Q: Auto Scaling is not working as expected, and new instances are not being launched within the required time frame. How would you troubleshoot and resolve this issue?

Review the Auto Scaling Group's Activity History. Look for failed launch attempts due to subnet IP exhaustion, account vCPU service quotas exceeded (`InsufficientInstanceCapacity`), or invalid Launch Template configurations.

aws autoscaling describe-scaling-activities \
  --auto-scaling-group-name asg-production-web \
  --max-records 10
# Check for: "Failed" activities and Description messages

Standard CloudWatch metrics report on a 5-minute schedule. If using Step Scaling or Simple Scaling with 5-minute metric periods and a 300-second cooldown, Auto Scaling can take 10 to 15 minutes to trigger the first new instance. Switch to 1-minute detailed monitoring and Target Tracking policies.

# Enable detailed 1-minute CloudWatch monitoring for ASG Launch Template
"Monitoring": { "Enabled": true }

Inspect the instance user-data bootstrap script (`/var/log/cloud-init-output.log`). If instances run `yum update -y`, download 2 GB Docker images, or compile binaries at launch, boot time will exceed 10 minutes. Pre-bake all dependencies, packages, and container images into a Golden AMI using HashiCorp Packer.

Enable ASG **Warm Pools**. A Warm Pool maintains pre-initialized, stopped EC2 instances that have already completed OS updates and bootstrap scripts. When scaling occurs, instances transition from `Stopped` to `Running` in under 30 seconds.