DOP-C02 Question 247
Select 3You are managing a web application that is deployed behind an Application Load Balancer (ALB) in your AWS environment. Recently, users have reported intermittent failures when accessing the application. You need to identify the root cause of these issues by analyzing CloudWatch metrics and logs. Which of the following metrics and logs would be most useful for diagnosing the issue?
- A
ALB Target Group HealthyHostCount metric
- B
ALB 5xxErrorCount metric
- C
EC2 CPUUtilization metric for the application instances
- D
Amazon RDS Queue Depth metric
- E
VPC Flow Logs for the subnets hosting the ALB
Show answer and explanation
Correct answers: A, B, C
Explanation
Intermittent failures behind an ALB can often be traced to unhealthy targets, server-side errors, or resource constraints on the application instances. Analyzing the ALB Target Group HealthyHostCount, ALB 5xxErrorCount, and EC2 CPUUtilization metrics provides insights into these potential causes. While RDS Queue Depth and VPC Flow Logs could be useful in other scenarios, they are not directly relevant to diagnosing this particular issue.
- A. Correct.
The ALB Target Group HealthyHostCount metric helps determine if there are any unhealthy targets (EC2 instances) behind the ALB, which could be causing intermittent failures.
- B. Correct.
The ALB 5xxErrorCount metric tracks server-side errors generated by the ALB, which can indicate misconfigurations or application issues.
- C. Correct.
The EC2 CPUUtilization metric is critical for understanding if high CPU load on the application instances is contributing to the intermittent failures.
- D. Incorrect.
The Amazon RDS Queue Depth metric is unrelated to the issue because it tracks the number of outstanding I/O requests to the database, which is not directly relevant to identifying failures in the ALB or application layer.
- E. Incorrect.
The VPC Flow Logs provide network-level information but are not the most useful initial diagnostic tool for this type of application issue. They are typically used for network troubleshooting rather than application-layer problems.