Oracle® Beehive Installation Guide Release 1 (1.2 and 1.3) for Microsoft Windows Part Number E10478-03 |
|
|
View PDF |
Beta Draft
This section describes Oracle Beehive database requirements, creating databases, and configuring database settings. It covers the following topics:
Note:
The installation scenario "Installing Oracle Beehive for Demilitarized Zone (DMZ)" described in "What You Should Know Before Installing Oracle Beehive" does not require a database. Refer to this section for more information about this scenario.Oracle Beehive requires an existing database with the following characteristics:
Uses Oracle Database 10g Release 2 (10.2.0.3) Enterprise Edition Patch Set 2 installed on any supported operating system.
Ensure that you have applied all the patches listed in the section "Patches".
Uses character set AL32UTF8 (Unicode)
Contains the following standard database components:
Oracle XML DB
Oracle interMedia
Oracle Text
Partitioning
Contains Rules Manager and Expression Filter (Rules Manager and Expression Filter are installed automatically with Oracle Database 10g Enterprise Edition.)
For Microsoft Windows (32-bit and 64-bit versions), install the latest patch set for Oracle Database 10g Release 2 (10.2.0.3) Patch Set 2.
For all other operating systems, install the following patches for Oracle Database 10g Release 2 (10.2.0.3) Patch Set 2:
Install the following patches if your database is installed on a 64-bit system:
5240469 GENOCCISH GENERATES LOTS OF ERRORS AND FAILS ON AMD64. This patch must be installed before any 64-bit patch. In addition, this patch is required only for Linux.
5252061 ORA-00600: INTERNAL ERROR CODE, ARGUMENTS: [17456], [0X69FCD9BC], [], [], [], []
4898608: OPATCH 10.2 ARU PLACEHOLDER
Merge label patches:
6124210: MERGE LABEL REQUEST ON TOP OF 10.2.0.3 FOR BUGS 5924208 5705795
6485648: MERGE LABEL REQUEST ON TOP OF 10.2.0.3 FOR BUGS 5196175 AND 5472702
6671491: MERGE LABEL REQUEST ON TOP OF 10.2.0.3 FOR BUGS 4033868 4554026 6121260
6800085: MERGE LABEL REQUEST ON TOP OF 10.2.0.3 FOR BUGS 5060402 6369463
6449407: MERGE LABEL REQUEST ON TOP OF 10.2.0.3 FOR BUGS 6349820 6121258
7031761: MERGE LABEL REQUEST ON TOP OF 10.2.0.3 FOR BUGS 6650096 6121248 4684070 5190392
7280499: MERGE LABEL REQUEST ON TOP OF 10.2.0.3 FOR BUGS 6725634 6667542
5837795: CONNECT BY PUMP JOINING TABLE SCAN CAUSES ORA-600
5890966: INTERMITTENT ORA-06502 DURING PEAK LOADING
6150438: ORA-00932: COLLECT() DOESN'T WORK CORRECTLY FOR ADTS RESIDING IN DIFF. SCHEMAS
4335601: VALGRIND ERROR OF LXSCNTBY.C
6376915: ENQ HW - CONTENTION WITH LOB SEGMENTS
5524408: WRONG RESULT WHEN IN LIST HAS MORE THAN 256 BIND VARIABLES
7150622: MLR BUG FOR 10.2.0.3 FOR CPUJUL2008
6081550: TRACKING BUG FOR INVENTORY OF DG LOGICAL/LOGMINER BUGS ON TOP OF 10.2.0.3
6653934: XF11.1SSHR - TRC - KDRWRIC
5485914:MUTEX REPORTED SELF DEADLOCK AFTER DBMS_MONITOR.SESSION_TRACE_ENABLE. This patch is required only for Solaris operating system.
