Section Nine: Going Crazy¶

Regions and geographic diversity¶

OpenStack supports the concepts of ‘Regions’ - ususally geographicaly separated installations that are all connected to the one service catalogue. This section explains how to expand the Fractal app to to use multiple regions for high availability.

Multiple clouds¶

Sometimes, you want to use multiple clouds, such as a private cloud inside your organisation and a public cloud. This section attempts to do exactly that.

High Availability¶

Using Pacemaker to look at the API.

conf.d, etc.d¶

Use conf.d and etc.d.

In earlier sections, the Fractal Application uses an install script, with parameters passed in from the metadata API, in order to bootstrap the cluster. Etcd is a “a distributed, consistent key value store for shared configuration and service discovery” that can be used for storing configuration. Updated versions of the Fractal worker component could be writted to connect to Etcd, or use Confd which will poll for changes from Etcd and write changes to a configuration file on the local filesystem, which the Fractal worker could use for configuration.

Using Swift instead of a database¶

We haven’t quite figured out how to do this yet, but the general steps involve changing the fractal upload code to store metadata with the object in swift, then changing the API code such as “list fractals” to query swift to retrieve the metadata. If you do this, you should be able to stop using a database.

Next Steps¶

Wow, if you’ve made it through this section, you know more about working with OpenStack clouds than the authors of this guide.

Perhaps you can contribute?