Installing, Configuring, and Troubleshooting Network Protocols

Installing, Configuring, and Troubleshooting Network Protocols

Windows 2000 supports a variety of network protocols, the most popular of which is the Transmission Control Protocol/Internet Protocol ( TCP / IP ) . The different protocols enable a Windows 2000 Server to interoperate in a number of network environments, and to support multiple client platforms, as well as different services and applications. Because network communication is dependent on network protocols, it's important to have the knowledge necessary to install, configure, and troubleshoot them. The following sections cover the fundamentals regarding the two main network protocols supported by Windows 2000: TCP/IP and NWLink.

Installing and Configuring TCP/IP

TCP/IP is an industry standard suite of protocols and utilities that enables communication between hosts on a network. Because of the increasing popularity of TCP/IP, it is quickly becoming the protocol of choice in many network environments. You will find that most operating systems today, including Windows 2000, provide support for TCP/IP.

How a suite of protocols performs is described through the Open Systems Interconnection (OSI) Model. The OSI Model consists of seven layers . The protocols within a protocol suite, such as TCP/IP, operate at different layers of the model, performing different functions to enable network communication. The seven layers of the OSI model are as follows :

• Application This is the top layer of the model. It defines how network applications perform network- related functions.

• Presentation Protocols operating at this layer are responsible for converting and encrypting data.

• Session This layer is responsible for establishing, maintaining, and ending communication sessions.

• Transport Protocols functioning at this layer provide flow control and error checking.

• Network This layer is responsible for addressing and routing.

• Data Link This layer controls the logical network topologies, physical protocol assigned to the data, as well as sequencing.

• Physical This layer defines the physical characteristics of the network.

The suite of protocols that make up TCP/IP also map to a model referred to as the Department of Defense (DoD) model. This model defines communication in four layers. Each of the four layers maps to the different layers within the OSI model and each layer of the model also defines a specific role or function. The different protocols making up the suite each function at a specific layer and work together to provide network communication. The four layers of the DoD model include:

• Application This is the top layer of the model where applications such as FTP, SMTP, and HTTP gain access to the network.

• Transport Protocols operating at this layer are responsible for establishing sessions between two hosts. The two protocols that function at this level include the Transmission Control Protocol (TCP) and the User Datagram Protocol (UDP). The main difference between the two protocols is that TCP is connection-orientated, which means that it provides reliable delivery, whereas UDP is connectionless and does not.

• Internet The main responsibilities of protocols operating at this layer are addressing and routing. The protocols working at this layer include the Internet Protocol (IP), which is responsible for addressing and routing, Address Resolution Protocol (ARP), which maps IP addresses to MAC addresses, Internet Control Message Protocol (IMCP), and Internet Group Management Protocol (IGMP).

• Network This is the bottom layer of the model and is responsible for sending and receiving frames over the physical medium.

IP Addressing

For packets to be routed on an IP network, every host requires a unique IP address (hosts can include workstations, servers, routers, printers, or any other device with a network interface card). The IP address is a 32-bit number, represented in decimal format, that identifies each host.

An IP address consists of two parts : the network ID and the host ID . The network ID is used to identify a specific network or subnet, whereas the host ID identifies the hosts on a given network or subnet. For example, an IP address of 132.10.26.2 has a network ID of 132.10 and a host ID of 26.2 .

The network ID determines whether a destination host is on the local network. If the network ID of the destination host does not correspond to the network ID of the local host, the packet is forwarded to the default gateway. From there, the default gateway uses the information in its routing tables to determine to which network or subnet the packet should be forwarded. This is called an adjacency test .

IP addresses are organized into different address classes that define the number of bits out of the 32 that are used to identify the network. These classes also identify the number of bits used to identify the hosts on a network. By examining the address classes, you can also determine the number of networks and the number of hosts. Table 5.1 summarizes the different address classes.

