Examkin

Soru

Zorluk: ZorCloudWatch Metrics and Alarms

A SysOps Administrator is managing an Amazon EC2 Auto Scaling group where the instances are launched with standard monitoring enabled by default. To react to sudden load spikes, the administrator configures an Amazon CloudWatch alarm to trigger a scale-out policy. The alarm is defined with the CPUUtilization metric, a period of 60 seconds (1 minute), and evaluation periods set to 3. During a load test, the actual CPU utilization on the instances stays at 90% for 15 minutes, but the alarm remains in the INSUFFICIENT_DATA state and fails to trigger the scaling policy. Which of the following is the root cause of this behavior?

  1. With standard monitoring enabled, Amazon EC2 metrics are only generated and sent to CloudWatch at 5-minute intervals, meaning there are no 1-minute data points available for the alarm to evaluate.Cevap
  2. B
    Standard monitoring collects CPU utilization data at 1-minute intervals but only publishes the aggregated data to CloudWatch once every 5 minutes, causing a delay in the alarm evaluation.
  3. C
    The Auto Scaling group is currently within a scale-out cooldown period, which automatically suspends CloudWatch metric evaluation and forces associated alarms into the INSUFFICIENT_DATA state.
  4. D
    The CloudWatch alarm is attempting to send the state change notification directly to the Auto Scaling group via an Amazon SQS queue, which is not a supported event flow target for Auto Scaling remediation.

Cevap

With standard monitoring enabled, Amazon EC2 metrics are only generated and sent to CloudWatch at 5-minute intervals, meaning there are no 1-minute data points available for the alarm to evaluate.
The correct answer is that with standard monitoring enabled, Amazon EC2 metrics are only generated and sent to CloudWatch at 5-minute intervals, meaning there are no 1-minute data points available for the alarm to evaluate. In standard monitoring, metric data is sent to CloudWatch at 5-minute intervals. If an alarm is configured with a 1-minute period, it expects metric data every 1 minute. Without detailed monitoring enabled, the lack of 1-minute data points causes the alarm to remain in the INSUFFICIENT_DATA state.

Adım Adım Çözüm

1
Analyze the monitoring configuration of the EC2 instances.
The instances have standard monitoring enabled, which collects and reports metrics at 5-minute intervals.
To understand the frequency at which CloudWatch receives data points for the CPUUtilization metric.
2
Review the alarm configuration parameters.
The alarm is configured with a period of 60 seconds (1 minute) and requires 3 evaluation periods.
To determine what data points the alarm expects in order to perform an evaluation.
3
Evaluate the state transition logic under the current metric frequency.
Because CloudWatch only receives data points every 5 minutes, there are no consecutive 1-minute data points to satisfy the 3-minute evaluation window, keeping the alarm in the INSUFFICIENT_DATA state.
To identify why the alarm failed to transition to the ALARM state despite the actual CPU utilization being high.

Anahtar Kavram

CloudWatch Standard vs Detailed Monitoring Intervals
Tahmini Süre:2m 0s
Bu soruyu puanla

Konular

CloudWatch Metrics and Alarms

Bu konudan daha fazla soru