Course contentsShow
← System Design
Lesson 1440 of 1,91935. Backup and Disaster RecoveryFree lesson

DNS and Traffic Management in DR

Use DNS failover, health checks, and low TTLs to redirect traffic to DR sites with minimal downtime.

DNS and Traffic Management in DR

What You'll Learn

You'll learn how DNS-based routing, health checks, and low TTL values work together to automatically redirect user traffic to a healthy disaster recovery site during outages.

The Problem: Switching Users to DR Sites

When your primary region fails, you need a way to reroute millions of user requests to your DR site without manual intervention. DNS (Domain Name System) becomes your traffic cop, translating api.yourapp.com to whichever data center IP is healthy.

How DNS Failover Works

DNS failover monitors your primary site's health and automatically updates DNS records when problems arise. Think of it like a GPS that detects a road closure and reroutes traffic instantly.

Here's the typical flow:

  1. Health Checks: Your DNS provider continuously pings your primary region's endpoints (e.g., every 10 seconds)
  2. Failure Detection: If health checks fail repeatedly, the DNS service marks that endpoint as unhealthy
  3. Automatic Update: DNS responses now return your DR site's IP address instead
  4. Traffic Shifts: New user requests resolve to the backup location

The TTL Trade-off

TTL (Time To Live) controls how long DNS resolvers cache your records. During normal operations, you might use TTL=3600 (1 hour) to reduce DNS query load. But for DR-ready systems, use low TTLs (60-300 seconds) so changes propagate quickly.

The trade-off: lower TTLs mean more DNS queries and slightly higher latency, but much faster failover.

Active Traffic Management

Modern DNS providers offer weighted routing (split traffic 90/10 between regions), geolocation routing (route EU users to EU-West), and latency-based routing. These enable gradual failover testing and sophisticated multi-region strategies.

Key Takeaway:
DNS failover with health checks and low TTLs (60-300s) enables automated traffic redirection to DR sites within minutes, making it a critical component of any multi-region disaster recovery strategy.