V11 SP8

Best Practices for Oracle

Use these guideline to effectively configure, backup and restore the Oracle database.

Running on UNIX Environments

If you run more than one database application on the same UNIX client, perform the following administration on the client before you install the Oracle Agent:

  • Create a new user group for the databases.
  • Add the users and owners of each database application to the user group you created.

Creating Subclients

  • Leave the default subclient contents intact to act as the catch-all entity for data. Synchronize the control file and recovery catalog, so that the control file contains the latest database metadata. See Synchronizing the Control File with the Recovery Catalog for Oracle.
  • Create separate subclients for data that needs frequently changes and needs a unique backup schedule.
  • Create separate subclients for data and log backups.
  • Use the above recommendations to distribute the data across subclients, which can help improve performance by distributing the workload.

Performing On-Demand Backups

  • Run data and log On-Demand backups in separate scripts.

Performing RMAN Command Line Operations

  • Verify the availability of the required resources before you run command line scripts to ensure the resources are online and not reserved by another backup or recovery process.
  • Use separate scripts for data and logs, because only one data type can be passed through the argument file. This system uses the data type to mark the archive files created by the backups as DATA or LOG in the CommServe database. If you have scripts containing both data and logs, the archive files are marked as either DATA or LOG, depending on the data type that is defined in the argument file.
  • The drive reservation retry mechanism works more efficiently when you use separate RMAN scripts for data and log backups.

Scheduling Log Backups

Use the archive log subclient that the Commvault software creates by default, to run archive logs backup. Take the following into consideration when plan your log backup schedule:

  • The number of transactions that the database generates
  • The disk size for the log file destination

After you have that information, configure an automatic schedule to back up the logs before the archive log destination fills up to 80% of the space (Backup Task Options dialog box, Schedule Pattern tab, Start backup when disk at log destination is box). For information on how to configure an automatic schedule, see Creating an Automatic Schedule for Database Log Backups.

Performing Restores

  • If you have exported the Recovery Catalog by using the exp command, you must import the recovery catalog into a new database by using the imp command.
  • When you restore tables, include the parent tables with all the dependent tables so that the reference constraints remain intact.

Performance Tuning

Oracle Parameters

Use the Oracle recommendations for the  following parameters.

This learn more about the values to use, see  the "Tuning RMAN Performance" section of the Oracle Database and Recover User's Guide on the Oracle documentation site.

Parameter Recommendation
LARGE_POOL_SIZE Set the LARGE_POOL_SIZE parameter by using the following formula:

LARGE_POOL_SIZE=number of allocated channels * (16 MB + ( 4 *  size of tape buffer))

If you do not set the LARGE_POOL_SIZE parameter, the Oracle software allocates the buffer from the shared pool.

TAPE_IO_SLAVES Set the TAPE_IO_SLAVES parameter to true so that RMAN uses the System Global Area (SGA) for the buffer allocation, which prevents resource contention.

Commvault Parameters

Use client side deduplication with compression on the client (ON).

Use 128K for the deduplication block size, but if the data rate changes, you can increase the size to 512K.

Obtaining the Instance Status for Browse Operations

In order to verify that the instance browse operation works with your SQL settings, Include the following contents in the ORACLE_HOME/sqlplus/admin/cvlogin.sql file.

set heading on;
set pause off;