instance_server: resizing root volume results in replacement

[Mika56]

Community Note

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• Please do not leave "+1" or other comments that do not add relevant new information or questions, they generate extra noise for issue followers and do not help prioritize the request
• If you are interested in working on this issue or have submitted a pull request, please leave a comment

Terraform Version

Terraform v1.8.4
on linux_amd64
+ provider registry.terraform.io/scaleway/scaleway v2.40.0

Affected Resource(s)

• scaleway_instance_server

Terraform Configuration Files

resource "scaleway_instance_ip" "public_ip" {}

resource "scaleway_instance_server" "web" {
  type = "DEV1-S"
  image = "debian_bullseye"
  ip_id = scaleway_instance_ip.public_ip.id

  root_volume {
    size_in_gb  = 40
    volume_type = "b_sdd"
  }
}

Plan Output

  # module.scaleway-<redacted>.scaleway_instance_server.server must be replaced
-/+ resource "scaleway_instance_server" "server" {
      ~ bootscript_id                    = "fdfe150f-a870-4ce4-b432-9f56b5b995c1" -> (known after apply)
      + cloud_init                       = (known after apply)
      - enable_ipv6                      = false -> null
      ~ id                               = "fr-par-1/22d88a83-511f-49ca-82b9-b6610256e832" -> (known after apply)
      ~ ip_ids                           = [
          ~ "d90cb452-95f1-4387-8adf-79ebd4488b91" -> "fr-par-1/d90cb452-95f1-4387-8adf-79ebd4488b91",
          ~ "c2ea6f43-32fb-473a-a8aa-48c6824abd3e" -> "fr-par-1/c2ea6f43-32fb-473a-a8aa-48c6824abd3e",
        ]
      + ipv6_address                     = (known after apply)
      + ipv6_gateway                     = (known after apply)
      ~ ipv6_prefix_length               = 0 -> (known after apply)
        name                             = "<redacted>"
      ~ organization_id                  = "0647354e-9577-4f39-914e-da1159a2cbe7" -> (known after apply)
      + placement_group_policy_respected = (known after apply)
      + private_ip                       = (known after apply)
      ~ public_ip                        = "<redacted>" -> (known after apply)
      ~ user_data                        = {} -> (known after apply)
        # (12 unchanged attributes hidden)

      - public_ips {
          - address = "<redacted>" -> null
          - id      = "fr-par-1/d90cb452-95f1-4387-8adf-79ebd4488b91" -> null
        }
      - public_ips {
          - address = "<redacted>" -> null
          - id      = "fr-par-1/c2ea6f43-32fb-473a-a8aa-48c6824abd3e" -> null
        }

      ~ root_volume {
          ~ name                  = "Debian Bookworm" -> (known after apply)
          ~ size_in_gb            = 40 -> 45 # forces replacement
          ~ volume_id             = "fr-par-1/d519ba07-0d33-4eab-8c4f-cf8a2185d57d" -> (known after apply)
            # (3 unchanged attributes hidden)
        }
    }

Expected Behavior

Root volume size is increased

Actual Behavior

Whole instance is replaced

Steps to Reproduce

1. Create a basic scaleway_instance_server resource
2. terraform apply
3. Increase its root_volume.size_in_gb value
4. terraform apply

Important Factoids

References

• Docs says that updating this field will recreate the resource (which is the case)
• Same docs say that updates to this field will be ignored after creation (which is false)
• Console allows for hot resizing of root volumes. Terraform should be able to do the same (reducing a volume would still recreate the server)