidation records mark a range of blocks as logically corrupt, since the redo data is not logged. Thus if you cannot afford to lose this index, it is important to take a backup after the NOLOGGING operation.  

 

If the database is run in ARCHIVELOG mode, media recovery from a backup taken before the LOGGING operation will re-create the index. However, media recovery from a backup taken before the NOLOGGING operation will not re-create the index.  

 

The logging attribute of the index is independent of that of its base table.  

 

If the [NO]LOGGING clause is omitted, the logging attribute of the index defaults to the logging attribute of the tablespace in which it resides.  

 

For more information about the LOGGING option and Parallel DML, see "NOLOGGING", Oracle8 Concepts and Oracle8 Parallel Server Concepts and Administration.  

GLOBAL  

specifies that the partitioning of the index is user defined and is not equipartitioned with the underlying table. By default, nonpartitioned indexes are global indexes.  

PARTITION BY RANGE  

specifies that the global index is partitioned on the ranges of values from the columns specified in column_list. You cannot specify this clause for a LOCAL index.  

(column_list)  

is the name of the column(s) of a table on which the index is partitioned. The column_list must specify a left prefix of the index column list.  

 

You cannot specify more than 32 columns in column_list, and the columns cannot contain the ROWID pseudocolumn or a column of type ROWID.  

LOCAL  

specifies that the index is range partitioned on the same columns, with the same number of partitions, and the same partition bounds as table. Oracle automatically maintains LOCAL index partitioning as the underlying table is repartitioned.  

PARTITION partition_name  

describes the individual partitions. The number of clauses determines the number of partitions. If the index is local, the number of index partitions must be equal to the number of the table partitions, and in the same order.  

 

The partition_name is the name of the physical index partition. If partition_name is omitted, Oracle generates a name with the form SYS_Pn.  

 

For local indexes, if partition_name is omitted, Oracle generates a name that is consistent with the corresponding table partition. If the name conflicts with an existing index partition name, the form SYS_Pn is used.

See also "Creating Partitioned Indexes".  

VALUES LESS THAN (value_list)  

specifies the (noninclusive) upper bound for the current partition in a global index. The value_list is a comma-separated, ordered list of literal values corresponding to column_list in the PARTITION BY RANGE clause. Always specify MAXVALUE as the value_list of the last partition.  

 

You cannot specify this clause for a local index.  

parallel_clause  

specifies the degree of parallelism for creating the index. See the PARALLEL clause.  

Creating Indexes

An index is an ordered list of all the values that reside in a group of one or more columns at a given time. Such a list makes queries that test the values in those columns vastly more efficient. However, indexes take up data storage space and must be changed whenever the data is changed. Therefore, you should make a cost-benefit analysis in each case to determine whether and how indexes should be used. Oracle can use indexes to improve performance when:

When you initially insert rows into a new table, it is generally faster to create the table, insert the rows, and then create the index. If you create the index before inserting the rows, Oracle must update the index for every row inserted.

Oracle recommends that you do not explicitly define UNIQUE indexes on tables; uniqueness is strictly a logical concept and should be associated with the definition of a table. Instead, define UNIQUE integrity constraints on the desired columns. Oracle enforces UNIQUE integrity constraints by automatically defining a unique index on the unique key. Exceptions to this recommendation are usually performance related. For example, using a CREATE TABLE ... AS SELECT with a UNIQUE constraint is very much slower than creating the table without the constraint and then manually creating the UNIQUE index.

If indexes contain NULLs, the NULLS generally are considered distinct values. There is, however, one exception: if all the non-NULL values in two or more rows of an index are identical, the rows are considered identical; therefore, UNIQUE indexes prevent this from occurring. This does not apply if there are no non-NULL values-in other words, if the rows are entirely NULL.


Note:

You cannot create an index on columns or attributes whose type is user-defined, LOB, or REF. The only exception is that Oracle supports creation of an index on REF type columns or attributes that have been defined with a SCOPE clause.

 

Index Columns

An index can contain a maximum of 32 columns. The index entry becomes the concatenation of all data values from each column. You can specify the columns in any order. The order you choose is important to how Oracle uses the index.

When appropriate, Oracle uses the entire index or a leading portion of the index. Assume an index named IDX1 is created on columns A, B, and C of table TAB1 (in the order A, B, C). Oracle uses the index for references to columns A, B, C (the entire index); A, B; or just column A. References to columns B and C do not use the IDX1 index. Of course, you can also create another index just for columns B and C.

Multiple Indexes Per Table

You can create unlimited indexes for a table provided that the combination of columns differs for each index. You can create more than one index using the same columns provided that you specify distinctly different combinations of the columns. For example, the following statements specify valid combinations:

CREATE INDEX emp_idx1 ON emp (ename, job);
CREATE INDEX emp_idx2 ON emp (job, ename);

You cannot create an index that references only one column in a table if another such index already exists.

Note that each index increases the processing time needed to maintain the table during updates to indexed data. Thus, updating a table with a single index will take less time than if the table had five indexes.

The NOSORT Option

The NOSORT option can substantially reduce the time required to create an index. Normal index creation first sorts the rows of the table based on the index columns and then builds the index. The sort operation is often a substantial portion of the total work involved. If the rows are already physically stored in ascending order (based on the indexed column values), then the NOSORT option causes Oracle to bypass the sort phase of the process.

You cannot use the NOSORT option to create a cluster index, partitioned index, or a bitmap index.

The NOSORT option also reduces the amount of space required to build the index. Oracle uses temporary segments during the sort. Since a sort is not performed, the index is created with much less temporary space.

To use the NOSORT option, you must guarantee that the rows are physically sorted in ascending order. However, you run no risk by trying the NOSORT option. If your rows are not in the ascending order, Oracle returns an error. You can issue another CREATE INDEX without the NOSORT option. Because of the physical data independence inherent in relational database management systems, especially Oracle, there is no way to force a physical internal order on a table. The CREATE INDEX command with the NOSORT option should be used immediately after the initial load of rows into a table.

NOLOGGING

The NOLOGGING option may substantially reduce the time required to create a large index. This feature is particularly useful after creating a large index in parallel. For backup and recovery considerations, see Oracle8 Backup and Recovery Guide and Oracle8 Administrator's Guide.

Example

To quickly create an index in parallel on a table that was created using a fast parallel load (so all rows are already sorted), you might issue the following statement:

CREATE INDEX i_loc
ON big_table (akey)
NOSORT
NOLOGGING
PARALLEL (DEGREE 5);

Nulls

Oracle does not index table rows in which all key columns are NULL, except in the case of bitmap indexes.

Example

Consider the following statement:

SELECT ename 
FROM emp 
WHERE comm IS NULL; 

The above query does not use an index created on the COMM column unless it is a bitmap index.

Creating Cluster Indexes

Oracle does not automatically create an index for a cluster when the cluster is initially created. Data manipulation language statements cannot be issued against clustered tables until a cluster index has been created.

Example

To create an index for the EMPLOYEE cluster, issue the following statement:

CREATE INDEX ic_emp 
ON CLUSTER employee 

Note that no index columns are specified, because the index is automatically built on all the columns of the cluster key. For cluster indexes, all rows are indexed.

Creating Partitioned Indexes

Indexes can be local prefixed (unique or nonunique), local nonprefixed (unique, but only when the partitioning key is a subset of the index key or nonunique), or global prefixed (unique or nonunique). Oracle does not support global nonprefixed indexes. Local indexes are always partitioned. Global indexes can be nonpartitioned or partitioned.

Index partitions must be listed in order. For a global index, this means that the partition bound of the first partition listed must be less than the partition bound of the second partition listed, and the partition bound of the second partition listed must be less than the third, and so on. For a local index, you must list the partitions in the same order as the partitions of the underlying table to which they correspond.

Example

The following statement creates a global prefixed index STOCK_IX on table STOCK_XACTIONS with two partitions, one for each half of the alphabet. The index partition names are system generated:

CREATE INDEX stock_ix ON stock_xactions
  (stock_symbol, stock_series)
   GLOBAL PARTITION BY RANGE (stock_symbol)
     (PARTITION VALUES LESS THAN ('N') TABLESPACE ts3,
      PARTITION VALUES LESS THAN (MAXVALUE) TABLESPACE ts4);

Creating Bitmap Indexes

Bitmap indexes store the ROWIDs associated with a key value as a bitmap. Each bit in the bitmap corresponds to a possible ROWID, and if the bit is set, it means that the row with the corresponding ROWID contains the key value. The internal representation of bitmaps is best suited for applications with low levels of concurrent transactions, such as data warehousing. See Oracle8 Concepts and Oracle8 Tuning for more information about using bitmap indexes.

Example

To create a bitmap partitioned index on a table with four partitions, issue the following statement:

CREATE BITMAP INDEX partno_ix
ON lineitem(partno)
TABLESPACE ts1
LOCAL (PARTITION quarter1 TABLESPACE ts2,
PARTITION quarter2 STORAGE (INITIAL 10K NEXT 2K),
PARTITION quarter3 TABLESPACE ts2,
PARTITION quarter4);

You cannot create bitmap indexes, unique bitmap indexes, or global partitioned indexes.

Creating Indexes on Nested Table Columns

Creating a table with nested table columns implicitly creates a storage table for each nested table column. The storage table stores the rows of the nested table values and the nested table identifier values assigned to each row. These identifier values are contained in a storage table pseudocolumn called NESTED_TABLE_ID.

You create an index on a nested table column by creating the index on the nested table storage table. You can include the NESTED_TABLE_ID pseudocolumn to create a UNIQUE index, which effectively ensures that the rows of a nested table value are distinct.

Example

In the following example, UNIQUE index UNIQ_PROJ_INDX is created on storage table NESTED_PROJECT_TABLE. Including pseudocolumn NESTED_TABLE_ID ensures distinct rows in nested table column PROJS_MANAGED:

CREATE TYPE proj_table_type AS TABLE OF proj_type;

CREATE TABLE employee ( emp_num NUMBER, emp_name CHAR(31),
projs_managed proj_table_type )
NESTED TABLE projs_managed STORE AS nested_project_table;

CREATE UNIQUE INDEX uniq_proj_indx
ON nested_project_table ( NESTED_TABLE_ID, proj_num);

Related Topics

ALTER INDEX
CREATE TABLE
"Index-Organized Tables"
DROP INDEX
CONSTRAINT clause
STORAGE clause

CREATE LIBRARY

Purpose

To create a schema object (library), which represents an operating-system shared library, from which SQL and PL/SQL can call external third-generation-language (3GL) functions and procedures. See "Examples".

Prerequisites

You must have CREATE ANY LIBRARY system privileges. To use the procedures and functions stored in the library, you must have EXECUTE object privileges on the library.

The CREATE LIBRARY command is valid only on platforms that support shared libraries and dynamic linking.

Syntax

filespec: See "Filespec".

Keywords and Parameters

OR REPLACE  

re-creates the library if it already exists. Use this option to change the definition of an existing library without dropping, re-creating, and regranting schema object privileges granted on it.

Users who had previously been granted privileges on a redefined library can still access the library without being regranted the privileges.  

libname  

is the name of the library (schema object) from which SQL and PL/SQL will call external 3GL functions and procedures.  

'filespec'  

is a non-zero-length string, enclosed in single quotes. The 'filespec' is not interpreted by PL/SQL.

The directory and filename specified in 'filespec' are not interpreted by PL/SQL; therefore the existence of the specification is not checked until procedure run time.  

Examples

Example I

The following statement creates library EXT_LIB:

CREATE LIBRARY ext_lib AS '/OR/lib/ext_lib.so';
Example II

The following example re-creates library EXT_LIB:

CREATE OR REPLACE ext_lib IS '/OR/newlib/ext_lib.so';

Related Topics

CREATE FUNCTION
CREATE PROCEDURE
PL/SQL User's Guide and Reference

CREATE PACKAGE

Purpose

To create the specification for a stored package. A package is an encapsulated collection of related procedures, functions, and other program objects stored together in the database. The specification declares these objects.

Prerequisites

Before a package can be created, the user SYS must run the SQL script DBMSSTDX.SQL. The exact name and location of this script depend on your operating system.

To create a package in your own schema, you must have CREATE PROCEDURE system privilege. To create a package in another user's schema, you must have CREATE ANY PROCEDURE system privilege.

To embed a CREATE PACKAGE statement inside an Oracle precompiler program, you must terminate the statement with the keyword END-EXEC followed by the embedded SQL statement terminator for the specific language.

For more information, see PL/SQL User's Guide and Reference.

Syntax

Keywords and Parameters

OR REPLACE  

re-creates the package specification if it already exists. Use this option to change the specification of an existing package without dropping, re-creating, and regranting object privileges previously granted on the package. If you change a package specification, Oracle recompiles it. For information on recompiling package specifications, see ALTER PROCEDURE.  

 

Users who had previously been granted privileges on a redefined package can still access the package without being regranted the privileges.  

schema  

is the schema to contain the package. If you omit schema, Oracle creates the package in your own schema.  

package  

is the name of the package to be created. See also "Packages".  

pl/sql_package_spec  

is the package specification. The package specification can declare program objects. Package specifications are written in PL/SQL. For information on PL/SQL, including writing package specifications, see PL/SQL User's Guide and Reference.  

Packages

A package is an encapsulated collection of related program objects stored together in the database. Program objects are: procedures, functions, variables, constants, cursors, and exceptions.

Using packages is an alternative to creating procedures and functions as standalone schema objects. Packages have many advantages over stand-alone procedures and functions. They:

For more information on these and other benefits of packages, see Oracle8 Application Developer's Guide.

How to Create Packages

To create a package, you must perform two distinct steps:

  1. Create the package specification with the CREATE PACKAGE command. You can declare program objects in the package specification. Such objects are called public objects. Public objects can be referenced outside the package as well as by other objects in the package.
  2. Create the package body with the CREATE PACKAGE BODY command. You can declare and define program objects in the package body:
    • You must define public objects declared in the package specification.
    • You can also declare and define additional package objects. Such objects are called private objects. Private objects are declared in the package body rather than in the package specification, so they can be referenced only by other objects in the package. They cannot be referenced outside the package.

See CREATE PACKAGE BODY.

The Separation of Specification and Body

Oracle stores the specification and body of a package separately in the database. Other schema objects that call or reference public program objects depend only on the package specification, not on the package body. This distinction allows you to change the definition of a program object in the package body without causing Oracle to invalidate other schema objects that call or reference the program object. Oracle invalidates dependent schema objects only if you change the declaration of the program object in the package specification.

