Go to main content

Administering the Disaster Recovery Framework for Oracle® Solaris Cluster 4.4

Exit Print View

Updated: June 2019
 
 

Overview of Oracle Solaris Cluster Administration Concepts

You must be an experienced Oracle Solaris Cluster administrator to administer the disaster recovery framework.

This section describes the following Oracle Solaris Cluster administration topics that you need to understand before you administer the disaster recovery framework:

Configuring Resources and Resource Groups

You use either Oracle Solaris Cluster commands or the Oracle Solaris Cluster Manager browser interface to create failover and scalable resource groups.

For more information about administering resources and resource groups in Oracle Solaris Cluster software, see the Planning and Administering Data Services for Oracle Solaris Cluster 4.4.

Configuring Logical Hostnames

The logical hostname is a special high-availability (HA) resource. The geoadm start command configures the logical hostname that corresponds to the cluster name. The IP address and host maps for the logical hostname must be set up before you run this command. Before assigning hostnames, familiarize yourself with the legal names and values that are described in Legal Names and Values of Disaster Recovery Framework Entities.

For more information about using the geoadm start command, see Enabling the Disaster Recovery Framework.

The following table lists the Oracle Solaris Cluster and disaster recovery framework components that require IP addresses. Add these IP addresses to the following locations:

  • All naming services that are being used

  • The local /etc/inet/hosts file on each cluster node, after you install the Oracle Solaris OS software

Table 2  IP Addresses Required by the Disaster Recovery Framework
Component
Number of IP Addresses Needed
Oracle Solaris Cluster administrative console
1 per subnet
IP Network Multipathing groups
  • Single-adapter groups – 1 primary IP address.

  • Multiple-adapter groups – 1 primary IP address plus 1 test IP address for each adapter in the group.

Cluster nodes
1 per node, per subnet
Domain console network interface
1 per domain
Console-access device
1
Logical addresses
1 per logical host resource, per subnet
Disaster recovery framework hostnames
1 logical IP address per cluster infrastructure. For example, if you have two clusters in your disaster recovery framework configuration, you need two IP addresses.

For more information about configuring the IP address and host maps during the installation of Oracle Solaris Cluster software, refer to Chapter 2, Installing Software on Global-Cluster Nodes in Installing and Configuring an Oracle Solaris Cluster 4.4 Environment.

Managing Device Groups

A device group is a hardware resource that is managed by the Oracle Solaris Cluster software. A device group is a type of global device that is used by the Oracle Solaris Cluster software to register device resources, such as disks. A device group can include the device resources of disks and Solaris Volume Manager disk sets.

For information about configuring device groups in Oracle Solaris Cluster software, refer to Chapter 5, Administering Global Devices, Disk-Path Monitoring, and Cluster File Systems in Administering an Oracle Solaris Cluster 4.4 Configuration.

The disaster recovery framework configures Oracle Solaris Cluster device groups to include replication.

For more information about configuring data replication in thedisaster recovery framework, see the Oracle Solaris Cluster documentation for the data replication product you use.

Disaster Recovery Framework Configuration and Administration Tasks

The following table lists administration tasks for your disaster recovery framework configuration.


Note -  For procedures to create partnerships, heartbeats, protection groups, sites, and multigroups, see Installing and Configuring the Disaster Recovery Framework for Oracle Solaris Cluster 4.4 and the Oracle Solaris Cluster guide for your data replication product.
Table 3  Disaster Recovery Framework Administration Tasks
Task
Description and Documentation
Switch over services to the partner cluster.
You can switch over individual protection groups or multigroups.
See the procedures to switch over a protection group from primary to secondary in the Oracle Solaris Cluster guide for your data replication product. You can also use the Oracle Solaris Cluster Manager browser interface.
To switch over multiple protections groups that are configured in a multigroup, see How to Switch Over a Multigroup. You can also use the Oracle Solaris Cluster Manager browser interface.
Take over services from primary to secondary during a disaster.
You can take over services for an individual cluster or for a multigroup.
See the procedures to force immediate takeover of services by a secondary cluster in the Oracle Solaris Cluster guide for your data replication product. You can also use the Oracle Solaris Cluster Manager browser interface.
To take over multiple protection groups that are configured in a multigroup, see How to Take Over a Multigroup. You can also use the Oracle Solaris Cluster Manager browser interface.
Recover from a takeover.
  • Perform data recovery and error repair outside of the disaster recovery framework. See the Oracle Solaris Cluster documentation for the data replication product you are using.

  • Resynchronize the partner clusters. See Resynchronizing a Protection Group. You can also use the Oracle Solaris Cluster Manager browser interface.

Start an individual protection group or a multigroup.
To start an individual protection group, see How to Activate a Protection Group or use the Oracle Solaris Cluster Manager browser interface.
To start a multigroup, see How to Start All Protection Groups in a Multigroup or use the Oracle Solaris Cluster Manager browser interface.
Stop an individual protection group or a multigroup.
To stop an individual protection group, see How to Deactivate a Protection Group. You can also use the Oracle Solaris Cluster Manager browser interface.
To stop a multigroup, see How to Stop All Protection Groups in a Multigroup. You can also use the Oracle Solaris Cluster Manager browser interface.
Delete a multigroup
See Deleting a Multigroup. You can also use the Oracle Solaris Cluster Manager browser interface.
Delete a site.
See Deleting a Site. You can also use the Oracle Solaris Cluster Manager browser interface.
Delete a protection group or a data replication component.
See Deleting Protection Groups and Data Replication Components. You can also use the Oracle Solaris Cluster Manager browser interface.
Delete a partnership.
See Leaving or Deleting a Partnership. You can also use the Oracle Solaris Cluster Manager browser interface.
Disable the disaster recovery framework.
Uninstall the disaster recovery framework software.