OSPF Facts

OSPF Facts

The Open Shortest Path First (OSPF) routing protocol is a robust link state routing protocol well-suited for large networks. OSPF:

• Is a public (non-proprietary) routing protocol.
• Is considered a classless routing protocol because it does not assume the default subnet masks are used. It sends the subnet mask in the routing update and supports manual route summarization and VLSM. OSPF does not perform automatic route summarization.
• Is not susceptible to routing loops. Instead, OSPF uses built-in loop avoidance techniques. Mechanisms such as holddown timers, split horizon, or poison reverse are not needed.
• Is scalable and does not have the 16 hop limitation of RIP.
• Uses multicasts to share routing information (using 224.0.0.5 and 224.0.0.6).
• Uses link costs (bandwidth) as a metric for determining best routes. The Shortest Path First (SPF) algorithm (also called the Dijkstra SPF algorithm) is used to identify and select the optimal route.
• Supports load balancing over equal-cost paths. Up to 16 equal-cost paths can be used (the default is 4).
• Uses hello packets to discover neighbor routers.
• Shares routing information through Link State Advertisements (LSAs). LSAs contain small bits of information about routes. (Unadvertised links save on IP space, but they cannot be pinged because they won't appear in an OSPF routing table.)
• Under normal conditions, OSPF only sends out updated information rather than exchanging the entire routing table.
• Converges faster than a distance vector protocol. Following convergence sends updates when routes change or every 30 minutes.
• Can require additional processing power (and therefore increased system requirements). Good design can minimize this impact.
• Maintains a logical topographical map of the network in addition to maintaining routes to various networks.
• Uses areas to subdivide large networks. Routers within an area share information about the area. Routers on the edge of areas (called Area Border Routers (ABR)) share summarized information between areas.
  • The backbone is a specialized area connected to all other areas. It contains networks not held within another area, and distributes routing information between areas. You can think of the backbone as the "master" or "root" area. Its address is always 0.0.0.0. All OSPF networks must have a backbone area.
  • A stub area is an area with a single path in to and out of the area.

To help minimize traffic caused by routing updates, OSPF defines the following router roles:

Role	Description
Designated Router (DR)	On each subnet, a single OSPF router is elected as the designated router (DR). The DR is responsible for coordinating routing table updates for all routers on the subnet. Routing information is sent from other routers to the DR. The DR manages the changes and forwards any necessary information to other routers on the subnet.
Backup Designated Router (BDR)	On each subnet, a single OSPF router is identified as the backup designated router (BDR). The BDR becomes the DR if the DR becomes unavailable.
DROTHER	Any other router that is not a DR or a BDR is called a DROTHER. This is simply a term used to describe a non-DR or non-BDR router. It is not technically an OSPF router role.

Be aware of the following facts about the DR and BDR:

• Based on the network link type, a DR/BDR might not be used. A DR/BDR is used on broadcast networks (like Ethernet) where multiple routers exist on the same subnet. For point-to-point networks, a DR/BDR is not used. By default, the network type is identified based on the media type used. You can manually configure the network type if desired.
• If the network type uses a DR/BDR, a single DR and a single BDR is identified for each subnet.
• When routers first come on line, they exchange hello packets. Part of this process is used to elect (identify) the DR and the BDR.
• The following values are used to elect the DR and BDR:
  • The router with the highest OSPF priority becomes the DR. The priority value is a number between 0-255. By default, all routers have a priority of 1.
  • If two or more routers have the same highest priority value, the router with the highest router ID becomes the DR. The router ID is a 32-bit number expressed in A.B.C.D format. The router ID for a specific router is chosen in the following order:
    • For a specific OSPF process, you can manually configure a router ID. If a router ID has been configured, that value is used.
    • If no router ID has been manually configured, the system uses the highest IP address assigned to a loopback address.
    • If the router does not have a loopback address, the router ID is the highest IP address assigned to any interface in the up state.

    Note: Using a loopback address is preferred over using the interface IP address because it allows you to control which router becomes the DR, and because loopback interfaces never go down. If an interface address is used for the router ID, the router ID might change if that interface goes down.

  • In most cases, the BDR is the router with the next highest priority or router ID.
• Configuring a priority of 0 for a router means that the router will never become the DR or BDR.
• Once a DR has been elected, it remains the DR, even if another router with a higher priority or router ID comes on line. You must clear or reset the OSPF process to force a new election.
• If the DR goes down, the BDR automatically becomes the DR. When the original DR comes back on line, it will not automatically resume the DR role unless a reset is performed.

OSPF routers share route information only with adjacent neighbor routers. The following conditions must be met for two routers to become fully adjacent:

• Both routers must be on the same subnet and use the same subnet mask.
• Both routers must have the same hello and dead intervals configured.
  • The hello interval identifies how frequently neighbor routers exchange hello packets.
  • The dead interval identifies the amount of time to allow without an expected hello packet. If a periodic hello packet has not been received within the dead interval, the router assumes that its neighbor has gone offline.
• Both routers must use the same OSPF area.
• If authentication is required, both routers must pass the authentication requirements.
• The stub area flag (value) for each router must match.