Table 5.1. TCP/IP Address Classes

Address Class	Range	Number of Network IDs	Number of Host IDs
Class A	1126	126	16,777,214
Class B	128191	16,834	65,534
Class C	192223	2,097,152	254

Table 5.1 summarizes three of the address classes. There are also Classes D and E. Class D is reserved for multicasting and Class E is reserved for testing purposes, but neither of these is relevant to Windows networking. You can determine the class of an IP address by mapping the first decimal value (the number before the first period) to one of the ranges outlined in Table 5.1. For example, 198.221.10.254 is a Class C address.

When assigning IP addresses, each host also requires a subnet mask , which determines which part of an IP address is used as the network ID and which is used to identify a host. For example, the default subnet mask for a Class C address is 255.255.255.0 , which means the first three decimal places (called octets) identify the network and the last octet identifies the host. The subnet mask is also used to determine whether the destination host is on the local or a remote subnet. The subnet mask of the local host is compared against the IP address of the destination host and, through a process known as "anding," it is determined whether the destination IP address is local or remote. (This is called an adjacency test .)

Installing TCP/IP

During the installation of Windows 2000, TCP/IP is automatically installed by default when a network adapter is detected , unless you override the default settings. If it wasn't installed during setup or if you need to add it again, you can do so through the Network and Dial-up Connections applet within the Control Panel. The following steps will help you install TCP/IP:

1. Right-click the Local Area Connection within the Network and Dial-up Connections applet and select Properties.

2. From the Properties window, click the Install button.

3. The Select Network Component Type window appears. Select Protocol from the list and click Add (see Figure 5.1).

   Figure 5.1. Installing additional protocols.

   

4. From the Select Network Protocol window, select Internet Protocol (TCP/IP). Click OK.

5. Click Close.

Configuring TCP/IP

As discussed in Chapter 2, "DNS," IP addresses can be assigned dynamically or statically. When TCP/IP is installed, it is automatically configured to use the Dynamic Host Configuration Protocol (DHCP). If you choose the have an IP address assigned dynamically, you can leave the default configuration as is.

Although not normally used, there is a third method of assigning IP addresses known as Automatic Private IP Address (APIPA) . In the event that a DHCP server is not available, a DHCP client will assign itself an IP address in the range of 169.254.0.1 169.254.255.154 . The address is used until a DHCP server becomes available.

However, for many servers, depending on the role configured, it is recommended or even required that you statically configure an IP address. For example, if you install Active Directory on a server with a dynamically assigned IP address, a warning will appear during the installation informing you that the server should be configured with a static IP address. Or if you are installing the DHCP service, the server must first be configured with a static IP address.

TCP/IP can be configured using the Properties window for the Local Area Connection by performing the following steps:

1. Right-click Local Area Connection within the Network and Dial-up Connections applet.

2. From the list of installed components , select Internet Protocol (TCP/IP) and click the Properties button.

3. The Internet Protocol (TCP/IP) Properties window appears, as shown in Figure 5.2. To override the default settings and statically configure an IP address, select Use The Following IP Address. Type a unique IP address, corresponding subnet mask, default gateway, and the IP addresses of the primary and alternate DNS servers.

   Figure 5.2. Statically configuring an IP address.

   

4. Click OK.

You can configure advanced settings by clicking the Advanced button shown in Figure 5.2. The Advanced TCP/IP Settings window appears, as shown in Figure 5.3. Using the TCP/IP Settings tab, you can assign additional IP addresses to the network connection. For example, a single connection can be assigned a private address for the internal network and a public address for the Internet. The Interface metric box allows you to associate a cost with the route.

Figure 5.3. Configuring advanced TCP/IP settings.

Testing TCP/IP

Once the necessary settings have been specified, the configuration can be tested using the ipconfig and ping commands. Using ipconfig , you can verify the settings that have been configured. To do so, open the Command Prompt window and type ipconfig (using the /all parameter brings up more detailed configuration information), as shown in Figure 5.4. Table 5.2 outlines some of the common parameters that can be used with the ipconfig command.

Figure 5.4. Verifying TCP/IP configuration using the ipconfig command.

Table 5.2. Parameters Used with the ipconfig Command

Parameter	Description
/all	Displays detailed IP configuration information.
/release	Releases the IP address for the specified adapter.
/renew	Renews the IP address for the specified adapter.
/flushDNS	Purges the entries in the DNS cache.
/registerDNS	Refreshes all leased IP addresses and re-registers DNS names .
/displayDNS	Displays the contents of the DNS cache.