Example

This SQL statement creates the specification of the EMP_MGMT package:

CREATE PACKAGE emp_mgmt AS 
   FUNCTION hire(ename VARCHAR2, job VARCHAR2, mgr NUMBER,
                 sal NUMBER, comm NUMBER, deptno NUMBER) 
      RETURN NUMBER; 
   FUNCTION create_dept(dname VARCHAR2, loc VARCHAR2) 
      RETURN NUMBER; 
   PROCEDURE remove_emp(empno NUMBER); 
   PROCEDURE remove_dept(deptno NUMBER); 
   PROCEDURE increase_sal(empno NUMBER, sal_incr NUMBER); 
   PROCEDURE increase_comm(empno NUMBER, comm_incr NUMBER); 
      no_comm EXCEPTION; 
      no_sal EXCEPTION; 
END emp_mgmt;

The specification for the EMP_MGMT package declares the following public program objects:

All of these objects are available to users who have access to the package. After creating the package, you can develop applications that call any of the package's public procedures or functions or raise any of the package's public exceptions.

Before you can call this package's procedures and functions, you must define these procedures and functions in the package body. For an example of a CREATE PACKAGE BODY statement that creates the body of the EMP_MGMT package, see CREATE PACKAGE BODY.

Related Topics

ALTER PACKAGE
CREATE FUNCTION
CREATE PROCEDURE
CREATE PACKAGE BODY
DROP PACKAGE

CREATE PACKAGE BODY

Purpose

To create the body of a stored package. A package is an encapsulated collection of related procedures, stored functions, and other program objects stored together in the database. The body defines these objects.

Packages are an alternative to creating procedures and functions as standalone schema objects. For a discussion of packages, including how to create packages, see CREATE PACKAGE. For some illustrations, see "Examples".

Prerequisites

Before a package can be created, the user SYS must run the SQL script DBMSSTDX.SQL. The exact name and location of this script depend on your operating system.

To create a package in your own schema, you must have CREATE PROCEDURE system privilege. To create a package in another user's schema, you must have CREATE ANY PROCEDURE system privilege.

To embed a CREATE PACKAGE BODY statement inside an Oracle precompiler program, you must terminate the statement with the keyword END-EXEC followed by the embedded SQL statement terminator for the specific language.

For more information, see PL/SQL User's Guide and Reference.

Syntax

Keywords and Parameters

OR REPLACE  

re-creates the package body if it already exists. Use this option to change the body of an existing package without dropping, re-creating, and regranting object privileges previously granted on it. If you change a package body, Oracle recompiles it. For information on recompiling package bodies, see ALTER PACKAGE  

 

Users who had previously been granted privileges on a redefined package can still access the package without being regranted the privileges.  

schema  

is the schema to contain the package. If you omit schema, Oracle creates the package in your current schema.  

package  

is the name of the package to be created.  

pl/sql_package_body  

is the package body. The package body can declare and define program objects. Package bodies are written in PL/SQL. For information on PL/SQL, including writing package bodies, see PL/SQL User's Guide and Reference.  

Examples

Example

This SQL statement creates the body of the EMP_MGMT package:

CREATE PACKAGE BODY emp_mgmt AS 
   tot_emps  NUMBER; 
   tot_depts NUMBER;

FUNCTION hire
   (ename VARCHAR2, 
    job VARCHAR2, 
    mgr NUMBER,
    sal NUMBER, 
    comm NUMBER, 
    deptno NUMBER)

RETURN NUMBER IS
   new_empno NUMBER(4);
BEGIN
   SELECT empseq.NEXTVAL
      INTO new_empno
      FROM DUAL;
   INSERT INTO emp
      VALUES (new_empno, ename, job, mgr, sal, comm, deptno, 
         tot_emps := tot_emps + 1;
   RETURN(new_empno);
END;

FUNCTION create_dept(dname VARCHAR2, loc VARCHAR2)
   RETURN NUMBER IS
      new_deptno NUMBER(4);
   BEGIN
      SELECT deptseq.NEXTVAL
         INTO new_deptno
         FROM dual;
      INSERT INTO dept
         VALUES (new_deptno, dname, loc);
            tot_depts := tot_depts + 1;
      RETURN(new_deptno);
   END;

PROCEDURE remove_emp(empno NUMBER) IS
   BEGIN
      DELETE FROM emp
      WHERE emp.empno = remove_emp.empno;
         tot_emps := tot_emps - 1;
   END; 

PROCEDURE remove_dept(deptno NUMBER) IS 
   BEGIN 
      DELETE FROM dept 
      WHERE dept.deptno = remove_dept.deptno; 
         tot_depts := tot_depts - 1; 
      SELECT COUNT(*) 
         INTO tot_emps 
         FROM emp; 
         /* In case Oracle deleted employees from the EMP table
         to enforce referential integrity constraints, reset
         the value of the variable TOT_EMPS to the total
         number of employees in the EMP table. */ 
   END; 

PROCEDURE increase_sal(empno NUMBER, sal_incr NUMBER) IS 
   curr_sal NUMBER(7,2); 
   BEGIN 
      SELECT sal 
      INTO curr_sal 
      FROM emp 
      WHERE emp.empno = increase_sal.empno; 
      IF curr_sal IS NULL 
         THEN RAISE no_sal; 
      ELSE 
         UPDATE emp 
         SET sal = sal + sal_incr 
         WHERE empno = empno; 
      END IF; 
   END; 

PROCEDURE increase_comm(empno NUMBER, comm_incr NUMBER) IS
      curr_comm NUMBER(7,2);
   BEGIN
      SELECT comm
      INTO curr_comm
      FROM emp
      WHERE emp.empno = increase_comm.empno
      IF curr_comm IS NULL
         THEN RAISE no_comm;
      ELSE 
         UPDATE emp
         SET comm = comm + comm_incr;
      END IF;
   END;

END emp_mgmt; 

This package body corresponds to the package specification in the example of the CREATE PACKAGE command earlier in this chapter. The package body defines the public program objects declared in the package specification:

These objects are declared in the package specification, so they can be called by application programs, procedures, and functions outside the package. For example, if you have access to the package, you can create a procedure INCREASE_ALL_COMMS separate from the EMP_MGMT package that calls the INCREASE_COMM procedure.

These objects are defined in the package body, so you can change their definitions without causing Oracle to invalidate dependent schema objects. For example, if you subsequently change the definition of HIRE, Oracle need not recompile INCREASE_ALL_COMMS before executing it.

The package body in this example also declares private program objects, the variables TOT_EMPS and TOT_DEPTS. These objects are declared in the package body rather than the package specification, so they are accessible to other objects in the package, but they are not accessible outside the package. For example, you cannot develop an application that explicitly changes the value of the variable TOT_DEPTS. However, the function CREATE_DEPT is part of the package, so CREATE_DEPT can change the value of TOT_DEPTS.

Related Topics

ALTER PACKAGE
CREATE FUNCTION
CREATE PROCEDURE
CREATE PACKAGE
DROP PACKAGE

CREATE PROCEDURE

Purpose

To create a standalone stored procedure or to register an external procedure. A procedure is a group of PL/SQL statements that you can call by name. An external procedure is a third-generation language (3GL) routine stored in a shared library which can be called from SQL or PL/SQL. To call an external procedure, you must provide information in your PL/SQL function about where to find the external procedure, how to call it, and what to pass to it. See also "Using Procedures".

For more information about registering external procedures, see the PL/SQL User's Guide and Reference.

Prerequisites

Before creating a procedure, the user SYS must run the SQL script DBMSSTDX.SQL. The exact name and location of this script depends on your operating system.

To create a procedure in your own schema, you must have CREATE PROCEDURE system privilege. To create a procedure in another schema, you must have CREATE ANY PROCEDURE system privilege. To replace a procedure in another schema, you must have ALTER ANY PROCEDURE system privilege.

To call an external procedure, you must have EXECUTE privileges on the callout library in which the procedure resides.

Syntax

Keywords and Parameters

OR REPLACE  

re-creates the procedure if it already exists. Use this option to change the definition of an existing procedure without dropping, re-creating, and regranting object privileges previously granted on it. If you redefine a procedure, Oracle recompiles it. For information on recompiling procedures, see ALTER PROCEDURE.  

 

Users who had previously been granted privileges on a redefined procedure can still access the procedure without being regranted the privileges.  

schema  

is the schema to contain the procedure. If you omit schema, Oracle creates the procedure in your current schema.  

procedure  

is the name of the procedure to be created.  

argument  

is the name of an argument to the procedure. If the procedure does not accept arguments, you can omit the parentheses following the procedure name.  

IN  

specifies that you must specify a value for the argument when calling the procedure.  

OUT  

specifies that the procedure passes a value for this argument back to its calling environment after execution.  

IN OUT  

specifies that you must specify a value for the argument when calling the procedure and that the procedure passes a value back to its calling environment after execution.  

 

If you omit IN, OUT, and IN OUT, the argument defaults to IN.  

datatype  

is the datatype of the argument. As long as no length specifier is used, an argument can have any datatype supported by PL/SQL. For information on PL/SQL datatypes, see PL/SQL User's Guide and Reference.  

 

Datatypes are specified without a length, precision, or scale. For example, VARCHAR2(10) is not valid, but VARCHAR2 is valid. Oracle derives the length, precision, and scale of an argument from the environment from which the procedure is called.  

IS pl/sql_subprogram_body  

is the definition of the procedure. Procedure definitions are written in PL/SQL. For information on PL/SQL, including how to write a PL/SQL subprogram body, see PL/SQL User's Guide and Reference.  

AS external_body  

identifies an external 3GL procedure stored in a sharable library. The AS external_body clause is the interface between PL/SQL and the external procedure.  

LIBRARY  

specifies the shared library in which the external procedure is stored. You must have EXECUTE privileges on the library. See CREATE LIBRARY for the syntax.  

library_name  

is a PL/SQL identifier. Enclosing library_name in double quotes makes it case sensitive, but quotes are not required.  

NAME external_procedure_name  

specifies the external procedure to be called. Enclosing external_procedure_name in double quotes makes it case sensitive, but quotes are not required. If you omit the name, it defaults to the PL/SQL subprogram (uppercase) name.  

LANGUAGE  

specifies the 3GL in which the external procedure was written. Currently, the only language name supported is C. If you omit the name, it defaults to C.  

CALLING STANDARD  

specifies the calling standard (C or PASCAL) under which the external procedure was compiled. If you omit the calling standard, it defaults to C.  

WITH CONTEXT  

specifies that a context pointer will be the first parameter passed to the external procedure. The context is opaque to the external procedure but is available to access functions called by the external procedure. For more information about the WITH CONTEXT clause, see the PL/SQL User's Guide and Reference.  

PARAMETERS  

specifies the positions and datatypes of parameters passed to the external procedure. It can also specify parameter properties such as current length and maximum length, and the preferred parameter passing method (by value or by reference). For more information about parameter passing, see the PL/SQL User's Guide and Reference.  

To embed a CREATE PROCEDURE statement inside an Oracle precompiler program, you must terminate the statement with the keyword END-EXEC followed by the embedded SQL statement terminator for the specific language.  

Using Procedures

A procedure is a group of PL/SQL statements that you can call by name. Stored procedures and stored functions are similar in many ways. This discussion applies to functions as well as to procedures. For information specific to functions, see CREATE FUNCTION.

With PL/SQL, you can group multiple SQL statements together with procedural PL/SQL statements similar to those in programming languages such as Ada and C. With the CREATE PROCEDURE command, you can create a procedure and store it in the database. You can call a stored procedure from any environment from which you can issue a SQL statement.

Stored procedures offer advantages in the areas of development, integrity, security, performance, and memory allocation. For more information on stored procedures, including how to call stored procedures, see Oracle8 Application Developer's Guide.

The CREATE PROCEDURE command creates a procedure as a standalone schema object. You can also create a procedure as part of a package. For information on creating packages, see CREATE FUNCTION.

Example I

The following statement creates the procedure CREDIT in the schema SAM:

CREATE PROCEDURE sam.credit (acc_no IN NUMBER, amount IN NUMBER) 
AS BEGIN 
UPDATE accounts 
SET balance = balance + amount 
WHERE account_id = acc_no; 
END; 

The CREDIT procedure credits a specified bank account with a specified amount. When you call the procedure, you must specify the following arguments:

ACC_NO  

is the number of the bank account to be credited. The argument's datatype is NUMBER.  

AMOUNT  

is the amount of the credit. The argument's datatype is NUMBER.  

The procedure uses an UPDATE statement to increase the value in the BALANCE column of the ACCOUNTS table by the value of the argument AMOUNT for the account identified by the argument ACC_NO.

Example II

In the following example, external procedure C_FIND_ROOT expects a pointer as a parameter. Procedure FIND_ROOT passes the parameter by reference using the BY REF phrase:

CREATE PROCEDURE
( x IN REAL ) AS
EXTERNAL
EXTERNAL
LIBRARY c_utils
NAME "c_find_root"
PARAMETERS
( x BY REF );

See the PL/SQL User's Guide and Reference for information about external procedures.

Related Topics

ALTER PROCEDURE
CREATE FUNCTION
CREATE LIBRARY
CREATE PACKAGE
DROP PROCEDURE
PL/SQL User's Guide and Reference


CREATE PROFILE

Purpose

To create a profile. A profile is a set of limits on database resources. If you assign the profile to a user, that user cannot exceed these limits.

Prerequisites

You must have CREATE PROFILE system privilege.

Syntax

Keywords and Parameters

profile  

is the name of the profile to be created. See also "Using Profiles".  

SESSIONS_PER_USER  

limits a user to integer concurrent sessions.  

CPU_PER_SESSION  

limits the CPU time for a session, expressed in hundredth of seconds  

CPU_PER_CALL  

limits the CPU time for a call (a parse, execute, or fetch), expressed in hundredths of seconds.  

CONNECT_TIME  

limits the total elapsed time of a session, expressed in minutes.  

IDLE_TIME  

limits periods of continuous inactive time during a session, expressed in minutes. Long-running queries and other operations are not subject to this limit.  

LOGICAL_READS_PER_SESSION  

specifies the number of data blocks read in a session, including blocks read from memory and disk.  

LOGICAL_READS_PER_CALL  

specifies the number of data blocks read for a call to process a SQL statement (a parse, execute, or fetch).  

