Rapid Recovery 6.1.3 - User Guide

Restoring clusters and cluster nodes

A restore is the process of restoring the volumes on a machine from recovery points. For a server cluster, you perform a restore at the node, or machine, level. This section provides guidelines for performing a restore for cluster volumes.

Performing a restore for CCR and DAG (Exchange) clusters

Complete the steps in this procedure to perform a restore for CCR and DAG (Exchange) clusters.

Performing a restore for SCC (Exchange, SQL) clusters

Complete the steps in this procedure to perform a restore for SCC (Exchange, SQL) clusters.

NOTE: You do not need to roll back the quorum disk. It can be regenerated automatically or by using cluster service functionality.

Restoring from an attached archive

When you attach an archive, it appears under Attached Archives on the Archives page of the Core Console, while the contents of the archive become accessible from the left navigation area. The contents appear under the name of the archive. Machines that were archived appear as recovery-points-only machines so that you can access the recovery points in the same way that you would for a currently protected machine: by mounting a recovery point, locating the item that you want to recover, and using Windows Explorer to copy and paste the item to your destination.

There are advantages to restoring from an attached archive rather than importing an archive to a repository.

Because these archived recovery points are likely the oldest items in the repository, they may be rolled up according to your retention policy during the next nightly job. (Although, this action does not delete them from the archive; you could re-import them the next day.)

You can remove the association by deleting the attachment.

To restore data from an attached archive, complete the following steps using the related links:

NOTE: The procedure for restoring from an attached archive assumes that you already have an archive of rolled-up recovery points.

Bare metal restore

This section describes how to restore a protected Windows machine from bare-metal similar or dissimilar hardware.

Bare metal restore for Windows machines

Servers, when operating as expected, perform the tasks they are configured to do. It is only when they fail that things change. When a catastrophic event occurs, rendering a server inoperable, immediate steps are needed to restore the full functionality of that machine.

Rapid Recovery provides the ability to perform a bare metal restore (BMR) for your Windows or Linux machines. BMR is a process that restores the full software configuration for a specific system. It uses the term “bare metal” because the restore operation recovers not only the data from the server, but also reformats the hard drive and reinstalls the operating system and all software applications. To perform a BMR, you specify a recovery point from a protected machine, and roll back (perform a restore) to the designated physical or virtual machine. If you are performing a restore to a system volume, this is considered a BMR. If you are performing a restore and require a boot CD, this is also considered a BMR. Other circumstances in which you may choose to perform a bare metal restore include hardware upgrade or server replacement In both of these cases, you perform a restore from a recovery point to the upgraded or replaced hardware.

Rapid Recovery supports Windows 8, 8.1 and Windows Server 2012, 2012 R2 operating systems that are booted from FAT32 EFI partitions are available for protection or recovery, as well as Resilient File System (ReFS) volumes.

NOTE: Bare metal restore of Storage Spaces disks configuration (a feature of Windows 8.1) is also not supported in this release. At present, only full restore and BMR are supported as a driver limitation exists with ReFS, so restore is implemented in user mode, VM export, and so on. If a Core is protecting at least one agent volume that contains the ReFS file system, it should be installed on a Window 8, Windows 8.1, Windows Server 2012, or Windows Server 2012 R2 machine, since these operating systems provides native support of ReFS. Otherwise, functionality will be limited and operations that involve such things as mounting a volume image will not work. The Rapid Recovery Core Console will present applicable error messages in these occurrences. Only supported Linux operating systems are available for protection or recovery. This includes Ubuntu®, Red Hat® Enterprise Linux® (RHEL), CentOS™, and SUSE® Linux Enterprise Server (SLES®). For details, see the Rapid Recovery Installation and Upgrade Guide.

Performing a BMR is possible for physical or virtual machines. As an added benefit, Rapid Recovery allows you to perform a BMR whether the hardware is similar or dissimilar. Performing a BMR on Rapid Recovery separates the operating system from a specific platform, providing portability.

Examples of performing a BMR for similar hardware include replacing the hard drive of the existing system, or swapping out the failed server with an identical machine.

Examples of performing a BMR for dissimilar hardware include restoring a failed system with a server produced by a different manufacturer or with a different configuration. This process encompasses creating a boot CD image, burning the image to disk, starting up the target server from the boot image, connecting to the recovery console instance, mapping volumes, initiating the recovery, and then monitoring the process. Once the bare metal restore is complete, you can continue with the task of loading the operating system and the software applications on the restored server, followed by establishing unique settings required for your configuration.

Bare metal restore is used not only in disaster recovery scenarios, but also to migrate data when upgrading or replacing servers.

Restoring virtual machines

While bare metal restore (BMR) is supported for virtual machines (VMs), it is also worth noting that it is easier to perform a Virtual Export for a VM than it is to perform a BMR as on a physical machine. For more information on performing a VM export for virtual machines, see .

To perform a BMR on a Windows machine, refer to the topic specific to Windows, including the prerequisites. For more information, see Performing a bare metal restore for Windows machines.

You can also perform a BMR from the Restore Machine Wizard. To do this, start with the procedure About restoring volumes from a recovery point and, when directed in that procedure, proceed to Performing a bare metal restore for Windows machines.

To perform a BMR on a Linux machine, see Performing a bare metal restore for Linux machines. In addition to performing a BMR using the command line local_mount utility, you can now perform a BMR from within the Core Console UI. The roadmap takes both approaches into account.