White Papers

46 TR1081 | Microsoft SQL Server Database Protection Using EqualLogic Auto-Snapshot Manager / Microsoft Edition
8.6 SQL Server Database Disaster Recovery with Smart Copy Replicas
The recommended method for restoring SQL Server Databases from Smart Copy replicas is to use the
Clone and Restore All as New option. This creates a clone of the Smart Copy replica, mounts the clone to
the host you choose, and restores the database online to the host. This process looks for the server name
in the backup document and times out. Add the new server name to the database server dialog box and
continue with the restore process.
Once the database is online, you can put it into production use and have it take over the production
database tasks. The Clone and Restore All as New option does not promote the entire replica set and
does not affect other replication processes. The Smart Copy replica clone looks like a regular volume in
the PS Series Group Manager GUI.
If the cloned replica was not used for production purposes you can clean up the replica smart copy using
ASM/ME. Use the Detach Database option within ASM/ME to detach the database from SQL Server and
log off using the same steps as earlier in the previous Data Mining section. You can also choose to delete
or detach the database from the SQL Server and disconnect the clone volume from the host using the
Microsoft iSCSI Initiator. Then, delete the clone from the PS Series Group Manager GUI.
If the clone replica is used for production purposes you may want to replicate the volumes back to the
primary site if/when it is available. To do this you will have to configure replication on the replica clone(s)
and replicate these back to the primary site so that the changes made at the secondary site will be
available to the primary site.
Note: There is no support for the PS Series failback capability with Clone and Restore as New. Once
production has been started on the remote group, the only way to propagate changes back to the
primary group is by replicating the clone volumes to the primary group when it is back up and
operational.
8.7 Disaster Recovery with Failback Capabilities
PS Series replication has the ability to failback only the changes from a promoted replica set back to the
primary site once the primary site is running again. This saves time in recovering the primary site when the
secondary site has been running production for some time. For more information on PS Series replication
and Failback features please refer to the PS Series Group Administrator Guide.
To enable the failback capabilities of PS Series replication with ASM/ME, the replica restore procedure is
completely manual. ASM/ME is used to create the Smart Copy replicas but manual recovery methods
must be used to preserve the failback capabilities. To utilize failback in the event of a disaster follow the
steps below to recover SQL database replicas.
1. Using the PS Series group manager, locate the replica sets on the PS Series group at the secondary
or DR site. Be sure to locate all the replica sets for the volumes that make up the SQL database.