Using Standby Clusters
Topic Last Modified: 2006-01-16
If you lose all the nodes of a Microsoft® Exchange cluster at the same time, you must recover the whole cluster. You can use a standby cluster to recover an entire cluster. The process for recovering a whole cluster includes many of the same procedures used for recovering stand-alone Exchange member servers. For detailed information about how to restore a member server, see "Exchange Member Server Recovery."
If you do not have the required full computer backups or Microsoft Windows® backups of the nodes in your cluster, you can still recover your whole cluster. To attempt this type of recovery, you must have backups of your Exchange databases (or your Exchange database files and transaction log files must be intact on one of your cluster's shared disk resources). You must also have sufficient informational records about your cluster configuration. For information about how to record cluster information, see "Maintaining Records About Your Server Clusters."
|To rebuild a whole cluster using your cluster's information records instead of restoring the quorum, contact Microsoft Help and Support. The procedures required in this type of recovery are for advanced-level administrators only. Additionally, advanced-level administrators should only consider this cluster recovery method if there is no alternative method available.|
When implementing a recovery strategy for a whole cluster, the first node that you recover (also known as the first node) must be the node that owned the quorum disk resource at the time that you created the backup sets for your nodes. After you recover the first node, make sure that all your cluster resources come online. After your cluster resources are online, you can insert new nodes, insert standby recovery nodes, or continue to restore or rebuild additional failed nodes.
The following information provides more explanation about using the "restore the server" and the "rebuild the server" methods to recover the first node of your cluster.
- Restoring the Server If you use the "restore the server" method to recover the first node, you might have to restore one or more of the cluster's shared disk resources (for example, the quorum disk resource or Exchange databases) after you restore the full computer backup set. For more information about how to restore these shared disk resources, see "Restoring Shared Disk Resources." After you ensure that the first node in the cluster can bring all the cluster resources online, you can insert new nodes, insert standby recovery nodes, or continue to restore nodes from your full computer backup sets.
- Rebuilding the Server If you use the "rebuild the server" method to recover the first node, you must recover the quorum disk resource (if necessary) after you restore the Windows backup set. After you restore your Windows backup set, the Cluster service starts, and the objects in your cluster should appear as they did before the disaster. To view your cluster information, use Cluster Administrator. After you install Exchange (do not use the /disasterrecovery switch, perform a normal install), restore your Exchange database backups (if applicable). After you ensure that the first node in your cluster can bring all the cluster resources online, you can insert new nodes, insert standby recovery nodes, or continue to rebuild nodes from your backup sets.
Important: When you install Exchange to a cluster node as part of a cluster node recovery, you must run Exchange Setup without the /disasterrecovery switch. The /disasterrecovery switch is not supported and cannot be used on clustered Exchange servers.
A standby Exchange cluster is a Windows Server cluster that:
Matches the production Exchange cluster in terms of hardware and software configuration, including Windows and Exchange versions and software updates.
Has Exchange program files installed on it, but is not yet configured with any Exchange Virtual Servers.
Can be used only when all Exchange Virtual Servers on the production cluster are offline.
A standby cluster can be used to recover from the loss of an entire Exchange cluster, or as a site resilience solution for Exchange clusters. When transferring Exchange Virtual Servers from a production cluster to a standby cluster, all of the Exchange Virtual Servers in the production cluster must be moved. No Exchange Virtual Server(s) should be running on the production cluster.
|This process is only supported for Exchange Server 2003 clusters running on Windows Server™ 2003. The process described in this topic cannot be applied to and is not supported for Exchange 2000 Server or Exchange Server 5.5.|
For more information about moving Exchange Virtual Servers from a production cluster to a standby cluster, see How to Move All Exchange Virtual Servers from a Production Exchange 2003 Cluster to a Standby Exchange 2003 Cluster.