Windows Server 2003 Manage TCP/IP Routing

  • Print Article |
  • Send to a Friend |
  • |
  • Add to Google |

Manage TCP/IP Routing: So far, we have explored what a routing interface is. As mentioned earlier, there are two other elements needed-the routing protocol and the routing table itself. A routing protocol is used whenever you want to maintain the routing tables dynamically, which is definitely preferred in most situations. Routing protocols allow routers to share routing information.

The two routing protocols supported by Windows Server 2003 are RIP and OSPF. RIP, or Routing Information Protocol, is used in smaller networks. It does not scale well, but is very simple to implement. RIP is available in two versions, v.1 and v.2. Both use a routing table. RIPv.1 broadcasts its full table to its neighboring routers, while RIPv.2 can be configured to multicast instead of broadcast. Windows Server 2003 only supports RIPv.2, so you will need to ensure that all other routers also support version 2 if you plan to use RIP. Typically, RIP is used in networks with less than 50 servers, and less than 15 connected networks.

OSPF, or Open Shortest Path First, is known as a link state protocol. OSPF maintains a database of link states rather than a routing table and, rather than advertising its full database to all neighbors, it only sends change messages. Because of this, it is a far more efficient routing protocol from the perspective of processor overhead and bandwidth utilization. Its main drawback is that it can be more complex to configure-though a default configuration on a network using entirely Windows Server 2003 can be quite effective. If, however, you have a mix of hardware routers and software routers running Windows 2000 Server or Windows Server 2003, you will need to be careful to configure all of the routers so that they properly communicate with each other.

Manage Routing Tables: The third element is the routing table itself. Every computer connected to a network has a routing table for that interface. This is true whether the interface is a modem or a T-3 connection.

You see the results of typing route print from the command console. This shows you the protocol version (in this case IPv4), the interfaces (MS Loopback, 3Com 3C918 and IBM 10/100 EtherJet), the MAC address in hexadecimal and then the routing table itself. The routing table section shows the active routes, with 5 columns of information. First is the destination network, followed by the netmask for that destination. Next you have the gateway, which is the IP address for the interface used to access that destination, followed by the IP address of the interface that is using that gateway address. Last you find the metric.

The metric is the value deducted from the time to live for the packet. Every packet has a TTL, or time to live, so that it does not just keep going. Every routing interface will have a cost or metric, which will be deducted from the TTL for the packet. If the TTL runs out before the packet reaches its destination, it will be destroyed and a time-out message will be returned to the sending host.

The purpose of a routing table is to allow the IP portion of the TCP/IP suite to determine the best path for a data transmission. Bear in mind that all IP hosts will have a routing table-not just routers. There are three types of entries in a routing table. These are the Network Route, Host Route and Default Route. A network route is a path to a specific network. For example, if there was a network which the server in Figure 7.3 was part of, it may have a network route of:

Destination Netmask Gateway Interface Metric: 2 This can be read as "to get to network with subnet mask, use address and deduct 2 from the TTL for the packet."A host route is an entry that shows the exact path to an IP host, rather than a network ID. In place of the network ID, the host address would be listed.

A default route is used when there are no entries in the routing table that match up to either the destination host or the destination network ID. Essentially, IP looks for either a host or network route, and if it does not find an entry, it sends the packet on to a default router, which hopefully has more information on how to get to the destination.

Uses for RRAS Information:
Windows Server 2003 RRAS is typically enabled on a multi-role server. In smaller networks, it may be a file and print server that is also routing between connected networks. It may also be used to connect remote offices over the Internet or via a demand-dial connection. Finally, RRAS is often used as a remote access server for end-user VPN access to the internal network over the Internet.

Enabling RRAS: Remember that RRAS is always present with Windows Server 2003-in order to use it, you need to enable it. All you need to do is choose Configure and Enable Routing and Remote Access from the Action menu of the RRAS management console. You will be presented with a splash screen, which is the beginning of the wizard from which you choose from one of several options When you first install Windows Server 2003, you will also be presented with a "Manage your Server" page, from which you can choose a variety of roles for a server-including one of several RRAS roles.

For the certification exam, you may want to get familiar with both methods of enabling and configuring RRAS. Once you have enabled RRAS, you can further configure it with routing protocols, packet filters and remote access policies. Remote Access Policies will be covered later in this section. First, however, we are going to look at how to add routing protocols and then packet filters.

There are 3 steps involved in configuring a routing protocol. First, you must add the protocol itself, and then you must add at least one interface for the protocol. You can add as many interfaces as you want to a given routing protocol, or at least as many as you have. Once you have added the interface or interfaces to the protocol, you then can perform any additional configuration necessary for your networking environment.
Let's start by looking at how you install a routing protocol. Starting in the RRAS Management console, let's configure this server for LAN routing. To do this, you would choose the custom configuration option.

On the next screen, you will choose the LAN Routing checkbox only. After Windows Server 2003 enables and configures the server as a LAN router, you will see the following: Now, to add a routing protocol, you click on General, then either right-click or click on the Action menu and choose New Routing ProtocolAs you can see, you have the option of installing and configuring the DHCP Relay Service, IGMP Router and Proxy (used for multicast), Open Shortest Path First (OSPF) and RIPv.2.

Once you have installed the routing protocols needed, they will be listed in the IP Routing section for that server. After they are installed, you will need to add an interface to the protocol, and may then need to perform additional configuration. Adding an interface is similar to the other actions you have performed. Either right-click on the routing protocol or select the protocol then click on the Action menu, selecting Add Interface. You will be presented with a dialog box that shows the names of the interfaces as they are listed in the Network Connections dialog box-which is why I suggest renaming them before you enable and configure RRAS!

After you add the interface or interfaces for the routing protocol, you will be presented with the properties page for that protocol. Here you can configure it as needed.

Deborah Timmons is a Microsoft Certified Trainer and Microsoft Certified Systems Engineer. She came into the Microsoft technical field after six years in the adaptive technology field, providing technology and training for persons with disabilities. She is the President and co-owner of Integrator Systems Inc.

Rate this Article:
  • Article Word Count: 1272
  • |
  • Total Views: 276
  • |
  • permalink
  • Print Article |
  • Send to a Friend |
  • |
  • Add to Google |