What do you need to do to ensure trial instances marked unhealthy by the ELB will be terminated and replaced?
Change the thresholds set on the Auto Scaling group health check
Add an Elastic Load Balancing health check to your Auto Scaling group
Increase the value for the Health check interval set on the Elastic Load Balancer
Change the health check set on the Elastic Load Balancer to use TCP rather than HTTP checks
Explanations:
Changing the thresholds for the Auto Scaling group’s health check would affect how Auto Scaling determines the health of instances based on its own checks, but it doesn’t resolve the issue of ELB health checks not triggering instance termination.
Adding an Elastic Load Balancing health check to the Auto Scaling group ensures that the health of the instances is determined based on both ELB and Auto Scaling checks, prompting Auto Scaling to terminate unhealthy instances as necessary.
Increasing the health check interval on the ELB would only affect how frequently the ELB checks the health of instances, but it wouldn’t cause Auto Scaling to terminate instances marked unhealthy by the ELB.
Changing the health check type to TCP on the ELB might alter how health is determined but does not directly solve the issue of ensuring that unhealthy instances are terminated by the Auto Scaling group.
I think the answer is:
Add an Elastic Load Balancing health check to your Auto Scaling group