Exchange 2007 supports Local Continuous Replication (LCR) and Cluster Continuous Replication (CCR). When either of these are enabled, the Microsoft Exchange Writer Replica instance of the VSS Writer is enabled.
There are limitations on VCS support for snapshots of the passive (replicated) copy of the storage group. The Quick Recovery Wizard and the VSS Scheduler Wizard do not support snapshots for the passive copy. Therefore, you cannot schedule snapshots of the passive copy.
However, VCS does support creating snapshots of the LCR or CCR passive storage group with the VSS Exchange Snapshot Wizard or with the vxsnap CLI command. Creating a snapshot of a passive storage group requires using the Microsoft Exchange Writer Replica instance to prepare the snapshot mirror and to create the snapshot. If replication is enabled:
The Microsoft Exchange Writer Replica instance is displayed in the tree view of the VEA under the Applications node. Right-clicking the Exchange Replica node displays a context menu that shows VSS Snapshot, VSS Snapback, and VSS Refresh selections.
Using the vxsnap CLI command, you can specify Microsoft Exchange Writer Replica (instead of Microsoft Exchange Writer) for the writer option in the required commands.
Although you cannot restore a snapshot to the passive copy, you can use the passive copy snapshot to restore a storage group or database after first manually performing a "Restore-StorageGroupCopy" on the storage group. However, note the following considerations:
The LCR copy is automatically disabled when running the Restore-StorageGroupCopy cmdlet.
If a snapshot schedule exists for the active storage group, running the Restore-StorageGroupCopy cmdlet makes the schedule invalid, and you must delete the invalid schedule before running the restore operation.
The cmdlet changes which volumes are the active copy. The previous passive volumes are now active, and vice versa. The scheduled snapshots point the active writer instance, Microsoft Exchange Writer, toward what are now passive volumes, so the snapshot operations fail. The schedule must be deleted so that the active writer can be pointed toward the new active volumes in order to do the restore operation.