Microsoft

Using OSPF on Windows 2000 Server

Jim Boyce previously showed you how to configure your Windows 2000 server as a router and how to use the RIP routing protocol. In this Daily Drill Down, he introduces you to OSPF, the other routing protocol that Windows 2000 supports.


In my previous Daily Drill Down, “Using RIP on Windows 2000 Server,” I introduced you to RIP, one of the routing protocols that Windows 2000 can use when you configure it to act as a router on your network. Windows 2000 Server also supports OSPF. In this Daily Drill Down, I’ll show you how OSPF works and how you can configure Windows 2000 to support it.

Configuring and using OSPF
The second routing protocol supported by Windows 2000 RRAS is Open Shortest Path First, or OSPF. Each OSPF router maintains a link-state database (LDB) that contains link-state advertisements (LSAs) from adjacent routers. The LSA contains information about a router, its connected networks, and configured costs. The cost is similar to a route metric used by RIP, in that it defines the relative cost of using the route. OSPF uses an algorithm to calculate the shortest path for routing based on the information contained in its LDB, making it a very efficient means of routing. Adjacent routers recalculate and synchronize their LDBs as network changes occur, such as network interfaces going down or coming on line.

OSPF is more complicated to configure than RIP. Its performance advantages are geared primarily toward very large networks, so if you’re setting up a router for a small or medium-size network, RIP is generally the better option. Where network size is a factor, however, OSPF is the better choice.

Adding OSPF
Adding OSPF requires essentially the same process as adding RIP, in that you first add the protocol and then specify the interface(s) on which OSPF will function. The settings for OSPF are quite different from RIP, however.

To add OSPF, open the RRAS console, open the IP Routing branch, right-click General, and choose New Routing Protocol. From the list of available protocols, select OSPF and click OK. You should now see an OSPF node under the IP Routing branch. As with RIP, you need to add interfaces for OSPF, so right-click OSPF and click New Interface. Select the desired interface and click OK. Windows 2000 opens the property sheet for OSPF, with which you configure the protocol. The configuration steps are explained next.

Configuring OSPF interface properties
Windows 2000 displays the OSPF property sheet when you add an interface, and you can access the properties by right-clicking the interface and choosing Properties. As with RIP, you need to configure OSPF’s properties on each interface for which OSPF is enabled (unless you use the default values for all settings).

Use the settings on the General page, shown in Figure A, to specify the IP address on which the router interface responds (if the interface has multiple addresses bound to it). The first option, Enable OSPF For This Address, has a twofold purpose. It lets you enable or disable OSPF on the selected interface. For interfaces bound by multiple IP addresses, it lets you specify on which address OSPF will operate.

Figure A
The General page for an OSPF interface


The Area ID property lets you identify the interface as a member of a specific OSPF area. OSPF areas enable the enterprise to be segregated into contiguous groups of networks, which allows the Link State Database (LSDB) to be smaller, requiring less overhead for routing table calculation. In effect, OSPF areas allow a router to concentrate only on those routers that fall within its area.

Although the area ID is expressed as a dotted decimal value, it has no relation to an IP address or network ID. Area IDs are an administrative tool only. However, it’s often most practical from an administrative standpoint to use network IDs for area IDs to make it easy to identify areas based on placement within the enterprise. By default, only the backbone interface of 0.0.0.0 exists. You can add other areas through the general properties for OSPF, as explained a little later.

The Router Priority value on the General page designates the relative priority of the local router, which enables you to control router adjacency. On a multi-access network where multiple OSPF routers are present, it’s possible that all OSPF routers could attempt to establish adjacency with every other router, resulting in a significant number of adjacencies. This would result in an excessive amount of traffic as all those routers tried to synchronize with one another.

To avoid this problem, a multi-access network elects a Designated Router (DR) that forms adjacencies with all other routers. The DR acts as a hub router for the others in distributing link-state data and performing LSDB synchronization.

The Router Priority value also specifies the priority of the local router. Set this value according to the needs of your network, but remember that at least one router must have a value of 1 or higher. If all routers have a priority of zero, none can become the DR, resulting in no adjacencies, no synchronization, and ultimately, no routing. However, you might want to assign a value of zero to any routers that you don’t want to be able to be elected as the DR. If multiple routers have the same router priority value, the one with the highest router ID becomes the DR.

The Cost option on the General property page specifies the relative cost (metric) for the interface. As with RIP, the route with the lowest cost is used when multiple routes exist for the same destination. Set the value according to your network’s structure.

Just as with RIP v2, you can use passwords to enable OSPF routers to identify themselves to other routers. The Password property on the General page specifies the password used by the local router in communicating with other routers. All routers in the same area must use the same password, and the default password is 12345678. This field is unavailable if plain text passwords are disabled for the selected area. If that’s the case and you want to enable the use of passwords for the area, open the general properties for OSPF and edit the area’s properties.

Use the Network Type radio buttons to specify the type of network to which the interface is connected. Select Broadcast if the interface is connected to a broadcast-type network, such as Ethernet, Token Ring, or FDDI. Use Point-to-Point if the interface is connected to a T1/E1, T3/E3, ISDN, or other point-to-point type network. Select Non-Broadcast Multiple Access (NBMA) if the interface is connected to a Frame Relay, X.25, or ATM network.

