Skip Headers
Oracle® Database Administrator's Reference
10g Release 2 (10.2) for hp OpenVMS

Part Number B25416-04
Go to Documentation Home
Home
Go to Book List
Book List
Go to Table of Contents
Contents
Go to Index
Index
Go to Master Index
Master Index
Go to Feedback page
Contact Us

Go to previous page
Previous
Go to next page
Next
PDF · Mobi · ePub

1 Administering Oracle Database 10g

This chapter provides information about administering Oracle Database 10g on OpenVMS. It contains the following sections:

1.1 Overview

You must set OpenVMS logical names, parameters, and user settings for Oracle Database to work. This chapter describes the various settings for Oracle Database on OpenVMS.

This section contains the following topics:

1.1.1 Oracle Database 10g Code

Oracle Database 10g code consists of several object libraries that are used to form the Oracle Database 10g image during installation. This code also consists of a shared Oracle Database Client image linked during installation.

See Also:

Oracle Database Installation Guide for more information about shared Oracle Database Client image

Oracle Database 10g code is built to use 64-bit pointers to support very large System Global Areas (SGAs). The code for client software, however, is built to use 32-bit pointers to maintain compatibility with existing client code. There are, therefore, both 32-bit and 64-bit versions of the installed object and sharable libraries. Oracle only supports 32-bit clients. Client applications may not be built with 64-bit pointers.

1.1.1.1 ORACLE.EXE File

When Oracle Database 10g Enterprise Edition is linked and installed, the image resides as sharable code in OpenVMS global memory.

You must relink the image when, for example, new code is distributed or when a new release of OpenVMS is installed.

1.1.2 Oracle Database 10g Instances

An Oracle Database 10g instance is a combination of Oracle Database processes and memory buffers.

Because many instances can exist on one system or in one OpenVMS Cluster, you must assign a unique, one-to-six character system ID (SID) to every instance. During the installation procedure, you create an instance when you create the initial database. The SID that you assign to this instance becomes the default value of logical ORACLE_SID.

Before performing any operations on the database instance, the correct Oracle environment must be set up, including setting of logical ORACLE_SID. To do this, run the following:

$ @device:[top_install_location]orauser SID 

This section contains the following topics:

1.1.2.1 System Global Area

An SGA is an area of shared memory that is allocated to each Oracle Database instance. All database operations use data stored in the SGA.

The size of the SGA is determined by parameters in the INIT.ORA file. These parameters determine:

  • Allocation of Oracle Database resources used by the processes that share the SGA

  • Amount of data that may be maintained in the SGA

After you create an instance, you can change the size of its SGA by shutting down the instance with the SQL*Plus utility and modifying the values set in the INIT.ORA file as needed.

Consequently, parameter settings also determine the memory space needed to support these requirements. Increasing the value of these parameters can improve performance, but performance may also decrease if the SGA consumes the system memory to the extent that the system is forced to page portions of processes in and out of memory.

Oracle Database 10g release 2 (10.2) includes support for the Very Large Memory (VLM) 64-bit feature. This allows a large SGA that is limited only by the amount of physical memory available.

Data retrieved or inserted by user transactions is temporarily buffered in the SGA. Because this data resides in an area of memory accessible by all Oracle processes, disk I/O is reduced and transaction time is significantly improved.

The following are the most significant structures in the SGA.

shared pool

The shared pool contains shared cursors, stored procedures, SQL, PL/SQL blocks, and trigger code. The size of the shared pool is specified by the initialization parameter SHARED_POOL_SIZE. Larger values of this parameter improve performance in multiuser systems. Smaller values use less memory. The limit for this parameter is determined by the size of the SGA. The shared pool must be at least 150 MB.

database buffer pool

Blocks of data retrieved by user transactions are read from the database file and then cached in the database buffer pool in the SGA. This data remains in the buffer pool (even after changes are committed) until more buffers are required. If the data has been modified, then it is written to the database files.

The number of blocks that can be maintained in the buffer pool is determined by the initialization parameters DB_CACHE_SIZE or DB_nnK_CACHE_SIZE.

redo log buffer

When data is modified, a record of the change (known as a redo entry) is generated in the redo log buffer. When changes to the data are committed, the redo entries in the buffer are written to the current redo log file. Redo log files provide for data recovery if media or system failure occurs before modified data is written from the database buffer to the database file.

