Raw Disk is not found during refresh because format is a driver instead of raw

[markush81]

Description

With vCenter a RAW disk is going to be created each time again, because FORMAT in TEMPLATE is vcenter instead of raw.

Terraform and Provider version

Terraform v1.9.3
on darwin_arm64

• provider registry.terraform.io/hashicorp/vsphere v2.7.0
• provider registry.terraform.io/opennebula/opennebula v1.4.0

Your version of Terraform is out of date! The latest version
is 1.9.5. You can update by downloading from https://www.terraform.io/downloads.html

Affected resources and data sources

opennebula_virtual_machine

Terraform configuration

disk {
      image_id = ...
      size            = 10240
      target          = "sda"
    }
    disk {
      volatile_format = "raw"
      volatile_type   = "fs"
      size            = 10240
      target          = "sdb"
    }

### Expected behavior

terraform apply -> Creation
terraform apply -> No Change

### Actual behavior

terraform apply -> Creation
terraform apply -> Change

  + disk {
      + image_id        = -1
      + size            = 10240
      + target          = "sdb"
      + volatile_format = "raw"
      + volatile_type   = "fs"
    }

### Steps to Reproduce

> 1x time `terraform apply`

### Debug output

Compile Plugin with certain additional output

DISK=[
ALLOW_ORPHANS="NO",
CLUSTER_ID="100",
DATASTORE="ssd-003279(SYS)",
DATASTORE_ID="109",
DEV_PREFIX="sd",
DISK_ID="1",
DISK_TYPE="FILE",
DRIVER="vcenter",
FORMAT="vcenter", <----------
SIZE="10240",
TARGET="sdb",
TM_MAD="vcenter",
TYPE="fs",
VCENTER_DS_REF="datastore-1056",
VCENTER_INSTANCE_ID="a59e28b4-7adc-4118-8338-e2a3462aa431" ]
...
2024-09-11T15:35:15.244+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 sdb sdb true: timestamp="2024-09-11T15:35:15.244+0200"
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 10240 10240 true: timestamp="2024-09-11T15:35:15.244+0200"
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 vcenter true: timestamp="2024-09-11T15:35:15.244+0200"
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 sd true: timestamp="2024-09-11T15:35:15.244+0200"
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 true: timestamp="2024-09-11T15:35:15.244+0200"
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 true: timestamp="2024-09-11T15:35:15.244+0200"
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 true: timestamp="2024-09-11T15:35:15.244+0200"
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 fs fs true: timestamp="2024-09-11T15:35:15.244+0200"
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 raw vcenter false: timestamp="2024-09-11T15:35:15.244+0200"
...
2024-09-11T15:35:15.245+0200 [INFO] provider.terraform-provider-opennebula: 2024/09/11 15:35:15 [WARN] Configuration for disk ID: 1 not found.: timestamp="2024-09-11T15:35:15.244+0200"
...
2024-09-11T15:35:15.248+0200 [WARN] Provider "registry.terraform.io/opennebula/opennebula" produced an unexpected new value for opennebula_virtual_machine.opennebula_vm during refresh.
- .disk: block count changed from 2 to 1

The part being response for the "wrong" FORMAT seems to be https://github.com/OpenNebula/one/blob/master/src/datastore/Datastore.cc#L218, also

TM_MAD_CONF = [
    NAME = "vcenter", LN_TARGET = "NONE", CLONE_TARGET = "SYSTEM", SHARED = "YES"
]

does not define a DRIVER.

It is unclear to me, to what exactly "system DS template" is referring to, but i haven't found any "DRIVER = ..." configuration at all, related to vcenter.

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.

[markush81]

still relevant and not only for vcenter, according to code

[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.

[markush81]

still relevant and not only for vcenter, according to code

[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.

[markush81]

keep open

[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.

[markush81]

still existent