Open Stack

Hi Everyone,

I’ll be straight to the point here.
We have a cloud with Fuel operating it and provisioning CentOS nodes. You
might know that Fuel, by default, creates mdraid if there is more than
three disks on server (3+). In the debug process we are commented out some
rows in one of the nailgun scripts, and not removed that ‘commenting’
symbols after the experiments. As a result, this brought us to the
situation when disk was not accepted and whole node provisioning was broken.

So, here is the question to discuss:
Can we implement some mechanism here that will check such behaviour and
warn user if he/she did wrong in configuring the system (nailgun). OSTF is
not correct place here, because it tests already provisioned nodes.

Thanks to everyone for reading this.

-- 

Alex Savatieiev | QA manager | DevOps | OpenStack Services | Mirantis

http://www.mirantis.com | cell: +380669077482 | skype: savexx
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150119/ba89d59c/attachment.html>