Disk could be defined as IMAGE in VM template, not only IMAGE_ID

[deynekoaa]

Description

Problem: Disk size not changed after creating VMs from first template

Goal:
terraform can not work with VM templates, that have disk info as IMAGE (i.e. name), not as IMAGE_ID (identifier)
First template

DISK = [
  IMAGE = "ubuntu-22.04-amd64.base_image.test_system.stable.b20230812-7.ceph-ssd",
  IMAGE_SIZE = "4096" ]

Output terraform plan:

      + disk                   = [
          + {
              + cache                    = null
              + computed_cache           = (known after apply)
              + computed_dev_prefix      = (known after apply)
              + computed_discard         = (known after apply)
              + computed_driver          = (known after apply)
              + computed_io              = (known after apply)
              + computed_size            = (known after apply)
              + computed_target          = (known after apply)
              + computed_volatile_format = (known after apply)
              + dev_prefix               = null
              + discard                  = null
              + disk_id                  = (known after apply)
              + driver                   = null
              + image_id                 = -1
              + io                       = null
              + size                     = 15360
              + target                   = null
              + volatile_format          = null
              + volatile_type            = null
            },
        ]

Second template:

DISK = [
  IMAGE_ID = "81726",
  IMAGE_SIZE = "4096" ]

Output terraform plan:

      + disk                   = [
          + {
              + cache                    = null
              + computed_cache           = (known after apply)
              + computed_dev_prefix      = (known after apply)
              + computed_discard         = (known after apply)
              + computed_driver          = (known after apply)
              + computed_io              = (known after apply)
              + computed_size            = (known after apply)
              + computed_target          = (known after apply)
              + computed_volatile_format = (known after apply)
              + dev_prefix               = null
              + discard                  = null
              + disk_id                  = (known after apply)
              + driver                   = null
              + image_id                 = 81726
              + io                       = null
              + size                     = 15360
              + target                   = null
              + volatile_format          = null
              + volatile_type            = null
            },
        ]

As result, disk size changed in VM from second template

Terraform and Provider version

# terraform -v
Terraform v1.5.5
on linux_amd64
...
- Installed opennebula/opennebula v1.3.0 (self-signed, key ID A0224DDC2BF90FA7)
...
OpenNebula 5.10.1

Affected resources and data sources

opennebula_virtual_machine

Terraform configuration

terraform {
  required_providers {
    opennebula = {
      source = "OpenNebula/opennebula"
      version = "~> 1.3"
    }
  }
}

locals {
  config  = yamldecode(file("configfile.yaml"))
}

provider "opennebula" {
  endpoint = local.config.one_endpoint
  username = local.config.one_username
  password = local.config.one_password
}

data "opennebula_templates" "one_templates" {
  name_regex = local.config.template_regex
  has_cpu    = true
  sort_on   = "register_date"
  order      = "ASC"
}

resource "opennebula_virtual_machine" "one_vm" {
  name = local.config.vm_name
  description = "terraform"

  template_id = data.opennebula_templates.one_templates.templates[0].id

  disk {
    size = 15360 # 15GB
  }
}

output "one_vm_ip" {
  description = "IP address of the opennebula VM"
  value       = resource.opennebula_virtual_machine.one_vm.ip
}

Expected behavior

Both VMs has 15Gb disk size

Actual behavior

Only VM from second template has 15Gb disk size

Steps to Reproduce

Just try to create VMs from two templates

Debug output

No response

Panic output

No response

Important factoids

No response

References

No response

[github-actions]

This issue is stale because it has been open for 30 days with no activity and it has not the 'status: confirmed' label or it is not in a milestone. Remove the 'status: stale' label or comment, or this will be closed in 5 days.

[deynekoaa]

Still actual

[treywelsh]

Running TF_LOG=TRACE TF_LOG_PATH=terraform.log terraform apply you'll have some more logs, it may help.
There's a chance to see this message in logs: "[INFO] ignore disk without image_id and type"

Digging the provider code and history a to understand:

The problem may be due to this line
The provider check if the image_id is not equal to -1 (a default value)

image_id value to -1 by default was added by PR #75 to solve issue #71, which could be discussed (is it legitimate etc.)

At first glance we could try to add a check on the image attribute, however, I'll need to make a bunch of test to check if it break something else

[github-actions]

This issue is stale because it has been open for 30 days with no activity and it has not the 'status: confirmed' label or it is not in a milestone. Remove the 'status: stale' label or comment, or this will be closed in 5 days.

[github-actions]

This issue is stale because it has been open for 30 days with no activity and it has not the 'status: confirmed' label or it is not in a milestone. Remove the 'status: stale' label or comment, or this will be closed in 5 days.