OS::Heat::InstanceGroup deploy Bare-metal instance

asked 2014-03-02 22:27:52 -0600

nora gravatar image

Hi, all. I recently use OS::Heat::InstanceGroup to deploy multiple Bare-metal instances. I Also hope to attach network for these Bare-metal instances. However, the resource OS::Heat::InstanceGroup and AWS::AutoScaling::LaunchConfiguration both has no network property.

What can I do to attach network for Bare-metal instance? I have already tried some way.

  1. Use AWS::AutoScaling::AutoScalingGroup instead of OS::Heat::InstanceGroup because AWS::AutoScaling::AutoScalingGroup has a property named "VPCZoneIdentifier". However, this method can only apply to non-Bare-metal instance. There would be error "no valid host found" if used to deploy Bare-metal instance. I checked the log and the error is "PortNotUseable" so that I suspect the error is related to my setting the property "VPCZoneIdentifier".

  2. I changed the code in /usr/lib/python2.7/dist-packages/heat/engine/resources/autoscaling.py. I added a property "SubnetId" in resource OS::Heat::InstanceGroup, and it works. I don't know if it is a proper way to solve this problem.

Could somebody help me?

edit retag flag offensive close merge delete


I think the second way looks more reasonable.

Qiming gravatar imageQiming ( 2015-02-12 23:14:23 -0600 )edit

1 answer

Sort by ยป oldest newest most voted

answered 2015-09-24 16:16:51 -0600

Steve Baker gravatar image

I would recommend using OS::Heat::AutoScalingGroup or OS::Heat::ResourceGroup if you can't express your networking configuration by scaling an OS::Nova::Server resource then you should write a template resource which contains the server plus the other resources required to scale (ports, pool members, etc)

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2014-03-02 22:27:52 -0600

Seen: 417 times

Last updated: Sep 24 '15