ECS Autoscaling Group and EC2 Instance not deleted by delete stack CDK typescript

Question

In my Stack I have a Autoscaling group with an EC2 Instance. If I delete the stack this two resource will stay and I have to delete them manualy. I expected that CDK Stack will get deleteled complete. The code for this inside my stack is:

```
		const ec2AutoScalingGroup = new AutoScalingGroup(this, 'todo-app-asg', {
			vpc: this.mainVpc,
			instanceType: new InstanceType('t3a.2xlarge'), 
			machineImage: EcsOptimizedImage.amazonLinux2(AmiHardwareType.STANDARD),
			desiredCapacity: 2,
			minCapacity: 1,
			maxCapacity: 3,
			spotPrice: '0.5',
		  });		
		  ec2AutoScalingGroup.protectNewInstancesFromScaleIn();
		  ec2AutoScalingGroup.connections.addSecurityGroup(securityGroup);

const capacityProviderDef = new AsgCapacityProvider(this, 'todo-app-capacity-provider', {
			autoScalingGroup: ec2AutoScalingGroup,
			enableManagedTerminationProtection: true,
			enableManagedScaling: true,			
		  });
	  		  
		  cluster.addAsgCapacityProvider(capacityProviderDef);
```

re:Post Agent · Answer

The issue you're experiencing with the EC2 Auto Scaling group and EC2 instances not being deleted when you delete the CDK stack is likely due to the configuration of your Auto Scaling group and capacity provider. There are a few factors contributing to this behavior:

1. Instance protection: You're using the `protectNewInstancesFromScaleIn()` method on your Auto Scaling group. This protects instances from being terminated during scale-in events, which can interfere with the deletion process.

2. Managed termination protection: In your capacity provider definition, you've set `enableManagedTerminationProtection: true`. This feature protects instances that are part of an ECS task from being terminated, which can prevent the Auto Scaling group from being deleted.

3. Spot instances: You're using spot instances (`spotPrice: '0.5'`), which may have different termination behaviors compared to on-demand instances.

To ensure that your Auto Scaling group and EC2 instances are properly deleted when you delete the stack, you can try the following modifications:

1. Remove the `protectNewInstancesFromScaleIn()` call if you don't need this protection.

2. Set `enableManagedTerminationProtection: false` in your capacity provider definition.

3. Consider setting a lower `desiredCapacity` or even setting it to 0 before deleting the stack.

4. If you're using the `--force-delete` option when deleting the stack, be aware that it might not work as expected with protected instances.

If you still encounter issues after making these changes, you may need to manually set the desired capacity of the Auto Scaling group to 0 before deleting the stack, or use the AWS CLI or SDK to force-delete the Auto Scaling group and terminate any remaining instances.

Remember that when working with Auto Scaling groups and capacity providers, it's important to carefully consider the protection and termination settings to ensure proper cleanup when resources are no longer needed.  
**Sources**  
[Delete your Auto Scaling infrastructure - Amazon EC2 Auto Scaling](https://docs.aws.amazon.com/autoscaling/ec2/userguide/as-process-shutdown.html)  
[DeleteAutoScalingGroup - Amazon EC2 Auto Scaling](https://docs.aws.amazon.com/autoscaling/ec2/APIReference/API_DeleteAutoScalingGroup.html)  
[How to use EC2 Auto Scaling with EC2 Spot Instances](https://aws.amazon.com/tutorials/ec2-auto-scaling-spot-instances/)

ECS Autoscaling Group and EC2 Instance not deleted by delete stack CDK typescript

Add your answer

Relevant content