Open Stack

hi, hao,

First, thanks for your bug and patch, I thinks it's really a problem
we should fix. I forward your suggestions about the workaround with
Dan Smith and Joe Gordon.to get more feedback.

1.Raise exception to make to set vm status to error,and user can
re-perform a delete action to clean up successfully.
2.Add full set of cinder admin credentials in nova.conf(just like what
neutron does with nova) and make sure nova can clean up in init_host



On Tue, Apr 28, 2015 at 8:21 PM, hao wang <sxmatch1986 at gmail.com> wrote:
> Hi, everyone
>
> There is a Nova bug:
>
> https://bugs.launchpad.net/nova/+bug/1408865.
> https://review.openstack.org/#/c/147042/
>
> The bug Scenario is:
>
> 1. create a vm using bootable volume.
> 2. delete this vm
> 3. restart service nova-compute when vm's task state is deleting.
>
> When nova-compute is up, vm became deleted successful, but the bootable
> volume is still in-use state and can't delete it using cinder delete volume.
>
> solve method:
>
> Add init=True in _delete_instance when init_host, and raise exception
> when EndpointNotFound exists. It will set vm's status to error and
> make user can re-issue a delete.
>
>
> Is this method ok? Or cinder could do something to fix this bug?
>
>
> I need your suggestion to push this work forward.
>
>
> Thanks.
>
> wanghao
>
>
> --
>
> Best Wishes For You!
>
>
> __________________________________________________________________________
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: OpenStack-dev-request at lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>



-- 
Regards!
-----------------------------------
Lingxian Kong