DNS failover in Route 53

#aws

Where to configure DNS failover

In AWS, DNS failover is configured in Amazon Route 53 — AWS’s DNS service.

How DNS failover works in AWS

Route 53 health checks

You create a health check for your primary site (the primary region’s load balancer).
Route 53 constantly monitors whether your primary site is healthy (by pinging an endpoint or checking HTTP status).

DNS records with failover routing policy

You create two DNS records for your domain:
- Primary record pointing to your primary region’s load balancer.
- Secondary (failover) record pointing to your DR region’s load balancer.
You assign a failover routing policy to the DNS records so Route 53 knows which one is primary and which one is secondary.

Automatic failover

If Route 53 detects your primary site is unhealthy via the health check, it automatically routes traffic to the secondary record (the DR region’s load balancer).

Step-by-step example

Let’s say your domain is example.com.

Step 1 — Set up Route 53 hosted zone

In AWS console, go to Route 53 → Hosted Zones.
Select your domain.

Step 2 — Create health check

Go to Route 53 → Health Checks.
Create a health check that monitors your primary region’s load balancer endpoint (using HTTP, HTTPS, or TCP check).
Set the check interval (default is every 30 seconds).

Step 3 — Create DNS records with failover

In your hosted zone, create two A or Alias records:

Primary record:

 * Type: A (Alias)
 * Alias target: Primary region’s load balancer.
 * Routing policy: Failover
 * Failover record type: Primary
 * Associate with health check you created.

Secondary record:

 * Type: A (Alias)
 * Alias target: DR region’s load balancer.
 * Routing policy: Failover
 * Failover record type: Secondary.

How it works in practice

Under normal conditions → Route 53 resolves your domain to the primary region’s load balancer IP.
If Route 53 health check fails → DNS automatically switches to the DR region’s load balancer.

Extra tip

DNS failover is not instantaneous — DNS changes depend on TTL (time-to-live) settings. You can reduce TTL (e.g., to 30 seconds) so Route 53 switches faster.

                         +--------------------+
                         |  Route 53 Hosted  |
                         |      Zone         |
                         +--------------------+
                                  |
                                  | DNS query for example.com
                                  |
             +-------------------------------------------+
             |                                           |
+----------------------------+              +----------------------------+
| Primary Region Load       |              | Disaster Recovery Region   |
| Balancer (ELB)            |              | Load Balancer (ELB)       |
+----------------------------+              +----------------------------+
             |                                           |
   +--------------------+                        +--------------------+
   | Auto Scaling Group|                        | Auto Scaling Group|
   | EC2 Instances     |                        | EC2 Instances     |
   +--------------------+                        +--------------------+
             |                                           |
   +--------------------+                        +--------------------+
   | DynamoDB Table    |<---------------------> | DynamoDB Global    |
   | (Primary Region)  |                        | Table (DR Region) |
   +--------------------+                        +--------------------+

Route 53 Health Check:
   - Monitors Primary Region Load Balancer
   - If unhealthy → Route 53 fails over to DR Region Load Balancer