Previously, utilizing DNS Failover with applications behind an ELB was challenging due to the inability to configure health checks on ELB endpoints, since these load balancers do not have fixed IP addresses.
What Sets Apart DNS Failover for ELB?
The complexity of determining the health of an ELB endpoint is greater than merely checking a single IP address. For instance, what if your EC2 instances are functioning properly, but the load balancer itself is unreachable? Or if both your load balancer and EC2 instances are operational, but a bug in your application code causes a crash? There is also the scenario where EC2 instances in one Availability Zone of a multi-AZ ELB face issues.
Route 53 DNS Failover adeptly manages these failure scenarios by seamlessly integrating with ELB. Once activated, it automatically configures and oversees health checks for individual ELB nodes. Additionally, Route 53 utilizes the EC2 instance health checks performed by ELB (for more details). By merging the results of health checks from both EC2 instances and ELBs, Route 53 DNS Failover can accurately evaluate the health of both the load balancer and the application hosted on the EC2 instances. Essentially, if any component of the system fails, Route 53 will detect it and reroute traffic away from the affected endpoint.
An added advantage is that you don’t need to create any health checks yourself. This means that the DNS Failover for ELB endpoints incurs no extra cost—there are no charges for health checks.
To set up DNS Failover for an ELB endpoint, simply set the “Evaluate Target Health” option to true—you won’t need to establish your own health check for this endpoint.
Possible Scenarios with DNS Failover
With Route 53 DNS Failover, you can operate your primary application across multiple AWS regions around the globe and switch between regions as needed. Users will be directed to the nearest healthy region based on latency. Route 53 automatically removes any region from service where your application is not available—it pulls an endpoint out of service if there’s a region-wide connectivity issue, if your application goes down in that area, or if your ELB or EC2 instances are not functioning.
Moreover, you can make use of a simple backup site hosted on Amazon S3, with Route 53 directing users there if your application becomes unavailable. In a previous blog post, we discussed how to create a simple backup website, which you can find at this link. If your primary site runs behind an ELB, simply skip creating a health check for your main site and create an Alias record pointing to your ELB instead, ensuring to enable the evaluate target health option (full documentation on using DNS Failover with ELB is accessible in the Route 53 Developer Guide).
Get Started and Learn More
Join the High Availability with Route 53 DNS Failover Webinar at 10:00 AM PDT on July 9, 2013, to delve deeper into DNS Failover and the high-availability architecture options it facilitates. To begin with DNS Failover for Route 53, visit the Route 53 product page or check our walkthrough in the Amazon Route 53 Developer Guide. Getting started with Route 53 is straightforward, and there are no upfront costs involved. For comprehensive details and pricing, see the Route 53 product page.
— Alex (with assistance from Sara Mitchell, Product Manager)
Leave a Reply