[FEAT] - Add feature to schedule snapshots of `mgc_virtual_machine_instances` (parity with console)

[antonionovaesjr]

Is your feature request related to a problem? Please describe.
Currently, the Magalu Cloud web console allows you to manually take snapshots at the instance (VM) level.

However, the Terraform provider doesn't have an equivalent feature. The only way to schedule snapshots via Terraform is using the mgc_block_storage_schedule_attach feature. The problem is that this feature operates at the individual volume level, requiring us to attach a schedule to each volume, one by one.

This creates two problems:

1. Lack of Parity: The Terraform provider can't automate functionality that already exists in the web console.

Describe the solution you'd like
I'd like to request a new feature (perhaps mgc_block_storage_schedule_instance_attach) that allows attaching an mgc_virtual_machine_instances.my_vm.id.

This feature would instruct the platform to apply the snapshot schedule to that instance, mirroring the behavior of the web console.

Intended usage example:

resource "mgc_virtual_machine_instances" "my_vm" {
name = "critical-server"
machine_type = "BV4-8-100"
image = "cloud-ubuntu-24.04 LTS"
# ... other volumes can be attached here
}

resource "mgc_block_storage_schedule" "daily_backup" {
name = "critical-vm-daily-backup"
snapshot_type = "instant"
policy_retention_in_days = 7
policy_frequency_daily_start_time = "03:00:00"
}

# --- THE NEW PROPOSED FEATURE ---
# Attaches the schedule directly to the instance, just like in the console
resource "mgc_block_storage_schedule_instance_attach" "attach_backup_to_the_vm" {
schedule_id = mgc_block_storage_schedule.backup_daily.id
instance_id = mgc_virtual_machine_instances.my_vm.id
}

Describe alternatives you've considered
Create a Python script and leave it inside the instance

Additional context
Implementing this feature would bring parity between the Terraform provider and the web console, drastically simplifying backup management and making automation with Terraform more robust and reliable.

[luizcavalcanti]

Hi Antonio!

Not sure if I got this.
On the Console you are allowed to take manual snapshots of VMs and volumes.
On the terraform provider you can take manual (VM snapshots). You can also take manual snapshots of volumes or schedule them.

But on the proposed resource you seem to be talking about volumes snapshots and automatically attaching them to running instances, which is something we don't support on any capacity on Console or on the API itself. Am I getting something wrong?

[antonionovaesjr]

Hi @luizcavalcanti , thanks for answer

The idea is to offer a full instance backup on a scheduled basis, similar to what's possible with volumes/block storage. In the console, I can take a snapshot of the entire instance. I can't schedule a backup of the instance's root volume.

[antonionovaesjr]

Hi @luizcavalcanti

Do you think it would be possible to add this feature?