Configure OSPF Areas

To prevent a large amount of database information from occupying too much CPU and memory, you can divide an OSPF AS into multiple areas. An area can be identified with a 32-bit area ID, a decimal number in the range of 0-4294967295, or an IP address in the range of 0.0.0.0-255.255.255.255. Area 0 or 0.0.0.0 represents an OSPF backbone area, while other non-zero areas are non-backbone areas. All routing information between areas must be forwarded through the backbone area. Non-backbone areas cannot directly exchange routing information.

OSPF defines several types of routers:

• Internal router: All interfaces belong to the devices in one area.
• Area Border Router (ABR): It is connected to devices from different areas.
• Autonomous System Boundary Router (ASBR): It is a device that introduces external routes to the OSPF AS.

Configuration Condition

Before configuring an OSPF area, ensure that:

• Interface IP addresses have been configured so that neighbor nodes are reachable at the network layer.
• Enable OSPF.

Configure an OSPF NSSA Area

A Not-So-Stub-Area (NSSA) does not allow injection of Type-5 Link State Advertisement (LSA) but it allows injection of Type-7 LSA. External routes can be introduced to an NSSA area through redistribution of configuration. The ASBR in the NSSA area generate Type-7 LSAs and flood LSAs to the NSSA area. The ABR in an NSSA area converts Type-7 LSAs into Type-5 LSAs, and floods the converted Type-5 LSAs into the entire AS.

The OSPF NSSA area that is configured by using the area area-id nssa no-summary command is called a totally NSSA area. An OSPF totally NSSA area does not allow cross-area routes to flood in the area. At this time, the ABR generates a default route and flood it into the NSSA area. The devices in the NSSA area access a network outside the area through the default route.

Table 7-3 Configuring an OSPF NSSA Area

• A backbone area cannot be configured as an NSSA area.
• All devices in one NSSA area must be configured as NSSA areas, because devices with different area types cannot form neighbor relations.

Configure an OSPF Stub Area

A Stub area does not allow external route outside an AS to flood in the area so as to reduce the size of the link status database. After an area is configured as a Stub area, the ABR which is located at the Stub border generates a default route and flood the route into the Stub area. The devices in the Stub area access a network outside the area through the default route.

The OSPF Stub area that is configured by using the area area-id stub no-summary command is called a totally Stub area. An OSPF totally Stub area does not allow inter-area routes and external routes to flood in the area. The devices in the area access a network outside the area and outside the OSPF AS through the default route.

Table 7-4 Configuring an OSPF Stub Area

• ---

  A backbone area cannot be configured as a Stub area.
• All devices in one Stub area must be configured as Stub areas, because devices with different area types cannot form neighbor relations.

  ---

Configure an OSPF Virtual Link

The non-backbone areas in OSPF must synchronize and exchange data through the backbone area. Therefore, all non-backbone areas must keep connected with the backbone area.

If the requirement fails to be met in certain cases, you can solve the problem by configuring a virtual link. After configuring a virtual link, you can configure an authentication mode for the virtual link and modify the Hello interval. The meanings of the parameters are the same as the meanings of the parameter of common OSPF interfaces.

Table 7-5 Configuring an OSPF Virtual Link

• ---

  A virtual link must be configured between two ABRs.
• Two ABRs on which the virtual link is configured must be in the same public area. This area is also called the transit area of the virtual link.
• The transit area of a virtual link must not be a Stub area or NSSA area.

  ---