Getting an error while detaching EBS volume(s) from instance(s) on pulumi.

aws:ec2:VolumeAttachment (ebsatt1-asg03):
    error: deleting urn:pulumi:asg-ex-03::asg::aws:ec2/volumeAttachment:VolumeAttachment::ebsatt1-asg03: 1 error occurred:
    	* Failed to detach Volume (vol-*********) from Instance (i-*******): IncorrectState: Volume 'vol-*******' is in the 'available' state.
    	status code: 400, request id: 80154710-********

While checking it from the Ui, it shows the volumes are already detached from the instances, also instances are terminated. I deleted the volumes manually, but showing the same error messages while running pulumi destroy. Can someone explain the issue here?

The main error is because Pulumi still thinks the volume is attached and gets an error back from AWS when it tries to detach an “available” volume. I’m not sure how it got in this state, but if you do a

pulumi refresh -y

that will synchronize Pulumi state with AWS and then allow you to do the

pulumi destroy

Well EBS state

available

means the EBS volumes are already detached from the instance and it’s

available

to use/attach to other instances. It make sense while pulumi prompts an error message of “failed to detach volume” while EBS is

in-use

state.

Right. That’s because Pulumi’s state model still has the volume as attached and is telling AWS to detach it. AWS is then coming back with an error message saying it’s already detached - that’ the

status code: 400...

bit. Did you try the

pulumi refresh

? That will update Pulumi’s state to synchronize it with AWS - essentially detaching the volume in the state model. That should then allow subsequent

pulumi up

or

pulumi destroy

to proceed as expected.

would you mind sharing your code? my suspicion here is that it's deleting the ASG before the detach call can happen