OSPF inter area route

[Manish Prasad]

How inter area route is calculated in OSPF routing process?

As per my understanding , when hello packet is exchanged between neighbors which contain a Area ID field.Through this field routes are determined whether they belong to which area.

If my understanding is correct then here is the scenario.

R1(area 0 and area 3)--R0(area 0 and area 3)----R3(Area 3)

R1

interface Loopback10

ip address 10.10.10.10 255.255.255.255

ip ospf 28 area 3

router ospf 28

router-id 2.2.2.2

log-adjacency-changes

network 10.10.0.2 0.0.0.0 area 0

network 10.10.2.1 0.0.0.0 area 0

network 10.10.10.10 0.0.0.0 area 3

R1 routing table

10.0.0.0/8 is variably subnetted, 10 subnets, 2 masks

C 10.10.10.10/32 is directly connected, Loopback10

R0 routing table

O IA 10.10.10.10/32 [110/65] via 10.10.0.2, 00:35:19, Serial0/0

R3 routing table

10.0.0.0/8 is variably subnetted, 10 subnets, 2 masks

O IA 10.10.10.10/32 [110/129] via 10.3.1.1, 00:28:57, Serial0/2

So my question is why R3 get route of 10.10.10.10 tagged with inter area(IA) as it only belongs to area 3 ?

Thanks

Prasad

[Giuseppe Larosa]

Hello Prasad,

>> So my question is why R3 get route of 10.10.10.10 tagged with inter area(IA) as it only belongs to area 3 ?

Because your area 3 is divided in more parts:

the only path(s) between R0 and R1 are in area 0.

Should a path in area 3 exist between R1 and R0 and between R0 and R3 both R0 and R3 would see an intra area route.

Notice that this would happen also if a parallel path in area 0 exists like the current one.

In more details:

R1 sends out an LSA type 3 for net 10.10.10.10 that is received by R0 and propagated to R3.

actually the LSA type 3 has no field that says what is the "original" area-id of the prefix R0 just knows that comes from a non-zero area.

see

http://www.ietf.org/rfc/rfc2328.txt

section 12.4.3. Summary-LSAs

Hope to help

Giuseppe

[Manish Prasad]

Thank you Giuseppe ,now i understand its all because the LSA type 3 message.

One more question..

R1(area 3-OSPF Process ID-12)---R2(area3-process ID-12)(area 0-process id 28)--R3(area 0-process id-28)

R2 acting as a ABR with its two interface running two different process ids.

As i know that OSPF process id contains different database table and they cant be synchronized .The only way to get route of two different process id is to do redistribution between two different process id.

So my question is, do i have to redistribute OSPF process id within each other on ABR(R2) to get the different areas route in its routing table?

As only one interface can talk with only one process id with its neighbor.

Thanks

Prasad

[Giuseppe Larosa]

Hello Prasad,

your understanding is correct:

actually R2 having one interface in area 3 process 12 and one in area 0 process 28 is not an ABR:

you can see with

sh ip ospf 12 database

sh ip ospf 28 database

formally R2 is an internal router in area 3 for process 12 and a backbone router for process 28.

the two databases are completely separated and are not communicating.

The only way to pass routes between the two processes is to use redistribution.

But if you do so R2 becomes an ASBR because it injects external routes in each OSPF domain (if redis is mutual = bidirectional)

The only way to have R2 to act as ABR is to have the two interfaces on the same process.

Notice that because OSPF version2 hello packets don't carry the OSPF process-id the same ip subnet/interface is not supported at the same time in the same area in both processes: if you try to do this you get that the prefix is not advertised in any OSPF process and you cannot build adjacencies over it.

Hope to help

Giuseppe

[Manish Prasad]

Do the routing table of router R2 will contain all the routes of both the OSPF process id which has been learned by both of its neighbor router? If yes then Why?

[Giuseppe Larosa]

Hello Prasad,

yes R2 IP routing table can contain IP prefixes provided by OSPF process 12 and OSPF process 28.

Each OSPF process is separated (ships in the night) and proposes IP prefixes to the IP routing table mantainer process.

The ip routing table mantainer accepts prefixes from multiple OSPF processes.

If there is an IP prefix proposed by both processes the first OSPF process to suggest it sees its prefix installed in the routing table regardless of OSPF route type: the mantainer looks for AD and metric only.

In this way multiple OSPF processes are supported on the same router and this can be used in some scenarios (for example two different customer networks or two different departments in an enterprise).

But a router connected to R2 in area 3 process 12 doesn't receive any information about OSPF 28 area 0 prefixes.

Hope to help

Giuseppe