Ask Your Question

Understanding OpenStack

asked 2015-03-29 18:59:23 -0500

Stefano Mtangoo gravatar image

I need to understand more complexity of OpenStack stuff. Suppose I want to have a Cloud supporting Tomcat 7, PostGreSQL and File Servers (NFS+Samba). Usually I would create 3 Virtual Machine (Let say KVM under ubuntu), do virtual network and route traffic to outside word. Now external users can happily use, for example file server, to save/retrieve their files. I would use KVM tools with SSH to manage the hypervisor instances.

Now the documents I read and video I see in youtube about openstack seems to give me impression that openstack Simplifies this process (with Juju). Is this correct? If yes how does it achieve it? If no any resource for beginner (I find their beginners guide hard to understand)

Sorry If I missed something, am new here!

edit retag flag offensive close merge delete


Is this quietness normal or something is not right in my post?

Stefano Mtangoo gravatar imageStefano Mtangoo ( 2015-03-30 16:23:41 -0500 )edit

So with openstack you can still use KVM and set everything up the same way, the big difference would be how the users interact with the available resources you give them. Also Juju is just one method to assist, another is puppet. Have you installed OS yet?

omar-munoz gravatar imageomar-munoz ( 2015-03-31 11:28:50 -0500 )edit

Thanks. What I don't know is whether OpenStack creates its own VM (thereby being hypervisor) or It just create KVM for example and then install services needed (MySQL et al). What do you mean by "Have you installed OS yet?", you mean openstack itself?

Stefano Mtangoo gravatar imageStefano Mtangoo ( 2015-03-31 11:49:09 -0500 )edit

2 answers

Sort by ยป oldest newest most voted

answered 2015-04-01 05:59:46 -0500

Charles Benon gravatar image

Hi Stefano,

To respond your first question, Openstack will have one node(s) (Compute) which will run VM (or instances).

So, to make it simple, here are the layers:

  1. Hardware
  2. Operating system (Ubuntu, Red Hat, Suse)
  3. Openstack packages (this will depend of the node you want to run: Controller, Network, Compute, Storage)

After, I would suggest to install it to really understand which packages is doing what. In this case, you could use one physical server and run each of the node as VMs:

  1. Hardware
  2. Operating system + KVM (create a VM for each of the node: Controller, Compute and Network - Keep it simple for now, and use just local storage)
  3. Follow the installation guide for each of you nodes (steps 1 to 7 included): (for example for Ubuntu)

Let us know if it is clear or still isn't.

edit flag offensive delete link more


Thank you. How do I do (2)? I want to try it. Also, can I use it in a Laptop (core i3) just to test or should I use dedicated server?

Stefano Mtangoo gravatar imageStefano Mtangoo ( 2015-04-02 04:08:34 -0500 )edit

Ubuntu installation:

i3 CPU may be a bit short, even just to run Openstack without instances. But, you can give a try, for sure, you won't run production on it ;) (you may go for single node deployment)

Charles Benon gravatar imageCharles Benon ( 2015-04-02 09:12:05 -0500 )edit

Thank you. Let me try it out. I will come back for any question!

Stefano Mtangoo gravatar imageStefano Mtangoo ( 2015-04-02 15:02:34 -0500 )edit

answered 2015-04-01 06:10:43 -0500

Please use the below link to install openstack and then learn about it.

You can, to get a hang of it , install it using packstack much simple way to install. Once install work with different api to get an idea of how things work.

edit flag offensive delete link more


I found it hard, since I lacked some basics of openStack itself!

Stefano Mtangoo gravatar imageStefano Mtangoo ( 2015-04-02 04:09:08 -0500 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2015-03-29 18:59:23 -0500

Seen: 134 times

Last updated: Apr 01 '15