Single Copy Cluster Resource Model
Applies to: Exchange Server 2007 SP3, Exchange Server 2007 SP2, Exchange Server 2007 SP1, Exchange Server 2007
Topic Last Modified: 2007-10-26
Microsoft Exchange Server 2007 clusters create logical servers that are referred to as clustered mailbox servers by using the Microsoft Windows Server Cluster service. Exchange 2007 only supports active/passive single copy cluster (SCC) configurations. In previous versions of Exchange, clustered mailbox servers were referred to as Exchange Virtual Servers. Unlike a stand-alone (non-clustered) Exchange 2007 Mailbox server, a clustered mailbox server is mobile and can be failed over if the server currently running the clustered mailbox server fails. When the computer that is actively servicing client requests for Exchange in the cluster fails, one of the remaining nodes in the cluster takes over for the failed clustered mailbox server, and clients can access this server by using the same Exchange Server name.
To create an Exchange 2007 cluster, you first create a Windows Server 2003 cluster and then install Exchange on each node of the cluster. It is also necessary to use Setup to create each clustered mailbox server in the cluster.
A clustered mailbox server is a resource group that requires, at a minimum, the following resources:
One or more physical disks for shared storage
Several Exchange-specific resources
Client computers connect to a clustered mailbox server the same way that they connect to a stand-alone computer running Exchange 2007. Exchange 2007 Setup automatically creates the IP address, network name, and Exchange-specific resources. It is your responsibility to populate the physical disks in the resource group.
Exchange 2007 supports only clusters that have at least one passive node, for example, two active nodes and a passive node. In active/passive clusters, the cluster includes at least one primary or active node and at least one secondary or passive node. The secondary nodes are idle until a failover occurs on a primary node. When the primary node in an active/passive cluster fails or is taken offline, the clustering feature in Windows takes over. The failed node is taken offline, and a secondary node takes over the operations of the failed node. It usually takes only a few minutes for the cluster to fail over to another node. As a result, the Exchange resources on your cluster are unavailable to users for only a brief period of time.
In active/passive clusters, the number of clustered mailbox servers in the cluster is always less than the number of physical nodes in the cluster.
In addition to the resource groups created for each clustered mailbox server, an SCC always has a resource group to represent the quorum of the cluster. This resource group, by default named Cluster Group, is created when the cluster is created. For an SCC, the default quorum strategy uses a shared disk. However, a Majority Node Set (MNS) quorum with file share witness is also supported for SCCs. In a shared disk quorum, the disk containing the quorum resource is called the quorum disk, and it must be a member of the default Cluster Group.
|Failure of the quorum disk takes the cluster offline. This stops all clustered mailbox servers that the cluster supports.|
The quorum disk maintains configuration data in the quorum log, cluster database checkpoint, and resource checkpoints. The quorum disk resource also provides persistent physical storage across system failures. Because the cluster configuration is kept on a quorum disk resource, all nodes in the cluster must be able to communicate with the node that owns it.
When a cluster is created or when network communication between nodes in a cluster fails, the quorum disk resource prevents the nodes from forming multiple clusters. To form a cluster, a node must arbitrate for, and gain ownership of, the quorum disk resource. For example, if a node cannot detect a cluster during the discovery process, the node attempts to form its own cluster by taking control of the quorum disk resource. However, if the node does not succeed in taking control of the quorum disk resource, it cannot form a cluster.
The quorum disk resource stores the most current version of the cluster configuration data. This data contains cluster configuration and state data for each individual node. When a node joins or forms a cluster, the Cluster service updates the node's individual copy of the configuration database. When a node joins an existing cluster, the Cluster service retrieves the configuration data from the other active nodes.
The Cluster service uses the quorum disk resource recovery logs to do the following:
Guarantee that only one set of active, communicating nodes can operate as a cluster.
Enable a node to form a cluster only if it can gain control of the quorum disk resource.
Allow a node to join or remain in an existing cluster only if it can communicate with the node that controls the quorum resource.
Note: You should never create a clustered mailbox server in the default Cluster Group. Clustered mailbox servers are supported only in dedicated resource groups.
SCCs are based on creating a logical cluster model, called the resource model, which abstracts the individual resources that make up a clustered mailbox server, such as the Microsoft Exchange Information Store service and the databases maintained by the clustered mailbox server. The resource model is represented as a tree to allow the software to define an order of actions when a piece of the model starts or stops.
The following are significant differences between the Exchange 2007 resource model and the Exchange Server 2003 resource model:
There is a new cluster resource called the Microsoft Exchange Database Instance.
Physical disk resources are now dependencies for database instances.
The Microsoft Exchange Information Store service no longer depends on the Microsoft Exchange System Attendant service. Thus, the Microsoft Exchange Information Store cluster resource does not depend on the Microsoft Exchange System Attendant resource.
Only the IP Address and Network Name resources should have the Affect the Group option selected.
Databases, the Microsoft Exchange Information Store service, and other resources can be brought online and offline manually by an administrator or automatically by the Cluster service. The term online generally refers to the process of starting a service and mounting a database. The term offline generally refers to the process of stopping a service, dismounting a database, or stopping an entire clustered mailbox server. The Cluster service tracks the online and offline states of resources in the resource model.
The resources are used as a focal point for managing the processes, databases, and network identities associated with the clustered mailbox server. The following is a brief summary of each resource type in an SCC:
- Microsoft Exchange Database Instance This resource represents a database that is hosted on the clustered mailbox server. When this resource is online, the database is mounted. When this resource is offline, the database is dismounted. By default, the Microsoft Exchange Information Store is a dependency of each database instance. In addition, you must manually make the appropriate disk dependencies of each database instance. All dependencies must be online before the database instance comes online. By default, the Affect the Group option is disabled for this resource.
- Microsoft Exchange Information Store This resource represents the clustered mailbox server. When this resource is online, the Microsoft Exchange Information Store service is started and capable of accepting MAPI traffic. The clustered mailbox server may or may not have mounted databases. When it is offline, the Microsoft Exchange Information Store service is stopped, and it is not capable of accepting MAPI traffic. By default, the Affect the Group option is disabled for this resource.
- Microsoft Exchange System Attendant This resource represents the Microsoft Exchange System Attendant service for the clustered mailbox server. When this resource is online, the Microsoft Exchange System Attendant service is started. When this resource is offline, the Microsoft Exchange System Attendant service is stopped. By default, the Affect the Group option is disabled for this resource.
- Network Name (Name) This resource represents the network name of the clustered mailbox server. When the Network Name resource is online, the name is associated with a network adapter on the specified computer. When the Network Name resource is offline, the name is not associated with a network adapter on the specified computer. By default, Kerberos authentication and successful Domain Name System (DNS) registration are required for this resource. In addition, by default, the Affect the Group option is enabled for this resource.
- IP Address (Name) This resource represents the IP address associated with the clustered mailbox server. This IP address is bound to the clustered mailbox server network name in DNS. When the IP Address resource is online, it is associated with a network adapter on the specified computer. When the IP Address resource is offline, it is not associated with a network adapter on the specified computer. By default, NetBIOS is enabled for this resource. In addition, by default, the Affect the Group option is enabled for this resource.
In addition to these resources, SCCs also contain physical disk resources that represent the disks containing the storage group and database files for the clustered mailbox servers in the SCC. When you add physical disks to the cluster group containing the clustered mailbox server, in addition to making the disks dependencies for the appropriate database, you must also clear the Affect the Group check box for each physical disk resource. If this check box is left selected, a data failure will cause failover, and this is contrary to the default behavior.