Why is my Auto Scaling group not launching new instances when the load increases?

There are several potential reasons why your Auto Scaling group is not launching new instances when the load increases, despite having a scaling policy in place. Here are some possible causes and troubleshooting steps:

1. Suspended processes: One or more Auto Scaling processes might be suspended, preventing the group from launching new instances. Check if any processes are suspended and resume them if necessary.

2. Launch failures: Your Auto Scaling group may be experiencing launch failures. This could be due to various reasons such as insufficient EC2 instance capacity in the Availability Zone, issues with encrypted EBS volumes, or inadequate IAM permissions. Review the Auto Scaling group's Activity history to identify any launch failures and their specific causes.

3. Insufficient instance capacity: There might not be enough available instance types in the relevant Availability Zone for either On-Demand or Spot Instances. To mitigate this:

   • Specify multiple instance types in your Auto Scaling group's configuration
   • Use attribute-based instance selection to widen the range of available instance types
   • Increase the number of active Availability Zones for your Auto Scaling group
   • Consider using On-Demand Capacity Reservations or Spot Instance allocation strategies

4. Incorrect health check settings: Ensure that your health check settings are properly configured, including the health check grace period and any custom health checks you may have implemented.

5. Security group or network configuration issues: Verify that your security groups and network settings allow the necessary traffic for your instances to function correctly.

6. CloudWatch metrics or alarms: Check if your CloudWatch alarms and metrics are correctly set up to trigger the scaling policy based on CPU utilization.

To troubleshoot and resolve the issue:

1. Review the Auto Scaling group's Activity history for any error messages or launch failures
2. Check the status of Auto Scaling processes and resume any suspended processes
3. Verify your launch configuration or launch template settings
4. Ensure that you have sufficient EC2 capacity in your chosen Availability Zones
5. Review your IAM roles and permissions
6. Double-check your CloudWatch alarms and metrics configuration

If you continue to experience issues after these steps, you may need to engage AWS support for further assistance in diagnosing the problem.
Sources
Troubleshoot Auto Scaling group stuck in "Updating Capacity" | AWS re:Post
Fix EC2 Auto Scaling group launch failure with capacity errors | AWS re:Post
Troubleshoot unhealthy instances in Amazon EC2 Auto Scaling - Amazon EC2 Auto Scaling