Skip Navigation Links | |
Exit Print View | |
Oracle Solaris Cluster Data Services Planning and Administration Guide Oracle Solaris Cluster 4.1 |
1. Planning for Oracle Solaris Cluster Data Services
Configuration Guidelines for Oracle Solaris Cluster Data Services
Identifying Data Service Special Requirements
Determining the Location of the Application Binaries
Verifying the nsswitch.conf File Contents
Planning the Cluster File System Configuration
Enabling Oracle Solaris SMF Services to Run Under the Control of Oracle Solaris Cluster
Relationship Between Resource Groups and Device Groups
Determining Whether Your Data Service Requires HAStoragePlus
Data Services With Nodes That Are Not Directly Connected to Storage
Considerations for Installing and Configuring a Data Service
Overview of the Installation and Configuration Process
Installation and Configuration Task Flow
Example of Configuring a Failover Data Service
Tools for Data Service Resource Administration
Oracle Solaris Cluster Maintenance Commands
Summary by Task of Tools for Administering Data Service Resources
Rules for Names Except Resource Type Names
The HAStoragePlus resource type can be used to configure the following options.
Coordinate the boot order of disk devices and resource groups. Other resources in the resource group that contains the HAStoragePlus resource are brought online only after the disk device resources become available.
With AffinityOn set to True, enforce collocation of resource groups and device groups on the same node. This enforced collocation enhances the performance of disk-intensive data services.
Monitor entities managed by the HASP resource, including global devices, file systems, and ZFS storage pools.
In addition, HAStoragePlus is capable of mounting local and global file systems. For more information, see Planning the Cluster File System Configuration.
Note - If the device group is switched to another node while the HAStoragePlus resource is online, AffinityOn has no effect. The resource group does not migrate with the device group. However, if the resource group is switched to another node, the setting of AffinityOn to True causes the device group to follow the resource group to the new node.
See Synchronizing the Startups Between Resource Groups and Device Groups for information about the relationship between device groups and resource groups.
See Enabling Highly Available Local File Systems for procedures for mounting of file systems such as Solaris ZFS in a local mode. The SUNW.HAStoragePlus(5) man page provides additional details.
The following types of data services require HAStoragePlus:
Data services with nodes that are not directly connected to storage
Data services that are disk intensive
Some nodes in the node list of a data service's resource group might not be directly connected to the storage. In this situation, you must coordinate the boot order between the storage and the data service. To meet this requirement, configure the resource group as follows:
Configure HAStoragePlus resources in the resource group.
Set the dependency of the other data service resources to the HAStoragePlus resource.
Some data services, such as Oracle Solaris Cluster HA for Oracle and Oracle Solaris Cluster HA for NFS are disk intensive. If your data service is disk intensive, ensure that the resource groups and device groups are colocated on the same node. To meet this requirement, perform the following tasks:
Add an HAStoragePlus resource to your data service resource group
Switch the HAStoragePlus resource online
Set the dependency of your data service resources to the HAStoragePlus resource
Set AffinityOn to True
Note - The failback settings must be identical for both the resource group and device groups.
Some data services are not disk intensive. For example, HA for DNS (which reads all of its files at startup) is not disk intensive. If your data service is not disk intensive, configuring the HAStoragePlus resource type is optional.