Oracle® Fusion Applications Enterprise Deployment Guide 11g Release 1 (11.1.1.5) Part Number E16684-02 |
|
|
View PDF |
This chapter provides information on how the database tier is implemented in an enterprise deployment topology.
This chapter includes the following topics:
Section 3.1, "Understanding the Database in the Enterprise Deployment Topology"
Section 3.4, "Loading the Oracle Fusion Applications Repository into the Oracle RAC Database"
The Oracle Fusion Applications reference enterprise deployment topology uses a single database for the following components:
Oracle Fusion Applications metadata
Oracle Fusion Applications transactional data
Oracle Transactional Business Intelligence data
Technology stack data, such as Oracle SOA Suite, Oracle Enterprise Manager Fusion Middleware Control, Oracle WebCenter, and Oracle Essbase.
Oracle Secure Enterprise Search data
Implementing Oracle Business Intelligence Data Warehouse requires a separate database for following components:
Data Warehouse Administration Console (DAC)
Informatica
a Data Warehouse
For the enterprise topology, Oracle Real Application Clusters (Oracle RAC) databases are highly recommended. You must set up these databases before you can install and configure the Oracle Fusion Applications components. You install the Oracle Fusion Applications and Oracle Fusion Middleware metadata repositories into existing databases using the Fusion Applications Repository Creation Utility (Fusion Applications RCU).
Before loading the metadata repository into your database, check that the database meets the requirements described in these sections:
Note:
When creating the database, ensure that the length of the Oracle System ID (SID) does not exceed eight (8) characters.For example:
SID: abcd12345 is invalid
SID: abcd123 is valid
Note the following requirements for the hosts FUSIONDBHOST1
and FUSIONDBHOST2
in the data tier:
Oracle Clusterware
For Oracle Database 11g Release 2 (11.2.0.2) for Linux, refer to the Oracle Database Installation Guide.
Oracle Real Application Clusters
For Oracle RAC 11g Release 2 (11.2.0.2) for Linux or Oracle Database 10g Release 2 (10.2) for Linux, refer to the Oracle Database Installation Guide.
Oracle Automatic Storage Management (optional)
Oracle ASM gets installed for the node as a whole. It is recommended that you install it in a separate Oracle Home from the Database Oracle Home. This option comes in at runInstaller. In the Select Configuration page, select the Configure Automatic Storage Management option to create a separate ASM home.
Oracle Fusion Applications requires the presence of a supported database and schemas. To check if your database is certified or to see all certified databases, refer to the Oracle Fusion Applications system requirements and supported platforms documentation.
To check the release of your database, you can query the PRODUCT_COMPONENT_VERSION
view as follows:
SQL> SELECT VERSION FROM SYS.PRODUCT_COMPONENT_VERSION WHERE PRODUCT LIKE 'Oracle%';
Notes:
The database you use as the Oracle Fusion Applications supporting database must support the AL32UTF8 character set.
When installing the database, please ensure that Oracle Label Security is enabled, as shown in Figure 3-1. In the case of an Oracle RAC installation, the Oracle Label Security should be enabled on all the nodes.
You enable label security in the Select Database Edition screen, shown in Figure 3-1, using Oracle Universal Installer.
Table 3-1 shows the recommended minimum init.ora parameters for an Oracle database. The database shipped with the Oracle Fusion Applications software contains this configuration. When you run the Fusion Applications RCU, its prerequisite check feature checks to see that the database meets these minimum requirements.
