ANS-C01 Question 469
Select 3A company runs a critical web application on Amazon EC2 instances behind an Application Load Balancer (ALB). The application is configured to scale automatically based on traffic. The company wants to be notified immediately if the average CPU utilization across the EC2 instances exceeds 80% for more than 5 minutes. Which combination of steps would meet this requirement?
- A
Create a CloudWatch alarm for the ALB with a threshold of 80% CPU utilization and a period of 5 minutes.
- B
Create a CloudWatch alarm for the EC2 Auto Scaling group with the metric 'Average CPU Utilization' and a threshold of 80%, with a period of 5 minutes.
- C
Set the alarm action to send a notification to an Amazon SNS topic.
- D
Create a CloudWatch alarm for individual EC2 instances with a threshold of 80% CPU utilization and a period of 1 minute.
- E
Subscribe an email endpoint to the Amazon SNS topic to receive the alarm notifications.
Show answer and explanation
Correct answers: B, C, E
Explanation
To meet the company's requirements, you need to monitor the average CPU utilization across the EC2 Auto Scaling group. This requires creating a CloudWatch alarm for the Auto Scaling group using the 'Average CPU Utilization' metric and setting the appropriate threshold and period. The alarm action should trigger an Amazon SNS topic, and subscribing an email endpoint to the SNS topic ensures that notifications are received. Monitoring individual instances or unrelated components like the ALB is not suitable for this requirement.
- A. Incorrect.
Incorrect. While the ALB can provide metrics such as request counts and latency, it does not directly measure the CPU utilization of the EC2 instances behind it.
- B. Correct.
Correct. The Auto Scaling group aggregates the CPU utilization of all its instances, and creating a CloudWatch alarm for this metric allows you to monitor the average CPU utilization across the group.
- C. Correct.
Correct. Setting the alarm action to trigger an Amazon SNS topic ensures that notifications can be sent when the alarm is breached.
- D. Incorrect.
Incorrect. Creating alarms for individual EC2 instances is not efficient in this scenario because the requirement is to monitor the average CPU utilization across the Auto Scaling group, not individual instances.
- E. Correct.
Correct. Subscribing an email endpoint to the SNS topic ensures that the notifications are delivered to the appropriate recipient as soon as the alarm is triggered.