A website runs on four EC2 instances behind an ALB. When the ALB marks an instance unhealthy, a CloudWatch alarm goes to ALARM and an operator manually adds a replacement instance. The company wants a highly available solution that automatically replaces failed EC2 instances while minimizing downtime during migration to the new design. Which sequence of steps should the solutions architect use?

The correct approach minimizes downtime by integrating the new Auto Scaling Group (ASG) with the existing infrastructure. First, create the ASG with a launch template to define the new instance configuration. Then, attach this ASG to the existing Application Load Balancer (ALB). This ensures continuous traffic flow through the current ALB. Finally, add the existing EC2 instances to the ASG. This allows the ASG to manage these instances, replacing them automatically if they fail, and smoothly transitioning to the new, highly available setup without requiring a full system rebuild or service interruption. Option 1 is incorrect because deleting the existing ALB would cause significant downtime. Option 3 is incorrect because deleting the existing ALB and EC2 instances would cause significant downtime. Option 4 is incorrect because the ALB does not automatically register existing instances with an ASG; they must be explicitly added.