Table 3-1 Minimum Requirements for Database Configuration
INST_ID Parameter | Value |
---|---|
_b_tree_bitmap_plans |
FALSE |
audit_trail |
NONE |
compatible |
11.2.0 |
db_files |
1024 |
db_recovery_file_dest_size |
2147483648 |
db_writer_processes |
1 |
disk_asynch_io |
FALSE |
fast_start_mttr_target |
3600 |
filesystemio_options |
Setall |
job_queue_processes |
10 |
log_buffer |
10485760 |
log_checkpoints_to_alert |
TRUE |
max_dump_file_size |
10M |
memory_target |
unset or N/A |
nls_sort |
BINARY |
open_cursors |
500 |
pga_aggregate_target |
>= 4294967296 (4GB) |
plsql_code_type |
NATIVE |
processes |
5000 |
session_cached_cursors |
500 |
sga_target |
>=9663676416 (9GB) |
trace_enabled |
FALSE |
undo_management |
AUTO |
For example, to use the SHOW PARAMETER
command using SQL*Plus to check the value of the initialization parameter:
As the SYS user, issue the SHOW PARAMETER
command as follows:
SQL> SHOW PARAMETER processes
Set the initialization parameter using the following commands:
SQL> ALTER SYSTEM SET processes=5000 SCOPE=SPFILE; SQL> ALTER SYSTEM SET open_cursors=500 SCOPE=SPFILE;
Restart the database.
Note:
The method that you use to change a parameter's value depends on whether the parameter is static or dynamic, and on whether your database uses a parameter file or a server parameter file. See the Oracle Database Administrator's Guide for details on parameter files, server parameter files, and how to change parameter values.Oracle recommends using the Oracle Enterprise Manager Fusion Middleware Control Cluster Managed Services screen or SQL*Plus to create database services that client applications will use to connect to the database.
To configure this using SQL*Plus:
Use the CREATE_SERVICE
subprogram to create the database service.
Log in to SQL*Plus as the sysdba user and run the following command:
SQL> EXECUTE DBMS_SERVICE.CREATE_SERVICE (SERVICE_NAME => 'crm.mycompany.com', NETWORK_NAME => 'crm.mycompany.com' );
Add the service to the database and assign it to the instances using srvctl
:
prompt> srvctl add service -d fusiondb -s crm.mycompany.com -r fusiondb1 fusiondb2
Start the service using srvctl
:
prompt> srvctl start service -d fusiondb -s crm.mycompany.com
Verify that the crm
service is running on instance(s) fusiondb1
and fusiondb2
:
prompt> srvctl status service -d fusiondb
Note:
For more information about thesrvctl
command, see the Oracle Real Application Clusters Administration and Deployment Guide.Oracle recommends that a specific database service be used for a product suite even when they share the same database. It is also recommended that the database service used is different than the default database service. For example, for Oracle Fusion Customer Relationship Management the database would be crmdb
.mycompany.com
and the default service is one with the same name. The Oracle Fusion Customer Relationship Management install is configured to use the service crm
.mycompany.com
.
This section describes how to update the kernel parameters for Linux before the database is installed.
To update the parameters:
Log in as root and add or edit the following values in the /etc/sysctl.conf
file:
fs.file-max = 6815744 kernel.shmall = 2097152 kernel.shmmax = 2147483648 kernel.shmmni = 4096 kernel.sem = 250 32000 100 128 net.core.rmem_default = 4194304 net.core.rmem_max = 4194304 net.core.wmem_default = 262144 net.core.wmem_max = 1048576 net.ipv4.ip_forward = 0 net.ipv4.conf.default.rp_filter = 1 tcp.ipv4.tcp_wmem = 262144 262144 262144 tcp.ipv4.tcp_rmem = 4194304 4194304 4194304 fs.aio-max-nr = 1048576 net.ipv4.ip_local_port_range = 9000 65000
Execute the following command to activate the changes:
/sbin/sysctl -p
Before loading the Oracle Fusion Applications repository into a database, you must apply database patch 10220058 in order to run the Oracle Fusion Applications Repository Creation Utility (Fusion Applications RCU) with Oracle Database 11g Enterprise Edition Release 11.2.0.2.0. To find the patch, go to My Oracle Support (https://support.oracle.com
) and click the Patches & Updates tab.
