proxmox_kvm: allow suspending and hibernation as vm state and resuming of vms

[ff05]

Summary

I try to bulk hibernate all VMs on a node, rather than shutting them down to reboot this node.
That would save the vmstates.

Issue Type

Feature Idea

Component Name

proxmox_kvm

Additional Information

    - name: hibernate all vms
      community.general.proxmox_kvm:
        api_host: "{{ proxmox_api_host }}"
        api_user: "{{ proxmox_api_user }}"
        api_token_id: "{{ proxmox_api_token_id }}"
        api_token_secret: "{{ proxmox_api_token_secret }}"
        vmid: "{{ vmid }}"
        state: hibernate

possible solution could be:

[...]
    def suspend_vm(self, vm, timeout, todisk):
        vmid = vm['vmid']
        proxmox_node = self.proxmox_api.nodes(vm['node'])
        taskid = proxmox_node.qemu(vmid).status.suspend.post(todisk=(1 if todisk else 0), timeout=timeout)
        if not self.wait_for_task(vm['node'], taskid):
            self.module.fail_json(msg='Reached timeout while waiting for suspending VM. Last line in task before timeout: %s' %
                                  proxmox_node.tasks(taskid).log.get()[:1])
            return False
        return True
    
    def resume_vm(self, vm):
        vmid = vm['vmid']
        proxmox_node = self.proxmox_api.nodes(vm['node'])
        taskid = proxmox_node.qemu(vmid).status.resume.post()
        if not self.wait_for_task(vm['node'], taskid):
            self.module.fail_json(msg='Reached timeout while waiting for suspending VM. Last line in task before timeout: %s' %
                                  proxmox_node.tasks(taskid).log.get()[:1])
            return False
        return True
[...]
def main():
    module_args = proxmox_auth_argument_spec()
    kvm_args = dict(
[...]
        state=dict(default='present', choices=['present', 'absent', 'stopped', 'started', 'restarted', 'current', 'template', 'paused', 'hibernated']),
[...]
    elif state == 'started':
        if not vmid:
            module.fail_json(msg='VM with name = %s does not exist in cluster' % name)

        status = {}
        try:
            vm = proxmox.get_vm(vmid)
            current = proxmox.proxmox_api.nodes(vm['node']).qemu(vmid).status.current.get()['status']
            status['status'] = current
            if current == 'running' and status['qmpstatus'] == 'running':
                module.exit_json(changed=False, vmid=vmid, msg="VM %s is already running" % vmid, **status)

            if status['qmpstatus'] == 'paused' or \
                (current == 'stopped' and status['qmpstatus'] == 'stopped' and status['lock'] == 'suspended'):
                proxmox.resume_vm(vm)
                module.exit_json(changed=True, vmid=vmid, msg="VM %s resumed" % vmid, **status)

            if proxmox.start_vm(vm):
                module.exit_json(changed=True, vmid=vmid, msg="VM %s started" % vmid, **status)
        except Exception as e:
            module.fail_json(vmid=vmid, msg="starting of VM %s failed with exception: %s" % (vmid, e), **status)
[...]
    elif state in ['paused', 'hibernated']:
        if not vmid:
            module.fail_json(msg='VM with name = %s does not exist in cluster' % name)

        status = {}
        try:
            vm = proxmox.get_vm(vmid)
            current = proxmox.proxmox_api.nodes(vm['node']).qemu(vmid).status.current.get()['status']
            status['status'] = current
            if current != 'running':
                module.exit_json(changed=False, vmid=vmid, msg="VM %s is not running" % vmid, **status)

            proxmox.suspend_vm(vm, force=module.params['force'], timeout=module.params['timeout'], todisk=(state == 'hibernated'))
            module.exit_json(changed=True, vmid=vmid, msg="VM %s is suspending" % vmid, **status)
        except Exception as e:
            module.fail_json(vmid=vmid, msg="suspending of VM %s failed with exception: %s" % (vmid, e), **status)

Code of Conduct

• I agree to follow the Ansible Code of Conduct

[ansibullbot]

Files identified in the description:

• plugins/modules/proxmox_kvm.py

If these files are incorrect, please update the component name section of the description or use the !component bot command.

click here for bot help

[ansibullbot]

cc @Ajpantuso @Thulium-Drake @UnderGreen @helldorado @joshainglis @karmab @krauthosting
click here for bot help

[russoz]

Hi @ff05 thanks for your contribution!

Would you be willing to raise a Pull Request with your suggested change?

[ff05]

Hi @russoz
I raised the request here #9653