Target-based auto-scaling policy in gamelift not working properly based on configured buffer of available game session

If your target-based auto scaling is not triggering new instance creation when the active game session limit reaches the desired threshold, there could be a few possible reasons for this behavior. Here are some guidelines to help you troubleshoot and handle this issue:

Check Auto Scaling Policies: Verify that the auto scaling policies are correctly configured. Ensure that you have set the scaling policy to trigger when the average number of active game sessions exceeds the desired threshold (e.g., 40 in your case). Review the scaling policies' settings and make sure they align with your requirements.

Evaluate Scaling Metrics: Assess the metrics used for scaling decisions. Determine whether the metrics being monitored accurately reflect the number of active game sessions. It's possible that the metric being used is not capturing the correct information, resulting in incorrect scaling actions. For example, if the scaling metric is based on CPU utilization, it may not correlate directly with the number of active game sessions.

Monitor Scaling Activities: Keep track of the scaling activities and check the CloudWatch metrics for the fleet. This will help you identify whether the scaling activities are being triggered and if they correspond to the expected behavior. Review the scaling events, timestamps, and associated metric values to gain insights into the scaling decisions.

Consider Scaling Cooldown Period: Verify that a scaling cooldown period is not impeding the scaling actions. The cooldown period is a configurable interval during which Auto Scaling waits before triggering additional scaling activities. If the cooldown period is set too long, it could delay the creation of new instances, even if the conditions are met.

Review Scaling Alarms: If you are using CloudWatch Alarms to trigger the scaling actions, verify that the alarm thresholds and settings are correctly configured. Ensure that the alarms are triggering at the desired thresholds based on the appropriate metrics.

Evaluate Scaling Policies: Assess the scaling policies and their configuration. Ensure that the scaling policies are properly adjusted to achieve the desired behavior. For example, if the scaling policy has a scaling adjustment value that is too small, it might not trigger new instance creation until a significant increase in load occurs.

Validate Fleet Target Configuration: Double-check the fleet target configuration to ensure it accurately reflects your desired scaling behavior. Ensure that the buffer values and target capacities are set correctly for your game session instances. If necessary, make adjustments to align the configuration with your requirements.