[openstack-dev] [heat] Deployment of Lock / Unlock Operation like MUTEX

Angus Salkeld asalkeld at mirantis.com
Sun Aug 30 22:47:37 UTC 2015


On Sat, Aug 29, 2015 at 10:33 AM Shinobu Kinjo <skinjo at redhat.com> wrote:

> Hello,
>
> Here is a situation which I faced, that can be reproduced.
>
> I did "heat resource-signal" multiple times simultaneously to scale
> instance.
> As a result 3 resources were made in the scaling group having max_size=2.
>
> "stack-list -n" showed me 3 stacks in one parent.
>
> Heat itself seems not to actually check resource data in the database.
>
> Is there any lock/unlock mechanism like mutex in heat implementation like
> locking
> database when auto-scaling feature is triggered.
>
>
Hi

In master there is such a check:
https://github.com/openstack/heat/blob/master/heat/scaling/cooldown.py#L33-L50

This makes sure that there are no concurrent scaling actions.

-Angus



> Or is there plan to deploy such a mutex mechanism.
> What I'm concerning about more is that cielometer also has some feature to
> triggering
> auto-scaling.
>
> So I would like to make sure that there is a mechanism to keep data
> consistency on
> each component.
>
> Please let me know, if I've missed anything.
>
> Shinobu
>
> __________________________________________________________________________
> 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
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150830/b9baa85a/attachment.html>


More information about the OpenStack-dev mailing list