Updating cluster in virtual machine results in an additional virtual machine with the exact same name

[Sedl]

Ansible NetBox Collection version

v3.14.0

Ansible version

ansible [core 2.15.4]
  config file = /Users/sedlmeier/ansible/venv/lib/python3.9/site-packages/ansible_collections/netbox/netbox/ansible.cfg
  configured module search path = ['/Users/sedlmeier/.ansible/plugins/modules', '/usr/share/ansible/plugins/modules']
  ansible python module location = /Users/sedlmeier/ansible/venv/lib/python3.9/site-packages/ansible
  ansible collection location = /Users/sedlmeier/.ansible/collections:/usr/share/ansible/collections
  executable location = /Users/sedlmeier/ansible/venv/bin/ansible
  python version = 3.9.16 (main, Dec  7 2022, 10:06:04) [Clang 14.0.0 (clang-1400.0.29.202)] (/Users/sedlmeier/ansible/venv/bin/python3.9)
  jinja version = 3.1.2
  libyaml = True

NetBox version

v3.5.1

Python version

3.9

Steps to Reproduce

We wanted to move an existing virtual machines to a different cluster. The task looks like this:

    - name: Updating NetBox
      delegate_to: localhost
      become: false
      netbox.netbox.netbox_virtual_machine:
        netbox_url: '{{ hostvars["localhost"].netbox_url }}'
        netbox_token: '{{ hostvars["localhost"].netbox_token }}'
        data:
          name: '{{ inventory_hostname }}'
          vcpus: 8
          memory: 1024
          disk: 128
          cluster: cluster1.example.name
        state: present

Expected Behavior

Move the existing VM to a different cluster.

Observed Behavior

A new VM with the exact same name and the given data is created on the given cluster.

[sc68cal]

I will have to look at the code but my guess is that name and cluster is the way that a specific VM is looked up when state: present - because you are changing the cluster parameter it is not matching the VM that exists on the old cluster and that is why it is creating a new one.

[Sedl]

Thank you for your answer, this was the hint i needed. I've added query_params to restrict the lookup to name and now it works as i intended:

    - name: Updating NetBox
      delegate_to: localhost
      become: false
      netbox.netbox.netbox_virtual_machine:
        netbox_url: '{{ hostvars["localhost"].netbox_url }}'
        netbox_token: '{{ hostvars["localhost"].netbox_token }}'
        query_params:
          - name
        data:
          name: '{{ inventory_hostname }}'
          vcpus: '{{ vcpus | int }}'
          memory: '{{ mem_mib | int }}'
          disk: '{{ disk_gib | int }}'
          cluster: '{{ _kvm_host }}'
        state: present

The question is if this is the intended default behaviour and if it makes sense to change that. This behaviour is probably not very obvious to most of the users. I think we should at least mention such things somewhere in the documentation.

[sc68cal]

This behavior matches what I have seen with other modules where name is not sufficient to find the object. For example, I had to use slug instead of the name. I believe it was for Site?

I have transferred this to a discussion in the hopes that if someone runs across this, they will find this. I am also fine with adding more documentation around query_params and how this can help.