3 web instances in different AZs.
-
For each instance create a health checks with the following properties:
-
Name it after the AZ where the instance resides
-
Failure threshold should be 5
-
Edit the security group of one of your instances and remove HTTP rules.
-
Did it change the status of the health check?
-
Go to Route 53
-
Click on "Health Checks" in the left-side menu
-
Click on "Create health check"
-
Insert the name: us-east-2
-
What to monitor: endpoint
-
Insert the IP address of the instance
-
Insert the endpoint /health if your web instance supports that endpoint
-
In advanced configuration, set Failure threshold to 5
-
Click on "next" and then on "Create health check"
-
Repeat steps 1-9 for the other two instances you have
-
Go to security group of one of your instances
-
Click on "Actions" -> Edit inbound rules -> Delete HTTP based rules
-
Go back to health checks page and after a couple of seconds you should see that the status becomes "unhealthy"