HP 3PAR Recovery Manager Software - Manual (Page 79)

Recovering from Logical Errors with Point-in-Time Recovery
Point-in-time recovery is useful for recovering from logical errors. For example, if you accidentally
altered an entry or deleted a table, you can use a virtual copy to restore your instance or database
back to a consistent, known prior state. Point-in-time recovery restores the instance or database to
the state it was in at the time the virtual copy was created.
To perform a point-in-time recovery, use Recovery Manager to perform a file copy restore or a
volume restore. The restore steps depend on whether SQL servers are in clustered environment or
non-clustered environment.
1.
Schedule downtime before performing the restore.
2. If you are performing a restore on a standalone SQL production server in a non-clustered
environment:
•
If you are performing a file copy or volume restore on an instance: Stop the SQL Server
service for the instance, perform a standard file copy restore or volume restore, and start
the SQL Server service.
•
If you are performing a volume restore on a database: Stop the SQL Server service,
perform a standard volume restore, and start the SQL Server service.
•
If you are performing a file copy restore on a database: Perform a standard file copy
restore. (Do NOT stop and start the SQL Server service.)
For more information about performing a standard restore using the Recovery Manager
GUI, see
“Using the Recovery Manager GUI” (page 34) . For more information about
performing a standard restore using the CLI HpRmCli Sql restore command, see
“Using the Recovery Manager CLI” (page 54) .
If you are performing the restore in Failover Cluster environment, you must follow the instructions
provided in one of the following sections, depending on the type of restore you are performing:
•
“Restoring an SQL Instance in an Failover Cluster Environment Using File Copy Restore”
(page 79)
•
“Restoring a Volume in Failover Cluster Environment Using Volume Restore” (page 79)
Restoring an SQL Instance in an Failover Cluster Environment Using File Copy Restore
To perform a file copy restore on an SQL instance in Failover Cluster environment:
1.
Schedule downtime. Recovery Manager automatically takes the SQL Server Agent and SQL
Instance Service offline.
2. Locate the instance-level virtual copy you want to use to restore the SQL instance.
3. Run the HpRmCli Sql restore command on this virtual copy. For more information on
Recovery Manager CLI commands, see
“Using the Recovery Manager CLI” (page 54) .
Restoring a Volume in Failover Cluster Environment Using Volume Restore
To perform a volume restore on an SQL server in Failover Cluster environment:
1.
Schedule downtime. Recovery Manager automatically takes the relevant SQL services offline.
2. Log onto your backup server and perform a volume restore. To perform a volume restore, use
either the HpRmCli Sql restore command or the Recovery Manager GUI.
For more information about performing a restore using the Recovery Manager GUI, see “Using
the Recovery Manager GUI” (page 34) . For more information about performing a restore
using the CLI HpRmCli Sql restore command, see “Using the Recovery Manager CLI”
(page 54) .
Recovering from Logical Errors with Point-in-Time Recovery 79