Crowbar OpenStack deployment video (15 mins): Diablo + Keystone + Dashboard October 3, 2011
Posted by Rob H in Crowbar, Open source, OpenStack, OpenStack Design Summit, Video.Tags: Crowbar, dashboard, Diablo, Keystone, OpenStack, Video
trackback
This week at the OpenStack Design Summit and Conference in Boston, my team unveiled the Diablo+ Crowbar deployments. The OpenStack deployment that’s included with Crowbar reflects a collaborative effort between Dell, Opscode, and Rackspace and pulls packages from the Rackspace repository. It was important for us to use the Rackspace repos so that we could include integrated Keystone and Dashboard components that were omitted from the Diablo (current) release. Our decision to include these Essex (coming) components is based on customer feedback.
Since some of you cannot make it to the show and see the demo in person, we’ve captured it as a video for your enjoyment. The OpenStack deployment is available in our open source distribution. We are currently in QA for the overall solution so expect additional refinement as we progress towards our next OpenStack solution release.
REMINDER: Dell Hardware is NOT required to use Crowbar for OpenStack. The open source version has everything you need – the BIOS and RAID barclamps are optional (but handy).
Can you do a demo on the API?
For example, if I want to bring up 100 physical or virtual machines, and programatically supply this list to Crowbar, where I say 20 physical machines should be X Role, 30 should be Y role and 50 VM should be XY combination?
And once they are up, how do I make changes to dozens of machines, programatically, through Crowbar?
Great question! yes. There is a rest api (and command line) that can do that. You use it to submit proposals.
Nice video !
Crowbar is really impressive, thank you so much for sharing it !
I guess that version you are using is not yet published (because mine has a darker design and doesn’t work that well)
Keep rocking, you are doing an amazing job !
The version I am showing is the Dell skin. The darker skin is the open source version. Same function. Different color.
Hey Rob,
Is there any way to PXE boot crowbar? From what I can tell the procedures require taking the sledgehammer netboot images and integrating the barclamps and other crowbar stuff into an ISO and then installing that way.
Secondly, I’m getting an error when trying to boot from the prebuilt ISO. It’s telling me no root partition is defined. I’m attempting to install it on a machine that already has an OS on it. I assumed it would overwrite the existing partition table, but perhaps that is the issue?
Thanks for all the great resources!
To update my own comment… I think the issue with the root partition is that this is HP gear using the cciss stuff, which isn’t in the kernel until 2.6.33 and I’d wager the drivers for it aren’t being installed in the default crowbar install
As for the PXE boot, I’d still be interested in knowing if there is a way to do that instead of mounting an ISO. Most of our stuff doesn’t have CD-rom drives so I have to mount the ISO via the iLO which isn’t super practical long term.
There is a way to PXE the initial install. We do it internally because most of our PEC servers don’t have DVDs. Not sure if that’s in the open source list…let me check.
[...] our team moved Crowbar development into the open. This change was reflected in our work on OpenStack Diablo (+ Keystone and Dashboard) with contributions by Opscode and Rackspace Cloud Builders. Rather than work internally and push [...]
[...] recently been moved out into the open. As Rob explains, This change was reflected in our work on OpenStack Diablo (+ Keystone and Dashboard) with contributions by Opscode and Rackspace Cloud Builders. Rather than work internally and push [...]
Hi Rob,
Is it correct to say, that applying a proposal does just the configuration for the next chef run? So if the flashing light turns to green this does not been that the actual provisioning with chef is already done (and successful). e.g. if I tail the chef-client log on a node while clicking apply and waiting for the light to turn green again chef does not start provisioning. It runs on the next cycle (the default is 15 min). right?
You’re right – clicking Apply puts the selected roles in the Nodes run list. The green light is set when the state flags are released from the node.
From Greg Althaus: We force a chef-client run to make sure it happens. If you want for the light to go green, a chef-cient run wil have been done. What we currently don’t do (and have bugs on) is returning the error status on that chef-client run. The logs for chef-clilent are not on the node, but on that admin node in /opt/dell/crowbar_framework/logs/production.log. We are going to split this out later.
Here’s the code that turns on the green/ready light:
bc = @item["deployment"][self.barclamp]if bc.nil?
"hold"
else
return "unready" if bc.has_key? "crowbar-committing" and bc["crowbar-committing"]
return "pending" if bc.has_key? "crowbar-queued" and bc["crowbar-queued"]
return "hold" if !bc.has_key? "crowbar-queued" and !bc.has_key? "crowbar-committing"
"ready"
end
[...] management and is represented by our project Crowbar. While Crowbar began life simply as a way to install Openstack on Dell hardware, it has expanded from [...]