A solutions architect deployed a web app accessible from two AWS Regions under a custom domain using Amazon Route 53 latency-based routing. For each Region, the architect created weighted record sets that point to a pair of web servers in different Availability Zones. During a disaster-recovery test, all web servers in one Region were stopped, but Route 53 did not automatically route traffic to the other Region. Which of the following could explain this behavior? (Choose two.)

For Route 53 to fail over traffic, it must be aware that the endpoints are unhealthy. The two correct options address this. First, if the weighted record sets for the stopped web servers do not have associated HTTP health checks, Route 53 will continue to consider them healthy, even though the servers are down. This prevents failover to other healthy endpoints within the same Region or to the other Region. Second, even if the individual weighted record sets have health checks, if the latency alias record set for the entire Region does not have "Evaluate Target Health" enabled, Route 53 will not consider the health of the underlying weighted records when determining the health of the alias. This means the alias record will always appear healthy, preventing failover to the other Region. Incorrect options: A larger weight would mean more traffic is sent to that Region, but it wouldn't prevent failover if health checks were properly configured. A failed health check in the fallback Region would reduce its availability but not prevent failover from the primary Region if the primary Region was correctly configured for health checks and evaluation. Latency-based and weighted resource record sets can be combined; this is a common pattern for advanced routing.