...

Link state database describes the routers and links that interconnect them and are appropriate for forwarding. It also contains the cost (metric) of each link. This metric is used to calculate the shortest path to the destination network.
Each router can advertise a different cost for the router's own link direction, making it possible to have asymmetric links (packets to the destination travel over one path, but the response travels a different path). Asymmetric paths are not very popular, because it makes it harder to find routing problems.
The Cost in RouterOS is set to 10 on all interfaces by default. Value value of the cost can be changed in the OSPF interface template configuration menu, for example, to add an ether2 interface with a cost of 100:

...

As you can see from the image above multiple shortest paths have been found to the 172.16.1.0 network, allowing load balancing of the traffic to that destination called equal-cost multipath (ECMP). After the shortest-path tree is built, a router starts to build the routing table accordingly. Networks are reached consequently to the cost calculated in the tree.Routing table calculation looks quite simple, however, when some of the OSPF extensions are used or OSPF areas are calculated, routing calculation gets more complicated.-path tree is built, a router starts to build the routing table accordingly. Networks are reached consequently to the cost calculated in the tree.

Routing table calculation looks quite simple, however, when some of the OSPF extensions are used or OSPF areas are calculated, routing calculation gets more complicated.

Forwarding Address

OSPF router can set the forwarding-address to something other than itself which indicates that an alternate next-hop is possible. Mostly forwarding address is set to 0.0.0.0 suggesting that the route is reachable only via the advertising router.

The forwarding address is set in LSA if the following conditions are met:

• OSPF must be enabled on the next-hop interface
• Next-hop address falls into the network provided by OSPF networks

A router that receives such LSA can use a forwarding address if OSPF is able to resolve the forwarding address. If forwarding address is not resolved directly - router sets nexthop for forwading address from LSA as a gateway, if forwarding address is not resolved at all - the gateway will be originator-id. Resolve happens only using OSPF instance routes, not the whole routing table.

Let's look at the example setup below:

Image Added

Router R1 has a static route to the external network 192.168.0.0/24. OSPF is running between R1, R2, and R3, and the static route is distributed across the OSPF network.

The problem in such a setup is obvious, R2 can not reach the external network directly. Traffic going to the LAN network from R2 will be forwarded over the router R1, but if we look at the network diagram we can see that more R2 can directly reach the router where the LAN network i located.

So knowing the forwarding address conditions, we can make router R1 to set the forwarding address. We simply need to add 10.1.101.0/24 network to OSPF networks in the router's R1 configuration:

/routing/ospf/interface-template add area=backbone_v2 networks=10.1.101.0/24

Now lets verify that forwarding address is actually working: 

[admin@r2] /ip/route> print where dst-address=192.168.0.0/24
Flags: D - DYNAMIC; A - ACTIVE; o, y - COPY
Columns: DST-ADDRESS, GATEWAY, DISTANCE
    DST-ADDRESS       GATEWAY            DISTANCE
DAo 192.168.0.0/24    10.1.101.1%ether1       110

On all OSPF routers you will see LSA set with forwarding address other than 0.0.0.0 

[admin@r2] /routing/ospf/lsa> print where id=192.168.0.0
Flags: S - self-originated, F - flushing, W - wraparound; D - dynamic 

 1  D instance=default_ip4 type="external" originator=10.1.101.10 id=192.168.0.0 
      sequence=0x80000001 age=19 checksum=0xF336 body=
        options=E
        netmask=255.255.255.0
        forwarding-address=10.1.101.1
        metric=10 type-1
        route-tag=0

Neighbour Relationship and Adjacency

...

...

Automatic neighbor discovery by multicasting or broadcasting Hello packets. Less bandwidth usage compared to other subnet types. On the broadcast segment, there are n*(n-1)/2 neighbor relations, but those relations are maintained by sending only n Hellos. If the broadcast has the multicast capability, then OSPF operates without disturbing non-OSPF nodes on the broadcast segment. If the multicast capability is not supported all routers will receive broadcasted Hello packet packets even if the node is not an OSPF router.

...

Non-broadcast multiaccess (NBMA) segments are similar to broadcast. Support more than two routers, the only difference is that NBMA does not support a data-link broadcast capability. Due to this limitation, OSPF neighbors must be discovered initially through configuration. On RouterOS static neighbor configuration is set in the /routing ospf static-neighbor menu. To reduce the amount of Hello traffic, most routers attached to the NBMA subnet should be assigned a Router Priority of 0 (set by default in RouterOS). Routers that are eligible to become Designated Routers should have priority values other than 0. It ensures that during the election of DR and BDR Hellos are sent only to eligible routers.

...

