[Gluster-infra] openstack "gluster cloud"

[Dan Lambright]

I have mentioned to a few people about an openstack cloud I've set up, which could assist in providing sundry cluster infra tasks and services.

It is at rhs-srv-06.gdev.lab.eng.bos.redhat.com. It is accessible via CLI. There is a rhel7 glance image, we could also put fedora or bsd, images there.

This server has 16 cores and just under 2TB of storage available. A "flavor" can be devised to give some amount of storage to each guest, I am using m1.small (20G) presently.

I expect we could easily run dozens on guests on this machine for any purpose. It is also possible to expand this "cloud" with more servers.

The process to set up guests is currently a little cumbersome as its via CLI, but this can be simplified / scripted. (boot image, assign floating IP address (10.19.101.161-192), login via pem file).

In short, this is an RFC to brainstorm ways to use our new "gluster cloud". 

Here are some ideas I've hard on how it could be used.

- engineering machines (dev, QE)

- expanding jenkins pool 

- certain scale-out tests