Network Admin Basics: Models, Roles, Tools & Open-Source Systems

1. What is Network Administration?

Network administration is the design, implementation, maintenance, and troubleshooting of computer networks to ensure reliable, secure, and efficient communication between devices. It involves managing hardware (routers, switches, cables), software (operating systems, protocols), and policies (security, access control).

Key Roles in Network Administration

Network administrators work alongside system administrators (sysadmins) and network engineers, but their focus differs:

2. The OSI and TCP/IP Models: How Networks Are Structured

Networks are organized into layers to simplify troubleshooting and standardization. Two models are critical:

A. OSI Model (7 Layers)

A theoretical framework for understanding how data moves through networks. Each layer has a specific function:

graph TD
    A["Application (Layer 7)"] --> B["Presentation (Layer 6)"]
    B --> C["Session (Layer 5)"]
    C --> D["Transport (Layer 4)"]
    D --> E["Network (Layer 3)"]
    E --> F["Data Link (Layer 2)"]
    F --> G["Physical (Layer 1)"]

Worked Example: Sending an Email from TU Student to PU Professor

1. Application Layer: Your email client (e.g., Gmail) uses SMTP to send data.
2. Transport Layer: TCP breaks the email into segments and assigns port 25 (SMTP).
3. Network Layer: IP adds a source (TU IP) and destination (PU IP).
4. Data Link Layer: Ethernet frames the packet with MAC addresses (e.g., 00:1A:2B:3C:4D:5E).
5. Physical Layer: The signal travels via fiber optic cables (NTC backbone) or Wi-Fi.

B. TCP/IP Model (4 Layers)

A practical model used in real-world networks (e.g., the Internet). It combines OSI layers:

graph TD
    A["Application"] --> B["Transport"]
    B --> C["Internet"]
    C --> D["Network Access"]

Why TCP/IP Matters in Nepal

• Nepal’s Internet (NTC, Worldlink): Uses TCP/IP for routing traffic between Kathmandu and Pokhara.
• eSewa Payments: Relies on TCP/IP for secure transactions between your phone and the bank’s server.

3. Network Administration Tools: Hardware and Software

Admins use a mix of hardware devices and software tools to manage networks.

A. Hardware Tools

B. Software Tools

Worked Example: Troubleshooting Slow Internet at TU

1. Ping Google: ping google.com → High latency? Check ISP (NTC) or local router.
2. Traceroute: traceroute google.com → Identifies where packets drop (e.g., between Kathmandu and Pokhara).
3. Wireshark: Captures packets to see if DNS queries are timing out (e.g., NTC’s DNS server issue).

4. Open-Source Network Monitoring: Why It’s Used

Open-source tools are free, customizable, and widely adopted by companies (even Google uses some!). In Nepal:

• eSewa uses open-source Nginx for load balancing.
• NTC monitors its backbone with Zabbix (open-source).
• TU’s IT team uses Wireshark for debugging.

Example: Monitoring Daraz’s Order Queue

Daraz’s servers use Nagios to:

1. Track HTTP response times (e.g., checkout page).
2. Alert admins if CPU > 90% (slow orders).
3. Log failed transactions (e.g., payment gateways).

Mermaid Diagram: Open-Source Monitoring Workflow

sequenceDiagram
    participant User
    participant Nagios
    participant Server
    participant Admin
    User->>Server: Places order (HTTP request)
    Server-->>Nagios: Sends metrics (CPU, latency)
    Nagios->>Admin: Alert if latency > 2s
    Admin->>Server: Restarts service

5. Network Topologies: How Devices Are Connected

The physical or logical arrangement of devices affects performance and fault tolerance.

Worked Example: Kathmandu Traffic Routes (Like Network Routing!)

• Star: All roads lead to Thamel (central hub). If Thamel is blocked, traffic jams everywhere.
• Mesh: Ring Road connects multiple routes (e.g., Swoyambhu → Budhanilkantha → Koteshwor). If one road is closed, traffic reroutes.

6. Network Administration vs. System Administration

When to Use Which?

• Network Admin: Fixing slow internet at home (router config).
• Sysadmin: Installing Linux on a new server for a bank’s database.

In the Real World

1. eSewa Payments

   • Uses TCP/IP for secure transactions between your phone and the bank’s server.
   • Relies on firewalls to block fraudulent requests.
   • Monitors performance with open-source tools like Nagios.

2. NTC’s Internet Backbone

   • Uses OSI Layer 3 (IP routing) to direct traffic between Kathmandu and Pokhara.
   • Deploys mesh topology for redundancy (if one fiber fails, traffic reroutes).
   • Admins use Wireshark to debug packet loss during peak hours (e.g., exam results day).

3. Pathao’s Delivery System

   • Star topology: All delivery partners connect to Pathao’s central servers.
   • Load balancing: Uses Nginx (open-source) to distribute orders across servers.
   • Security: Firewalls block unauthorized access to driver locations.

Exam Tip

1. OSI vs. TCP/IP: Always map layers correctly. In exams, questions may ask:

   • "Which OSI layer handles MAC addressing?" → Data Link (Layer 2).
   • "Where does TCP operate?" → Transport Layer (Layer 4).

2. Tools and Commands:

   • Know ping, traceroute, Nmap, Wireshark for troubleshooting.
   • Example question: "How would you diagnose a slow connection to NEPSE’s website?" Answer:
     • ping nepse.com → Check connectivity.
     • traceroute nepse.com → Find where packets drop.
     • Wireshark → Analyze DNS delays.

3. Open-Source Monitoring:

   • Expect questions like: "Explain how Nagios monitors a web server’s uptime." Answer:
     • Nagios polls the server every minute.
     • If HTTP response time > 2s, it sends an alert to admins.
     • Admins can then restart the service or scale up servers.

4. Topologies:

   • Compare star vs. mesh in terms of cost and fault tolerance.
   • Example: "Why does NTC use a mesh topology for its backbone?" Answer: To ensure redundancy—if one fiber fails, traffic reroutes automatically.

5. Real-World Scenarios:

   • Always tie answers to Nepalese examples (e.g., NTC, eSewa, TU network).
   • Example: "How would you apply the OSI model to troubleshoot Daraz’s slow loading?"
     • Application Layer: Check if Daraz’s servers are down.
     • Transport Layer: Is TCP retransmitting packets? (Use Wireshark.)
     • Network Layer: Is there IP routing congestion? (Check NTC’s backbone.)

Final Note: This unit is foundational. Master the OSI/TCP layers, tools, and real-world applications (eSewa, NTC, TU network). Always draw diagrams in exams—even if not asked, they boost your score!