The following patches are recommended for Oracle Database 10g Release 2 (10.2.0.3) Patch Set 2:
5907779: "CURSOR: PIN S WAIT ON X" RUNNING DBMS_STATS.GATHER_DATABASE_STATS_JOB_PROC
5140631: SQL_ADDRESS AND SQL_HASH_VALUE NOT SET
5165885: STAR NT RAC ORA-00600 [KCLCLS_8]
6358116: MERGE LABEL REQUEST ON TOP OF 10.2.0.3 FOR BUGS 5983683 6121250
6897966: JOBS AREN'T STARTED: SERVICE NAME IS CASE SENSITIVE
The database requires some initialization parameters to have a minimum value, as specified in the following table:
Table 2-1 Minimum Values of Oracle Database Initialization Parameters
Property | Value |
---|---|
compatible |
10.2.0.0.0 Note: Do not modify this parameter if it is already set to a higher value. |
db_block_size |
Either this parameter must be set to 8192, or one of the following parameters have to be set:
The db_cache_size parameter specifies in bytes of the cache of standard block size buffers. Oracle Beehive will use this value for the block size of its tablespaces. If any of the db_nk_cachce_size parameters are set, then Oracle Beehive will use the lowest value that is equal or greater than 8192. Refer to "Specifying Database Block Sizes" in "Oracle Database Administrator's Guide" for more information. |
java_pool_size |
50M (52428800) |
job_queue_processes |
10 |
processes |
150 |
undo_retention |
3600 |
Note:
You may have to increase the values of processes and sessions if you are using Oracle RAC.To see all parameters (in alphabetical order) along with their current values, use the following SQL*Plus command:
SQL> SHOW PARAMETERS
You may display parameters that contain a specified string. For example, to display all parameters having COMPATIBLE
in their names, use the following command:
SQL> SHOW PARAMETERS COMPATIBLE
Set initialization parameters with the ALTER SYSTEM
command. For example, to set the COMPATIBLE
parameter with the value 10.2.0.3.0
, use the following command:
SQL> ALTER SYSTEM SET COMPATIBLE = '10.2.0.3.0' SCOPE = SPFILE;
Note:
After setting initialization parameters, restart the database.The archive log mode of the database must be ARCHIVELOG
, which enables the archiving of the redo logs. Oracle Beehive uses Change Data Capture (CDC), which requires that the archive log mode be turned on. Use the following command to determine the database's archive log mode:
SQL> SELECT LOG_MODE FROM SYS.V$DATABASE;
To set the archive log mode to ARCHIVELOG
, use the following commands:
SQL> shutdown immediate SQL> startup mount exclusive SQL> alter database archivelog; SQL> alter database open;
You may use Database Configuration Assistant (DBCA) to create and configure a database. For more information about this tool, refer to "Using DBCA to Create and Configure a Database" in Chapter 2, "Installing Oracle and Building the Database" in Oracle Database 2 Day DBA.
Tip:
When creating a database for Oracle Beehive with DCBA, you may choose any template. With DCBA, you may also turn on the archive log mode (by selecting the option Enable Archiving), set the character set to AL32UTF8, and set initialization parameters listed in "Initialization Parameters".Refer to "Configuring and Installing Oracle Beehive for Oracle RAC" for pre-installation and post-installation steps required for Oracle Beehive deployments that use Oracle Real Application Clusters (RAC) databases.
You may customize the layout of Oracle Beehive tablespaces with the script <Oracle home>
\beehive\db\framework\beehive_custom_ts.sql
. You may change tablespace names, initial size of the datafile, block size, and datafile autoextend options.
By default, Oracle Beehive uses eleven tablespaces. The beehive_custom_ts.sql
script maps each tablespace to a placeholder. The Oracle Beehive Install Wizard uses these placeholders (instead of the real tablespace name) for configuration.
For example, in the beehive_custom_ts.sql
script, the tablespace BEE_DATA
is mapped to the placeholder ts_data
:
REM Tables containing transactional data DEFINE ts_data = BEE_DATA DEFINE ts_data_path = '' DEFINE ts_data_sz = 512M DEFINE ts_data_bsz = 8192 DEFINE ts_data_ae = ON DEFINE ts_data_ae_nx = 128M
You must specify the mapping between the placeholders, such as ts_data
and ts_index
, and the corresponding tablespaces. If one tablespace is to be shared across multiple placeholders, adjust the tablespace size accordingly.
This script creates a special view named bee_tablespaces
, which then is used by the install scripts. You may drop the view after a successful installation. Since the tablespace mapping information can be reused for subsequent product reinstallations, the installation script does not drop it.
The script also produces an output file with the actual CREATE TABLESPACE commands. The output file name is create_beehive_tablespaces.sql
After making your changes to the script, run it with SYS privileges.
If you want to use a database that uses raw storage, customize the script <Oracle home>
/beehive/db/framework/beehive_custom_ts.sql
. Refer to "Customizing Oracle Beehive Tablespaces Layout" for more information about this script.
To use a raw device (which you have created and properly configured) for a particular tablespace, specify the full path of its datafile in the appropriate variable. For example, if you want to use a raw device for the BEE_DATA
tablespace, specify the full path of its datafile in the ts_data_path
variable.
By default, Oracle Beehive uses eleven tablespaces. The beehive_custom_ts.sql
script maps each tablespace to a placeholder. The Oracle Beehive Install Wizard uses these placeholders (instead of the real tablespace name) for configuration.
To use a raw device (which you have created and properly configured) for a particular tablespace, specify the full path of its datafile in the appropriate variable. For example, if you want to use a raw device for the BEE_DATA
tablespace, specify the full path of its datafile in the ts_data_path
variable.
In addition, set the variable APPEND_FILE_NAME
to N
.
After making your changes to the script, run it with SYS privileges.