Examkin

Question

Difficulty: Very hardCloudWatch Metrics and Alarms

A SysOps Administrator is managing an application running on a fleet of Amazon EC2 instances within an Auto Scaling group. The application publishes a custom metric named `TransactionLatency` to Amazon CloudWatch at a standard resolution of 6060 seconds. The administrator has configured a CloudWatch alarm to monitor the 99th99\text{th} percentile (p99p_{99}) of `TransactionLatency` over a period of 11 minute. The alarm is designed to trigger an AWS Systems Manager (SSM) Automation document to reboot an unhealthy instance if the latency exceeds 800 ms800\text{ ms} for 33 consecutive evaluation periods.

During testing, the administrator observes two issues:
1. The alarm frequently transitions to `INSUFFICIENT_DATA` when traffic is very low and no transactions are processed, triggering false notifications.
2. When the latency threshold is breached, the SSM Automation document is never executed, even though the SSM execution role has the correct trust policy.

Which TWO configuration changes should the SysOps Administrator make to address these issues?

  1. Configure the CloudWatch alarm's missing data treatment to treat missing data as notBreaching.Answer
  2. Create an Amazon EventBridge rule that detects the CloudWatch alarm state change to ALARM and targets the Systems Manager Automation document.Answer
  3. C
    Enable detailed monitoring on the EC2 instances in the Auto Scaling group to increase the custom metric reporting frequency.
  4. D
    Add the Amazon Resource Name (ARN) of the Systems Manager Automation document directly to the CloudWatch alarm actions list under the ALARM state.
  5. E
    Set the Auto Scaling group's default cooldown period to zero to prevent metric aggregation delays during instance reboots.

Answer

Configure the CloudWatch alarm's missing data treatment to treat missing data as notBreaching, and create an Amazon EventBridge rule that detects the CloudWatch alarm state change to ALARM and targets the Systems Manager Automation document.
To resolve the false notifications caused by the alarm entering the INSUFFICIENT_DATA state during low-traffic periods, the administrator should set the missing data treatment to treat missing data as notBreaching. This ensures the alarm stays in or returns to the OK state when no transactions occur. To resolve the issue with the Systems Manager Automation document not triggering, the administrator must use an Amazon EventBridge rule. CloudWatch alarms cannot directly trigger SSM Automation documents; instead, an EventBridge rule must be configured to match the alarm state transition to ALARM and target the SSM Automation document.

Step-by-Step Solution

1
Analyze the root cause of the false notifications due to missing metric data.
Determine that during low-traffic periods, no transaction latency data points are published, causing CloudWatch to evaluate the alarm state as INSUFFICIENT_DATA.
To prevent false alarms, the missing data behavior must be configured to transition to a safe state when data is absent.
2
Configure the missing data treatment for the CloudWatch alarm.
Set the missing data behavior to treat missing data as notBreaching.
Treating missing data as notBreaching ensures the alarm transitions to or stays in the OK state when no transactions are occurring, rather than entering INSUFFICIENT_DATA.
3
Analyze the root cause of the SSM Automation document execution failure.
Identify that CloudWatch alarms cannot directly target Systems Manager Automation documents from the ALARM actions configuration.
CloudWatch alarms are restricted to routing actions to SNS, EC2 resource actions, Auto Scaling policies, and OpsCenter/Incident Manager.
4
Establish the correct event routing path for automated remediation.
Create an Amazon EventBridge rule that monitors CloudWatch alarm state transitions to ALARM and triggers the SSM Automation document.
EventBridge is designed to capture CloudWatch alarm state change events and natively route them to Systems Manager Automation for automated remediation.

Key Concept

CloudWatch Alarm Actions and Missing Data Treatment
Estimated Time:3m 0s
Rate this question

Topics

CloudWatch Metrics and Alarms

More questions from this topic