The number of bytes that can be maintained in the redo log buffer is determined by the initialization parameter LOG_BUFFER.

1.1.2.2 Storing Data

Data is stored in database files. Each database must have at least one database file. Whenever you create a database, an initial database file is also created for the database.

During the installation procedure, if you choose to create a database, then create one database file, typically in the ORA_ROOT:[ORADATA.dbname] directory, where dbname is the name you assign to the database. You can specify any directory for the system data file, and this directory does not necessarily need to be under ORACLE_HOME.

This initial file contains the data dictionary tables and all data entered by Oracle users (until you expand the database by creating tablespaces and adding data files).

Oracle recommends that the cluster size on the disk drive that will contain the database files be an integer multiple of the Oracle Database 10g Enterprise Edition block size. For example, if the blocks are 2 KB, then the cluster size should be 2 KB, 8 KB, 12 KB, and so on. Keep in mind, though, that cluster sizes are specified in terms of disk blocks (where one block = 512 bytes). So, a 2 KB cluster is a 4-block cluster.

A disk cluster size is the minimum unit of disk allocation. You determine the size when you initialize a disk.

1.1.2.3 Storing Changes to the Database

Changes made to the database are logged in the database buffer pool and in a file called a redo log. The changes recorded in the redo log provide for data recovery if media, software, or system failure occurs before the database buffers are written to the database files. Every database must have at least two redo log files so that another redo log is available when the current log is filled.

Modified data is written from the database buffer pool to the database files when the current redo log is full or when the number of blocks in the redo log equals the value set by the INIT.ORA parameter LOG_CHECKPOINT_INTERVAL. Any event that causes the database buffers to be written is known as a checkpoint.

The default value of the LOG_CHECKPOINT_INTERVAL parameter is 10,000 disk blocks (5MB). If required, you can specify a different value. The values assigned to this parameter must be in multiples of physical block size. By setting the LOG_CHECKPOINT_INTERVAL initialization parameter to zero, you can eliminate interval checkpoints. This reduces the checkpoint frequency and optimizes run-time performance.

1.1.2.4 Using Redo Log Files

You can specify one of two modes for writing redo log files: ARCHIVELOG and NOARCHIVELOG. Using the redo logs in ARCHIVELOG mode allows data recovery in the event of media, software, or system failure.

Caution:

If you are using the NOARCHIVELOG mode when a media failure occurs, then you will not be able to perform media recovery. You must use the ARCHIVELOG mode to recover from media failure.

When a redo log file is full, the DBA must archive the log file to an offline file before the redo log file can be reused. If it is not archived by the time all other redo log files are filled, then database operations are suspended until archiving is completed. The DBA can archive the redo logs either manually or automatically.

In the NOARCHIVELOG mode, data in the log file is overwritten when a redo log file must be reused. However, data is never overwritten until data in the database buffer has been written to the database file. Using the redo log files in the NOARCHIVELOG mode ensures data recovery only for software and system failure.

During the Oracle Database installation procedure, redo log files are created with the default names REDO01.LOG, REDO02.LOG and REDO03.LOG. By default, these files are saved in the ORA_DB directory, but you can choose an alternative directory. These log files are used in NOARCHIVELOG mode by default. You can change the mode to ARCHIVELOG. These files are also 100 MB each, by default. You can alter this size and specify different file names during the installation procedure.

1.1.2.5 Using Logical Names

You can use logical names to specify the names of the database, redo log, and control files. Oracle recommends that you use system-level logical names to name the devices where the database and redo log files reside, and that you specify full directory and file name paths for these files.

Control files store logical file names as their translated equivalents, but do not translate concealed logical names.

Caution:

Never use process-level concealed logical names to name any Oracle Database, redo log, or control file. Read the information given in Oracle Database Administrator's Guide before renaming these files.

You can rename these files by using the ALTER DATABASE and ALTER TABLESPACE commands.

1.1.3 Logical Names Data Files: Locations and Identifying

Oracle data files may be placed in any location on any disk, subject to the following restrictions:

  • The data files or the directory that contains the data files cannot be owned by anyone with a group equal to or less than MAXSYSGROUP.

  • The Oracle Database 10g account must have write access to the location of the data files.

  • Data files cannot be saved in the root-level directory of a disk.

If the directory is not specified, then the default location for created data files is the ORA_DB directory.