Before database synchronization can begin, a hierarchy order of exchanging information must be established, which determines which router sends Database Descriptor (DD) packets first (Master). The master router is elected based on the highest priority and if priority is not set then the router ID will be used. Note that it is a router priority-based relation to arranging the exchanging data between neighbors which does not affect DR/BDR election (meaning that DR does not always have to be Master).

...

The image on the right illustrates adjacency formations on broadcast subnets. Routers R1 and R2 are Designated Routers and Backup Designated routers respectively. For example, if R3 wants to flood Link State Update (LSU) to both R1 and R2, a router sends LSU to the IP multicast address AllDRouters (224.0.0.6) and only DR and BDR listen to this multicast address. Then Designated Router sends LSU addressed to AllSPFRouters, updating the rest of the routers.

...

Database synchronization on NBMA networks is similar to that on broadcast networks. DR and BDR are elected, databases initially are exchanged only with DR and BDR routers and flooding always goes through the DR. The only difference is that Link State Updates must be replicated and sent to each adjacent router separately.

...

A distinctive feature of OSPF is the possibility to divide AS into multiple routing Areas which contain their own set of neighbors.
Imagine a large network with 300+ routers and multiple links between them. Whenever link flaps or some other topology change happens in the network, this change will be flooded to all OSPF devices in the network resulting in a quite heavy load on the network and even downtime since network convergence may take some time for such a large network. 

A large single-area network can produce serious issues:

...

OSPF area has unique 32-bit identification (Area ID) and the area with an Area ID of 0.0.0.0 (called the Backbone area) is the main one where any other area should connect. Routers that connect to more than one area are called ABR (Area Border Routers), and their main responsibility is summarization and update suppression between connected areas. The router connecting to another routing domain is called ASBR (Autonomous System Boundary Router).

Each area has its own link-state database, consisting of router-LSAs and network-LSAs describing how all routers within that area are interconnected. Detailed knowledge of the area's topology is hidden from all other areas; router-LSAs and network-LSAs are not flooded beyond the area's borders. Area Border Routers (ABRs) leak addressing information from one area into another in OSPF summary-LSAs. This allows one to pick the best area border router when forwarding data to destinations from another area and is called intra-area routing.

Routing information exchange between areas is essentially a Distance Vector algorithm and to prevent algorithm convergence problems, such as counting to infinity, all areas are required to attach directly to the backbone area making a simple hub-and-spoke topology. AreaThe area-ID of the backbone area is always 0.0.0.0 and can not be changed.

...

Simple multi-area network using default area. In this example, all networks from area1 are flooded to the backbone and all networks from the backbone are flooded to area1.
R1:

/ip address add address=10.0.3.1/24 interface=ether1
/ip address add address=10.0.2.1/24 interface=ether2
/routing ospf instance
add name=v2inst version=2 router-id=1.0.0.1
/routing ospf area
add name=backbone_v2 area-id=0.0.0.0 instance=v2inst
add name=area1 area-id=1.1.1.1 type=default instance=v2inst
/routing ospf interface-template
add networks=10.0.2.0/24 area=backbone_v2
add networks=10.0.3.0/24 area=area1

...

The main purpose of stub areas is to keep such areas from carrying external routes. Routing from these areas to the outside world is based on a default route. Stub A stub area reduces the database size inside an area and reduces the memory requirements of routers in the area.

...

Since redistribution is controlled by "originate-default" and "redistribute" parameter, it introduces some corner-cases for default route filtering. 

• if redistribute is enabled, then pick all routes matching redistribute parameters
• If originate-default=never, a default route will be rejected
• run selected routes through out-select-chain (if configured)
• run selected routes through out-filter-chain (if configured)
• if originate-default is set to always or if-installed:
  • OSPF creates a fake default route without attributes;
  • runs this route through out-filter-chain where attributes can be applied, but action is ignored (always accept);

...

It is better to summarise in the direction of the backbone. That way the backbone receives all the aggregated routes and injects them into other areas already summarisedsummarized. There are two types of summarization: inter-area and external route summarization.

...

Inter-area route summarization can be configured from from the OSPF area range menu.

Let's consider that we have two areas backbone and area1, area1 has several /24 routes from the 10.0.0.0/16 range and there is no need to flood the backbone area with each /24 subnet if it can be summarisedsummarized. On the router connecting area1 with the backbone we can set up the area range:

/routing ospf area range
add prefix=10.0.0.0/16 area=area1 advertise=yes cost=10

...

There are two common scenarios when virtual links can be used:

• to glue together the fragmented backbone area
• to connect remote are without direct connection to the backbone

...

Sub-menu: /routing/ospf/instance

Notes

OSPF protocol supports two types of metrics:

• type1 - OSPF metric is the sum of the internal OSPF cost and the external route cost
• type2 - OSPF metric is equal only to the external route cost.

Area

Sub-menu: /routing/ospf/area

...

...

...

...

...