PRIVATE_SGA  

specifies the amount of private space a session can allocate in the shared pool of the system global area (SGA), expressed in bytes. You can use K or M to specify this limit in kilobytes or megabytes. This limit applies only if you are using multithreaded server architecture. The private space for a session in the SGA includes private SQL and PL/SQL areas, but not shared SQL and PL/SQL areas.  

FAILED_LOGIN_ATTEMPTS  

specifies the number of failed attempts to log in to the user account before the account is locked.  

PASSWORD_LIFE_TIME  

limits the number of days the same password can be used for authentication. The password expires if it is not changed within this period, and further connections are rejected. See also "Fractions in Dates".  

PASSWORD_REUSE_TIME  

specifies the number of days before which a password cannot be reused. If you set PASSWORD_REUSE_TIME to an integer value, then you must set PASSWORD_REUSE_MAX to UNLIMITED.  

PASSWORD_REUSE_MAX  

specifies the number of password changes required before the current password can be reused. If you set PASSWORD_REUSE_MAX to an integer value, then you must set PASSWORD_REUSE_TIME to UNLIMITED.  

PASSWORD_LOCK_TIME  

specifies the number of days an account will be locked after the specified number of consecutive failed login attempts.  

PASSWORD_GRACE_TIME  

specifies the number of days after the grace period begins during which a warning is issued and login is allowed. If the password is not changed during the grace period, the password expires.  

PASSWORD_VERIFY_FUNCTION  

allows a PL/SQL password complexity verification script to be passed as an argument to the CREATE PROFILE command. Oracle provides a default script, but you can create your own routine or use third-party software instead.  

 

function  

is the name of the password complexity verification routine.  

 

NULL  

indicates that no password verification is performed.  

 

DEFAULT  

omits a limit for this resource in this profile. A user assigned this profile is subject to the limit on the resource specified in the default profile.  

COMPOSITE_LIMIT  

specifies the total resources cost for a session, expressed in service units. Oracle calculates the total service units as a weighted sum of CPU_PER_SESSION, CONNECT_TIME, LOGICAL_READS_PER_SESSION, and PRIVATE_SGA.

For information on how to specify the weight for each session resource, see ALTER RESOURCE COST.  

UNLIMITED  

indicates that a user assigned this profile can use an unlimited amount of this resource.  

DEFAULT  

omits a limit for this resource in this profile. A user assigned this profile is subject to the limit for this resource specified in the DEFAULT profile. See also "The DEFAULT Profile".  

Using Profiles

A profile is a set of limits on database resources. You can use profiles to limit the database resources available to a user for a single call or a single session. Oracle enforces resource limits in the following ways:

Fractions in Dates

You can use fractions of days for all parameters, with days as units. Fractions are expressed as x/y. For example, 1 hour is 1/24 and 1 minute is 1/1440.

For a detailed description and explanation of how to use password management and protection, see the Oracle8 Administrator's Guide .

To specify resource limits for a user, you must perform both of the following operations:

Enable resource limits

You can enable resource limits in one of two ways:

Specify resource limits

To specify a resource limit for a user, you must perform following steps:

  1. Create a profile that defines the limits using the CREATE PROFILE command.
  2. Assign the profile to the user using the CREATE USER or ALTER USER command.

Note that you can specify resource limits for users regardless of whether resource limits are enabled. However, Oracle does not enforce these limits until you enable them.

The DEFAULT Profile

Oracle automatically creates a default profile named DEFAULT. This profile initially defines unlimited resources. You can change the limits defined in this profile with the ALTER PROFILE command.

Any user who is not explicitly assigned a profile is subject to the limits defined in the DEFAULT profile. Also, if the profile that is explicitly assigned to a user omits limits for some resources or specifies DEFAULT for some limits, the user is subject to the limits on those resources defined by the DEFAULT profile.

Example I

The following statement creates the profile SYSTEM_MANAGER:

CREATE PROFILE system_manager 
LIMIT SESSIONS_PER_USER    UNLIMITED 
CPU_PER_SESSION            UNLIMITED 
CPU_PER_CALL               3000 
CONNECT_TIME               45 
LOGICAL_READS_PER_SESSION  DEFAULT 
LOGICAL_READS_PER_CALL     1000 
PRIVATE SGA                15K 
COMPOSITE_LIMIT            5000000; 

If you then assign the SYSTEM_MANAGER profile to a user, the user is subject to the following limits in subsequent sessions:

Example II

The following example creates the profile PROF:

CREATE PROFILE prof
  LIMIT PASSWORD_REUSE_MAX DEFAULT
        PASSWORD_REUSE_TIME UNLIMITED;
Example III

The following example creates profile MYPROFILE with password profile limits values set:

CREATE PROFILE myprofile LIMIT
FAILED_LOGIN_ATTEMPTS 5
PASSWORD_LIFE_TIME 60
PASSWORD_REUSE_TIME 60
PASSWORD_REUSE_MAX UNLIMITED
PASSWORD_VERIFY_FUNCTION verify_function
PASSWORD_LOCK_TIME 1/24
PASSWORD_GRACE_TIME 10;

Related Topics

ALTER PROFILE
ALTER RESOURCE COST
ALTER SYSTEM
ALTER USER
DROP PROFILE

CREATE ROLE

Purpose

To create a role. A role is a set of privileges that can be granted to users or to other roles. See also "Using Roles".

For a detailed description and explanation of using global roles, see Oracle8 Distributed Database Systems.

Prerequisites

You must have CREATE ROLE system privilege.

Syntax

Keywords and Parameters

role  

is the name of the role to be created. Oracle recommends that the role contain at least one single-byte character regardless of whether the database character set also contains multibyte characters. See also "Roles Predefined by Oracle".  

NOT IDENTIFIED  

indicates that this role is authorized by the database and that no password is required to enable the role.  

IDENTIFIED  

indicates that a user must be authorized by the specified method before the role is enabled with the SET ROLE command:  

 

BY password  

The user must specify the password to Oracle when enabling the role. The password can contain only single-byte characters from your database character set regardless of whether this character set also contains multibyte characters.  

 

EXTERNALLY  

indicates that a user must be authorized by an external service (such as an operating system or third-party service) before enabling the role.  

 

 

Depending on the operating system, the user may have to specify a password to the operating system before the role is enabled. For more information about third-party service, see Oracle Security Server Guide.  

 

GLOBALLY  

indicates that a user must be authorized to use the role by the Oracle Security Service before the role is enabled with the SET ROLE command, or at login.  

If you omit both the NOT IDENTIFIED option and the IDENTIFIED clause, the role defaults to NOT IDENTIFIED.  

Using Roles

A role is a set of privileges that can be granted to users or to other roles. You can use roles to administer database privileges. You can add privileges to a role and then grant the role to a user. The user can then enable the role and exercise the privileges granted by the role. For information on enabling roles, see ALTER USER.

A role contains all privileges granted to the role and all privileges of other roles granted to it. A new role is initially empty. You add privileges to a role with the GRANT command.

When you create a role, Oracle grants you the role with ADMIN OPTION. The ADMIN OPTION allows you to

Roles Predefined by Oracle

Some roles are defined by SQL scripts provided on your distribution media. The following roles are predefined:

The CONNECT, RESOURCE, and DBA roles are provided for compatibility with previous versions of Oracle. You should not rely on these roles; rather, Oracle recommends that you to design your own roles for database security. These roles may not be created automatically by future versions of Oracle.

The SELECT_CATALOG_ROLE, EXECUTE_CATALOG_ROLE, and DELETE_CATALOG_ROLE roles are provided for accessing exported data dictionary views and packages. For more information on these roles, see Oracle8 Application Developer's Guide.

The EXP_FULL_DATABASE and IMP_FULL_DATABASE roles are provided for convenience in using the Import and Export utilities.

Oracle also creates other roles that authorize you to administer the database. On many operating systems, these roles are called OSOPER and OSDBA. Their names may be different on your operating system.

Example I

The following example creates global role VENDOR:

CREATE ROLE vendor IDENTIFIED GLOBALLY;
Example II

The following statement creates the role TELLER:

CREATE ROLE teller 
IDENTIFIED BY cashflow; 

Users who are subsequently granted the TELLER role must specify the password CASHFLOW to enable the role with the SET ROLE command.

Related Topics

ALTER ROLE
DROP ROLE
GRANT (System Privileges and Roles)
REVOKE (System Privileges and Roles)
SET ROLE

CREATE ROLLBACK SEGMENT

Purpose

To create a rollback segment. A rollback segment is an object that Oracle uses to store data necessary to reverse, or undo, changes made by transactions.

Prerequisites

You must have CREATE ROLLBACK SEGMENT system privilege. Also, you must have either space quota on the tablespace to contain the rollback segment or UNLIMITED TABLESPACE system privilege.

Syntax

storage_clause: See the STORAGE clause.

Keyword and Parameters

PUBLIC  

specifies that the rollback segment is public and is available to any instance. If you omit this option, the rollback segment is private and is available only to the instance naming it in its initialization parameter ROLLBACK_SEGMENTS.  

rollback_segment  

is the name of the rollback segment to be created.  

TABLESPACE  

identifies the tablespace in which the rollback segment is created. If you omit this option, Oracle creates the rollback segment in the SYSTEM tablespace. See also "Rollback Segments and Tablespaces".  

storage_clause  

specifies the characteristics for the rollback segment. See the STORAGE clause.

Note: The PCTINCREASE option of the storage_clause is not permitted with CREATE ROLLBACK SEGMENT.  

OPTIMAL  

This part of the STORAGE clause specifies an optimal size in bytes for a rollback segment. You can use K or M to specify this size in kilobytes or megabytes. Oracle tries to maintain this size for the rollback segment by dynamically deallocating extents when their data is no longer needed for active transactions. Oracle deallocates as many extents as possible without reducing the total size of the rollback segment below the OPTIMAL value.  

 

NULL  

specifies no optimal size for the rollback segment, meaning that Oracle never deallocates the rollback segment's extents. This is the default behavior.  

 

The value of this parameter cannot be less than the space initially allocated for the rollback segment specified by the MINEXTENTS, INITIAL, and NEXT parameters. The maximum value depends on your operating system. Oracle rounds values to the next multiple of the data block size.  

Rollback Segments and Tablespaces

The tablespace must be online for you to add a rollback segment to it.

When you create a rollback segment, it is initially offline. To make it available for transactions by your Oracle instance, you must bring it online using one of the following:

For more information on creating rollback segments and making them available, see Oracle8 Administrator's Guide.

A tablespace can have multiple rollback segments. Generally, multiple rollback segments improve performance.

Example

The following statement creates a rollback segment with default storage values in the system tablespace:

CREATE ROLLBACK SEGMENT rbs_2
TABLESPACE system; 

The above statement is equivalent to the following:

CREATE ROLLBACK SEGMENT rbs_2
TABLESPACE system
STORAGE
( INITIAL 10 K
  NEXT 10 K
  MAXEXTENTS UNLMIITED); 

Related Topics

CREATE TABLESPACE
CREATE TABLESPACE
ALTER ROLLBACK SEGMENT
DROP ROLLBACK SEGMENT
STORAGE clause

CREATE SCHEMA

Purpose

To create multiple tables and views and perform multiple grants in a single transaction. See also "Creating Schemas".

Prerequisites

The CREATE SCHEMA statement can include, CREATE TABLE, CREATE VIEW, and GRANT statements. To issue a CREATE SCHEMA statement, you must have the privileges necessary to issue the included statements.

Syntax

Keyword and Parameters

schema  

is the name of the schema. The schema name must be the same as your Oracle username.  

CREATE TABLE command  

is a CREATE TABLE statement to be issued as part of this CREATE SCHEMA statement. See the CREATE TABLE  

CREATE VIEW command  

is a CREATE VIEW statement to be issued as part of this CREATE SCHEMA statement. See the CREATE VIEW.  

GRANT command  

is a GRANT statement (Object Privileges) to be issued as part of this CREATE SCHEMA statement. See GRANT (Object Privileges).  

 

The CREATE SCHEMA statement supports the syntax of these commands only as defined by standard SQL, rather than the complete syntax supported by Oracle.  

Creating Schemas

With the CREATE SCHEMA command, you can issue multiple data definition language (DDL) statements in a single transaction. To execute a CREATE SCHEMA statement, Oracle executes each included statement. If all statements execute successfully, Oracle commits the transaction. If any statement results in an error, Oracle rolls back all the statements.

Terminate a CREATE SCHEMA statement just as you would any other SQL statement using the terminator character specific to your tool. For example, if you issue a CREATE SCHEMA statement in SQL*Plus or Server Manager, terminate the statement with a semicolon (;). Do not separate the individual statements within a CREATE SCHEMA statement with the terminator character.

The order in which you list the CREATE TABLE, CREATE VIEW, and GRANT statements is unimportant:

The statements within a CREATE SCHEMA statement can also reference existing objects:

Example

The following statement creates a schema named BLAIR for the user BLAIR, creates the table SOX, creates the view RED_SOX, and grants SELECT privilege on the RED_SOX view to the user WAITES.

CREATE SCHEMA AUTHORIZATION blair 
CREATE TABLE sox 
(color VARCHAR2(10)  PRIMARY KEY, quantity NUMBER) 
CREATE VIEW red_sox 
AS SELECT color, quantity FROM sox WHERE color = 'RED' 
GRANT select ON red_sox TO waites; 

Related Topics

CREATE TABLE
CREATE VIEW
GRANT (Object Privileges)

CREATE SEQUENCE

Purpose

To create a sequence. A sequence is a database object from which multiple users may generate unique integers. You can use sequences to automatically generate primary key values. See also "Using Sequences", "Sequence Defaults", and "Accessing Sequence Values".

Prerequisites

To create a sequence in your own schema, you must have CREATE SEQUENCE privilege.

To create a sequence in another user's schema, you must have CREATE ANY SEQUENCE privilege.

Syntax

Keywords and Parameters

schema  

is the schema to contain the sequence. If you omit schema, Oracle creates the sequence in your own schema.  

sequence  

is the name of the sequence to be created.  

INCREMENT BY  

specifies the interval between sequence numbers. This integer value can be any positive or negative integer, but it cannot be 0. This value can have 28 or fewer digits. The absolute of this value must be less than the difference of MAXVALUE and MINVALUE. If this value is negative, then the sequence descends. If the increment is positive, then the sequence ascends. If you omit this clause, the interval defaults to 1. See also "Incrementing Sequence Values".  

