You have configured Cluster Server (VCS) service groups for the database on each cluster. Each cluster requires an additional virtual IP address associated with the cluster for cross-cluster communication. The VCS installation and creation of the ClusterService group typically involves defining this IP address.
Configure a global cluster by setting:
Heartbeat
Wide area cluster (wac)
GCO IP (gcoip)
remote cluster resources
Table: Tasks for configuring a parallel global cluster
Task |
Description |
---|---|
Prepare to configure global parallel clusters |
Before you configure a global cluster, review the following requirements:
|
Configure a global cluster using the global clustering wizard. |
See “To modify the ClusterService group for global clusters using the global clustering wizard”. |
Define the remote global cluster and heartbeat objects |
|
Configure global service groups for database resources |
See “To configure global service groups for database resources”. |
Start replication between the sites. |
For software-based replication using Volume Replicator (VVR): See About configuring a parallel global cluster using Volume Replicator (VVR) for replication. For replication using Oracle Data Guard see the Data Guard documentation by Oracle. For replication using hardware-based replication see the replicated agent guide for your hardware. See the Cluster Server Bundled Agents Guide |
Test the HA/DR configuration before putting it into production |
The global clustering wizard completes the following tasks:
Validates the ability of the current configuration to support a global cluster environment.
Creates the components that enable the separate clusters, each of which contains a different set of GAB memberships, to connect and operate as a single unit.
Creates the ClusterService group, or updates an existing ClusterService group.
To modify the ClusterService group for global clusters using the global clustering wizard
# /opt/VRTSvcs/bin/gcoconfig
The wizard starts running commands to create or update the ClusterService group. Various messages indicate the status of these commands. After running these commands, the wizard brings the ClusterService failover group online on any one of the nodes in the cluster.
After configuring global clustering, add the remote cluster object to define the IP address of the cluster on the secondary site, and the heartbeat object to define the cluster-to-cluster heartbeat. Heartbeats monitor the health of remote clusters. VCS can communicate with the remote cluster only after you set up the heartbeat resource on both clusters.
To define the remote cluster and heartbeat
# haconf -makerw
In this example, the remote cluster is clus2 and its IP address is 10.11.10.102:
# haclus -add clus2 10.11.10.102
In this example, the primary cluster is clus1 and its IP address is 10.10.10.101:
# haclus -add clus1 10.10.10.101
# hahb -add Icmp
Define the following attributes for the heartbeat resource:
ClusterList lists the remote cluster.
Arguments enable you to define the virtual IP address for the remote cluster.
# hahb -modify Icmp ClusterList clus2 # hahb -modify Icmp Arguments 10.11.10.102 -clus clus2
# haconf -dump -makero
# hatype -modify IP OnlineRetryLimit 1
# hatype -modify IP OfflineWaitLimit 1
# hastatus -sum
# haclus -list clus1 clus2
Example of heartbeat additions to the main.cf file on the primary site:
. . remotecluster clus2 ( Cluster Address = "10.11.10.102" ) heartbeat Icmp ( ClusterList = { clus2 } Arguments @clus2 = { "10.11.10.102" } ) system sys1 ( ) . .
Example heartbeat additions to the main.cf file on the secondary site:
. . remotecluster clus1 ( Cluster Address = "10.10.10.101" ) heartbeat Icmp ( ClusterList = { clus1 } Arguments @clus1 = { "10.10.10.101" } ) system sys3 ( ) . .
See the Cluster Server Administrator's Guide for more details for configuring the required and optional attributes of the heartbeat object.
To configure global service groups for database resources
Configure VCS service groups at both sites.
Configure the replication agent at both sites.
For SF Oracle RAC, make the Oracle RAC service group a global service group, enabling failover across clusters.
For SF Sybase CE, make the database service group (sybasece) a global service group, enabling failover across clusters.
For example:
See Modifying the Cluster Server (VCS) configuration on the primary site.
For example:
See “To migrate the role of primary site to the remote site”.
See “To migrate the role of new primary site back to the original primary site”.
For more information about VCS replication agents:
See the Cluster Server Bundled Agents Guide
For complete details on using VVR in a shared disk environment:
See the Veritas InfoScale™ Replication Administrator's Guide.