Route53 Health Checks

Health Checks¶

Why use health checks in Route53

Suppose we have ALBs in two different regions and we are using Route 53 latency based routing policy to route users to the ALBs.
Health checks help us in avoiding routing to ALBs which are down.

Health checks are only for public resources. It is possible to monitor private resources using CW metrics.
All the health checkers are hosted outside the VPC. So they cannot access private end points.
Health checks enable automated DNS failover.
We can do Route53 Routing Policies ⮕ Active Active Failover failovers using health checks.
Health checks can be different types:
- Health checks that monitor an endpoint: application, server, other AWS resource
- Health checks that monitor other health checks: Calculated Health Checks
- Health checks that monitor CloudWatch Alarms: full control e.g., throttles of Dynamo DB, alarms on RDS, custom metrics, (helpful for private resources)

Endpoint¶

Health checker comes from all the end points. There are 15 global health checkers.
We can choose which health checkers we want out of 15.
Supports HTTP, HTTPS, TCP.
We can change the threshold for healthy/unhealthy check (default is 3)
Interval is 30s but we can opt for 10s health check known as faster health check but it is costlier.
If health checkers > 18% then Route53 considers it healthy. Otherwise unhealthy.
Health checks can only pass if the status code is 2xx or 3xx.
If it is a text based response it can check first 5120 bytes of the response.

A SysOps Administrator created an Amazon Route 53 health check using a String Matching condition. The String Matching condition is searching for /html at the end of the page to ensure it fully loads. After enabling the health check, the administrator received an alert stating that the health check failed. However, the page loads successfully when navigating directly to it. What is the MOST likely cause of the health check failure?

The search string is not in the first 5120 bytes of the response body.

The security groups in front of the ALBs must allow the IP ranges of Route53 health checkers from which the health checks are coming in.

Calculated health checks¶

Combines the results of multiple health checks into a single health check.
Use case is to perform maintenance of website without causing all health checks to fail.
We can use AND, OR and NOT.
Can monitor up to 256 Child Health Checks
Specify how many of the health checks need to pass to make the parent pass

CW metrics¶

Generally used for private hosted zones or if you want full control over you health checks.

Why do we need CW metrics for private hosted zones?

Route 53 health checkers are outside the VPC. They can't access private endpoints (private VPC or on-premises resource)

You can create a CloudWatch Metric and associate a CloudWatch Alarm, then create a Health Check that checks the alarm itself.

Last updated: 2023-03-14