Use the ping command to verify connectivity with other hosts on a TCP/IP network. Connectivity on the network is verified by sending Internet Control Message Protocol (ICMP) echo requests and replies. When the ping command is issued, the source computer sends echo requests messages to another TCP/IP host. The remote host, if reachable , then responds with four echo replies. The ping command is also issued at the command prompt along with the TCP/IP address or domain name of the other TCP/IP host, as follows:

 C:> ping 124.120.105.110 

or

 C:> ping www.bayside.net 

To determine whether TCP/IP is initialized on the local computer, issue the ping command and specify the loopback address of 127.0.0.1 .

The general steps for troubleshooting TCP/IP using the ping command are as follows:

1. ping the loopback address of 127.0.0.1 to ensure TCP/IP is initialized on the local computer.

2. If successful, ping the IP address assigned to the local computer.

3. Next, ping the IP address of the default gateway. If this fails, verify that the IP address of the default gateway is correct and that the gateway is operational.

4. Next, ping the IP address of a host on a remote network. If this is unsuccessful , verify that the remote host is operational, verify the IP address of the remote host, and verify that all routers and gateways between the local computer and remote computer are operational.

A quick way of verifying TCP/IP connectivity is to complete step 4 in the preceding steps. If you can successfully ping the IP address of a remote host, steps 1 through 3 will be successful.

Two other utilities that you can use for TCP/IP troubleshooting are tracert and pathping . The tracert command determines the route that is taken to a specific destination. You may want to use the tracert command if you are not able to successfully ping the IP address of a remote host. The results of the tracert command indicate if there is a problem with a router or gateway between the local computer and the remote destination.

The pathping command is basically a combination of the ping and tracert commands. When the command is issued, packets are sent to each router between the local computer and a remote computer. The results determine which routers and gateways may be causing problems on the network.

Installing the NWLink Protocol

If there are NetWare servers hosting resources on the network, you'll want to install the NWLink protocol. This is Microsoft's 32-bit version of Novell's Internetwork Packet Exchange/Sequences Packet Exchange (IPX/SPX) protocol used to communicate with NetWare servers.

NWLink is not installed by default like TCP/IP, but can be installed using the same steps outlined in the section on installing TCP/IP, only choose to install the NWLink IPX/SPX NetBIOS Compatible Transport Protocol instead. Once it's installed, it can be configured using the Properties window of the local area connection.

Configuring NWLink

There are two configuration settings for the NWLink protocol. The first is the internal network number ; the second is the frame type . From the Properties window of the local area connection, select NWLink IPX/SPX/NetBIOS Compatible Transport Protocol (see Figure 5.5).

Figure 5.5. Configuring the NWLink protocol.

The internal network number is used for internal routing. By using the internal network number, virtual networks can be created. If there are multiple network adapters, information can be more efficiently routed to the services running on a computer.

The default internal network number is automatically set to all zeros. The following list describes certain instances in which the number must be manually configured:

• The computer is running File and Print Services for NetWare and there are multiple frame types configured.

• The Windows 2000 computer is configured as an IPX router.

• The Windows 2000 computer is running a program that uses NetWare Service Advertising Protocol (SAP).

The frame type defines how a computer running Windows 2000 and NWLink formats data being sent on the network ( specifically the header and footer information). To communicate with a NetWare server, the computer running Windows 2000 must be using the same frame type.

If there is only one frame type being used on the network, such as 802.2 for example, you can leave the frame type to autodetect. Once NWLink is installed, the frame type will be automatically detected. If there are multiple frame types detected, NWLink will default to 802.2.

You may need to configure the frame type if you have more than one NetWare server on the network using different frame types, because the frame types are not compatible. To manually configure the frame type, select Manual Frame Type Detection and click the Add button from the Properties window for NWLink. When configuring the frame types, keep the following points in mind:

• If there are NetWare 3.12 or later servers running on an Ethernet network, configure the frame type as 802.2.