You can identify data files by logical names rather than fully qualified file names in the CREATE DATABASE or ALTER TABLESPACE statements. However, these logical names must be defined at the GROUP level or higher, preferably at the SYSTEM level. Logical names at the PROCESS or JOB level cannot be used to identify data files. If you identify the data files by logical names, then ensure that these logical names are defined during system startup before calling any procedure to restart the database.

1.2 Initialization Parameters

The following sections provide information about Oracle Database initialization parameters:

1.2.1 Maximum Value of DB_BLOCK_SIZE Initialization Parameter

The DB_BLOCK_SIZE initialization parameter specifies the standard block size for the database. This block size is used for the SYSTEM tablespace and by default in other tablespaces.

You can set the DB_BLOCK_SIZE parameter value to a maximum of 32 KB.

Note:

You cannot change the value of the DB_BLOCK_SIZE parameter after you create a database.

1.2.2 CLUSTER_INTERCONNECTS Initialization Parameter

In an Oracle RAC environment, you can use the CLUSTER_INTERCONNECTS initialization parameter to specify an alternative interconnect for the private network.

The CLUSTER_INTERCONNECTS parameter requires the IP address of the interconnect instead of the device name. Oracle RAC network traffic is distributed among all the specified IP addresses.

The CLUSTER_INTERCONNECTS parameter is useful only in an Oracle RAC environment where UDP IPC is enabled. It enables users to specify an interconnect for all IPC traffic that includes Oracle Global Cache Service (GCS), Global Enqueue Service (GES), and Interprocessor Parallel Query (IPQ).

Overall cluster stability and performance may improve when you force Oracle GCS, GES, and IPQ over a different interconnect by setting the CLUSTER_INTERCONNECTS parameter. For example, to use the network interface whose IP address is 129.34.137.212 for all GCS, GES, and IPQ IPC traffic, set the CLUSTER_INTERCONNECTS parameter as follows:

CLUSTER_INTERCONNECTS=129.34.137.212

With Hewlett-Packard TCP/IP for OpenVMS, run the following:

@SYS$MANAGER:TCPIP$DEFINE_COMMANDS

Use the ifconfig command to display the IP address of a device. This command displays device names and their IP addresses. For example, to determine the IP address of a device on OpenVMS, enter the following command:

CPQ024> ifconfig -a
IE0: flags=c43<UP,BROADCAST,RUNNING,MULTICAST,SIMPLEX>
   *inet 144.25.70.16 netmask fffffc00 broadcast 144.25.71.255 ipmtu 1500

LO0:  
flags=100c89<UP,LOOPBACK,NOARP,MULTICAST,SIMPLEX,NOCHECKSUM> 
      inet 127.0.0.1 netmask ff000000 ipmtu 4096 

TN0: flags=80<NOARP> 

TN1: flags=80<NOARP>

In the preceding example, the interface IE0: has an IP address of 144.25.70.16.

Note the following points when using the CLUSTER_INTERCONNECTS initialization parameter:

  • The IP addresses specified for the different instances of the same database on different nodes should belong to network adapters that connect to the same network. If you do not follow this rule, then internode traffic may pass through bridges and routers or there may not be a path between the two nodes at all.

  • Specify the CLUSTER_INTERCONNECTS parameter in the parameter file, setting a different value for each database instance.

  • Set a value for CLUSTER_INTERCONNECTS only when you have an alternative network address to use for the specific Oracle RAC instance within an Oracle RAC environment.

    The minimum requirement for a Cluster Interconnect is 1 GB. You would typically use an alternative network address when the default OpenVMS Cluster network address is insufficient to meet the bandwidth requirements of large Oracle RAC databases.

    On OpenVMS Galaxy systems, this can be a shared memory network address.

Note:

If you specify an OpenVMS shared memory network address, then all Oracle RAC instances which are part of an Oracle RAC environment must be within the same OpenVMS Galaxy and all of them must be using shared memory network addresses. Contact the HP OpenVMS ambassador to set up shared memory.

1.3 Database Limits

Refer to Table A-1 for information about the default and maximum values for parameters in a CREATE DATABASE or CREATE CONTROLFILE statement.

Note:

Interdependencies among these parameters may affect allowable values.

Table 1-1 lists the Oracle Database 10g file size limits specific to OpenVMS.

