Routing
What is a Router?
A router is a device that forwards packets between different networks. It works at Layer 3 (Network Layer) of the OSI model and makes decisions based on the destination IP address of each packet. In simple terms, a router connects separate networks, such as LANs, WANs, or the Internet, and finds the best way for data to travel. Router core functions is:
Packet forwarding: Routes traffic from one network to another.
Routing table management: Keeps track of network paths and their metrics.
NAT/PAT: Maps private IPs to public IPs in situations with limited IPv4 addresses.
Security: Applies ACLs, firewall rules, or filters out malicious traffic.
Dynamic adjustment: Changes routing in real time if the network topology changes.
Diagram
This shows the main role of a router: connecting different networks.
+--------------------+ +----------+ +----------+ +-----------------+
| LAN |----------| Router A |----------| Internet |----------| Router B |
| 192.168.1.0/24 | | | | | | LAN 10.0.0.0/24 |
+--------------------+ +----------+ +----------+ +-----------------+Static vs Dynamic Routing
The main difference is how the router learns and keeps track of the paths in its routing table.
Static Routing
Static routes are set up manually by a network administrator. A static route defines the destination network, subnet mask, and the next-hop IP address or outgoing interface.
Advantages:
Predictable, no protocol overhead
Simple to troubleshoot in small networks
Secure (cannot be modified by other routers)
Disadvantages:
Manual updates required for topology changes
Not scalable for large networks
Can lead to human errors
Example Static Routing Table:
192.168.2.0
255.255.255.0
192.168.1.2
eth0
10.0.0.0
255.0.0.0
192.168.1.3
eth1
0.0.0.0
0.0.0.0
192.168.1.1
eth0
Dynamic Routing
Dynamic routing uses protocols to automatically share network information. Routers communicate to discover routes, share updates, and adapt to network failures.
Advantages:
Scales well in large networks
Changes automatically with topology modifications
Lowers administrative workload
Disadvantages:
Uses CPU, memory, and bandwidth for routing updates
Can share misconfigurations if not well designed
Requires careful protocol selection for stability
Routing Protocols
Routers use specific protocols to exchange routing information automatically
RIP
Distance-vector
Hop count (max 15)
Small networks
OSPF
Link-state
SPF algorithm, fast convergence
Enterprise networks
BGP
Path-vector
Policies & AS paths
Internet backbone
EIGRP
Hybrid
Bandwidth, delay, reliability
Cisco enterprise networks
RIP (Routing Information Protocol)
Type: Distance-vector protocol.
Operation: Sends its entire routing table to neighbors every 30 seconds.
Metric: Uses hop count (the number of routers to cross).
Limitation: Maximum of 15 hops; slow to respond and prone to loops. Best for small, simple networks.
OSPF (Open Shortest Path First)
Type: Link-state protocol.
Operation: Routers exchange LSAs (Link-State Advertisements) to create a complete map of the network topology in a Link-State Database.
Algorithm: Uses Dijkstra's SPF algorithm to find the shortest path to each network.
Benefit: Fast response and efficient path selection, making it good for medium and large networks.
BGP (Border Gateway Protocol)
Type: Path-vector protocol.
Operation: Used to connect different Autonomous Systems (AS) on the Internet. An AS is a network under a single administrative control (e.g., an ISP).
Decision Basis: Routing is based on policies, rules, and paths, not just simple metrics.
Role: The main protocol that connects the global Internet.
Routing Tables
A router's routing table contains the information needed for forwarding decisions.
Destination network: Where the packet should go
Subnet mask: Network part of the IP
Next hop: Router to send traffic to
Interface: Outgoing port
Metric: Cost of route (lower values are preferred)
Example Routing Table:
192.168.1.0
255.255.255.0
0.0.0.0
eth0
0
192.168.2.0
255.255.255.0
192.168.1.2
eth1
10
10.0.0.0
255.0.0.0
192.168.1.3
eth1
5
0.0.0.0
0.0.0.0
192.168.1.1
eth0
100
The route 0.0.0.0/0 is the default route, covering all traffic to unknown destinations, usually directing it to the Internet.
Packet Path Example:
A sends a packet to PC B. It sees that 10.0.0.5 is not part of its local network (192.168.1.0/24), so it forwards the packet to its default gateway, Router A.
Router A checks its routing table. It finds an entry for the network 10.0.0.0/8 and sees it can reach that via Router B. It forwards the packet to Router B.
Router B gets the packet, looks at its own table, and sees that the specific PC (10.0.0.5) is on a network directly connected to it. It delivers the packet straight to PC B.
OSPF Routing Example
Scenario: 3 routers forming a backbone
Router A: 192.168.1.0/24
Router B: 192.168.2.0/24
Router C: 10.0.0.0/24
OSPF Steps:
Neighbor Discovery: Routers exchange Hello packets to discover OSPF neighbors on their directly connected networks
Link-State Advertisement (LSA): Routers send LSAs to describe their connected networks and the state of their links
Link-State Database (LSDB): Each router gathers all LSAs into a synchronized LSDB, forming a complete map of the OSPF area
SPF Calculation: Each router runs the Dijkstra SPF algorithm on the LSDB to independently calculate the shortest paths to every network
Routing Table Update: The best paths from the SPF tree are added to the router's IP routing table.
Router B Table:
192.168.1.0
Router A
1
192.168.2.0
Local
0
10.0.0.0
Router C
1
Routing Loops and Convergence
Routing Loops: Happen when a packet is endlessly passed between routers in a cycle, never reaching its destination. Prevention methods are important:
TTL (Time-to-Live): A counter in the IP header that decreases with each hop; the packet is discarded if TTL reaches zero
Split Horizon: A rule that stops a route from being advertised back out the interface it was learned from
Route Poisoning: Marking a failed route as "unreachable" and advertising it to prevent other routers from using it
Convergence Delay: The time taken for all routers in a network to update their routing tables after a topology change. Fast convergence is vital for network stability. OSPF converges faster than RIP because it uses a more efficient update process.
Traceroute Example
Command: traceroute 10.0.0.5
This shows the path a packet takes and is useful for troubleshooting routing issues.
Hybrid Routing Networks
Real-world networks often mix static and dynamic routing to take advantage of both approaches:
Static Routes are used for important, stable paths (e.g., to a central server farm or a specific security subnet), where predictability is essential
Dynamic Routes (OSPF, EIGRP) serve the core network and inter-office links to provide automatic adjustment and scalability
This hybrid setup minimizes protocol overhead when possible while ensuring reliability and flexibility across the network.
Hybrid Routing Example
In this example, a static route may be used inside the local LAN for a specific server, while BGP is used to exchange routes with the Internet and OSPF is used within the larger corporate network.
Last updated