Network Configuration: IP, Subnetting, Routing, Topologies & Troubleshooting

TAKEAWAYS:

• Understand IPv4/IPv6 addressing, subnetting (CIDR, VLSM), and how to calculate subnets, hosts, and default gateways.
• Differentiate between static routing (manual) and dynamic routing (RIP, OSPF, BGP) and their use cases.
• Recognize network topologies (star, bus, mesh, hybrid) and their pros/cons in real-world deployments (e.g., NTC’s fiber backbone vs. Pathao’s star topology).
• Master troubleshooting commands (ping, traceroute, nslookup, ipconfig) and interpret their outputs.
• Configure Linux/Windows interfaces (IP, subnet mask, gateway) and verify connectivity.
• Apply NAT (Network Address Translation) to conserve public IPs in scenarios like Daraz’s e-commerce servers.

1. IP Addressing: IPv4 and IPv6

1.1 IPv4 Basics

IPv4 uses 32-bit addresses divided into network and host portions. The division is defined by the subnet mask (e.g., 255.255.255.0 for /24).

• Private vs. Public IPs:
  • Private: 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16 (used internally, e.g., home networks, corporate LANs).
  • Public: Assigned by IANA (e.g., 8.8.8.8 for Google DNS).
• Example: A bank like Nabil Bank uses private IPs (192.168.x.x) for internal systems but maps them to public IPs for internet-facing services (e.g., online banking).

classDiagram
    class IPv4 {
        +32-bit address
        +Divided by subnet mask
        +Private ranges: 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16
        +Public ranges: Assigned by IANA
    }
    class IPv6 {
        +128-bit address
        +Hexadecimal notation (e.g., 2001:0db8::1)
        +No NAT needed (plentiful addresses)
        +Autoconfiguration (SLAAC)
    }
    IPv4 --> "Uses" SubnetMask
    IPv6 --> "Replaces" IPv4

1.2 IPv6: The Future

IPv6 uses 128-bit addresses, written in hexadecimal (e.g., 2001:0db8:85a3::8a2e:0370:7334).

• Advantages:
  • No NAT required (eliminates address exhaustion).
  • Built-in security (IPsec).
  • Simplified header (faster routing).
• Example: Nepal Telecom (NTC) is gradually migrating its backbone to IPv6 to support future growth and IoT devices.

1.3 Subnetting: Dividing Networks

Subnetting splits a large network into smaller subnets for efficiency.

• Key Terms:
  • CIDR Notation: /24 means 24 network bits (e.g., 192.168.1.0/24).
  • Subnet Mask: Binary representation of network bits (e.g., /24 = 11111111.11111111.11111111.00000000).
  • Boradcast Address: All host bits set to 1 (e.g., 192.168.1.255).
  • Default Gateway: Router interface connecting subnets (e.g., 192.168.1.1).

Worked Example: Subnetting 192.168.1.0/24

Problem: Subnet 192.168.1.0/24 into 4 equal subnets for a university’s 4 departments. Solution:

1. Borrow 2 bits (since $2^2=4$ subnets).
2. New prefix: /26 (24 + 2).
3. Subnets:
   • 192.168.1.0/26 (dept A)
   • 192.168.1.64/26 (dept B)
   • 192.168.1.128/26 (dept C)
   • 192.168.1.192/26 (dept D)
4. Hosts per subnet: $2^{10}-2=1022$ (10 host bits).

Real-World Tie: Tribhuvan University’s central IT department uses subnetting to isolate networks for faculties, libraries, and admin offices, improving security and performance.

2. Network Topologies

Topologies define how devices connect in a network. Choose based on cost, scalability, and fault tolerance.

graph LR
    A["Star Topology"] --> B["Central Switch"]
    C["Device 1"] --> B
    D["Device 2"] --> B
    E["Device 3"] --> B

Real-World Example:

• Daraz’s Order Processing: Uses a hybrid topology—star for internal servers (e.g., inventory, payments) and mesh for data centers across Kathmandu/Lalitpur for redundancy.
• Nepal’s Traffic Routes: Think of Kathmandu’s ring road as a ring topology (all roads connect in a loop), but with star connections (flyovers acting as hubs for local roads).

