Archive for the ‘MCSA to MCSE-2003’ category

What is a Site?

A site is a collection of one or more well-connected (usually a local area network) TCP/IP subnets. The network between the subnets must be highly reliable and fast (512 Kbps and higher). Although the sites are generally defined on the basis of location, they can be spanned over more than one location. A site structure corresponds to the physical environment, whereas a domain is the logical environment of the network. A site can contain single or multiple domains, and a domain can contain single or multiple sites.

The sites are created to physically group the computers and resources to optimize network traffic. Administrators can configure Active Directory access and replication technology to take advantage of the physical network by configuring sites. When a user logs on to the network, the authentication request searches for the domain controllers in the same site as the user. A site prevents the network traffic from traveling on slow wide area network (WAN) links.

What are Directory Tree, Directory Partition, and Replica?

Directory tree is a hierarchy of objects and containers of Active Directory, which represents all the objects in the forest. Each domain controller stores a copy of a specific part of the directory tree, called a directory partition (sometimes called naming context). The copy of the directory partition is called a replica. A replica contains all attributes for each directory partition object. Each domain controller in the forest stores a replica.

What is replication?

Replication is a process through which the changes made to a replica on one domain controller are synchronized to replicas on all the other domain controllers in the network. Each domain controller stores three types of replicas:

• Schema partition: This partition stores definitions and attributes of objects that can be created in the forest. The changes made in this partition are replicated to all the domain controllers in all the domains in the forest.
• Configuration partition: This partition stores the logical structure of the forest deployment. It includes the domain structure and the replication topology. The changes made in this partition are replicated to all the domain controllers in all the domains in the forest.
• Domain partition: This partition stores all the objects in a domain. Changes made in this partition are replicated to all the domain controllers within the domain.

Note: Windows Server 2003 supports a new type of directory partition named Application directory partition. This partition is available only to Windows 2003 domain controllers. The applications and services use this partition to store application-specific data.

Creating, modifying, moving, and deleting an object trigger a replication between domain controllers. Replications are of two types:

• Intrasite: An intrasite (within a site) replication mostly uses LAN connections. As intrasite replication does not compress data, it saves a computer’s CPU time. In an intrasite replication, the replication partners poll each other periodically and notify each other when changes need to be replicated, and then pull the information for processing. Active Directory uses a remote procedure call (RPC) transport protocol for intrasite replication.
• Intersite: As an intersite (between sites) replication uses WAN connections, a large amount of data is compressed to save WAN bandwidth. For the same reason, the replication partners do not notify each other when changes need to be replicated. Instead, administrators configure the replication schedule to update the information. Active Directory uses an IP or SMTP protocol for intersite replication.

For intrasite replication to take place, connection objects are required. The Active Directory automatically creates and deletes connection objects as and when required. Connection objects can be created manually to force replication.

What are Site Links?

Site links are logical, transitive connections between two or more sites. For intersite replication to take place, site links are required to be configured. Once a site link has been configured, the knowledge consistency checker (KCC) then automatically generates the replication topology by creating the appropriate connection objects. Site links are used to determine the paths between two sites. They must be created manually.

Site links are transitive in nature. For example, if Site 1 is linked with Site 2 and Site 2 is linked with Site 3, then Site 1 and Site 3 are linked transitively. The administrators can control transitivity of the site link. By default, transitivity is enabled. Site link transitivity can be enabled or disabled through a bridge.

What is Site Link Bridge?

A site link bridge is created to build a transitive and logical link between two sites that do not have an explicit site link. The site link bridge is created only when the transitivity of the site link is disabled.

What is Site Link Cost?

Site link cost is an attribute of a site link. Each site link has been assigned a default cost of 100. The knowledge consistency checker (KCC) uses the site link cost to determine which site links should be preferred for replication. It should be remembered that the lower the site link cost, the more preferred is the link.

For example, an administrator has to configure the site link cost of links between Site 1 and Site 2. There are two site links available as shown in the image below:

S1S2 is a T1 site link that uses T1 lines for replication, whereas S1S2DU uses a dial-up connection for replication. If the administrator requires that the KCC should prefer the S1S2 site link to the S1S2DU site link for replication, he will have to configure the SIS2 link with a lower cost than that of the S1S2DU link. Any site link configured with the site link cost of one (1) will always get preference over the other site links with a higher cost.

What is Bridgehead Server?

A bridgehead server is a domain controller in each site, which is used as a contact point to receive and replicate data between sites. For intersite replication, KCC designates one of the domain controllers as a bridgehead server. In case the server is down, KCC designates another one from the domain controller. When a bridgehead server receives replication updates from another site, it replicates the data to the other domain controllers within its site.

What is Preferred Bridgehead Server?

A preferred bridgehead server is a domain controller in a site, specified by an administrator, to act as a bridgehead server. Administrators can specify more than one preferred bridgehead server, but only one server is active at a time in a site. A preferred bridgehead server is designated to take advantage of a certain domain controller having the appropriate bandwidth to transmit and receive information.

Note: If only one preferred bridgehead server is configured in a site, and it fails, then no replication will take place for that site.