[Openstack-operators] NOVA build timeout: Time to fail a vm from build state

David Medberry openstack at medberry.net
Wed Jun 15 17:51:00 UTC 2016


Ack. I'm picking my worst case with a 2T volume create and then doubling.

On Wed, Jun 15, 2016 at 11:25 AM, Matt Riedemann <mriedem at linux.vnet.ibm.com
> wrote:

> On 6/15/2016 12:09 PM, David Medberry wrote:
>
>> So, there is a nova.conf setting:
>>
>> instance_build_timeout (default to 0, never timeout)
>>
>> Does anyone have a "good" value they use for this? In my mind it falls
>> very much into the specific-to-your-cloud-implementation bucket but just
>> wondered what folks were usign for this setting (if any).
>>
>> 10 minutes would be way longer than I'd want to wait for a build to fail
>> but that's probably what we will set this to....
>>
>>
>> _______________________________________________
>> OpenStack-operators mailing list
>> OpenStack-operators at lists.openstack.org
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>>
>>
> Hmm, yeah definitely deployment specific. Also, if you're doing boot from
> volume where nova is creating the volume, that's going to add additional
> time depending on the size of the volume, whether the image is cached, etc.
> And there are separate timeouts in the compute manager for waiting for the
> volume to be available for attaching to the server.
>
> --
>
> Thanks,
>
> Matt Riedemann
>
>
> _______________________________________________
> OpenStack-operators mailing list
> OpenStack-operators at lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20160615/5ce76ada/attachment.html>


More information about the OpenStack-operators mailing list