3. Routing: Static vs. Dynamic

Routing determines how data travels between networks.

3.1 Static Routing

• Manual configuration of routes by an admin.
• Use Case: Small networks (e.g., home router with 1-2 connections).
• Example Command (Linux):

  ip route add 192.168.2.0/24 via 192.168.1.2
  

• Disadvantage: Must update manually if topology changes.

3.2 Dynamic Routing

Protocols automatically exchange routes between routers.

• RIP (Routing Information Protocol): Simple, but slow (max 15 hops).
• OSPF (Open Shortest Path First): Fast, hierarchical, used in large networks.
• BGP (Border Gateway Protocol): Used on the internet (e.g., between ISPs like NTC and Ncell).

sequenceDiagram
    participant RouterA as Router A
    participant RouterB as Router B
    RouterA->>RouterB: Hello (OSPF)
    RouterB-->>RouterA: Link State Advertisement (LSA)
    RouterA->>RouterB: ACK
    RouterA->>RouterB: Updated Routing Table
    Note over RouterA,RouterB: Dynamic routes calculated automatically.

Real-World Example:

• Nepal’s Internet Backbone: NTC and Ncell use BGP to exchange routes between their ISP networks, ensuring seamless connectivity across Nepal.
• WhatsApp’s Data Centers: Use OSPF internally to route messages between servers in Singapore, London, and São Paulo with minimal delay.

4. Network Configuration Commands

Master these commands to configure and troubleshoot networks.

Worked Example: Troubleshooting a Daraz Order Failure Scenario: A customer in Pokhara reports that their Daraz order isn’t processing. Steps:

1. Ping the Daraz server:

   ping daraz.com.np
   

   • If no reply, check internet connection or ISP (NTC/Ncell).
   • If reply, proceed to step 2.
2. Traceroute to identify bottlenecks:

   traceroute daraz.com.np
   

   • Output shows delays at 10.0.0.1 (NTC router in Kathmandu). This suggests a congestion issue on NTC’s backbone.
3. Check DNS resolution:

   nslookup daraz.com.np
   

   • If DNS fails, the issue is with Cloudflare’s DNS servers (Daraz’s DNS provider).

Solution: Contact NTC support or switch to Ncell if NTC’s backbone is saturated.

5. NAT (Network Address Translation)

NAT allows multiple devices on a private network to share a single public IP.

• Types:
  • PAT (Port Address Translation): Maps private IPs to public IP + port (used in home routers).
  • Static NAT: One-to-one mapping (e.g., a web server’s private IP to a public IP).
• Example: Your home router uses PAT to let 10 devices (e.g., phones, laptops) access the internet via one public IP from NTC.

graph TD
    A["Private LAN\n(192.168.1.x)"] -->|"PAT"| B["Router"]
    B -->|"Public IP: 203.123.45.67"| C["Internet"]

Real-World Example:

• Khalti’s Payment Gateway: Uses static NAT to map its internal servers (e.g., 10.0.0.10) to a fixed public IP (203.123.45.78) for secure transactions.
• Your Smartphone: Uses PAT when connecting to Wi-Fi—your private IP (192.168.1.10) is translated to your ISP’s public IP.

6. Configuring Interfaces (Linux/Windows)

Linux: Using ip or nmcli

# Assign IP to eth0
sudo ip addr add 192.168.1.100/24 dev eth0
sudo ip link set eth0 up

# Verify
ip a show eth0

Output:

2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    inet 192.168.1.100/24 brd 192.168.1.255 scope global dynamic eth0

Windows: Using ipconfig and GUI

1. Open Network Connections (ncpa.cpl).
2. Right-click adapter → Properties → IPv4 → Enter IP, subnet, gateway.
3. Verify:

   ipconfig /all
   

Real-World Example:

• Nepal Stock Exchange (NEPSE): Configures static IPs on trading servers (e.g., 10.10.10.1/24) with a gateway to the internet (203.123.45.1) to ensure low-latency trading.

