This procedure configures Availability Suite volumes in an Oracle Solaris Cluster environment. These volumes can be Solaris Volume Manager volumes or raw device volumes.
The volumes are encapsulated at the Oracle Solaris Cluster device-group level. The Availability Suite feature interacts with the Solaris Volume Manager disk sets or raw device through this device group interface. The path to the volumes depends on the volume type, as described in the following table.
|
For example, if you configure the volume by using a raw device, choose a raw device group, dsk/d3, on cluster-paris and cluster-newyork.
The Availability Suite feature requires a dedicated bitmap volume for each data volume to track which modifications to the data volume when the system is in logging mode.
If you use a raw device to configure the volumes, create two partitions, /dev/did/rdsk/d3s3 and /dev/did/rdsk/d3s4, on the /dev/did/rdsk/d3 device on cluster-paris.
If you use a raw device to configure the volumes, create two partitions, /dev/did/rdsk/d3s5 and /dev/did/rdsk/d3s6, on the /dev/did/rdsk/d3 device on cluster-paris.
You can optionally create two additional volumes on each cluster for each data volume for which a fallback snapshot will be created, as described in Availability Suite Volume Sets. The compact dependent shadow volume can normally be 10% of the size of the volume it will protect. The bitmap shadow volume is sized according to the rules described in the Availability Suite Point-in-Time Copy Software Administration Guide and the iiadm (1M) man page. The volumes used by the fallback snapshot must be in the same device group as the replicated volume they protect.