You can use an existing SF Oracle RAC cluster or you can install a new SF Oracle RAC cluster for your primary site.
See About planning for the installation.
If you have an existing SF Oracle RAC cluster, and you want to set up a global cluster, skip the steps below and proceed to configure your secondary cluster.
If you do not have an existing cluster and you are setting up two new sites for an SF Oracle RAC global cluster, follow the steps below.
To set up the cluster and database at the primary site
If you are using hardware-based replication, install the sofware for managing your array.
Verify that you have the correct installation options enabled, whether you are using keyless licensing or installing keys manually. You must have the GCO option for a global cluster. If you are using VVR for replication, you must have it enabled.
Install and configure SF Oracle RAC. Prepare for your installation according to your configuration needs, including fencing.
See About preparing to install and configure SF Oracle RAC.
See Options for installing and configuring SF Oracle RAC.
After verifying a successful SF Oracle RAC installation:
Verify the CVM group is online on all nodes in the primary cluster:
# hagrp -state cvm
Prepare systems and storage for a global cluster. Identify the hardware and storage requirements before installing Oracle RAC Clusterware and RDBMS software.
Install and configure the Oracle RAC binaries:
See About preparing to install Oracle RAC.
See About installing Oracle RAC.
After successful Oracle RAC installation and configuration, verify that CRS daemons and resources are up on all nodes.
$ crs_stat -t
Modify the $ORACLE_HOME/dbs/ init_ora_db1.ora file by adding the following lines:
remote_listener= <SCAN_NAME>:1521 SPFILE='<SPFILE NAME PATH>'
Identify the disks that will be replicated, create the required CVM disk group, volume, and file system.
Create the database on the file system you created in the previous step.
Configure the VCS service groups for the database.