7. Network Troubleshooting Flowchart

flowchart TD
    A["Network Issue Detected"] --> B["Is device powered on?"]
    B -->|"No"| C["Power on device"]
    B -->|"Yes"| D["Check physical connections"]
    D -->|"Cable loose?"| E["Reseat cable"]
    D -->|"Yes"| F["Check IP configuration"]
    F -->|"Wrong IP/subnet?"| G["Reconfigure IP"]
    F -->|"Correct"| H["Ping gateway"]
    H -->|"No reply"| I["Check router/firewall"]
    H -->|"Reply"| J["Ping external host"]
    J -->|"No reply"| K["Check DNS/ISP"]
    J -->|"Reply"| L["Issue resolved"]

Example: Your laptop can’t access YouTube.

1. Ping 8.8.8.8: Works → ISP is fine.
2. Ping youtube.com: Fails → DNS issue.
3. Use 8.8.8.8 as DNS: Works → Fix your router’s DNS settings.

In the Real World

1. eSewa’s Payment Routing:

   • Uses static routing to direct transactions between its servers in Kathmandu and payment gateways (e.g., IME Pay, Global IME Bank). If a route fails, admins manually update tables.
   • NAT: Maps eSewa’s internal servers (10.0.x.x) to a public IP for secure HTTPS connections.

2. Pathao’s Driver Dispatch System:

   • Star Topology: All driver apps connect to central Pathao servers in Thapathali. If the central server fails, no drivers can dispatch orders (single point of failure).
   • Dynamic Routing: Uses OSPF-like logic to reroute orders to the nearest available server if a data center goes down.

3. NTC’s Fiber Backbone:

   • Mesh Topology: NTC’s fiber optic cables connect major cities (Kathmandu, Pokhara, Biratnagar) in a mesh for redundancy. If one link fails (e.g., landslide in Dhulikhel), traffic reroutes automatically.
   • IPv6 Deployment: NTC is rolling out IPv6 to support IoT devices (e.g., smart meters) without NAT overhead.

Exam Tip

1. Subnetting Questions:

   • Always show binary conversion and subnet/broadcast/host calculations.
   • Example: Given 172.16.0.0/20, calculate subnets and usable hosts. Draw a table:

     Subnet   | Range               | Usable Hosts
     --------------------------------------------
     172.16.0.0/20 | 172.16.0.1–172.16.15.254 | 4094
     

2. Routing Protocols:

   • RIP: Max 15 hops, uses hop count as metric.
   • OSPF: Uses cost (based on link speed), hierarchical (areas).
   • BGP: Used between autonomous systems (e.g., NTC ↔ Ncell).

3. Troubleshooting:

   • Step-by-step approach: Device → Cable → IP → Gateway → DNS → ISP.
   • Commands: Know when to use ping, traceroute, nslookup, and netstat.

4. NAT/PAT:

   • PAT: Many private IPs → one public IP + ports (e.g., home router).
   • Static NAT: One-to-one (e.g., web server).

5. Topologies:

   • Star: Easy to manage (e.g., corporate LANs).
   • Mesh: Redundant (e.g., NTC backbone).
   • Hybrid: Balanced (e.g., Daraz’s servers).

Common Pitfalls:

• Forgetting to exclude network/broadcast addresses when counting hosts.
• Confusing subnet mask (255.255.255.0) with wildcard mask (0.0.0.255).
• Misconfiguring default gateways (must be in the same subnet).

Pathao’s central server architecture. (Image: Umapathy, CC BY-SA 3.0, via Wikimedia Commons) NTC’s fiber mesh connecting major cities. (Image: Simon Villeneuve, CC BY-SA 3.0, via Wikimedia Commons)