OSPF (Part 2)
Day 27
Last updated
Day 27
Last updated
OSPF's metric is known as cost. It is automatically calculated by dividing a reference bandwidth (default is 100 Mbps) value by the interface's bandwidth. All values less than 1 are converted to 1. To change the default reference bandwidth, enter the OSPF config mode and enter auto-cost reference-bandwidth followed by the value in Mbps. The reference bandwidth on all OSPF routers in the network should be the same.
The OSPF cost to a destination is the total cost of the exit interfaces along the path. Loopback interfaces have a cost of 1. To manually configure the cost of an interface, enter the interface config mode and use the command ip ospf cost followed by the cost. Another way to change the cost is to change the interface bandwidth. It doesn't change the actual speed at which the interface operates. It is just used to calculate the OSPF cost. This method is not recommended. The command is bandwidth followed by the value in Kbps which is entered from the interface config mode.
The default hello timer is 10 seconds on an Ethernet connection. Hello messages are multicast to 224.0.0.5. OSPF messages are encapsulated in an IP header with a value of 89 in the Protocol field. For OSPF routers to become neighbours, they have to go through several states.
OSPF is activated on R1's G0/0 interface
It sends an OSPF Hello message to 224.0.0.5
It doesn't know about any neighbours yet
When R2 receives the Hello packet, it adds an entry for R1
Hello packet is received but R2's ID is not in the Hello packet
R1 still doesn't know about R2 yet
Both routers receive Hello messages with their own RID included
If both routers reach this state, it means they are ready to share LSAs to build a common LSDB
DR (Designated Router) and BDR (Backup Designated Router) are elected at this point
Both routers prepare to exchange information about their LSDB
The router with the higher RID becomes the Master and initiates the exchange. The router with the lower RID becomes the Slave
To decide between the Master and Slave, they exchange DBD (Database Description) packets
The routers exchange DBDs which contain the list of LSAs in their LSDB
These DBDs do not contain detailed information about LSAs, just basic information
The routers compare the information they receive with their own LSDB to determine which LSA they need to request
Routers send LSR (Link State Request) messages to request any LSAs they don't have
LSAs are sent in LSU (Link State Update) messages
The routers then send LSAck (Link State Acknowledgment) messages to acknowledge that they received the LSAs
The routers have a full OSPF adjacency and identical LSDBs
They continue to send Hello packets (every 10 seconds by default) to maintain the neighbour adjacency
Every time a Hello packet is received, a Dead timer (40 seconds by default) is reset
If no Hello packet is received and Dead timer reaches 0 (zero), the neighbour is removed
The routers keep sharing LSAs as the network changes
There are 5 message types in OSPF:
1 - Hello
2 - DBD
3 - LSR
4 - LSU
5 - LSAck
To activate OSPF on a particular interface, enter the interface config mode and use the command ip ospf followed by the process ID followed by area and area number.
To configure all interfaces as passive by default, enter the command passive-interface default from the OSPF config mode. To activate the interface, just enter no passive-interface followed by the interface number.
