Continuing networking fundamentals. Today: how DNS resolution actually works, load balancers, firewalls, and the protocols/ports you'll run into constantly in AWS and DevOps work.
How DNS Resolution Works
When you type a URL in your browser, two things are needed: a Hostname (e.g., google.com) and an IP Address (e.g., 142.250.xxx.xxx).
DNS (Domain Name Server) stores and remembers IP addresses and hostnames — it keeps track of all hostname-to-IP mappings and converts IP to hostname and hostname to IP.
Resolution order:
- Browser first hits your Local DNS (stored in /etc/host on your machine)
- If Local DNS has the IP, it sends it straight back to the browser
- If Local DNS has no IP, it talks to the RNS (Root Name Server)
- RNS identifies the TLD (Top-Level Domain) — .com, .in, .org, etc.
- TLD points to the correct Name Server, which identifies the domain
- The Name Server has an SOA (Start of Authority) record — the record that actually holds the IP for that domain
- The IP is passed back down the chain: Name Server → RNS → Local DNS → Browser
Handy command: nslookup <domain> (run in cmd) — shows you the resolved IP address for any domain directly.
How Data Travels — Data Packets
Data travels across a network in the form of data packets. Each packet contains the actual data plus routing info.
Simplified request flow: Browser → Firewall → Load Balancer → Web Server → App Server → Database Server
Load Balancers
A load balancer takes an incoming request and distributes traffic across multiple servers, instead of overloading a single one.
Round Robin method — the most common technique — equally distributes traffic across all available servers in rotation.
Firewalls
A firewall stops unauthorized access to the network. It decides what traffic is allowed or denied based on defined rules.
Common Protocols and Ports
| Protocol | Port | Purpose |
|---|---|---|
| HTTP | 80 | Not secure — standard web traffic |
| HTTPS | 443 | Secure — encrypted web traffic (SSL/TLS certificates) |
| SSH | 22 | Connects to Linux machines |
| RDP | 3389 | Connects to Windows machines |
HTTP over TCP/IP
HTTP (HyperText Transfer Protocol) transfers data from browser to server and back. HTTP itself is not secure; HTTPS (HTTP + SSL/TLS) is secure — data is encrypted using certificates.
TCP (Transmission Control Protocol) establishes a reliable connection between two hosts — think of it as a bridge or messenger that tracks the connection.
UDP (User Datagram Protocol) is TCP's faster, connectionless counterpart — but not reliable, since it doesn't guarantee delivery the way TCP does.
Common HTTP Status Codes
| Code | Meaning |
|---|---|
| 200 | Page Found / Success |
| 401 | Unauthorized |
| 404 | Page Not Found |
| 500 | Internal Server Error |
| 503 | Service Unavailable |
Where to read & follow
- GitHub: https://github.com/sr-palatasingh/AWS-DevOps-Blog/tree/main/posts
- Hashnode: https://sr-palatasingh.hashnode.dev/series/aws-devops-blog
- LinkedIn: https://www.linkedin.com/in/soumyaranjan-palatasingh/
Coming up next: Some networking questions and key concepts — testing everything covered so far.
Top comments (0)