Exam AWS Certified Solutions Architect - Professional topic 1 question 453 discussion

C looks to be the best answer IMO. This is a capacity issue. Losing one EC2 instance resulted in reduced performance. What's needed is the benefits of an Auto Scaling group. You can scale out when needed and scale in as well (during off-peak hours, for example). Also, changing the size of the server can help with costs. You can scale out and in with smaller increments of capacity at a time. So I don't see an issue with the instance size change. Nothing in this question points to an issue with CLB. Yes, AWS is encouraging people to move to ALB, but CLB is not the cause of the issue here. Having an ALB would not have avoided the issue being experienced.

By default, connection draining is enabled for Application Load Balancers but must be enabled for Classic Load Balancers. When Connection Draining is enabled and configured, the process of deregistering an instance from an Elastic Load Balancer gains an additional step. For the duration of the configured timeout, the load balancer will allow existing, in-flight requests made to an instance to complete, but it will not send any new requests to the instance. During this time, the API will report the status of the instance as InService, along with a message stating that “Instance deregistration currently in progress.” Once the timeout is reached, any remaining connections will be forcibly closed. https://docs.aws.amazon.com/autoscaling/ec2/userguide/attach-load-balancer-asg.html https://aws.amazon.com/blogs/aws/elb-connection-draining-remove-instances-from-service-with-care/

C looks ok. I got it, It's a performance issue, so only ALB will solve this!

C looks good. Q says "maximize system reliability.", the issue is with one of the instances failing, and we know autoscaling will immediately remediate this when one instance is lost

Based on all comments

@sb333's explanation clearly shows that the answer should be C.

B SURE

what i do not get for C is how come a m4.large is more capable than m4.xlarge? I meant, I have chosen B. Then again, B isn't the best, either. This entire question has a lot of assumptions in it. If there is a capacity issue by loosing a m4.xlarge, by replacing the entire fleet of m4.xlarge with something less capable, wouldn't cause similar problem, too?

The scaling issue has nothing to do with intermittent session issues or users being logged off after an instance fails. The answer has to be C, not B.

requests processing slow don't have any relation with ALB vs Classic LB, it is capacity issue so C is right answer

It's C

C Think of it this way. If 2 instances fail, processing is gonna get slower. The CLB will still direct traffic to the instances. If it was latency or SNI needed than, an ALB would satisfy the answer

CCC ---

I'll go with C

I think the answer is C. Because the use of Classic Load Balancer is a pre-requisite, which means the application is using EC2-Classic Network instead of VPC, hence this rule out the use of ALB.

B,C correct

key word reduce cost and improve reability, both direct me to go for C.