Table 1-1 File Size Limits

File Type Maximum Size

Data File

4,194,303 multiplied by the value of the DB_BLOCK_SIZE parameter

Import file

Unlimited (limited by the OS file size limit)

Export file

Unlimited (limited by the OS file size limit)

SQL*Loader file

Unlimited (limited by the OS file size limit)


1.4 Operating System Accounts and Groups

Special operating system accounts and groups are required by Oracle Database 10g.

Table 1-2 describes the oracle and system operating system accounts.

Table 1-2 OpenVMS Accounts

Name Description

oracle

This account, called the Oracle software owner, is the operating system account that owns the Oracle Database 10g software. The remainder of this document refers to this account simply as the oracle user.

system

The system user is a special operating system account with maximum privileges. This account is used to configure and install networking software, and create user accounts. This user is not to be confused with the Oracle account SYSTEM.

ORA_NOBODY

This account, which echoes the Unix 'nobody' account, is the operating system account through which 'external jobs' are executed. Section 1.5.6 of this document discusses this account in detail.


Note:

It is possible for different accounts to be used to perform Oracle Cluster Ready Services operations in the CRS home. However, all such accounts must be in the same UIC group as the account that was used to install and startup Oracle Cluster Ready Services in the CRS home.

1.5 Security

Oracle Database 10g uses several features of the OpenVMS operating system to provide a secure environment for users. These features include file ownership, group accounts, and the ability of a program to change its user ID when it is run.

The two-task architecture of Oracle Database 10g improves security by dividing work (and address space) between the user program and the oracle server process. All database access is achieved using the shadow process and special authorizations in the oracle server process.

This section covers the following topics:

See Also:

Oracle Database Administrator's Guide for more information about security issues

1.5.1 Security for SQL*Plus Command

Only the Oracle software owner and database administrator should have system privileges and the requirements for STARTUP, SHUTDOWN, and CONNECT AS SYSDBA.

1.5.2 Security for Database Files

The oracle user should own the database files. Set the permissions on these files to read/write for the owner and no permission for System, Group, and World.

The oracle user should also own the directories containing the database files. For additional security, you can curtail visibility of the datafiles. To do this, remove any directory permission for System, Group, and World.

1.5.3 Customizing the Initialization File

The default initialization file (init.ora) is provided with the Oracle Database 10g software. All Oracle Database 10g instances assume these values if you do not specify different values for them in the initsid.ora file. Oracle recommends that you include only those parameters in the initsid.ora file that differ from the default initialization parameter values.

Use the SQL*Plus command SHOW PARAMETERS to display the current values of these parameters on the system.

Caution:

Ensure that your database uses the following values for the listed parameters.
  • disk_asynch_io=false

  • tape_asynch_io=false

  • backup_tape_io_slaves=false

  • db_Writer_processes=1

By default, Database Configuration Assistant creates the two asynch parameters with true values.

Table 1-3 lists default initialization parameter values on OpenVMS and their default values and range of values.

Table 1-3 Initialization Parameters

Parameters Default Value Range of Values

BITMAP_MERGE_AREA_SIZE

1048576

From 65536 onward (no upper limit)

COMMIT_POINT_STRENGTH

1

0 to 255

CONTROL_FILES

none

none

CREATE_BITMAP_AREA_SIZE

8388608

From 65536 onward (no upper limit)

DB_nnK_CACHE_SIZE

0

2 KB to 32 KB

DB_BLOCK_SIZE

8192

2 KB to 32 KB

DB_CACHE_SIZE

0

From 8 MB onward (no upper limit)

DB_FILES

200

1 to 2000000

DB_FILE_MULTIBLOCK_READ_COUNT

16

1 to the lower of the following:

  • The value of the DB_nnK_CACHE_SIZE parameter divided by 4

  • 1048576 divided by the value of the DB_BLOCK_SIZE parameter

DISTRIBUTED_TRANSACTIONS

1/4 TRANSACTIONS

From 0 onward (no upper limit)

HASH_AREA_SIZE

2*SORT_AREA_SIZE

From 0 onward (no upper limit)

JAVA_POOL_SIZE

210 MB

Between 1000000 and 1000000000

LOG_BUFFER

512 KB or 128 KB multiplied by the value of the CPU_COUNT parameter, which ever is higher

66560 to unlimited

LOG_CHECKPOINT_INTERVAL

0

0 to unlimited

DISPATCHERS

5

Between MAX_DISPATCHERS and PROCESSES

MAX_SHARED_SERVERS

2 multiplied by the value of the SHARED_SERVERS parameter, if the value of the SHARED_SERVERS parameter is greater than 20; otherwise, 20

