Unable to evacuate all VMs after compute node failure on OSP 13

Posted on

Hi,

We have an OSP 13 environment and have had a compute node failure where the compute node has lost power.

I have attempted to evacuate all of the VMs to an available host with the below:

$ nova host-evacuate --target_host "target" "source"

It looks like this has been successful for all but 1 of the VMs which is attempting to power off:

| | False | Error while evacuating instance: Cannot 'evacuate' instance "uuid" while it is in task_state powering-off
(HTTP 409) (Request-ID: |

Any power on or power off tasks return a similar response that
$ openstack server stop "uuid"
Cannot 'stop' instance "uuid" while it is in task_state powering-off (HTTP 409) (Request-ID: req-213a57a9-cbbf-4784-bd90-00b8edaa89f4)

How can I force this VM to power off and migrate?

Regards,
Cameron.

Responses