This topic describes the cooldown time in Auto Scaling.

Cooldown time refers to a period during which Auto Scaling cannot execute any new scaling activity after one scaling activity is executed successfully in a scaling group. You can configure the cooldown time in either of the following ways:

Note If you configure the cooldown time in both the scaling group and scaling rule, Auto Scaling gives preferences to the one configured in the scaling rule.

Cooldown time rules

After a scaling group successfully executes a scaling activity, Auto Scaling starts to calculate the cooldown time. If multiple Elastic Compute Service (ECS) instances are added to or removed from the scaling group in the scaling activity, the cooldown time is calculated since the last instance is added to or removed from the scaling group. For more information, see Examples. If no ECS instance is successfully added to or removed from the scaling group during the scaling activity, the cooldown time is not calculated.

Within the cooldown time, the scaling group rejects all scaling activity requests triggered by alarm tasks from CloudMonitor. However, the scaling activity triggered by other types of tasks (manually triggered tasks and scheduled tasks) can be executed immediately without being limited by the cooldown time.

If you disable a scaling group and then enable it again, the cooldown time becomes invalid. For more information, see Example 2.

Note The cooldown time only locks the scaling activities in the same scaling group. It does not affect the scaling activities in other scaling groups.

Examples

Example 1

You have a scaling group named asg-uf6f3xewn3dvz4bs****. The default cooldown time is 10 minutes, and a scaling rule add3 exists in the scaling group with a cooldown time of 15 minutes.

After a scaling activity is successfully executed based on scaling rule add3, three ECS instances are added to the scaling group. The cooldown time is calculated since the third instance is added. Within 15 minutes, scaling activity requests triggered by alarm tasks from CloudMonitor are rejected.

Example 2

You have a scaling group named asg-m5efkz67re9x7a57****. The default cooldown time is 10 minutes, and a scaling rule remove1 exists in the scaling group without cooldown time set.

After a scaling activity is successfully executed based on scaling rule remove1 at 18:00, one ECS instance is removed from the scaling group. Normally, scaling activity requests triggered by alarm tasks from CloudMonitor are rejected before 18:10. If you disable the scaling group and then enable it again at 18:05, the cooldown time becomes invalid. If a scaling activity request is triggered by an alarm task from CloudMonitor from 18:05 to 18:10, the scaling group accepts and executes the request.