[Rdo-list] [RDO-Manager] Working with the overcloud
marius at remote-lab.net
Fri Nov 6 16:12:49 UTC 2015
On Thu, Nov 5, 2015 at 4:04 PM, Ignacio Bravo <ibravo at ltgfederal.com> wrote:
> After jumping some hoops and with the help of the usual suspects on IRC, I
> was able to deploy an overcloud with HA, Network isolation and Ceph. Great!
> Now I want to focus on what’s next, or how to manage this environment going
> forward. Let me give you a couple of examples:
> After the installation of the overcloud, I was hit with the cinder bug
> described here: https://bugzilla.redhat.com/show_bug.cgi?id=1272572 The
> issue is that the cinder.conf file needs to replace ‘localhost’ with the ip
> of the public keystone. What I did, based on the bugzilla, was to log in to
> each controller node and then update the value inside the cinder.conf file.
> Is this one off the proper way to patch and keep the environment updated? I
> mean, one week from now we will find that a particular RPM needs to be
> updated, how do you handle this? I thought that the proper way was to
> recreate the TripleO image and redeploy.
Updating the overcloud nodes packages should be done via update stack:
> Or another example is Ceph. Currently Tripleo installs version 0.8 and I
> want to install version 9 Inferno. What is the correct path to achieve this?
> Additionally, let’s say that I want to install, say: CloudKitty (choose your
> alternate, non mainstream openstack project here)
> Do we recreate the images and redeploy, or do a puppet run after they have
> been installed with a tool like Foreman/Katello?
You could try doing it via a post deploy script:
> Ignacio Bravo
> LTG Federal, Inc
> Rdo-list mailing list
> Rdo-list at redhat.com
> To unsubscribe: rdo-list-unsubscribe at redhat.com
More information about the rdo-list