ebs_block_volume causing a refresh

[mzupan]

I have the following

resource "aws_instance" "ci" {
  ami = "${lookup(var.ami, var.env)}"
  instance_type = "${lookup(var.instance-type-ci, var.env)}"
  key_name = "${lookup(var.key, var.env)}"
  vpc_security_group_ids = [
      "${aws_security_group.internal-default.id}",
      "${aws_security_group.internal-web.id}",

  ]
  subnet_id = "${aws_subnet.mgmt.2.id}"

  iam_instance_profile = "${aws_iam_instance_profile.ci.id}"

  tags {
      CostCenter = "web"
      Name = "WEB ci.${var.env}"
      Roles = "ci"
  }
  user_data = "ci.${var.env}.urthecast.com"
  private_ip = "${lookup(var.vpc-subnet, var.env)}.2.20"


  ebs_block_device = {
    device_name = "/dev/xvdj"
    volume_size = "200"
    delete_on_termination = false
    volume_type = "gp2"
  }
}

When I do a plan I get

-/+ aws_instance.ci
    ami:                                               "ami-c4a1b1ac" => "ami-c4a1b1ac"
    availability_zone:                                 "us-east-1d" => "<computed>"
    ebs_block_device.#:                                "2" => "1"
    ebs_block_device.3871590789.delete_on_termination: "false" => "0" (forces new resource)
    ebs_block_device.3871590789.device_name:           "/dev/xvdj" => "/dev/xvdj" (forces new resource)
    ebs_block_device.3871590789.encrypted:             "false" => "<computed>"
    ebs_block_device.3871590789.iops:                  "600" => "<computed>"
    ebs_block_device.3871590789.snapshot_id:           "" => "<computed>"
    ebs_block_device.3871590789.volume_size:           "200" => "200" (forces new resource)
    ebs_block_device.3871590789.volume_type:           "gp2" => "gp2" (forces new resource)

its an EBS backed instance.. here's my state

                "aws_instance.ci": {
                    "type": "aws_instance",
                    "depends_on": [
                        "aws_iam_instance_profile.ci",
                        "aws_security_group.internal-default",
                        "aws_security_group.internal-web",
                        "aws_subnet.mgmt"
                    ],
                    "primary": {
                        "id": "i-479604b8",
                        "attributes": {
                            "ami": "ami-c4a1b1ac",
                            "availability_zone": "us-east-1d",
                            "ebs_block_device.#": "2",
                            "ebs_block_device.3871590789.delete_on_termination": "false",
                            "ebs_block_device.3871590789.device_name": "/dev/xvdj",
                            "ebs_block_device.3871590789.encrypted": "false",
                            "ebs_block_device.3871590789.iops": "600",
                            "ebs_block_device.3871590789.snapshot_id": "",
                            "ebs_block_device.3871590789.volume_size": "200",
                            "ebs_block_device.3871590789.volume_type": "gp2",
                            "ebs_block_device.630982455.delete_on_termination": "true",
                            "ebs_block_device.630982455.device_name": "/dev/sda",
                            "ebs_block_device.630982455.encrypted": "false",
                            "ebs_block_device.630982455.iops": "60",
                            "ebs_block_device.630982455.snapshot_id": "snap-314ff043",
                            "ebs_block_device.630982455.volume_size": "20",
                            "ebs_block_device.630982455.volume_type": "gp2",

Terraform v0.5.2

[phinze]

Hi @mzupan,

Just tested with a new instance to check if there was a perpetual refresh issue, which I'm not seeing over here. So I'm wondering if you're asking something different - did you add the ebs_block_device to the existing instance?

Modifying any ebs_block_device on an instance currently requires resource replacement [1]. If you'd like to add an EBS volume to an existing instance you'll need to use the ebs_volume resource along with the forthcoming volume_attachment #2050

Let me know if I'm missing something here - happy to dig in further.

[1] See Block Devices section here: https://terraform.io/docs/providers/aws/r/instance.html

[mzupan]

i'm gonna close it since i'm using that PR now and its working as expected.

[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.