START WITH  

specifies the first sequence number to be generated. Use this option to start an ascending sequence at a value greater than its minimum or to start a descending sequence at a value less than its maximum. For ascending sequences, the default value is the sequence's minimum value. For descending sequences, the default value is the sequence's maximum value. This integer value can have 28 or fewer digits.  

MAXVALUE  

specifies the maximum value the sequence can generate. This integer value can have 28 or fewer digits. MAXVALUE must be equal to or less than START WITH and must be greater than MINVALUE.  

NOMAXVALUE  

specifies a maximum value of 10^27 for an ascending sequence or -1 for a descending sequence. This is the default.  

MINVALUE  

specifies the sequence's minimum value. This integer value can have 28 or fewer digits. MINVALUE must be less than or equal to START WITH and must be less than MAXVALUE.  

NOMINVALUE  

specifies a minimum value of 1 for an ascending sequence or -(10^26) for a descending sequence. This is the default.  

CYCLE  

specifies that the sequence continues to generate values after reaching either its maximum or minimum value. After an ascending sequence reaches its maximum value, it generates its minimum value. After a descending sequence reaches its minimum, it generates its maximum.  

NOCYCLE  

specifies that the sequence cannot generate more values after reaching its maximum or minimum value. This is the default.  

CACHE  

specifies how many values of the sequence Oracle preallocates and keeps in memory for faster access. This integer value can have 28 or fewer digits. The minimum value for this parameter is 2. For sequences that cycle, this value must be less than the number of values in the cycle. You cannot cache more values than will fit in a given cycle of sequence numbers; thus, the maximum value allowed for CACHE must be less than the value determined by the following formula:  

 

(CEIL (MAXVALUE - MINVALUE)) / ABS (INCREMENT)

See also "Caching Sequence Numbers".  

NOCACHE  

specifies that values of the sequence are not preallocated.  

If you omit both the CACHE parameter and the NOCACHE option, Oracle caches 20 sequence numbers by default.  

ORDER  

guarantees that sequence numbers are generated in order of request. You may want to use this option if you are using the sequence numbers as timestamps. Guaranteeing order is usually not important for sequences used to generate primary keys.  

NOORDER  

does not guarantee sequence numbers are generated in order of request.  

If you omit both the ORDER and NOORDER options, Oracle chooses NOORDER by default. Note that the ORDER option is necessary only to guarantee ordered generation if you are using Oracle with the Parallel Server option in parallel mode. If you are using exclusive mode, sequence numbers are always generated in order.  

Using Sequences

You can use sequence numbers to automatically generate unique primary key values for your data, and you can also coordinate the keys across multiple rows or tables.

Values for a given sequence are automatically generated by special Oracle routines and, consequently, sequences avoid the performance bottleneck that results from implementation of sequences at the application level. For example, one common application-level implementation is to force each transaction to lock a sequence number table, increment the sequence, and then release the table. Under this implementation, only one sequence number can be generated at a time. In contrast, Oracle sequences permit the simultaneous generation of multiple sequence numbers while guaranteeing that every sequence number is unique.

When a sequence number is generated, the sequence is incremented, independent of the transaction committing or rolling back. If two users concurrently increment the same sequence, the sequence numbers each user acquires may have gaps because sequence numbers are being generated by the other user. One user can never acquire the sequence number generated by another user. Once a sequence value is generated by one user, that user can continue to access that value regardless of whether the sequence is incremented by another user.

Sequence numbers are generated independently of tables, so the same sequence can be used for one or for multiple tables. It is possible that individual sequence numbers will appear to be skipped, because they were generated and used in a transaction that ultimately rolled back. Additionally, a single user may not realize that other users are drawing from the same sequence.

Sequence Defaults

The sequence defaults are designed so that if you specify none of the clauses, you create an ascending sequence that starts with 1 and increases by 1 with no upper limit. Specifying only INCREMENT BY -1 creates a descending sequence that starts with -1 and decreases with no lower limit.

Incrementing Sequence Values

You can create a sequence so that its values increment in one of following ways:

To create a sequence that increments without bound, omit the MAXVALUE parameter or specify the NOMAXVALUE option for ascending sequences or omit the MINVALUE parameter or specify the NOMINVALUE for descending sequences.

To create a sequence that stops at a predefined limit, specify a value for the MAXVALUE parameter for an ascending sequence or a value for the MINVALUE parameter for a descending sequence. Also specify the NOCYCLE option. Any attempt to generate a sequence number once the sequence has reached its limit results in an error.

To create a sequence that restarts after reaching a predefined limit, specify values for both the MAXVALUE and MINVALUE parameters. Also specify the CYCLE option. If you do not specify MINVALUE, then it defaults to NOMINVALUE; that is, the value 1.

The value of the START WITH parameter establishes the initial value generated after the sequence is created. Note that this value is not necessarily the value to which an ascending cycling sequence cycles after reaching its maximum or minimum value.

Caching Sequence Numbers

The number of values cached in memory for a sequence is specified by the value of the sequence's CACHE parameter. Cached sequences allow faster generation of sequence numbers. A cache for a given sequence is populated at the first request for a number from that sequence. The cache is repopulated every CACHE requests. If there is a system failure, all cached sequence values that have not been used in committed data manipulation language (DML) statements are lost. The potential number of lost values is equal to the value of the CACHE parameter.

A CACHE of 20 future sequence numbers is the default.

Accessing Sequence Values

Once a sequence is created, you can access its values in SQL statements with the following pseudocolumns:

CURRVAL  

returns the current value of the sequence.  

NEXTVAL  

increments the sequence and returns the new value.  

For more information on using the above pseudocolumns, see the section "Pseudocolumns".

Example

The following statement creates the sequence ESEQ:

CREATE SEQUENCE eseq 
  INCREMENT BY 10  

The first reference to ESEQ.NEXTVAL returns 1. The second returns 11. Each subsequent reference will return a value 10 greater than the one previous.

Related Topics

ALTER SEQUENCE
DROP SEQUENCE

CREATE SNAPSHOT

Purpose

To create a snapshot. A snapshot is a table that contains the results of a query of one or more tables, often located on a remote database.

Prerequisites

The following prerequisites apply to creating snapshots:

By default, Oracle creates all new snapshots as primary key snapshots. To create a snapshot:

When you create a snapshot, Oracle creates one table, one view, and at least one index in the schema of the snapshot. Oracle uses these objects to maintain the snapshot's data. You must have the privileges necessary to create these objects. For information on these privileges, see CREATE TABLE, CREATE VIEW, and CREATE INDEX.

For complete information about the prerequisites that apply to creating snapshots, see Oracle8 Replication.

Syntax

physical_attributes_clause: See ALTER TABLE.

parallel_clause: See the PARALLEL clause.

index_physical_attributes_clause: See ALTER INDEX.

select_command: See SELECT.

LOB_storage_clause: See CREATE TABLE.

table_partition_clause: See CREATE TABLE.

Keywords and Parameters

schema  

is the schema to contain the snapshot. If you omit schema, Oracle creates the snapshot in your schema.  

snapshot  

is the name of the snapshot to be created. Oracle generates names for the table, view, and indexes used to maintain the snapshot by adding a prefix or suffix to the snapshot name. Oracle recommends that you limit your snapshot names to 19 bytes, so that the Oracle-generated names will be 30 bytes or less and will contain the entire snapshot name. See also "About Snapshots".  

 

physical_attributes_clause  

establishes values for the PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters (or, when used in the USING INDEX clause, for the INITRANS and MAXTRANS parameters only) and the storage parameters for the internal table Oracle uses to maintain the snapshot's data.

For information on the PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters, see CREATE TABLE. For information, about the STORAGE clause, see the STORAGE clause.  

TABLESPACE  

specifies the tablespace in which the snapshot is to be created. If you omit this option, Oracle creates the snapshot in the default tablespace of the owner of the snapshot's schema.  

LOB_storage_clause  

specifies the LOB storage characteristics. For detailed information about specifying the parameters of the LOB storage clause, see CREATE TABLE.  

STORAGE  

establishes storage characteristics for the table Oracle uses to maintain the snapshot's data.  

CLUSTER  

creates the snapshot as part of the specified cluster. Since a clustered snapshot uses the cluster's space allocation, do not use the physical_attributes_clause or the TABLESPACE option with the CLUSTER option.  

table_partition_clause  

specifies that the table is partitioned on specified ranges of values. For detailed information about specifying the parameters of the table partition clause, see CREATE TABLE. See also "Partitioned Snapshots".  

USING INDEX  

specifies parameters for the index Oracle creates to maintain the snapshot. See physical_attributes_clause, above.  

USING ROLLBACK SEGMENT  

specifies the local snapshot and/or remote master rollback segments to be used during snapshot refresh.  

 

rollback_segment  

is the name of the rollback segment to be used.  

 

DEFAULT  

specifies that Oracle will choose which rollback segment to use.  

 

MASTER  

specifies the rollback segment to be used at the remote master for the individual snapshot.  

 

LOCAL  

specifies the rollback segment to be used for the local refresh group that contains the snapshot.  

 

If you do not specify MASTER or LOCAL, Oracle uses LOCAL by default. If you do not specify rollback_segment, Oracle automatically chooses the rollback segment to be used. If you specify DEFAULT, you cannot specify rollback_segment. See also "Specifying Rollback Segments".  

REFRESH  

specifies how and when Oracle automatically refreshes the snapshot:  

 

FAST  

specifies a fast refresh, or a refresh using only the updated data stored in the snapshot log associated with the master table.  

 

COMPLETE  

specifies a complete refresh, or a refresh that reexecutes the snapshot's query.  

 

FORCE  

specifies a fast refresh if one is possible or complete refresh if a fast refresh is not possible. Oracle decides whether a fast refresh is possible at refresh time.  

 

If you omit the FAST, COMPLETE, and FORCE options, Oracle uses FORCE by default. See also "Refreshing Snapshots".  

 

START WITH  

specifies a date expression for the first automatic refresh time.  

 

NEXT  

specifies a date expression for calculating the interval between automatic refreshes.  

 

Both the START WITH and NEXT values must evaluate to a time in the future. If you omit the START WITH value, Oracle determines the first automatic refresh time by evaluating the NEXT expression when you create the snapshot. If you specify a START WITH value but omit the NEXT value, Oracle refreshes the snapshot only once. If you omit both the START WITH and NEXT values, or if you omit the REFRESH clause entirely, Oracle does not automatically refresh the snapshot.  

 

WITH PRIMARY KEY  

specifies that primary-key snapshots are to be created. Primary-key snapshots allow snapshot master tables to be reorganized without affecting the snapshot's ability to continue to fast refresh.  

 

 

Primary-key snapshots can also be defined as simple snapshots with subqueries.  

 

WITH ROWID  

specifies that ROWID snapshots are to be created.  

 

 

ROWID snapshots provide compatibility with Oracle7 Release 7.3 master tables.  

 

If you omit both WITH PRIMARY KEY and WITH ROWID, Oracle creates primary-key snapshots by default. See also "Specifying Primary-Key or ROWID Snapshots".  

FOR UPDATE  

allows a simple snapshot to be updated. When used in conjunction with the Replication Option, these updates will be propagated to the master. For more information, see Oracle8 Replication.  

AS select_command  

specifies the snapshot query. When you create the snapshot, Oracle executes this query and places the results in the snapshot. This query is any valid SQL query, but not all queries are fast refreshable. See also "Types of Snapshots".  

About Snapshots

A snapshot is a table that contains the results of a query of one or more tables, often located on a remote database. The tables in the query are called master tables. The databases containing the master tables are called the master databases. Note that a snapshot query cannot select from tables or views owned by the user SYS.

Snapshots are useful in distributed databases. Snapshots allow you to maintain read-only copies of remote data on your local node. You can select data from a snapshot as you would from a table or view.

Oracle recommends that you qualify each table and view in the FROM clause of the snapshot query with the schema containing it. For some additional caveats, see "The View Query" (in the context of the CREATE VIEW command). The same recommendations apply to creating snapshots.

Snapshots cannot contain long columns.

For more information on snapshots, see Oracle8 Replication.

Types of Snapshots

You can create two types of snapshots: simple and complex.

A simple snapshot is based on a single remote table, or is defined on multiple tables using restricted types of subqueries. For more information about simple snapshots with subqueries, see Oracle8 Replication.

Simple snapshots do not contain any of the following items in the snapshot query (select_command_clause):

A complex snapshot is one in which the snapshot query contains one or more of the constructs not allowed in the query of a simple snapshot. A complex snapshot can be based on multiple master tables on multiple master databases.

Refreshing Snapshots

A snapshot's master tables can be modified, so the data in a snapshot must be updated occasionally to ensure that the snapshot accurately reflects the data currently in its master tables. The process of updating a snapshot for this purpose is called refreshing the snapshot. With the REFRESH clause of the CREATE SNAPSHOT command, you can schedule the times and specify the mode for Oracle to refresh the snapshot automatically.

After you create a snapshot, you can subsequently change its automatic refresh mode and time with the REFRESH clause of the ALTER SNAPSHOT command. You can also refresh a snapshot immediately with the DBMS_SNAPSHOT.REFRESH() procedure.

Specifying Refresh Modes

Use the FAST or COMPLETE options of the REFRESH clause to specify the refresh mode.

Fast

To perform a fast refresh, Oracle updates the snapshot with the changes to the master table recorded in its snapshot log. For more information on snapshot logs, see CREATE SNAPSHOT LOG.

Oracle can only perform a fast refresh if all of the following conditions are true:

If you specify a fast refresh and all of above conditions are true, then Oracle performs a fast refresh. If any of the conditions are not true, Oracle returns an error at refresh time and does not refresh the snapshot.

Complete

To perform a complete refresh, Oracle reexecutes the snapshot query and places the results in the snapshot. If you specify a complete refresh, Oracle performs a complete refresh regardless of whether a fast refresh is possible.

A fast refresh is often faster than a complete refresh because it sends less data from the master database across the network to the snapshot's database. A fast refresh sends only changes to master table data since the last refresh, while a complete refresh sends the complete result of the snapshot query.

You can also use the FORCE option of the REFRESH clause to allow Oracle to decide how to refresh the snapshot at the scheduled refresh time. If a fast refresh is possible based on the fast refresh conditions, then Oracle performs a fast refresh. If a fast refresh is not possible, then Oracle performs a complete refresh.