Between SHARED_SERVERS and PROCESSES

SHARED_SERVERS

1, if DISPATCHERS is specified, else 0

Between 1 and PROCESSES

NLS_LANGUAGE

AMERICAN

Valid language names

NLS_TERRITORY

AMERICA

Valid territory names

OBJECT_CACHE_MAX_SIZE_PERCENT

10

0 to unlimited

OBJECT_CACHE_OPTIMAL_SIZE

100 KB

10 KB to unlimited

OPEN_CURSORS

300

1 to unlimited

OS_AUTHENT_PREFIX

OPS$

Arbitrary string


The following two parameters have been desupported:


DB_FILE_DIRECT_IO_COUNT
HASH_MULTIBLOCK_IO_COUNT

See Also:

Refer to Oracle Database Installation Guide for a complete list of desupported initialization parameters

1.5.4 Running the orapwd Utility

You can use a password file to identify users who can use the SYSDBA and SYSOPER privileges when connecting to the database. To create the password file:

  1. Log in as the oracle user.

  2. Use the ORACLE_HOME:[BIN]orawpd utility, which has the following syntax:

    $ ORAPWD FILE=filename PASSWORD=password ENTRIES=max_users 
    

    Table 1-4 describes the filename, password, and max_users syntax for running the orapwd utility.

Table 1-4 Syntax for Running the orapwd Utility

Variable Description

filename

Name of the file where password information is written. The name of the file must be orapwsid, and you must supply the full path name. Its contents are encrypted and not user-readable. This parameter is mandatory. The password file is typically created in the ora_db directory.

password

This parameter sets the password for the SYS user. If you use an ALTER USER statement to change the password for the SYS user after you connect to the database, then both the password stored in the data dictionary and the password stored in the password file are updated. This parameter is mandatory.

max_users

Maximum number of entries that you require the password file to accept.


See Also:

Oracle Database Administrator's Guide for more information about using the orapwd utility

1.5.5 Password Management

For security reasons, Oracle Database Configuration Assistant locks most Oracle user accounts after it creates the database. It does not lock the SYS, SYSTEM, or SCOTT accounts. You must unlock the accounts that are locked and change their passwords before logging into them. Use SQL*Plus to connect to the database as SYSDBA and enter the following command:

SQL> ALTER USER username IDENTIFIED BY passwd ACCOUNT UNLOCK;

1.5.6 Security for External Jobs

To securely execute external jobs, for example using DBMS_SCHEDULER.CREATE_JOB, it is necessary to create a minimally privileged OpenVMS account called ORA_NOBODY. The external jobs will run under this account. The account should be created by the OpenVMS System Administrator. For example:

$ SET DEFAULT SYS$SYSTEM
$ RUN SYSGEN
SYSGEN> SHOW MAXSYSGROUP

Make a note of the current value, for example, 8.

SYSGEN> EXIT
$ RUN AUTHORIZE
UAF> LIST * /BRIEF
%UAF-I-LSTMSG1, writing listing file
%UAF-I-LSTMSG2, listing file SYSUAF.LIS complete
UAF> EXIT
$ SORT/KEY=(POSITION=35,SIZE=12) SYSUAF.LIS SYSUAF.TMP
$ EDIT/READ SYSUAF.TMP

Determine a suitable new UIC group which can be used for the new ORA_NOBODY account. There should currently be no other accounts in this group and the group must be greater than MAXSYSGROUP. For example:

Greg Mayhew          GMAYHEW         [240,22]     24959    Normal
Gordon Brown         GBROWN          [240,23]     24959    Normal
Jeffrey Archer       JARCHER         [240,24]     24959    Normal
Charles Windsor      CWINDSOR        [240,25]     24959    Normal
Oracle DBA1          DBA1            [244,1]      27335    All
Oracle DBA2          DBA2            [244,2]      27335    All
Oracle Admin         ORA_ADMIN       [244,3]      27335    Devour
I

In the example, it could be determined that the UIC group 241 could be used

Exit from the editor.

