# How DNS Resolution Works ### ❀ Introduction : ⤷ Every time you type [**google.com**](http://google.com) in your browser, something important happens in the background. Your computer does **not understand website names**. It only understands **IP addresses**. ⤑ DNS (Domain Name System) helps convert: ```plaintext google.com → IP address ``` ⤑ That is why DNS is called the **internet’s phonebook**. ⤑ In this blog, we will understand **how DNS works step by step** using a tool called `dig`. ### ❋ What is DNS and Why Do We Need It? ⤑ Humans like easy names: * [google.com](http://google.com) * [amazon.com](http://amazon.com) * [github.com](http://github.com) ⤑ Computers like numbers: * 142.250.185.46 ⤑ Without DNS: * You would need to remember IP addresses * Websites changing servers would break bookmarks ⤑ DNS solves this by: * Mapping domain names to IP addresses * Working in a **distributed way**, not from one single server * Making the internet **fast, scalable, and reliable** ### ❋ What is `dig` and Why Is It Used? ⤑ `dig` stands for **Domain Information Groper**. ⤑ It is a command-line tool used to: * Check DNS records * Understand how DNS resolution works * Debug DNS problems * Learn DNS deeply Basic command: ```plaintext dig google.com ``` Unlike browsers (which hide DNS details), `dig` shows **exactly what is happening**. ### ❋ DNS Works in Layers (Hierarchy) DNS is not one server. It works in **three main layers**: ![](https://cdn.hashnode.com/res/hashnode/image/upload/v1769256793541/78a966a6-cd37-4eec-8721-39e8459311f2.png align="center") Each layer only knows **the next layer**, not everything. ### ❋ Layer 1 : Root Name Servers (`dig . NS`) Command: ```plaintext dig . NS ``` This asks: Who manages the root of DNS? **What root servers do:** * They are the top of DNS * They do NOT know website IPs * They only know where TLD servers are Example response idea: I don’t know [google.com](http://google.com), but I know who manages `.com` Important points: * There are **13 root server names** * They are spread across the world * Very reliable and fast ### ❋ Layer 2 : TLD Name Servers (`dig com NS`) Command: ```plaintext dig com NS ``` This asks: Who manages the `.com` domain? **What TLD servers do:** * They manage domains like `.com`, `.org`, `.net` * They do NOT know IP addresses * They point to **authoritative servers** Example response idea: *I don’t know google’s IP, but I know Google’s name servers* ### ❋ Layer 3 : Authoritative Name Servers (`dig` [`google.com`](http://google.com) `NS`) Command: ```plaintext dig google.com NS ``` This asks: *Who is responsible for* [*google.com*](http://google.com)*?* **Authoritative servers:** * Are owned by the domain owner (Google) * Store real DNS records * Are the **source of truth** They contain: * A records (IP address) * MX records (email) * TXT records (verification) Example: ```plaintext ns1.google.com ns2.google.com ``` These servers finally know: *Yes, this is the IP address of* [*google.com*](http://google.com) ### ❋ Full DNS Resolution (`dig` [`google.com`](http://google.com)) Command: ```plaintext dig google.com ``` This gives: * IP address * TTL (how long it can be cached) * Which DNS server answered ### ⤑ What happens behind the scenes: 1. Check cache 2. Ask root server 3. Root points to `.com` 4. `.com` points to Google servers 5. Google server returns IP 6. Result is cached TTL example: ```plaintext 300 seconds ``` Means the result can be reused for 5 minutes. ### ❋ What Are NS Records and Why Are They Important? ⤷ **NS (Name Server) records** tell: * Who controls a domain * Where DNS queries should go next ⤑ They are important because: * DNS is distributed * No single server controls everything * Multiple servers give backup (redundancy) If one server fails, another works. ### ❋ What is a Recursive Resolver? ⤷ Your computer does not talk to root servers directly. ⤑ Instead, it uses a **recursive resolver**, such as: * 8.8.8.8 (Google DNS) * 1.1.1.1 (Cloudflare DNS) * ISP DNS Flow: ```plaintext Your PC → Recursive Resolver → Root → TLD → Authoritative ``` ⤑ The resolver: * Does all DNS work for you * Caches results * Returns final IP ### ❋ DNS and Real Browser Requests ⤷ When you open a website: 1. Browser checks cache 2. OS checks cache 3. Recursive resolver is asked 4. DNS resolution happens 5. IP address is returned 6. Browser connects using HTTP/HTTPS ⤑ DNS usually takes: * **20–120 ms** if not cached * **<10 ms** if cached That’s why DNS feels instant. ### ❋ DNS from a System Design View ### ⤑ *DNS is designed to be:* * Distributed * Fast * Fault-tolerant * Scalable ⤑ Important ideas: * Caching improves speed * Multiple name servers improve reliability * DNS helps CDNs route users to nearby servers ### *You can watch for System Design View of DNS :*