[Labs-l] Upgrade complete

[Ryan Lane]

Ah. I should mention: I'm forcing everyone to re-authenticate to
labsconsole. Since the switch to keystone (and the use of the
nova-api) the way authentication works between labsconsole and
openstack has changed. This requires everyone to re-login.

On Mon, Aug 27, 2012 at 8:28 PM, Ryan Lane <rlane at wikimedia.org> wrote:
> The upgrade is complete.
>
> All instances should be up, but it's possible that some have failed to
> boot for one reason or another. deployment-prep, for example, has some
> instances down because some instances depend on other instances to
> start properly, which means the instances must be booted in a specific
> order. If your instance is not accessible for some reason, please let
> me know and I'll work with you to bring it up.
>
> There's a couple interface changes that come along with this upgrade:
>
> 1. There's no availability zones anymore. There are now regions. At
> this moment there's only a single region (pmtpa), but soon we'll be
> adding the eqiad region as well. Because of this, you'll notice that
> there's no zone selection on the instance creation page. Similarly,
> there's no choice for dns domain on the instance creation page. These
> are set for you automatically.
> 2. No more global groups. I doubt anyone will notice this.
>
> There's a few places the interface is showing the instance's openstack
> id rather than its name or its ec2-style ID. These will be fixed as I
> get a chance to get to them.
>
> Overall, the interface should be faster than it was pre-upgrade.
> There's still a few places where I can be memcaching things. I'll be
> taking care of that soon as well. Additionally, there should be no
> more 500 errors.
>
> If you run into any bugs, please submit them to bugzilla.
>
> - Ryan