Rubick/Service architecture
< Rubick
Revision as of 09:04, 23 October 2013 by Oleg Gelbukh (talk | contribs) (Add service architecture description)
Design & Architecture
This page describes design and architecture of OpenStack Diagnostics (Rubik) service. It is applicable to MVP1 version of the project.
Components
Service includes the following components:
- openstack.model is an OpenStack architecture model representation. It is a common format used by components of the system to exchange configuration of the inspected environment
- Rubick API is a web service which implements APIs to rules, inspections and OpenStack architecture model
- Rule engine is a logic which performs inspections on the data model. Rule engine will have an interface to the ruleset store in future.
- Config data store is a storage for architecture models
- Config data extractor creates OpenStack model based on data collected from different sources, implemented as pluggable back-ends
- Heat metadata plugin extracts configration metadata from Heat stacks created by TripleO/Tuskar service
- SSH metadata plugin (Joker) extracts configuration metadata from actual nodes of OpenStack cloud via secure SSH connection