Example

The following statement creates the simple snapshot EMP_SF that contains the data from SCOTT's employee table in New York:

CREATE SNAPSHOT emp_sf 
PCTFREE 5 PCTUSED 60 
TABLESPACE users 
STORAGE INITIAL 50K NEXT 50K
REFRESH FAST NEXT sysdate + 7 
AS 
SELECT * FROM scott.emp@ny; 

The statement does not include a START WITH parameter, so Oracle determines the first automatic refresh time by evaluating the NEXT value using the current SYSDATE. Provided a snapshot log currently exists for the employee table in New York, Oracle performs a fast refresh of the snapshot every 7 days, beginning 7 days after the snapshot is created.

The above statement also establishes for the table storage characteristics that Oracle uses to maintain the snapshot.

Specifying Automatic Refresh Times

To cause Oracle to refresh a snapshot automatically, you must perform the following tasks:

  1. Specify the START WITH and NEXT parameters in the REFRESH clause of the CREATE SNAPSHOT statement. These parameters establish the time of the first automatic refresh and the interval between automatic refreshes.
  2. Enable one or more job queue processes using the initialization parameters SNAPSHOT_REFRESH_PROCESSES, SNAPSHOT_REFRESH_INTERVAL, SNAPSHOT_REFRESH_KEEP_CONNECTIONS. The job queue processes then examine the automatic refresh time of each snapshot in the database. For each snapshot that is scheduled to be refreshed at or before the current time, one job queue process performs the following operations:
    • reevaluates the snapshot's NEXT value to determine the next automatic refresh time
    • refreshes the snapshot
    • stores the next automatic refresh time in the data dictionary

For more information on these initialization parameters, see Oracle8 Reference.

Example

The following statement creates the complex snapshot ALL_EMPS that queries the employee tables in Dallas and Baltimore:

CREATE SNAPSHOT all_emps 
   PCTFREE 5 PCTUSED 60 
   TABLESPACE users 
   STORAGE INITIAL 50K NEXT 50K 
   USING INDEX STORAGE (INITIAL 25K NEXT 25K)
   REFRESH START WITH ROUND(SYSDATE + 1) + 11/24 
   NEXT NEXT_DAY(TRUNC(SYSDATE, 'MONDAY') + 15/24 
AS 
   SELECT * FROM fran.emp@dallas 
      UNION
   SELECT * FROM marco.emp@balt; 

Oracle automatically refreshes this snapshot tomorrow at 11:00 am and subsequently every Monday at 3:00 pm. This command does not specify either fast or complete refreshes, so Oracle must decide how to refresh the snapshot. Since ALL_EMPS is a complex snapshot, Oracle must perform a complete refresh.

The above statement also establishes storage characteristics for both the table and the index that Oracle uses to maintain the snapshot:

Specifying Rollback Segments

You can specify the rollback segments to be used during a refresh for both the master site and the local site.

The local snapshot rollback segment is stored at the refresh group level. If the auto-refresh parameters are specified, a new refresh group is automatically created to refresh the snapshot with a background process. The local rollback segment, if specified, is associated with this new refresh group. An error is raised if you specify a local rollback segment but do not specify the auto-refresh parameters.

The master rollback segment is stored on a per-snapshot basis. The master rollback segment is validated during snapshot creation and refresh. If the snapshot is complex, the master rollback segment, if specified, is ignored.


Note:

Specifying DEFAULT is most useful with ALTER SNAPSHOT. See ALTER SNAPSHOT.

 

To direct Oracle to select the rollback segment automatically after one has been specified using CREATE SNAPSHOT or ALTER SNAPSHOT, specify DEFAULT with ALTER SNAPSHOT.

Example

The following example creates snapshot SALE_EMP with rollback segment MASTER_SEG at the remote master and rollback segment SNAP_SEG for the local refresh group that contains the snapshot:

CREATE SNAPSHOT sales_emp 
REFRESH FAST START WITH SYSDATE NEXT SYSDATE + 7
USING MASTER ROLLBACK SEGMENT master_seg 
LOCAL ROLLBACK SEGMENT snap_seg  
AS SELECT * FROM bar;  

The following statement is incorrect and generates an error because it specifies a segment name with a DEFAULT rollback segment:

CREATE SNAPSHOT bogus 
REFRESH FAST START WITH SYSDATE NEXT SYSDATE + 7
USING DEFAULT ROLLBACK SEGMENT snap_seg 
AS SELECT * FROM faux; 

Specifying Primary-Key or ROWID Snapshots

To create a primary-key snapshot you must:

To fast refresh primary-key snapshots, you must first create a snapshot master log specifying WITH PRIMARY KEY. The snapshot master log can also record ROWIDs.

Primary-key snapshots are the default if the WITH clause is not specified.

The above conditions must be met in order to create a primary-key snapshot.

Example I

The following example creates primary-key snapshot HUMAN_GENOME:

CREATE SNAPSHOT human_genome   
   REFRESH FAST START WITH SYSDATE NEXT  SYSDATE + 1/4096 
   WITH PRIMARY KEY 
AS SELECT * FROM genome_catalog; 
Example II

The following example creates a ROWID snapshot:

CREATE SNAPSHOT emp_data WITH ROWID 
AS SELECT * FROM emp_table73; 

Partitioned Snapshots

Partitioned snapshots are the same as partitioned tables, because snapshots are basically tables. The options have the same syntax and semantics as the partitioned table options for CREATE TABLE and ALTER TABLE. The only difference is that the following operations are not allowed on snapshots and snapshot logs:

You cannot perform bulk deletions by dropping or truncating partitions on master tables. Thus, after dropping or truncating a partition, all snapshots must be manually refreshed. A fast refresh will probably produce incorrect results, but Oracle will not raise an error.

Related Topics

ALTER SNAPSHOT
CREATE SNAPSHOT LOG
DROP SNAPSHOT
SELECT

CREATE SNAPSHOT LOG

Purpose

To create a snapshot log. A snapshot log is a table associated with the master table of a snapshot. Oracle stores changes to the master table's data in the snapshot log and then uses the snapshot log to refresh the master table's snapshots. See also "Using Snapshot Logs".

Prerequisites

The privileges required to create a snapshot log directly relate to the privileges necessary to create the underlying objects associated with a snapshot log.

In either case, the owner of the snapshot log must have sufficient quota in the tablespace intended to hold the snapshot log.

For detailed information about the prerequisites for creating a snapshot log, see Oracle8 Replication.

Syntax

parallel_clause: See PARALLEL clause.

storage_clause: See STORAGE clause.

LOB_storage_clause: See CREATE TABLE.

table_partition_clause: See CREATE TABLE.

physical_attributes_clause: See CREATE TABLE.

Keywords and Parameters

schema  

is the schema containing the snapshot log's master table. If you omit schema, Oracle assumes the master table is contained in your own schema. Oracle creates the snapshot log in the schema of its master table. You cannot create a snapshot log for a table in the schema of the user SYS.  

table  

is the name of the master table for which the snapshot log is to be created. You cannot create a snapshot log for a view.  

WITH  

specifies whether the snapshot log should record the primary key, ROWID, or both primary key and ROWID when rows in the master are updated.  

 

This clause also specifies whether the snapshot log records filter columns-non-primary-key columns referenced by snapshots defined as simple snapshots with subqueries. See also "Recording Primary Keys, ROWIDs, and Filter Columns".  

 

PRIMARY KEY  

specifies that the primary key of all rows updated should be recorded in the snapshot log.  

 

ROWID  

specifies that the ROWID of all rows updated should be recorded in the snapshot log.  

 

filter_column  

is a comma-separated list that specifies the list of filter columns to be recorded in the snapshot log.  

 

Oracle records the primary key of all rows updated in the master by default.  

physical_attributes_clause  

establishes values for the PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters and the storage characteristics for the snapshot log. See the descriptions of these parameters in CREATE TABLE and STORAGE clause.  

TABLESPACE  

specifies the tablespace in which the snapshot log is to be created. If you omit this option, Oracle creates the snapshot log in the default tablespace the owner of the snapshot log's schema.  

STORAGE  

establishes storage characteristics for the snapshot log. See the STORAGE clause.  

LOB_storage_clause  

specifies the LOB storage characteristics. For detailed information about specifying the parameters of the LOB storage clause, see STORAGE clause.  

table_partition_clause  

specifies that the table is partitioned on specified ranges of values. For detailed information about specifying the parameters of the table partition clause, see CREATE TABLE.  

Using Snapshot Logs

A snapshot log is a table associated with the master table of a snapshot. When changes are made to the master table's data, Oracle adds rows describing these changes to the snapshot log. Later Oracle can use these rows to refresh snapshots based on the master table. This process is called a fast refresh. Without a snapshot log, Oracle must reexecute the snapshot query to refresh the snapshot. This process is called a complete refresh. Usually, a fast refresh takes less time than a complete refresh.

A snapshot log is located in the master database in the same schema as the master table. You need only a single snapshot log for a master table. Oracle can use this snapshot log to perform fast refreshes for all simple snapshots based on the master table. For more information on snapshots, including how Oracle refreshes snapshots, see CREATE SNAPSHOT and Oracle8 Replication.

Example

The following statement creates a snapshot log on an employee table that records only primary-key values:

CREATE SNAPSHOT LOG ON emp 
PCTFREE 5 
TABLESPACE users 
STORAGE (INITIAL 10K NEXT 10K PCTINCREASE 50); 

Oracle can use this snapshot log to perform a fast refresh on any simple primary key snapshot subsequently created on the EMP table.

Recording Primary Keys, ROWIDs, and Filter Columns

For Oracle to perform primary-key snapshots, the primary key of updated rows in the master table must be recorded in the snapshot log. Similarly, for ROWID snapshots, the ROWID must be recorded in the snapshot log. Both primary keys and ROWIDs can be recorded to support configurations with both primary-key and ROWID snapshots.

For primary-key snapshots defined as simple snapshots with subqueries, all filter columns referenced by the defining subquery must be recorded in the snapshot log.

Example I

The following examples create snapshot logs that record only the primary keys of updated rows:

CREATE SNAPSHOT LOG ON emp; 
CREATE SNAPSHOT LOG ON emp WITH PRIMARY KEY; 
Example II

The following example creates a snapshot log that records both primary keys and ROWIDs of updated rows:

CREATE SNAPSHOT LOG ON sales WITH ROWID, PRIMARY KEY; 
Example III

The following example creates a snapshot log that records primary keys and updates to the filter column ZIP:

CREATE SNAPSHOT LOG ON address WITH (zip);

Related Topics

ALTER SNAPSHOT LOG
CREATE SNAPSHOT
CREATE TABLE
DROP SNAPSHOT

CREATE SYNONYM

Purpose

To create a synonym. A synonym is an alternative name for a table, view, sequence, procedure, stored function, package, snapshot, or another synonym. See also "Using Synonyms".

Prerequisites

To create a private synonym in your own schema, you must have CREATE SYNONYM system privilege.

To create a private synonym in another user's schema, you must have CREATE ANY SYNONYM system privilege.

To create a PUBLIC synonym, you must have CREATE PUBLIC SYNONYM system privilege.

Syntax

Keywords and Parameters

PUBLIC  

creates a public synonym. Public synonyms are accessible to all users. If you omit this option, the synonym is private and is accessible only within its schema.  

schema  

is the schema to contain the synonym. If you omit schema, Oracle creates the synonym in your own schema. You cannot specify schema if you have specified PUBLIC. See also "Scope of Synonyms".  

synonym  

is the name of the synonym to be created.  

FOR  

identifies the object for which the synonym is created. If you do not qualify object with schema, Oracle assumes that the schema object is in your own schema. The schema object can be of the following types:

  • table
  • object table
  • view
  • object view
  • sequence
  • stored procedure, function, or package
  • snapshot
  • synonym

You can create a synonym for an object table or an object view, but not for object types.  

 

The schema object cannot be contained in a package.  

 

Note that the schema object need not currently exist and you need not have privileges to access the object.  

dblink  

You can use a complete or partial dblink to create a synonym for a schema object on a remote database where the object is located. For more information on referring to database links, see "Referring to Objects in Remote Databases". If you specify dblink and omit schema, the synonym refers to an object in the schema specified by the database link. Oracle recommends that you specify the schema containing the object in the remote database.  

 

If you omit dblink, Oracle assumes the object is located on the local database.  

Using Synonyms

You can use a synonym to stand for its base object in any of the following statements:

DML Statements   DDL Statements  

SELECT  

AUDIT  

INSERT  

NOAUDIT  

UPDATE  

GRANT  

DELETE  

REVOKE  

EXPLAIN PLAN  

COMMENT  

LOCK TABLE  

 

Synonyms are used for security and convenience. Creating a synonym for an object allows you to:

Synonyms provide both data independence and location transparency; synonyms permit applications to function without modification regardless of which user owns the table or view and regardless of which database holds the table or view.

Scope of Synonyms

A private synonym name must be unique in its schema. Oracle attempts to resolve references to objects at the schema level before resolving them at the PUBLIC synonym level. Oracle uses a public synonym only when resolving references to an object if both of the following cases are true:

For example, assume the schemas SCOTT and BLAKE each contain tables named DEPT and the user SYSTEM creates a PUBLIC synonym named DEPT for BLAKE.DEPT. If the user SCOTT then issues the following statement, Oracle returns rows from SCOTT.DEPT:

SELECT * 
   FROM dept;

To retrieve rows from BLAKE.DEPT, the user SCOTT must preface DEPT with the schema name:

SELECT *
   FROM blake.dept;

If the user ADAM's schema does not contain an object named DEPT, then ADAM can access the DEPT table in BLAKE's schema by using the public synonym DEPT:

SELECT * 
   FROM dept;
Example I

To define the synonym MARKET for the table MARKET_RESEARCH in the schema SCOTT, issue the following statement:

CREATE SYNONYM market 
   FOR scott.market_research;
Example II

To create a PUBLIC synonym for the EMP table in the schema SCOTT on the remote SALES database, you could issue the following statement:

CREATE PUBLIC SYNONYM emp 
   FOR scott.emp@sales;

Note that a synonym may have the same name as the base table, provided the base table is contained in another schema.

Related Topics

CREATE DATABASE LINK
CREATE TABLE
CREATE VIEW

CREATE TABLE

Purpose

To create a table, the basic structure to hold user data, specifying the following information:

For illustrations of some of these purposes, "Examples".


Note:

Descriptions of commands and clauses preceded by are available only if the Oracle objects option is installed on your database server.

 

Use CREATE TABLE to create an object table or a table that uses an object type for a column definition. An object table is a table explicitly defined to hold object instances of a particular type.