The Fusion Applications RCU components are included in the zipped Fusion Applications RCU file delivered in the provisioning framework. Unzip the file to the RCU_HOME
location on the FUSIONDBHOST1
machine. For example, ORACLE_BASE
/rcu
.
Once you have the Fusion Applications RCU installed, do the following:
Copy all the required dump files locally on FUSIONDBHOST1
(for example, to /tmp
):
FUSIONDBHOST1> cd RCU_HOME/rcu/integration/fusionapps FUSIONDBHOST1> cp export_fusionapps_dbinstall.zip /tmp FUSIONDBHOST1> cd RCU_HOME/rcu/integration/biapps/schema FUSIONDBHOST1> cp otbi.dmp /tmp FUSIONDBHOST1> cd /tmp FUSIONDBHOST1> unzip export_fusionapps_dbinstall.zip
Note:
When running Fusion Applications RCU for Oracle RAC, you must copy theexport_fusionapps_dbinstall.zip
file as well as otbi.dmp
to all the nodes of the Oracle RAC.Create the incident_logs
directory on FUSIONDBHOST1
:
FUSIONDBHOST1> cd ORACLE_HOME FUSIONDBHOST1> mkdir incident_logs
Start Fusion Applications RCU from the /bin
directory in the Fusion Applications RCU home directory:
cd RCU_HOME/bin
./rcu
In the Welcome screen (if displayed), click Next.
In the Create Repository screen, shown in Figure 3-2, select Create to load component schemas into a database. Click Next.
In the Database Connection Details screen, shown in Figure 3-3, enter connect information for your database:
Database Type: Select Oracle Database
Host Name: Specify the name of the node on which the database resides. For the Oracle RAC database, specify the VIP name or one of the node names as the host name: FUSIONDBHOST1
-VIP
Port: Specify the listen port number for the database; for example 1521
Service Name: Specify the service name of the database (crm
.mycompany.com
)
Username: Specify the name of the user with DBA or SYSDBA privileges: SYS
Password: Enter the password for the SYS user
Role: Select the database user's role from the list: SYSDBA
(required by the SYS user)
Click Next.
The Repository Creation Utility selects the required components automatically, as shown in Figure 3-4.
Click Next. The Repository Creation Utility checks the prerequisites, as shown in Figure 3-5.
Click OK.
In the Schema Passwords screen, shown in Figure 3-6, enter passwords for the main and additional (auxiliary) schema users, and click Next.
Note:
For increased security, do the following:Specify different schema passwords for all schemas
Ensure that all paswords are more than eight (8) characters in length.
In the Custom Variables screen, shown in Figure 3-7, enter the required values:
DUMP FILE LOCATION - /tmp
INCIDENT LOG LOCATION - ORACLE_HOME
/incident_logs
Note:
When specifying an Oracle Business Intelligence Enterprise Edition (OBIEE) backup directory, set it to be a shared directory with read/write permissions for both DAC database hosts.In the Map Tablespaces screen, choose the tablespaces for the selected components, and click Next.
In the Summary screen, click Create.
In the Completion Summary screen, click Close.
Note:
If you encounter any issues while using the Repository Creation Utility, check the logs atRCU_HOME
/rcu/log
.Note:
Oracle recommends using the database used for identity management to store the Oracle WSM policies. It is therefore expected to use the IM database information for the OWSM MDS schemas, which will be different from the one used for the rest of SOA schemas. To create the required schemas in the database, repeat the steps above using the IM database information, but select only "AS Common Schemas: Metadata Services" in the Select Components screen (Step 7).After you have loaded the metadata repository in your database, you should make a backup.
Backing up the database is for the explicit purpose of quick recovery from any issue that may occur in the further steps. You can choose to use your backup strategy for the database for this purpose or simply make a backup using operating system tools or RMAN for this purpose. It is recommended that you use Oracle Recovery Manager for the database, particularly if the database was created using Oracle ASM. If possible, a cold backup using operating system tools such as tar can also be performed.