yahoo-eng-team team mailing list archive

[Tao Li <litao3721@xxxxxxx>]

Public bug reported:

Description
===========

Currently， the compute manager does not update the instance_info_caches
of the instance again after the instance spawning successfully. so the
status of port is also down in this time. Depending on the period task
_heal_instance_info_cache to synchronize the port status. If the
instances in the host are many, maybe the time for synchronizing the
status of port is long. If evacuate operation is triggered for the
instance， the network-vif-plugged event will send to the source host,
then the new instance will spawn failed in destination host because of
waiting for network-vif-plugged event timeout. The probability of this
question is big relatively.

Steps to reproduce
==================

1. Booting 5 or more instance in one host.
2. After instances were spawning successfully， power off the host at once.
3. Then evacuate the instance in this host.
4. execute 'nova migration-list' to search failed records

Expected result
===============
instances will be evacuated successfully.

Actual result
=============
Some instance will be evacuated failed because of waiting for network-vif-plugged event timeout.

** Affects: nova
     Importance: Undecided
         Status: New

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1652735

Title:
  we should update instance_info_caches after the instance spawning
  successfully

Status in OpenStack Compute (nova):
  New

Bug description:
  Description
  ===========

  Currently， the compute manager does not update the
  instance_info_caches of the instance again after the instance spawning
  successfully. so the status of port is also down in this time.
  Depending on the period task _heal_instance_info_cache to synchronize
  the port status. If the instances in the host are many, maybe the time
  for synchronizing the status of port is long. If evacuate operation is
  triggered for the instance， the network-vif-plugged event will send to
  the source host, then the new instance will spawn failed in
  destination host because of waiting for network-vif-plugged event
  timeout. The probability of this question is big relatively.

  Steps to reproduce
  ==================

  1. Booting 5 or more instance in one host.
  2. After instances were spawning successfully， power off the host at once.
  3. Then evacuate the instance in this host.
  4. execute 'nova migration-list' to search failed records

  Expected result
  ===============
  instances will be evacuated successfully.

  Actual result
  =============
  Some instance will be evacuated failed because of waiting for network-vif-plugged event timeout.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1652735/+subscriptions