You can also create an object type and then use it in a column when creating a relational table. For more information about creating objects, see Oracle8 Application Developer's Guide and CREATE TYPE.

Prerequisites

To create a relational table in your own schema, you must have CREATE TABLE system privilege. To create a table in another user's schema, you must have CREATE ANY TABLE system privilege. Also, the owner of the schema to contain the table must have either space quota on the tablespace to contain the table or UNLIMITED TABLESPACE system privilege.

In addition to the table privileges above, to create a table that uses types, the owner of the table must have the EXECUTE object privilege in order to access all types referenced by the table, or you must have the EXECUTE ANY TYPE system privilege. These privileges must be granted explicitly and not acquired through a role.

Additionally, if the table owner intends to grant access to the table to other users, the owner must have been granted the EXECUTE privileges to the referenced types with the GRANT OPTION, or have the EXECUTE ANY TYPE system privilege with the ADMIN OPTION. If not, the table owner has insufficient privileges to grant access on the table to other users.

For more information about the privileges required to create tables using types, see Oracle8 Application Developer's Guide.

Syntax

Relational table definition ::=

,

Object table definition ::=

column_ref_clause::=

table_ref_clause::=

segment_attributes_clause::=

physical_attributes_clause::=

storage_clause: See the STORAGE clause.

disable_clause: See the DISABLE clause.

enable_clause: See the ENABLE clause.

parallel_clause: See the PARALLEL clause

storage_clause: see STORAGE clause

subquery: See "Subqueries"

Keywords and Parameters

schema  

is the schema to contain the table. If you omit schema, Oracle creates the table in your own schema.  

table  

is the name of the table (or object table) to be created. A partitioned table cannot be a clustered table or an object table.  

OF object_type  

explicitly creates an object table of type object_type. The columns of an object table correspond to the top-level attributes of type object_type. Each row will contain an object instance, and each instance will be assigned a unique, system-generated object identifier (OID) when a row is inserted. If you omit schema, Oracle creates the object table in your own schema. For more information about creating objects, see CREATE TYPE. See also "Object Tables".  

column  

specifies the name of a column of the table. A table can have up to 1000 columns. You may omit column definitions only when using the AS subquery clause. See also "LOB Column Example".  

attribute  

specifies the qualified column name of an item in an object.  

datatype  

is the datatype of a column. Oracle-supplied datatypes are defined in "Datatypes". You can omit the datatype only if the statement also designates the column as part of a foreign key in a referential integrity constraint. Oracle automatically assigns to the column the datatype of the corresponding column of the referenced key of the referential integrity constraint.

Object types, REF object_type, VARRAYs, and nested tables are valid datatypes. See also "REFs".  

DEFAULT  

specifies a value to be assigned to the column if a subsequent INSERT statement omits a value for the column. The datatype of the expression must match the datatype of the column. The column must also be long enough to hold this expression. For the syntax of expr, see "Expressions". A DEFAULT expression cannot contain references to other columns, the pseudocolumns CURRVAL, NEXTVAL, LEVEL, and ROWNUM, or date constants that are not fully specified.  

column_ref_clause  

lets you further specify a column of type REF:  

 

WITH ROWID

 

stores the ROWID and the REF value in column or attribute. Storing a REF value with a ROWID can improve the performance of dereferencing operations, but will also use more space. Default storage of REF values is without ROWIDs.  

 

SCOPE IS scope_table_name  

restricts the scope of the column REF values to scope_table_name. The REF values for the column must come from REF values obtained from the object table specified in the clause. You can only specify one scope table per REF column.  

 

 

The scope_table_name is the name of the object table in which object instances (of the same type as the REF column) are stored. The values in the REF column point to objects in the scope table.

You must have SELECT privileges on the table or SELECT ANY TABLE system privileges.  

column_constraint  

defines an integrity constraint as part of the column definition. See the syntax description of column_constraint in the CONSTRAINT clause.  

table_constraint  

defines an integrity constraint as part of the table definition. See the syntax description of table_constraint in the CONSTRAINT clause.  

table_ref_clause  

 

SCOPE FOR... IS...  

restricts the scope of the REF values in ref_column or ref_attribute to scope_table_name. The REF values for the column or attribute must come from REF values obtained from the object table specified in the clause.  

 

 

The ref_column or ref_attribute is the name of a REF column in an object table or an embedded REF attribute within an object column of a relational table. The values in the REF column or attribute point to objects in the scope table.  

 

REF  

is a reference to a row in an object table. You can specify either a REF column name of an object or relational table (ref_column) or an embedded REF attribute within an object column (ref_attribute).  

OIDINDEX  

specifies an index on the hidden object identifier column and/or the storage specification for the index. Either index or storage_specification must be specified.  

 

index  

is the name of the index on the hidden object identifier column. If not specified, Oracle generates a name.  

physical_attributes_clause  

specifies the value of the PCTFREE, PCTUSED, INITRANS, and MAXTRANS parameters and the storage characteristics of the table.

Note: For a nonpartitioned table, each parameter and storage characteristic specified is the actual physical attribute of the segment associated with the table. For partitioned tables, the value specified for the parameter or storage characteristic is the default physical attribute of the segments associated with all partitions specified in the CREATE statement (and in subsequent ALTER TABLE ADD PARTITION statements), unless you explicitly override that value in the PARTITION clause of this command.  

PCTFREE  

specifies the percentage of space in each data block of the table, object table OIDINDEX, or partition reserved for future updates to the table's rows. The value of PCTFREE must be a value from 0 to 99. A value of 0 allows the entire block to be filled by inserts of new rows. The default value is 10. This value reserves 10% of each block for updates to existing rows and allows inserts of new rows to fill a maximum of 90% of each block.  

 

PCTFREE has the same function in the PARTITION description clause and in the commands that create and alter clusters, indexes, snapshots, and snapshot logs. The combination of PCTFREE and PCTUSED determines whether new rows will be inserted into existing data blocks or into new blocks.  

PCTUSED  

specifies the minimum percentage of used space that Oracle maintains for each data block of the table, object table OIDINDEX, or index-organized table overflow data segment. A block becomes a candidate for row insertion when its used space falls below PCTUSED. PCTUSED is specified as a positive integer from 1 to 99 and defaults to 40.  

 

PCTUSED has the same function in the PARTITION description clause and in the commands that create and alter clusters, snapshots, and snapshot logs.  

 

PCTUSED is not a valid table storage characteristic for an index-organized table (ORGANIZATION INDEX).  

 

The sum of PCTFREE and PCTUSED must be less than 100. You can use PCTFREE and PCTUSED together to utilize space within a table more efficiently. For information on the performance effects of different values PCTUSED and PCTFREE, see Oracle8 Tuning.  

INITRANS  

specifies the initial number of transaction entries allocated within each data block allocated to the table, object table OIDINDEX, partition, LOB index segment, or overflow data segment. This value can range from 1 to 255 and defaults to 1. In general, you should not change the INITRANS value from its default.  

 

Each transaction that updates a block requires a transaction entry in the block. The size of a transaction entry depends on your operating system.  

 

This parameter ensures that a minimum number of concurrent transactions can update the block and helps avoid the overhead of dynamically allocating a transaction entry.  

 

The INITRANS parameter serves the same purpose in the PARTITION description clause, clusters, indexes, snapshots, and snapshot logs as in tables. The minimum and default INITRANS value for a cluster or index is 2, rather than 1.  

MAXTRANS  

specifies the maximum number of concurrent transactions that can update a data block allocated to the table, object table OIDINDEX, partition, LOB index segment, or index-organized overflow data segment. This limit does not apply to queries. This value can range from 1 to 255 and the default is a function of the data block size. You should not change the MAXTRANS value from its default.  

 

If the number concurrent transactions updating a block exceeds the INITRANS value, Oracle dynamically allocates transaction entries in the block until either the MAXTRANS value is exceeded or the block has no more free space.  

 

The MAXTRANS parameter serves the same purpose in the PARTITION description clause, clusters, snapshots, and snapshot logs as in tables.  

storage_clause  

specifies the storage characteristics for the table, object table OIDINDEX, partition, LOB storage, LOB index segment, or index-organized table overflow data segment. This clause has performance ramifications for large tables. Storage should be allocated to minimize dynamic allocation of additional space. See the STORAGE clause.  

TABLESPACE  

specifies the tablespace in which Oracle creates the table, object table OIDINDEX, partition, LOB storage, LOB index segment, or index-organized table overflow data segment. If you omit this option, then Oracle creates that item in the default tablespace of the owner of the schema containing the table.  

 

For nonpartitioned tables, the value specified for TABLESPACE is the actual physical attribute of the segment associated with the table. For partitioned tables, the value specified for TABLESPACE is the default physical attribute of the segments associated with all partitions specified in the CREATE statement (and on subsequent ALTER TABLE ADD PARTITION statements), unless you specify TABLESPACE in the PARTITION description clause.  

LOGGING/NOLOGGING  

specifies whether the creation of the table (and any indexes required because of constraints), partition, or LOB storage characteristics will be logged in the redo log file. It also specifies that subsequent Direct Loader (SQL*Loader) and direct-load INSERT operations against the table, partition, or LOB storage are logged (LOGGING) or not logged (NOLOGGING).

If you omit the LOGGING/NOLOGGING clause, the logging attribute of the table or table partition defaults to the logging attribute of the tablespace in which it resides. For LOBs, if you omit the LOGGING/NOLOGGING clause,

  • if you specify CACHE, then LOGGING is used (because you cannot have CACHE NOLOGGING);
  • otherwise, the logging attribute defaults to the logging attribute of the tablespace in which it resides.

For nonpartitioned tables, the value specified for LOGGING is the actual physical attribute of the segment associated with the table. For partitioned tables, the logging attribute value specified is the default physical attribute of the segments associated with all partitions specified in the CREATE statement (and in subsequent ALTER TABLE ADD PARTITION statements), unless you specify LOGGING/NOLOGGING in the PARTITION description clause.  

 

In NOLOGGING mode, data is modified with minimal logging (to mark new extents invalid and to record dictionary changes). When applied during media recovery, the extent invalidation records mark a range of blocks as logically corrupt, because the redo data is not fully logged. Therefore, if you cannot afford to lose this table, you should take a backup after the NOLOGGING operation.  

 

The size of a redo log generated for an operation in NOLOGGING mode is significantly smaller than the log generated with the LOGGING option set.

If the database is run in ARCHIVELOG mode, media recovery from a backup taken before the LOGGING operation restores the table. However, media recovery from a backup taken before the NOLOGGING operation does not restore the table.  

 

The logging attribute of the table is independent of that of its indexes.  

 

NOLOGGING is not a valid keyword for creating index-organized tables.  

 

For more information about the LOGGING option and Parallel DML, see Oracle8 Concepts and Oracle8 Administrator's Guide.  

 

Note: In future versions of Oracle, the LOGGING keyword will replace the RECOVERABLE option. RECOVERABLE is still available as a valid keyword in Oracle when creating nonpartitioned tables, however, it is not recommended.  

ORGANIZATION INDEX  

specifies that table is created as an index-organized table. In an index-organized table, the data rows are held in an index defined on the primary key for the table. See also "Index-Organized Tables".  

ORGANIZATION HEAP  

specifies that the data rows of table are stored in no particular order. This is the default.  

index_organized_table_clause  

PCTTHRESHOLD integer  

specifies the percentage of space reserved in the index block for an index-organized table row. Any portion of the row that exceeds the specified threshold is stored in the area. PCTTHRESHOLD must be a value from 0 to 50.  

 

OVERFLOW  

specifies that index-organized table data rows exceeding the specified threshold are placed in the data segment listed in this clause. If OVERFLOW is not specified, then rows exceeding the PCTTHRESHOLD limit are rejected.  

 

INCLUDING column_name  

specifies a column at which to divide an index-organized table row into index and overflow portions. All columns that follow column_name are stored in the overflow data segment. A column_name is either the name of the last primary-key column or any nonprimary-key column.  

RECOVERABLE  

is a deprecated option. RECOVERABLE is not a valid keyword for creating partitioned tables or LOB storage characteristics.  

UNRECOVERABLE  

is a deprecated option. It specifies that the creation of the table (and any indices required because of constraints) will not be logged in the redo log file.

This keyword can only be specified with the AS subquery clause. UNRECOVERABLE is not a valid keyword for creating partitioned or index-organized tables.  

 

Note: In future versions of Oracle, the LOGGING keyword will replace the RECOVERABLE option. RECOVERABLE is still available as a valid keyword in Oracle when creating nonpartitioned tables, however, it is not recommended.  

LOB_storage_clause  

LOB  

specifies the LOB storage characteristics. For detailed information about LOBs, see Oracle8 Application Developer's Guide.  

 

lob_item  

is the LOB column name or LOB object attribute for which you are explicitly defining tablespace and storage characteristics that are different from those of the table.  

STORE AS  

lob_segname  

specifies the name of the LOB data segment. You cannot use lob_segname if you specify more than one lob_item.  

lob_parameters  

ENABLE STORAGE IN ROW  

specifies that the LOB value is stored in the row (in-line) if its length is less than approximately 4000 bytes minus system control information. This is the default.  

 

DISABLE STORAGE IN ROW  

specifies that the LOB value is stored outside of the row regardless of the length of the LOB value.  

 

Note that the LOB locator is always stored in the row regardless of where the LOB value is stored. You cannot change the value of STORAGE IN ROW once it is set.  

 

CHUNK integer  

specifies the number of bytes to be allocated for LOB manipulation. If integer is not a multiple of the database block size, Oracle rounds up (in bytes) to the next multiple. For example, if the database block size is 2048 and integer is 2050, Oracle allocates 4096 bytes (2 blocks). The maximum value is 32768 (32K), which is the largest Oracle block size allowed.

Note: The value of CHUNK must be less than or equal to the values of both INITIAL and NEXT (either the default values or those specified in the storage clause). If CHUNK exceeds the value of either INITIAL or NEXT, Oracle returns an error.  

 

PCTVERSION integer  

is the maximum percentage of overall LOB storage space used for creating new versions of the LOB. The default value is 10, meaning that older versions of the LOB data are not overwritten until 10% of the overall LOB storage space is used.  

 

INDEX lob_index_name  

is the name of the LOB index segment. You cannot specify lob_index_name if you specify more than one lob_item in the associated lob_item list. Note that you cannot alter the LOB index through the ALTER INDEX statement. You can alter a LOB index specification only through the ALTER TABLE statement (see ALTER TABLE).

