[scaleway provider] scaleway_volume_attachment does not get destroyed and thus can't migrate from C2S to C2M

[polymeris]

Terraform Version

v0.9.5

Affected Resource(s)

• scaleway_volume_attachment
• scaleway_server

Terraform Configuration Files

C2S

{
	"provider": {
		"scaleway": {
			"region": "par1"
		}
	},
	"resource": {
		"scaleway_volume": {
			"data": {
				"name": "test-data",
				"size_in_gb": 25,
				"type": "l_ssd"
			}
		},
		"scaleway_volume_attachment": {
			"data": {
				"server": "${scaleway_server.host.id}",
				"volume": "${scaleway_volume.data.id}"
			}
		},
		"scaleway_server": {
			"host": {
				"name": "test",
				"type": "C2S",
				"image": "d0c50b37-f0f5-4924-97b0-1fd40c457efe"
			}
		},
		"scaleway_ip": {
			"ip": {
				"server": "${scaleway_server.host.id}"
			}
		}
	}
}

Steps to Reproduce

Please list the steps required to reproduce the issue, for example:

0. Create the above file
1. terraform apply
2. Edit the file, change the type from C2S to C2M
3. terraform apply

Expected Behavior

• A new C2M server gets created
• It is attached to the existing "test-data" volume
• Reserved IP is associated with it
• Both the original C2S server and its main volume get destroyed.

Actual Behavior

[...]
Error applying plan:

1 error(s) occurred:

* scaleway_volume_attachment.data (destroy): 1 error(s) occurred:

* scaleway_volume_attachment.data: timeout while waiting for state to become 'success' (timeout: 5m0s)
[...]

None of the resources (server, volumes, volume_attachments, ips) get destroyed.

I suspect the issue is with the server not being stopped before trying to detach the volume. (EDIT: maybe not, manually stopping the server does not help)

Edit: logs

https://gist.github.com/polymeris/4b74548ee854709ff12a2ea185c1fbce

[nicolai86]

Thank you for the bug report, @polymeris .

In this case the best we can probably do is to increase the timeout.
To change volumes on a scaleway server the server needs to be stopped, and to stop a server all data is transfered from the attached volumes. there are some cases in the scaleway community forum where this apparently took more than 24 hours.

From your logs I can see that terraform only waited 5 minutes which is far too short. I'll look into this.

[polymeris]

Is that just for "baremetal" servers?

I tried stopping the server manually before terraform apply and still get the same issue, btw.

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.