Question

Difficulty: MediumAuto Scaling and Elasticity

A SysOps administrator is managing an Amazon EC2 Auto Scaling group (ASG) configured to scale out based on CPU utilization using a dynamic scaling policy. The administrator observes that although the associated Amazon CloudWatch alarm has transitioned to the ALARM state, the ASG does not launch any new instances. The maximum capacity of the ASG is set to 1010, the minimum is 22, and the current desired capacity is 22. Which two troubleshooting steps should the administrator perform to identify the root cause of this behavior? (Select TWO.)

  1. Check if the Launch process is suspended on the Auto Scaling group.Answer
  2. Check if the AlarmNotification process is suspended on the Auto Scaling group.Answer
  3. C
    Enable detailed monitoring on the launch template, as dynamic scaling policies require 11-minute metric intervals to trigger.
  4. D
    Increase the default cooldown period to ensure that the CloudWatch alarm has enough time to pass the evaluation period.
  5. E
    Verify that the SysOps administrator's IAM user policy contains the iam:PassRole permission for the CloudWatch service.

Answer

The correct troubleshooting steps are to check if the Launch process is suspended on the Auto Scaling group and to check if the AlarmNotification process is suspended on the Auto Scaling group.
The correct steps are to check the suspended processes on the Auto Scaling group. If the Launch process is suspended, the Auto Scaling group will not launch new instances even if a scaling action is requested. If the AlarmNotification process is suspended, the Auto Scaling group will not receive the notification from the CloudWatch alarm, preventing the scaling policy from executing.

Step-by-Step Solution

1
Review the suspended processes list for the Auto Scaling group using the AWS Management Console or the DescribeAutoScalingGroups API call.
Identify if processes like Launch or AlarmNotification are currently in a suspended state.
Suspended processes halt specific operations. The Launch process stops new instances from starting, while the AlarmNotification process stops the ASG from reacting to CloudWatch alarms.
2
Verify if the Auto Scaling service-linked role is configured correctly and has not been deleted or modified.
Ensure the service-linked role has permissions to launch instances and respond to alarms.
If the service-linked role is missing or misconfigured, it will prevent scaling actions, but checking suspended processes is the primary step when alarms are triggered and capacity settings are correct.

Key Concept

Auto Scaling suspended processes prevent scaling activities from executing even when CloudWatch alarms trigger policies.
Rate this question