Note also that a user cannot drop the LOB index. It is a system index created and maintained by the system.  

NESTED TABLE ... STORE AS ...  

specifies storage_table as the name of the storage table in which the rows of all nested_item values reside. You must include this clause when creating a table with columns or column attributes whose type is a nested table. See also "Nested Table Storage"  

 

  • nested_item is the name of a column or a column-qualified attribute whose type is a nested table.
  • storage_table is the name of the storage table. The storage table is created in the same schema and the same tablespace as the parent table.
 

CLUSTER  

specifies that the table is to be part of the cluster. The columns listed in this clause are the table columns that correspond to the cluster's columns. Generally, the cluster columns of a table are the column or columns that make up its primary key or a portion of its primary key.  

 

Specify one column from the table for each column in the cluster key. The columns are matched by position, not by name.

A clustered table uses the cluster's space allocation. Therefore, do not use the PCTFREE, PCTUSED, INITRANS, or MAXTRANS parameters, the TABLESPACE option, or the STORAGE clause with the CLUSTER option.  

 

Object tables cannot be part of a cluster.  

table_partitioning_clause:  

PARTITION BY RANGE  

specifies that the table is partitioned on ranges of values from column_list. See also "Partitioned Tables".  

 

column_list  

is an ordered list of columns used to determine into which partition a row belongs. You cannot specify more than 16 columns in column_list. The column_list cannot contain the ROWID pseudocolumn or any columns of datatype ROWID or LONG.  

 

PARTITION partition_name  

specifies the physical partition clause. If partition_name is omitted, Oracle generates a name with the form SYS_Pn for the partition. The partition_name must conform to the rules for naming schema objects and their part as described in "Schema Object Naming Rules".  

 

VALUES LESS THAN  

specifies the noninclusive upper bound for the current partition.  

 

value_list  

is an ordered list of literal values corresponding to column_list in the PARTITION BY RANGE clause. You can substitute the keyword MAXVALUE for any literal in value_list. MAXVALUE specifies a maximum value that will always sort higher than any other value, including NULL.

Specifying a value other than MAXVALUE for the highest partition bound imposes an implicit integrity constraint on the table. See Oracle8 Concepts for more information about partition bounds.  

parallel_clause  

specifies the degree of parallelism for creating the table and the default degree of parallelism for queries on the table once created.

This is not a valid option when creating index-organized tables. For more information, see PARALLEL clause.  

enable_clause  

enables an integrity constraint. See the ENABLE clause.  

disable_clause  

disables an integrity constraint. See the DISABLE clause.  

 

Constraints specified in the ENABLE and DISABLE clauses of a CREATE TABLE statement must be defined in the statement. You can also enable and disable constraints with the ENABLE and DISABLE keywords of the CONSTRAINT clause. If you define a constraint but do not explicitly enable or disable it, Oracle enables it by default.  

 

You cannot use the ENABLE and DISABLE clauses in a CREATE TABLE statement to enable and disable triggers.  

AS subquery  

inserts the rows returned by the subquery into the table upon its creation. See "Subqueries".

Note: This subquery is not supported for index-organized tables with overflow.  

 

The number of columns in the table must equal the number of expressions in the subquery. The column definitions can specify only column names, default values, and integrity constraints, not datatypes. Oracle derives datatypes and lengths from the subquery. Oracle also follows the following rules for integrity constraints:  

 

  • Oracle automatically defines any NOT NULL constraints on columns in the new table that existed on the corresponding columns of the selected table if the subquery selects the column rather than an expression containing the column.
  • A CREATE TABLE statement cannot contain both an AS clause and a referential integrity constraint definition.
  • If a CREATE TABLE statement contains both the AS clause and a CONSTRAINT clause or an ENABLE clause with the EXCEPTIONS option, Oracle ignores the EXCEPTIONS option. If any rows violate the constraint, Oracle does not create the table and returns an error message.
 

 

If all expressions in the subquery are columns, rather than expressions, you can omit the columns from the table definition entirely. In this case, the names of the columns of table are the same as the columns in the subquery.  

 

For object tables, subquery can contain either one expression corresponding to the table type, or the number of top-level attributes of the table type.  

CACHE  

for data that will be accessed frequently, specifies that the blocks retrieved for this table are placed at the most recently used end of the LRU list in the buffer cache when a full table scan is performed. This option is useful for small lookup tables.  

 

As a parameter in the LOB storage clause, CACHE specifies that Oracle preallocates and retains LOB data values in memory for faster access.

This is not a valid keyword when creating index-organized tables.  

NOCACHE  

for data that will not be accessed frequently, specifies that the blocks retrieved for this table are placed at the least recently used end of the LRU list in the buffer cache when a full table scan is performed. For LOBs, the LOB value either is not brought into the buffer cache or is brought into the buffer cache and placed at the least recently used end of the LRU list.  

 

This is the default behavior except when creating index-organized tables. This is not a valid keyword when creating index-organized tables.  

 

As a parameter in the LOB storage clause, NOCACHE specifies that LOB values are not preallocated in memory. This is the LOB storage default.  

Examples

Tables are created with no data unless a query is specified. You can add rows to a table with the INSERT command.

After creating a table, you can define additional columns, partitions, and integrity constraints with the ADD clause of the ALTER TABLE command. You can change the definition of an existing column or partition with the MODIFY clause of the ALTER TABLE command. To modify an integrity constraint, you must drop the constraint and redefine it.

Example I

To define the EMP table owned by SCOTT, you could issue the following statement:

CREATE TABLE scott.emp 
   (empno     NUMBER        CONSTRAINT pk_emp PRIMARY KEY, 
    ename     VARCHAR2(10)  CONSTRAINT nn_ename NOT NULL 
                            CONSTRAINT upper_ename 
CHECK (ename = UPPER(ename)), 
    job        VARCHAR2(9), 
    mgr      NUMBER         CONSTRAINT fk_mgr 
                            REFERENCES scott.emp(empno), 
    hiredate  DATE          DEFAULT SYSDATE, 
    sal       NUMBER(10,2)  CONSTRAINT ck_sal 
CHECK (sal > 500), 
       comm      NUMBER(9,0)   DEFAULT NULL, 
       deptno    NUMBER(2)     CONSTRAINT nn_deptno NOT NULL 
                               CONSTRAINT fk_deptno 
                               REFERENCES scott.dept(deptno) ) 
PCTFREE 5 PCTUSED 75; 

This table contains 8 columns. The EMPNO column is of datatype NUMBER and has an associated integrity constraint named PK_EMP. The HIRDEDATE column is of datatype DATE and has a default value of SYSDATE, and so on.

This table definition specifies a PCTFREE of 5 and a PCTUSED of 75, which is appropriate for a relatively static table. The definition also defines integrity constraints on some columns of the EMP table.

Example II

To define the sample table SALGRADE in the HUMAN_RESOURCE tablespace with a small storage capacity and limited allocation potential, issue the following statement:

CREATE TABLE salgrade 
   ( grade  NUMBER  CONSTRAINT pk_salgrade 
                    PRIMARY KEY 
                    USING INDEX TABLESPACE users_a, 
     losal  NUMBER, 
     hisal  NUMBER ) 
TABLESPACE human_resource 
STORAGE (INITIAL     6144  
         NEXT        6144 
         MINEXTENTS     1  
         MAXEXTENTS     5 
         PCTINCREASE    5); 

The above statement also defines a PRIMARY KEY constraint on the GRADE column and specifies that the index Oracle creates to enforce this constraint is created in the USERS_A tablespace.

For more examples of defining integrity constraints, see the CONSTRAINT clause. For examples of enabling and disabling integrity constraints, see the ENABLE clause and the DISABLE clause.

Example III

When using parallel query, the fastest way to create a table that has the same columns as the EMP table, but only for those employees in department 10, is to issue a command similar to the following:

CREATE TABLE emp_tmp
   NOLOGGING
   PARALLEL (DEGREE 3)
   AS SELECT * FROM emp WHERE deptno = 10; 

Using parallelism speeds up the creation of the table because Oracle uses three processes to create the table. After the table is created, querying the table is also faster, because the same degree of parallelism is used to access the table.

LOB Column Example

The following example creates table LOB_TAB with two LOB columns and specifies the LOB storage characteristics:

CREATE TABLE lob_tab (col1 BLOB, col2 CLOB)
STORAGE (INITIAL 256 NEXT 256)
LOB (col1, col2) STORE AS
   (TABLESPACE lob_seg_ts
    STORAGE (INITIAL 6144 NEXT 6144)
    CHUNK 4
    NOCACHE LOGGING
       INDEX (TABLESPACE lob_index_ts
       STORAGE (INITIAL 256 NEXT 256)
   )
);

Index-Organized Tables

Index-organized tables are special kinds of tables that keep data sorted on the primary key and are therefore best suited for primary key-based access and manipulation.

An index-organized table is an alternative to

Index-organized tables differ from other kinds of tables in that Oracle maintains the table rows in a B*-tree index built on the primary key. However, the index row contains both the primary key column values and the associated non-key column values for the corresponding row.

You must specify a primary key for an index-organized table, because the primary key uniquely identifies a row. Use the primary key instead of the ROWID for directly accessing index-organized rows.

Example

The following statement creates an index-organized table:

CREATE TABLE docindex
  ( token CHAR(20),
    doc_oid INTEGER,
    token_frequency SMALLINT,
    token_occurrence_data VARCHAR(512),
       CONSTRAINT pk_docindex PRIMARY KEY (token, doc_oid) )
  ORGANIZATION INDEX TABLESPACE text_collection
  PCTTHRESHOLD 20 INCLUDING token_frequency
  OVERFLOW TABLESPACE text_collection_overflow;

Partitioned Tables

A partitioned table consists of a number of pieces all of which have the same logical attributes. For example, all partitions share the same column and constraint definitions.

You can create a partitioned table with just one partition. Note, however, that a partitioned table with one partition is different from a nonpartitioned table. For instance, you cannot add a partition to a nonpartitioned table.

Example

The following example creates a table with three partitions:

CREATE TABLE stock_xactions
   (stock_symbol CHAR(5),
    stock_series CHAR(1),
    num_shares NUMBER(10),
    price NUMBER(5,2),
    trade_date DATE)
STORAGE (INITIAL 100K NEXT 50K) LOGGING
PARTITION BY RANGE (trade_date)
  (PARTITION sx1992 VALUES LESS THAN (TO_DATE('01-JAN-1993','DD-MON-YYYY'))
     TABLESPACE ts0 NOLOGGING,
   PARTITION sx1993 VALUES LESS THAN (TO_DATE('01-JAN-1994','DD-MON-YYYY'))
     TABLESPACE ts1,
   PARTITION sx1994 VALUES LESS THAN (TO_DATE('01-JAN-1995','DD-MON-YYYY'))
     TABLESPACE ts2);

For information about partitioned table maintenance operations, see the Oracle8 Administrator's Guide.

Object Tables

In order to have Oracle assign an object identifier to an object, the object must reside in a special kind of table called an object table. Objects residing in an object table are referenceable. For more information about using REFs, see "User-Defined Types", "User Functions", "Expressions", CREATE TYPE, andOracle8 Administrator's Guide.

The columns of an object table correspond to the top-level attributes of the corresponding type. Each row will contain an object instance, and each instance will be assigned a unique, system-generated object identifier (OID) when a row is inserted.

Example I

For example, consider object type DEPT_T:

CREATE TYPE dept_t AS OBJECT
( dname VARCHAR2(100), 
  address VARCHAR2(200) ); 

Object table DEPT holds department objects of type DEPT_T:

CREATE TABLE dept OF dept_t; 
Example II

The following example creates object table SALESREPS with a user-defined object type, SALESREP_T:

CREATE OR REPLACE TYPE salesrep_t AS OBJECT
  ( repId NUMBER,
    repName VARCHAR2(64));
CREATE TABLE salesreps OF salesrep_t;

Nested Table Storage

Creating a table with columns of type TABLE implicitly creates a storage table for each nested table column. The storage table is created in the same tablespace as its parent table (using the default storage characteristics) and stores the nested table values of the column for which it was created.

You cannot query or perform DML statements on the storage table directly, but you can modify the nested table column storage characteristics by using the name of storage table in an ALTER TABLE statement. For information about modifying nested table column storage characteristics, see ALTER TABLE.

Example

The following example creates relational table EMPLOYEE with a nested table column PROJECTS:

CREATE TABLE employee (empno NUMBER, name CHAR(31),
projects PROJ_TABLE_TYPE)
NESTED TABLE projects STORE AS nested_proj_table;

REFs

A REF value is a reference to a row in an object table. A table can have top-level REF columns or REF attributes embedded within an object type column. In general, if a table has a REF column, each REF value in the column could reference a row in a different object table. A SCOPE clause restricts the scope of references to a single table.

For example, if you create an object table DEPT which stores all the departments in an organization, you could then create table EMP that contains a REF column (E_DEPT) to point to the department in which each employee works. Because all employees work in some department stored in the DEPT table, a scope clause can be specified on the E_DEPT column of EMP to restrict the scope of references to the DEPT table.

You can increase the performance of queries with dereference operations and decrease the amount of storage needed for REF values by using the scope clause. Note that a SCOPE clause does not have the same semantics as referential constraints. Referential constraints do not allow dangling references. Also, referential constraints do not necessarily restrict the scope of references to a single table (one can specify multiple referential constraints on the same foreign key, with each one of them pointing to a different table).

You can also store REF values with or without ROWIDs. Storing REF values WITH ROWID can enhance the performance of dereference operations, but takes up more space. The default behavior is to store REF values without the ROWID.

You cannot specify REF clauses on REF columns in nested tables using the CREATE TABLE statement. To specify REF clauses on REF columns in nested tables, use the ALTER TABLE to modify the nested table 's storage table.

Example

The following example creates object type DEPT_T and object table DEPT to store instances of all departments. A table with a scoped REF is then created.

CREATE TYPE dept_t AS OBJECT
( dname VARCHAR2(100),
  address VARCHAR2(200) );

CREATE TABLE dept OF dept_t;

CREATE TABLE emp
( ename VARCHAR2(100),
  enumber NUMBER,
  edept REF dept_t SCOPE IS dept );

Constraints on Object Type Columns

You can create UNIQUE, PRIMARY KEY, and REFERENCES constraints on scalar attributes of object type columns. You can also create NOT NULL constraints on object type columns, and CHECK constraints that reference object type columns or any attribute of an object type column.

