EC2 autoscale creates instances exceeding maximum number, falling into an infinite loop.

Hello.

You can contact AWS Support from the URL below.
https://console.aws.amazon.com/support

Since the status of EC2 is all "Terminated", does that mean that you have stopped and restarted it repeatedly?
Also, can I check how many EC2 units were started at the same time from the metrics of the AutoScaling group?

If the ECS task placement strategy is set to "spread", such a behavior may occur.
https://docs.aws.amazon.com/AmazonECS/latest/developerguide/task-placement-strategies.html

Thank Riku_Kobayashi,

Since the status of EC2 is all "Terminated", does that mean that you have stopped and restarted it repeatedly? -> The auto-scaling group handled it. I stopped instances in the morning when the instance count was over 31.

If the ECS task placement strategy is set to "spread", such a behavior may occur. -> I'm not sure. In the morning, the issue was still there, after the VPC Route tables were fixed. I just stopped the redundant instances and changed the desired capacity in the Auto Scaling Group, and now the issue has disappeared.

It's very strange!

Anyway, thank you very much. I will contact AWS for this case.