The NBMA Neighbors page of the interface’s properties lets you specify the address and priority of neighboring routers if the local router is configured for NBMA in the Network Type group on the General page. If the interface has multiple IP addresses, select the appropriate one from the drop-down list, then enter the IP address and router priority value for all other neighboring routers.

The Advanced page of the Interface’s properties controls a range of transmission and polling properties for the interface. These settings include:
  • IP address: If the interface has more than one IP, select the address you want to manage.
  • Transit delay: This is the estimated time in seconds required to transmit a link-state update packet over the selected interface. Specify a value based on the transmission and reliability of the selected interface.
  • Retransmit interval: Use this value to specify the interval in seconds between retransmission of link state advertisements for adjacent routers. You should set the value to be higher than the total round-trip time between any two routers on the network to avoid excessive retransmissions. Determine the round-trip time based on network structure, number of routers, and so on.
  • Hello interval: OSPF uses Hello packets as a means for discovering neighboring routers. This value specifies the interval between Hello packet broadcasts by the local router. You can decrease the value to allow quicker router discovery, but doing so increases the amount of OSPF traffic on the network. The value must be the same for all routers on the common network. Choose a value based on the network structure and type. Thirty seconds is a reasonable value for X.25, and 10 seconds is a good choice for a LAN.
  • Dead interval: This value specifies the period the local router can be down before adjacent routers will consider it dead (unavailable). Adjacent routers determine the state based on the amount of time that passes between issuing a Hello packet and receiving a response. If a response isn’t received in the specified time, the adjacent routers assume the router is down. All routers in the common network should have the same Dead interval, and it should be a multiple of the Hello interval. Four times the Hello interval is a good rule of thumb.
  • Poll interval: Use this setting in conjunction with the Dead interval setting. The Poll interval specifies the interval between polls to determine if a dead neighbor has come back on line. Make the Poll interval at least twice the Dead interval.
  • MTU size: This value specifies the maximum transmission unit size, which is the maximum IP datagram size that can be sent without fragmentation. The default for Ethernet is 1,500.

Configuring OSPF global properties
After you configure the OSPF interface properties, you also need to configure or at least review some global properties. To configure global properties, open the RRAS console, expand the IP Routing branch, right-click OSPF, and choose Properties.

The options on the General page configure the router’s logging options and address identity. The Router Identification lets you specify a 32-bit number in dotted decimal format to uniquely identify the router. While this doesn’t have to be the router’s IP address, using the address is the best way to ensure a unique value. The Enable Autonomous System Boundary Router option, if selected, enables the router to advertise routing data gained through other sources, such as RIP and static routes. The logging options are self-explanatory.

Use the OSPF Area Configuration page, shown in Figure B, to add and configure OSPF areas, which were discussed earlier. Recall that OSPF areas enable the enterprise to be segregated into contiguous groups of networks, which allows the Link State Database (LSDB) to be smaller, requiring less overhead for routing table calculation. The settings you can use to configure an area are:
  • Area ID: The area ID is a 32-bit number in dotted decimal format that uniquely identifies the area. The value 0.0.0.0 is reserved for the backbone. The number doesn’t have to correlate to a network ID, but using the network ID is the easiest way to recognize the area’s physical location based on its area ID.
  • Enable plaintext password: Use this option to require that routers in the area use the password specified in the interface properties to identify themselves. All routers in an area on the same network segment must use the same password, but routers on other segments can use other passwords.
  • Stub area: Use this option to configure the area as a stub area, which is an OSPF area that accepts external routes from other OSPF routers but doesn’t broadcast external routes to other OSPF routers.
  • Stub metric: This value sets the metric for the summary default route advertised into the stub area.
  • Import summary advertisements: If deselected, all non-intra-area routes are based on a single default route. If selected, inter-area routes are imported into the stub area.
  • Ranges: The Ranges page lets you define the ranges of IP addresses that belong to the selected OSPF area. OSPF uses the ranges to summarize the routes within the area.

Figure B
Create an OSPF area through the OSPF Area Configuration page.


The Virtual Interfaces tab lets you create virtual links between a backbone area router and an area border router that can’t be physically connected to the backbone area. The virtual link transmits routing data between the two routers. Specify the transit delay, retransmit interval, hello interval, dead interval, and plain text password for each virtual link when you add the link.

The options on the External Routing page are available only if you select Enable Autonomous System Boundary Router on the General page for OSPF. These options enable you to define the sources from which the local router will accept routes. You can configure the router to either accept or ignore routes based on several criteria. Click Route Filters to configure filters to accept or ignore routes based on their destination addresses.

Conclusion
In this Daily Drill Down, I gave you an overview of OSPF. I also showed you how to configure a Windows 2000 router to support OSPF. If you need more background information on OSPF, you’ll find additional information in Microsoft TechNet and in the Microsoft Windows 2000 Resource Kit online documentation.
The authors and editors have taken care in preparation of the content contained herein but make no expressed or implied warranty of any kind and assume no responsibility for errors or omissions. No liability is assumed for any damages. Always have a verified backup before making any changes.

Editor's Picks