Example

CREATE TYPE address AS OBJECT
  ( hno NUMBER,
    street VARCHAR2(40),
    city VARCHAR2(20),
    zip VARCHAR2(5),
    phone VARCHAR2(10) );

CREATE TYPE person AS OBJECT
  ( name VARCHAR2(40),
    dateofbirth DATE,
    homeaddress address,
    manager REF person );

CREATE TABLE persons OF person
  ( homeaddress NOT NULL
      UNIQUE (homeaddress.phone),
      CHECK (homeaddress.zip IS NOT NULL),
      CHECK (homeaddress.city <> 'San Francisco') );

Related Topics

CREATE TYPE
CREATE CLUSTER
CREATE CLUSTER
CREATE TABLESPACE
CREATE TYPE
DROP TABLE
CONSTRAINT clause
DISABLE clause
ENABLE clause
PARALLEL clause
STORAGE clause

CREATE TABLESPACE

Purpose

To create a tablespace. A tablespace is an allocation of space in the database that can contain schema objects. See also "About Tablespaces".

Prerequisites

You must have CREATE TABLESPACE system privilege. Also, the SYSTEM tablespace must contain at least two rollback segments including the SYSTEM rollback segment.

Syntax

filespec: See "Filespec".

autoextend_clause::=

storage_clause: See STORAGE clause.

Keywords and Parameters

tablespace

DATAFILE filespec  

is the name of the tablespace to be created.  

specifies the datafile or files to make up the tablespace. See "Filespec".  

 

 

autoextend_clause  

enables or disables the automatic extension of the datafile.  

 

OFF  

disables autoextend if it is turned on. NEXT and MAXSIZE are set to zero. Values for NEXT and MAXSIZE must be respecified in further ALTER TABLESPACE AUTOEXTEND commands.  

 

ON  

enables autoextend.  

 

NEXT  

specifies the disk space to allocate to the datafile when more extents are required.  

 

MAXSIZE  

specifies the maximum disk space allowed for allocation to the datafile.  

 

UNLIMITED  

sets no limit on allocating disk space to the datafile.  

MINIMUM EXTENT integer  

controls free space fragmentation in the tablespace by ensuring that every used and/or free extent size in a tablespace is at least as large as, and is a multiple of, integer. For more information about using MINIMUM EXTENT to control space fragmentation, see Oracle8 Administrator's Guide.  

LOGGING/NOLOGGING  

specifies the default logging attributes of all tables, index, and partitions within the tablespace. LOGGING is the default.  

 

The tablespace-level logging attribute can be overridden by logging specifications at the table, index, and partition levels.

Only the following operations support the NOLOGGING mode:

DML:

  • direct-load INSERT (serial or parallel)
  • Direct Loader (SQL*Loader)

DDL:

  • CREATE TABLE ... AS SELECT
  • CREATE INDEX
  • ALTER INDEX ... REBUILD
  • ALTER INDEX ... REBUILD PARTITION
  • ALTER INDEX ... SPLIT PARTITION
  • ALTER TABLE ... SPLIT PARTITION
  • ALTER TABLE ... MOVE PARTITION
 

 

In NOLOGGING mode, data is modified with minimal logging (to mark new extents invalid and to record dictionary changes). When applied during media recovery, the extent invalidation records mark a range of blocks as logically corrupt, because the redo data is not logged. Therefore, if you cannot afford to lose the object, you should take a backup after the NOLOGGING operation.  

DEFAULT storage_clause  

specifies the default storage parameters for all objects created in the tablespace. For information on storage parameters, see the STORAGE clause.  

ONLINE  

makes the tablespace available immediately after creation to users who have been granted access to the tablespace.  

OFFLINE  

makes the tablespace unavailable immediately after creation.  

 

If you omit both the ONLINE and OFFLINE options, Oracle creates the tablespace online by default. The data dictionary view DBA_TABLESPACES indicates whether each tablespace is online or offline.  

PERMANENT  

specifies that the tablespace will be used to hold permanent objects. This is the default.  

TEMPORARY  

specifies that the tablespace will be used only to hold temporary objects-for example, segments used by implicit sorts to handle ORDER BY clauses.  

About Tablespaces

A tablespace is an allocation of space in the database that can contain any of the following segments:

All databases have at least one tablespace, SYSTEM, which Oracle creates automatically when you create the database.

When you create a tablespace, it is initially a read-write tablespace. After creating the tablespace, you can subsequently use the ALTER TABLESPACE command to take it offline or online, add datafiles to it, or make it a read-only tablespace.

Many schema objects have associated segments that occupy space in the database. These objects are located in tablespaces. The user creating such an object can optionally specify the tablespace to contain the object. The owner of the schema containing the object must have space quota on the object's tablespace. You can assign space quota on a tablespace to a user with the QUOTA clause of the CREATE USER or ALTER USER commands.


warning:

For operating systems that support raw devices, be aware that the STORAGE clause REUSE keyword has no meaning when specifying a raw device as a datafile in a CREATE TABLESPACE command; such a command will always succeed even if REUSE is not specified.

 

Example I

This command creates a tablespace named TABSPACE_2 with one data file:

CREATE TABLESPACE tabspace_2 
   DATAFILE 'diska:tabspace_file2.dat' SIZE 20M 
   DEFAULT STORAGE (INITIAL 10K NEXT 50K 
                    MINEXTENTS 1 MAXEXTENTS 999 
                    PCTINCREASE 10) 
   ONLINE; 
Example II

This command creates a tablespace named TABSPACE_3 with one data file; when more space is required, 50 kilobyte extents will be added up to a maximum size of 10 megabytes:

CREATE TABLESPACE tabspace_5 
   DATAFILE 'diskb:tabspace_file3.dat' SIZE 500K REUSE
   AUTOEXTEND ON NEXT 500K MAXSIZE 10M;
Example III

This command creates tablespace TABSPACE_5 with one data file and allocates every extent as a multiple of 64K:

CREATE TABLESPACE tabspace_3 
   DATAFILE 'tabspace_file5.dbf' SIZE 2M
   MINIMUM EXTENT 64K
   DEFAULT STORAGE (INITIAL 128K NEXT 128K)
   LOGGING;

Related Topics

ALTER TABLESPACE
DROP TABLESPACE
"Filespec"

CREATE TRIGGER

Purpose

To create and enable a database trigger. A database trigger is a stored PL/SQL block associated with a table. Oracle automatically executes a trigger when a specified SQL statement is issued against the table. See also "Using Triggers".


Note:

Descriptions of commands and clauses preceded by are available only if the Oracle objects option is installed on your database server.

 

Prerequisites

Before a trigger can be created, the user SYS must run the SQL script DBMSSTDX.SQL. The exact name and location of this script depend on your operating system.

To issue this statement, you must have one of the following system privileges:

CREATE TRIGGER  

lets you create a trigger in your own schema on a table in your own schema.  

CREATE ANY TRIGGER  

lets you create a trigger in any user's schema on a table in any schema.  

If the trigger issues SQL statements or calls procedures or functions, then the owner of the schema to contain the trigger must have the privileges necessary to perform these operations. These privileges must be granted directly to the owner, rather than acquired through roles.

Syntax

Keywords and Parameters

OR REPLACE  

re-creates the trigger if it already exists. Use this option to change the definition of an existing trigger without first dropping it.  

schema  

is the schema to contain the trigger. If you omit schema, Oracle creates the trigger in your own schema.  

trigger  

is the name of the trigger to be created. See also"Conditional Predicates", "Parts of a Trigger", and "Types of Triggers".  

BEFORE  

causes Oracle to fire the trigger before executing the triggering statement. For row triggers, this is a separate firing before each affected row is changed.  

 

You cannot specify a BEFORE trigger on a view or an object view.  

AFTER  

causes Oracle to fire the trigger after executing the triggering statement. For row triggers, this is a separate firing after each affected row is changed. See also "Snapshot Log Triggers".  

 

You cannot specify an AFTER trigger on a view or an object view.  

INSTEAD OF  

causes Oracle to fire the trigger instead of executing the triggering statement. By default, INSTEAD OF triggers are activated for each row. See also "INSTEAD OF Triggers".  

 

INSTEAD OF is a valid option only for views. You cannot specify an INSTEAD OF trigger on a table.  

DELETE  

causes Oracle to fire the trigger whenever a DELETE statement removes a row from the table.  

INSERT  

causes Oracle to fire the trigger whenever an INSERT statement adds a row to table.  

UPDATE  

causes Oracle to fire the trigger whenever an UPDATE statement changes a value in one of the columns specified in the OF clause. If you omit the OF clause, Oracle fires the trigger whenever an UPDATE statement changes a value in any column of the table.  

 

You cannot specify an OF clause with an INSTEAD OF trigger. Oracle fires INSTEAD OF triggers whenever an UPDATE changes a value in any column of the view.  

 

You cannot specify nested table or LOB columns in the OF clause. See also "User-Defined Types, LOB, and REF Columns".  

ON  

specifies the schema and table or view name of the of one of the following on which the trigger is to be created:

  • table
  • object table
  • view
  • object view

If you omit schema, Oracle assumes the table is in your own schema. You can create triggers on index-organized tables. You cannot create a trigger on a table in the schema SYS. See also "User-Defined Types, LOB, and REF Columns".  

table  

is the name of a table or an object table.  

view  

is the name of a view or an object view.  

REFERENCING  

specifies correlation names. You can use correlation names in the PL/SQL block and WHEN clause of a row trigger to refer specifically to old and new values of the current row. The default correlation names are OLD and NEW. If your row trigger is associated with a table named OLD or NEW, use this clause to specify different correlation names to avoid confusion between the table name and the correlation name.  

 

If the trigger is defined on an object table or view, OLD and NEW refer to object instances.  

FOR EACH ROW  

designates the trigger to be a row trigger. Oracle fires a row trigger once for each row that is affected by the triggering statement and meets the optional trigger constraint defined in the WHEN clause.

Except for INSTEAD OF triggers, if you omit this clause, the trigger is a statement trigger. Oracle fires a statement trigger only once when the triggering statement is issued if the optional trigger constraint is met.

INSTEAD OF trigger statements are implicitly activated for each row.  

WHEN (condition)  

specifies the trigger restriction-a SQL condition that must be satisfied for Oracle to fire the trigger. See the syntax description of condition in "Conditions". This condition must contain correlation names and cannot contain a query.

You can specify a trigger restriction only for a row trigger. Oracle evaluates this condition for each row affected by the triggering statement.

You cannot specify trigger restrictions for INSTEAD OF trigger statements.

You can reference object columns or their attributes, VARRAY, nested table, or LOB columns. You cannot invoke PL/SQL functions or methods in the trigger restriction.  

pl/sql_block  

is the PL/SQL block that Oracle executes to fire the trigger. For information on PL/SQL, including how to write PL/SQL blocks, see PL/SQL User's Guide and Reference.  

 

Note: The PL/SQL block of a trigger cannot contain transaction control SQL statements (COMMIT, ROLLBACK, SAVEPOINT, and SET CONSTRAINT).  

Using Triggers

Oracle automatically fires, or executes, a trigger when a triggering statement is issued. You can use triggers for the following purposes:

For more information on how to design triggers for the above purposes, see Oracle8 Application Developer's Guide.

An existing trigger must be in one of the following states:

When you create a trigger, Oracle enables it automatically. You can subsequently disable and enable a trigger with the DISABLE and ENABLE options of the ALTER TRIGGER command or the ALTER TABLE command.

For information on how to enable and disable triggers, see ALTER TRIGGER, ALTER TABLE, the ENABLE clause, and the DISABLE clause.

Before Release 7.3, Oracle parsed and compiled a trigger whenever it was fired. From Release 7.3 onward, Oracle stores a compiled version of a trigger in the data dictionary and calls this compiled version when the trigger is fired. This feature provides a significant performance improvement for applications that use many triggers.

If a trigger produces compilation errors, it is still created, but it fails on execution. This means it effectively blocks all triggering DML statements until it is disabled, replaced by a version without compilation errors, or dropped.

To embed a CREATE TRIGGER statement inside an Oracle precompiler program, you must terminate the statement with the keyword END-EXEC followed by the embedded SQL statement terminator for the specific language.

Conditional Predicates

When you create a trigger for more than one DML operation, you can use conditional predicates within the trigger body to execute specific blocks of code, depending on the type of statement that fires the trigger. Conditional predicates are evaluated as follows:

INSERTING  

returns true if the trigger fires for an INSERT statement.  

DELETING  

returns true if the trigger fires for a DELETE statement.  

UPDATING  

returns true if the trigger fires for an UPDATE statement.  

UPDATING (column_name)  

returns true if the trigger fires for an UPDATE statement and column_name is updated.

Note: You cannot specify an object attribute as column_name.  

For more information about creating and using conditional predicates in trigger bodies, see Oracle8 Application Developer's Guide.

Example

The following example uses conditional predicates to provide information about which DML statement fires trigger AUDIT_TRIGGER:

CREATE TRIGGER audit_trigger BEFORE INSERT OR DELETE OR UPDATE
  ON classified_table FOR EACH ROW
  BEGIN
    IF INSERTING THEN
       INSERT INTO audit_table
          VALUES (USER || ' is inserting' ||
                    ' new key: ' || :new.key);
    ELSIF DELETING THEN
       INSERT INTO audit_table
          VALUES (USER || ' is deleting' ||
                    ' old key: ' || :old.key);
    ELSIF UPDATING('FORMULA') THEN
       INSERT INTO audit_table
          VALUES (USER || ' is updating' ||
                    ' old formula: ' || :old.formula ||
                    ' new formula: ' || :new.formula);
    ELSIF UPDATING THEN
       INSERT INTO audit_table
          VALUES (USER || ' is updating' ||
                    ' old key: ' || :old.key ||
                    ' new key: ' || :new.key);
    END IF;
  END;

Parts of a Trigger

The syntax of the CREATE TRIGGER statement includes the following parts of the trigger:

Triggering statement

The definition of the triggering statement specifies what SQL statements cause Oracle to fire the trigger.

DELETE
INSERT
UPDATE  

You must specify at least one of these commands that causes Oracle to fire the trigger. You can specify as many as three.  

ON  

You must also specify the table with which the trigger is associated. The triggering statement is one that modifies this table. You can define a trigger on an index-organized table.  

Trigger restriction

The trigger restriction specifies an additional condition that must be satisfied for a row trigger to