Review Questions | 1. | You are the network administrator for a small company with several sites, each with host computers. You are in the process of migrating to Windows Server 2003, but you want to be sure that all your legacy Windows NT 4 computers can browse the network. What should you do? | | 2. | You have successfully deployed WINS in your environment, and it is working. Your users can browse the network. How does the WINS database get populated to make this possible? | | | | 3. | After the WINS is fully implemented and the database is populated, how are entries deleted from the database? | | 4. | What is the major difference between push replication and pull replication? | | 5. | You're the administrator of Little Faith Enterprise's Windows Server 2003 servers, and you are considering setting up WINS. You feel you need a better understanding of how NetBIOS works and how the various node types work. What are the four NetBIOS node types, and how do they work? |
Answers to Review Questions | 1. | You need to deploy WINS in your environment. Depending on your environment, you might be able to deploy redundant WINS servers in your central site, or you might need to deploy WINS servers to each location. In either event, you must configure replication between the servers. For more information, see the section "Installing and Configuring WINS." | | 2. | The WINS database is populated by the WINS client computer. When the WINS client computer comes on the network, it registers its name and address with the WINS server automatically. After the client computer is registered, it receives a time to live for its registration. For more information, see the section "Configuring WINS." | | 3. | There are two automated mechanisms for removing entries. When a client computer shuts down gracefully, it sends a release request to the WINS server. The entries are also removed when the time to live for that client computer's registration expires. For more information, see the section "Configuring WINS Replication." | | 4. | The main difference between push and pull replication (besides the direction the database is replicated) is the trigger for the event. In the case of a push replication, the trigger is event based. When a specified number of changes are made to the database, the replication is triggered. A pull replication is triggered by the time configured for the replication. This is user configured. For more information, see the section "Configuring WINS Replication." | | 5. | There are four node types, and the main differentiator between the types is the methods they use for name resolution (broadcast versus direct connection). They are as follows: b-node (broadcast node) Relies exclusively on broadcast messages and is the oldest NetBIOS name resolution mode. A host needing to resolve a name request sends a message to every host within earshot, requesting the address associated with a host name. b-node has two shortcomings: Broadcast traffic is undesirable and becomes a significant user of network bandwidths, and TCP/IP routers don't forward broadcast messages, which restricts b-node operation to a single network segment. p-node (point-to-point node) Relies on WINS servers for NetBIOS name resolution. Client computers register themselves with a WINS server when they come on the network. They then contact the WINS server with NetBIOS name resolution requests. WINS servers communicate using directed messages, which can cross routers, so p-node can operate on large networks. Unfortunately, if the WINS server is unavailable or if a node isn't configured to contact a WINS server, p-node name resolution fails. m-node (modified node) It's a hybrid mode that first attempts to resolve NetBIOS names using the b-node mechanism. If that fails, an attempt is made to use p-node name resolution. m-node was the first hybrid mode put into operation, but it has the disadvantage of favoring b-node operation, which is associated with high levels of broadcast traffic. h-node (hybrid node) It's also a hybrid mode that favors the use of WINS for NetBIOS name resolution. When a computer needs to resolve a NetBIOS name, it first attempts to use p-node resolution to resolve a name via WINS. Only if WINS resolution fails does the host resort to b-node to resolve the name via broadcasts. Because it typically results in the best network utilization, h-node is the default mode of operation for Windows clients configured to use WINS for name resolution. Microsoft recommends leaving TCP/IP client computers in the default h-node configuration.
For more information, see the section "Configuring NetBIOS Name Resolution." |
|
