# DNS and AWS Route 53 ## 1\. First, what is DNS (simple but deep) DNS = **Domain Name System** Humans remember names Machines understand IP addresses Example You type: `www.myapp.com` Browser needs: `52.66.123.10` DNS is the system that converts **domain name → IP address** Think of DNS like **phone contacts** * Name = www.myapp.com * Number = server IP Without DNS, you would have to remember IPs for every website. --- ## 2\. What is AWS Route 53 **Route 53 is AWS’s DNS service** It does 3 main things: 1. **Domain registration** 2. **DNS management** 3. **Traffic routing + health checks** Why name Route 53? * DNS uses **port 53** * “Route” means routing traffic --- ## 3\. Why companies use Route 53 (real reasons) Companies don’t use Route 53 just because it’s AWS. They use it because: * Highly available (AWS global infrastructure) * Very fast DNS resolution * Easy integration with AWS services * Advanced traffic routing (huge deal in DevOps) * Health checks + failover If your app is down, **DNS can redirect traffic automatically**. --- ## 4\. Core components of Route 53 (must know) ### 4.1 Domain Example: ```plaintext mycompany.com ``` You can: * Buy domain from Route 53 * Or buy from GoDaddy and manage DNS in Route 53 --- ### 4.2 Hosted Zone Hosted Zone = **DNS records container for a domain** Two types: 1. **Public Hosted Zone** * For internet facing apps * Example: myapp.com 2. **Private Hosted Zone** * Only inside VPC * Used for internal services * Example: database.internal Think: ```plaintext Hosted Zone = DNS configuration file ``` --- ### 4.3 DNS Records (very important) These tell DNS **where to send traffic** Common record types: #### A Record * Domain → IPv4 ```plaintext myapp.com → 13.234.56.78 ``` #### AAAA Record * Domain → IPv6 #### CNAME * Domain → another domain ```plaintext www.myapp.com → myapp-alb.amazonaws.com ``` #### Alias Record (AWS special) * Points to AWS resources * No extra cost * Faster than CNAME Used with: * ALB * CloudFront * S3 static website * API Gateway This is what you’ll use most in AWS. #### MX Record * Email routing #### TXT * Verification, SPF, DKIM --- ## 5\. How Route 53 actually works (flow) When user opens your website: 1. User types `www.myapp.com` 2. Browser asks local DNS resolver 3. Resolver queries Route 53 4. Route 53 checks hosted zone records 5. Returns IP or ALB endpoint 6. Browser connects to server This whole process happens in milliseconds. --- ## 6\. When to use Route 53 (real scenarios) ### Scenario 1: EC2 hosted website * EC2 has public IP * Create **A record** * Map domain → EC2 IP But problem: * EC2 IP can change Better: * Put ALB in front * Use Alias record → ALB --- ### Scenario 2: Load Balanced application (most common) Architecture: ```plaintext User → Route 53 → ALB → EC2 / ECS / EKS ``` Why: * DNS handles routing * ALB handles load balancing --- ### Scenario 3: High availability (Multi-region) Example: * App running in Mumbai and Singapore Route 53: * Checks health * If Mumbai fails * Automatically routes to Singapore No manual work. --- ### Scenario 4: Blue-Green deployment * Old version → 90% traffic * New version → 10% traffic Change DNS routing policy Gradually shift traffic --- ## 7\. Routing policies in Route 53 (very important for interviews) ### 7.1 Simple Routing * One record * One resource * No health check Used for basic apps. --- ### 7.2 Weighted Routing * Split traffic by percentage ```plaintext Version A → 80% Version B → 20% ``` Used in: * Canary releases * A/B testing --- ### 7.3 Latency-based Routing * Route user to **nearest region** * Based on AWS latency data Used for global apps. --- ### 7.4 Failover Routing * Primary + Secondary * Health check decides Used for disaster recovery. --- ### 7.5 Geolocation Routing * Based on user location * Country or continent Used for: * Legal compliance * Content control --- ### 7.6 Geoproximity Routing * Route traffic based on distance * Can bias traffic Advanced global routing. --- ### 7.7 Multivalue Answer * Returns multiple healthy IPs * Simple load balancing at DNS level --- ## 8\. Health Checks (power feature) Route 53 can: * Ping your endpoint * Check HTTP status * Monitor specific path If health check fails: * Route 53 stops sending traffic there This is DNS-level auto healing. --- ## 9\. Route 53 with AWS services (practical use) ### With ALB * Alias record → ALB DNS name ### With CloudFront * Domain → CDN * Best for performance ### With S3 static website * Direct domain mapping * No EC2 needed ### With API Gateway * Clean API URLs --- ## 10\. Security aspects * IAM controls who can edit DNS * Supports DNSSEC * Private hosted zones for internal apps --- ## 11\. Cost (basic idea) * Hosted zone: low monthly cost * Queries: pay per million * Alias records: free Usually very cheap unless massive traffic. --- ## 12\. Real DevOps engineer usage As a DevOps engineer, you will: * Create hosted zones * Manage records during deployments * Configure failover * Set up health checks * Integrate DNS with CI/CD * Troubleshoot DNS issues DNS mistakes can **take entire production down**, so this is taken seriously. --- ## 13\. Common mistakes beginners make * Using EC2 IP instead of ALB * Using CNAME at root domain * Forgetting TTL impact * No health checks * Manual DNS changes without planning --- ## 14\. Interview-level one-liners * Route 53 is a highly available DNS and traffic routing service. * Alias records are AWS specific and better than CNAME. * Route 53 supports multiple routing policies. * Health checks enable DNS-based failover. # Common DNS debugging commands ## 1\. `nslookup` (basic and fast) ### What it does Queries DNS and returns IP for a domain. ### When to use * Quick check * Is DNS resolving or not * First step in debugging ### Command ```plaintext nslookup google.com ``` ### Sample output meaning * **Server** → DNS resolver used * **Address** → IP returned ### Check specific DNS server ```plaintext nslookup google.com 8.8.8.8 ``` Use this to compare Route 53 vs public DNS. --- ## 2\. `dig` (most important command) This is the **industry standard DNS debugging tool**. ### Basic query ```plaintext dig google.com ``` ### Key sections to read * **QUESTION SECTION** → what you asked * **ANSWER SECTION** → actual DNS record * **AUTHORITY SECTION** → who controls domain * **Query time** → DNS latency * **SERVER** → which DNS replied --- ### Check specific record type ```plaintext dig google.com A dig google.com MX dig google.com TXT ``` Very useful when email or verification fails. --- ### Check using specific DNS (Route 53 name server) ```plaintext dig myapp.com @ns-123.awsdns-45.com ``` Used when: * DNS changes not reflecting * Want to confirm Route 53 config is correct --- ## 3\. `dig +short` (clean output) ### Why When you just want the answer, not full details. ```plaintext dig +short myapp.com ``` Output: ```plaintext 13.235.10.22 ``` Used in scripts and quick checks. --- ## 4\. `host` (simple and readable) ### What it does Simpler version of dig. ```plaintext host google.com ``` Good for: * Quick human readable checks * Less noisy output --- ## 5\. Check CNAME chain (very important in AWS) ```plaintext dig www.myapp.com ``` Look for: ```plaintext www.myapp.com. CNAME myapp-alb.amazonaws.com. ``` If CNAME is broken: * Website won’t open * SSL may fail --- ## 6\. Trace full DNS path (advanced but powerful) ```plaintext dig myapp.com +trace ``` What this shows: * Root servers * TLD servers (.com) * Authoritative DNS (Route 53) Used when: * Domain not resolving globally * Nameserver misconfiguration --- ## 7\. TTL debugging (cache issues) ### Why TTL matters DNS changes don’t apply instantly because of caching. Check TTL: ```plaintext dig myapp.com ``` Look at number like: ```plaintext 300 ``` Means: * DNS cached for 300 seconds Low TTL = faster changes, slightly more queries. --- ## 8\. Check Route 53 Alias behavior Alias records don’t show as CNAME. ```plaintext dig myapp.com ``` You’ll see: ```plaintext A dualstack.my-alb-123.us-east-1.elb.amazonaws.com ``` This is normal for ALB alias. --- ## 9\. Debug health check based routing If failover not working: ```plaintext dig myapp.com ``` * If primary unhealthy * Route 53 stops returning it Use AWS Console to confirm health check status. --- ## 10\. Check from different locations (real production issue) DNS can differ globally. Use: ```plaintext dig myapp.com @1.1.1.1 dig myapp.com @8.8.8.8 ``` Helps when: * Works in office but not for users * ISP cache issues --- ## 11\. Flush local DNS cache (local system issue) ### Linux ```plaintext sudo systemctl restart systemd-resolved ``` ### Ubuntu (older) ```plaintext sudo service dns-clean restart ``` ### Windows ```plaintext ipconfig /flushdns ``` ### Mac ```plaintext sudo dscacheutil -flushcache; sudo killall -HUP mDNSResponder ``` --- ## 12\. Common real-world DNS problems + command ### Website not opening ```plaintext dig myapp.com ``` Check: * IP returned? * Correct ALB? --- ### SSL certificate error ```plaintext dig www.myapp.com ``` Check: * CNAME mismatch * Wrong CloudFront distribution --- ### Email not working ```plaintext dig mydomain.com MX ``` Check: * Correct mail servers * Priority values --- ### Route 53 change not reflecting ```plaintext dig myapp.com @ns-xxx.awsdns-xx.com ``` If correct here but not globally → cache issue. --- ## 13\. DevOps interview gold lines * “I start DNS debugging with dig, not nslookup.” * “I check authoritative nameservers directly.” * “TTL explains why DNS changes take time.” * “Route 53 alias records behave differently from CNAME.” --- ## 14\. Must-practice mini lab (do this) 1. Create hosted zone in Route 53 2. Add A record → ALB 3. Run: ```plaintext dig yourdomain.com dig +trace yourdomain.com ``` This will make DNS crystal clear.