Instances stop after Instance-HA due to kernel panic

Issue

• To collect vmcore, delay of fence_ipmilan is set.

  # pcs config
         :
   Resource: stonith-fence_ipmilan-aaaaaaaaaaaa (class=stonith type=fence_ipmilan)
    Attributes: delay=120 ipaddr=10.0.0.1 lanplus=true login=user passwd=******** pcmk_host_list=compute-0 privlvl=administrator
                ^^^^^^^^^(*) delay is set to 120 seconds in this example.
    Operations: monitor interval=60s (stonith-fence_ipmilan-aaaaaaaaaaaa-monitor-interval-60s)
         :
  

• After instance-HA due to a kernel panic, instances which was running on the Compute node stopped.

  # Before instance HA
  $ openstack server show aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa
      :
  | OS-EXT-STS:power_state              | Running                                                          |
  | OS-EXT-STS:task_state               | None                                                             |
  | OS-EXT-STS:vm_state                 | active                                                           |
      :
  | status                              | ACTIVE                                                           |
  
  # After instance HA
  $ openstack server show aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa
      :
  | OS-EXT-STS:power_state              | Shutdown                                                         |
  | OS-EXT-STS:task_state               | None                                                             |
  | OS-EXT-STS:vm_state                 | stopped                                                          |
      :
  | status                              | SHUTOFF                                                          |