instance turned to "error" state after reboot compute node

[tanghang]

Description
===========
The instance turned to "error" state after reboot compute node. And I tried the "nova reset-state <id> --active" command, the instance turned into active state. But, when I login the compute node and use "virsh list" command, there are no instance on the compute node.Then, I ran the "openstack server stop" and "openstack server start" command.After that, the server recovered.

Steps to reproduce
==================
1. Create a instance
2. Confirm the compute node the instance running on
3. Reboot the compute node

===============
After the execution of the steps above, what should have
happened if the issue wasn't present?

the server shoud migrated to other compute nodes or should recover automatically after the conpute node reboot.
Actual result
=============
the server turned into "error" state

Environment
===========
1. Exact version of OpenStack you are running. See the following
  list for all releases: http://docs.openstack.org/releases/
       18.2.3

2. Which hypervisor did you use?
   Libvirt + KVM

2. Which storage type did you use?
   Ceph

3. Which networking type did you use?
   Neutron with OpenVSwitch

[Balazs Gibizer (balazs-gibizer)]

Could you please provide the nova logs from the compute host where the instance was running at step #2 in your reproduction?

[Launchpad Janitor (janitor)]

[Expired for OpenStack Compute (nova) because there has been no activity for 60 days.]