$ DELETE SYSUAF.LIS;0
$ DELETE SYSUAF.TMP;0
$ RUN AUTHORIZE
UAF> ADD ORA_NOBODY /ACCOUNT="Nobody"/UIC=[241,1]/PASSWORD=PRIVATE1 -
     /NOPWDEXP/PGFLQUOTA=200000/BYTLM=4000000/BIOLM=100/DIOLM=100 -
     /ASTLM=100/ENQLM=200/TQELM=100/FILLM=100/PRCLM=10/JTQUOTA=512000 -
     /WSDEF=512000/WSQUO=512000/WSEXTENT=1024000/PRIV=(NETMBX,TMPMBX) -
     /DEFPRIV=(NETMBX,TMPMBX)/DEVICE=DKA100:/DIR=[ORA_NOBODY] -
     /FLAGS=(DISCTLY,DISMAIL)

Note:

The values are just examples.
UAF> EXIT
$ CREATE/DIR/LOG/OWNER=ORA_NOBODY DKA100:[ORA_NOBODY]      !  Example

1.6 Using Trace and Alert Files

This section describes the trace (or dump) and alert files that Oracle Database 10g creates to diagnose and resolve operating problems, and includes:

1.6.1 Trace Files

Each server and background process can write to an associated trace file. When a process detects an internal error, it writes information about the error to its trace file. The file name format of a trace file is:

nodename_sid_processtype_processname_number.trc, 

In this file name format:

  • nodename is the name of the system where an instance is running

  • sid is the instance system identifier

  • processtype is FG (foreground) or (background)

  • processname is the process that generates it

  • number is a unique 3-digit numeric identifier

A sample trace file name is:

ORA_ROOT:[ADMIN.dbname.BDUMP]NODEA_PROD_BG_PMON_002.TRC

All trace files for background processes are written to the destination directory specified by the BACKGROUND_DUMP_DEST initialization parameter. All trace files for server processes are written to the destination directory specified by the USER_DUMP_DEST initialization parameter.

Set the MAX_DUMP_FILE_SIZE initialization parameter to unlimited or 100m to ensure that the trace file is large enough to store error information.

1.6.2 Alert Files

The nodename_sid_alert.log file stores significant database events and messages. Anything that affects the database instance or global database is recorded in this file. This file is associated with a database and is located in the directory specified by the BACKGROUND_DUMP_DEST initialization parameter.

1.7 PL/SQL Gateway

The mod_plsql module is a PL/SQL gateway running within an Apache module in the middle tier server. It runs PL/SQL procedures in a back-end Oracle Database using OCI. The mod_plsql module currently supports only stateless PL/SQL Web applications.

See Also:

Oracle Application Server mod_plsql User's Guide for information about developing Web applications using PL/SQL

The PL/SQL Gateway is installed as part of the Apache Install on OpenVMS.

See Also:

The README_MODPLSQL.TXT file in the Apache directory

1.8 Oracle HTTP Server Powered by Apache

Refer to Oracle Database Installation Guide (B13681-01) for information about installing and configuring Oracle HTTP Server powered by Apache.

1.9 Granting Access to Oracle Database 10g Users

To grant users access to Oracle Database 10g:

Note:

The ORAUSER.COM script must be located in the top-level Oracle home directory. Do not move this script. The definitions of Oracle logicals are created from the top-level Oracle home directory. If you run the ORAUSER.COM script from any other location, then Oracle Database 10g will not work correctly.
  1. You can define a symbol in the system-wide login procedure (typically, SYLOGIN.COM) that runs a particular ORAUSER.COM file. This method may be more useful if users access multiple instances and, therefore, need to run a database-specific ORAUSER file with the proper parameters. For example:

    $ go_prod:== @DISK$DISK1:[ORACLE10g]ORAUSER PROD
    
  2. Ensure that each user's OpenVMS account meets at least the minimum requirements for ASTLM, BYTLM, ENQLM, WSDEFAULT, WSEXTENT, WSQUOTA, and PGFLQUO.

    See Also:

    Oracle Database Installation Guide for more information about account quotas
  3. Create the Oracle Database 10g user accounts with the CREATE USER and ALTER USER commands. Use the GRANT command to grant the required database privileges or roles as documented in Oracle Database Administrator's Guide.

  4. To enable users to use the SQL*PLUS utility to start up or shut down an Oracle Database 10g instance, use the OpenVMS AUTHORIZE utility to add an ORA_sid_DBA or ORA_DBA process rights identifier to the user's OpenVMS account from the OpenVMS rights database.