• If there are running versions of NetWare earlier than 3.12 on an Ethernet network, configure the frame type as 802.3.

• For token-ring networks, the frame type must be set to 802.5.

On Ethernet networks, the standard frame type for Netware 2.2 and 3.11 is 802.3. Starting with Netware 3.12, the default frame type became 802.2.

Communicating with NetWare Servers

Installing the NWLink protocol does not in itself enable a computer to access a NetWare server; additional software must be installed.

Client Service for NetWare ( CSNW ) enables a computer to directly access resources on a NetWare server. Each client requiring access to resources on the NetWare server must have a user account.

Gateway Service for NetWare ( GSNW ) enables a Windows 2000 server to act as a gateway for clients when accessing resources on a NetWare server, eliminating the need for clients to be running NetWare client software. Once GSNW is installed on a Windows 2000 server, Microsoft clients can gain access to NetWare resources through the gateway. Installing GSNW reduces administration because the gateway is the only computer that requires a user account on the NetWare server. You don't have to install any additional software on the workstations.

The software you choose to install depends on the network environment. If your network consists of both Windows 2000 and NetWare servers, consider installing CSNW on all clients so they can directly access the NetWare servers. If you intend to migrate the NetWare servers to Windows 2000, consider installing GSNW to decrease administration.

To install CSNW on a Windows 2000 Professional workstation, perform the following steps:

1. Click Start, point to Settings, and click Control Panel.

2. Open the Network and Dial-up Connections applet.

3. Right-click the Local Area Connection and click Properties.

4. From the properties window, click the Install button. Select Client and click Add.

5. Select Client Service for NetWare and click OK. If NWLink is not already present, it will be installed when CSNW is installed.

GSNW is installed on a Windows 2000 server using the preceding process. The difference being when you select to Add a new client, you must install Gateway (and Client) Service for NetWare. During the installation of GSNW, NWLink and CSNW are installed and a Gateway Service for NetWare applet is added to the Control Panel.

The GSNW applet is used to configure the gateway for NetWare connectivity. If you are working within a Novell NDS environment, specify the default tree and context. This identifies the position of the user object within the NDS tree that the GSNW server logs on with. If you are not in an NDS environment, specify the preferred server. The preferred server is the NetWare server you are automatically connected to when you log on.

You must also create a group called NTGATEWAY on the NetWare server. Within the group, a user account must be added. This is the user account the GSNW server uses to access resources on the NetWare server.

The final step is to actually enable the gateway. Again, you can do this by using the GSNW applet within the Control Panel. Simply open the applet and click the Gateway button. Select the option to Enable Gateway and type the username and password for the account on the NetWare server. You can then use the Add button to configure the NetWare resources that will be available to clients through the gateway.

Configuring Network Bindings

Most servers have a number of protocols and services installed as well as multiple network adapters. Bindings determine the protocols and services that are available to a particular network adapter card as well as the order in which they are used. For example, a local network may use TCP/IP as its primary communication protocol but require NWLink for occasional access to a NetWare server. Because the server attempts any network connectivity using the first protocol bound to the network adapter card, the most frequently used protocol should be listed first. Another example is installing WINS but unbinding it from the external interface connected to the Internet for security purposes.

Configuring the bindings can also aid in reducing network traffic. If a network rarely uses NWLink except for occasional access to a NetWare server and it is listed first in the binding order, the server will attempt to use this protocol first for any network communication, thereby generating unnecessary network traffic.

To configure the bindings for a network connection, perform the following steps:

1. Within the Network and Dial-up Connections applet, click the Advanced menu and select Advanced Settings (see Figure 5.6).

   Figure 5.6. Configuring network bindings.

   

2. The Connections box lists the available network connections and the order in which they are used by network services.

3. In the Bindings for this connection box, select and deselect the services you want bound to the network interface card and the protocols bound to each of the services installed.

4. Use the up and down arrows to change the binding order of protocols so the most frequently used protocol is listed first.

5. Click OK.

Be prepared to encounter exam questions regarding network bindings. Know what network bindings are and how they can be configured to optimize network traffic and communication.