Open Stack

Vote for '--force'. Status update operation smells an overkill in this
case.

- Qiming
On Wed, Sep 13, 2017 at 01:37:44PM +0800, Lee Yi wrote:
> --force will be better, in my opinion.
> 
> On Wed, Sep 13, 2017 at 1:23 PM, <liu.xuefeng1 at zte.com.cn> wrote:
> 
> > You killed mysql? Okay ...
> >
> > Seems we need to add a "--force" parameter to the delete operation.
> >
> > - Qiming
> >
> > On Tue, Sep 12, 2017 at 09:14:25PM +0800, Lee Yi wrote:
> > > When deleting a cluster named 'server_cluster', I
> > shutdown mysql service.
> > > And then the status of cluster will be 'DELETING' forever, like follow:
> > >
> > > {0}nbsp;openstack cluster list
> > > +----------+----------------+----------+--------------------
> > > --+----------------------+
> > > | id       | name           | status   | created_at
> >     | updated_at
> > >           |
> > > +----------+----------------+----------+--------------------
> > > --+----------------------+
> > > | 672d043f | server_cluster | DELETING | 2017-09-12T10:52:16Z |
> > > 2017-09-12T11:16:15Z |
> > > +----------+----------------+----------+--------------------
> > > --+----------------------+
> > >
> > > I can't  delete it again, because:
> > >
> > > {0}nbsp;openstack cluster delete server_cluster
> > > Are you sure you want to delete this cluster(s) [y/N]?y
> > >  server_cluster: failed due to 'HttpException: The
> > cluster 'server_cluster'
> > > is in status DELETING. (HTTP 409) (Request-ID:
> > > req-42eeda10-9055-40c2-b690-5627ac555922),
> > > The cluster 'server_cluster' is in status DELETING.'
> > >
> > > And the action is in WAITING status:
> > >
> > > {0}nbsp;openstack cluster action show d36fc6b5-ac5e-
> > 460e-85dc-afdb680f3242
> >
> > > +---------------+--------------------------------------+
> > > | Field         | Value                                |
> > > +---------------+--------------------------------------+
> > > | action        | CLUSTER_DELETE                       |
> > > | cause         | RPC Request                          |
> > > | created_at    | 2017-09-12T11:16:15Z                 |
> > > | depended_by   |                                      |
> > > | depends_on    | d0f6e1c0-491f-4318-9cc2-41ae2028b99f |
> > > | domain_id     | None                                 |
> > > | end_at        | None                                 |
> > > | id            | d36fc6b5-ac5e-460e-85dc-afdb680f3242 |
> > > | inputs        | {}                                   |
> > > | interval      | -1                                   |
> > > | location      | None                                 |
> > > | name          | cluster_delete_672d043f              |
> > > | outputs       | {}                                   |
> > > | owner_id      | 9dd42d67-f0c6-4199-8c4f-15f3cd88683b |
> > > | project_id    | 3adeed30eb694bdd98257b294120e595     |
> > > | start_at      | 1505214976.0                         |
> > > | status        | WAITING                              |
> > > | status_reason | Waiting for depended actions.        |
> > > | target_id     | 672d043f-ef7b-4a7a-a41a-176a9929296e |
> > > | timeout       | 3600                                 |
> > > | updated_at    | None                                 |
> > > | user_id       | 4b805d08ec1644ef8f69078ddd59941c     |
> > > +---------------+--------------------------------------+
> > >
> > > What can I do about this cluster, and how to delete it again.
> > >
> > > Obviously, I can modify the code (
> > > https://github.com/openstack/senlin/blob/master/
> > senlin/engine/service.py#L878-L881)
> > > to support to delete the cluster again, but I don't
> > think it's the best way.
> > >
> > > The same problems exist  when do other actions.
> > >
> > > Any suggestion?
> >
> > > __________________________________________________________
> > ________________