Red Hat Training

A Red Hat training course is available for Red Hat OpenStack Platform

Chapter 1. Overview

The sample HA deployment used for this document was created using the following guides as reference:

Figure 1.1, “OpenStack HA environment deployed through director” shows the particular configuration that was built specifically to test the high availability features described here. For details on how to recreate this setup so you can try the steps yourself, refer to Appendix A, Building the Red Hat OpenStack Platform 10 HA Environment.

Figure 1.1. OpenStack HA environment deployed through director

computers network osp7 ha 4

1.1. Managing High Availability Services

In a High Availability (HA) deployment, there are three types of services: core, active-passive and systemd. Core and active-passive services are launched and managed by Pacemaker, with all the other services managed directly by systemd controlled with the systemctl command. The core OpenStack services (Galera, RabbitMQ and Redis) run on all the controller nodes and require a specific management for start, stop and restart actions.

Active-passive services only run on a single controller node at a time (for example, openstack-cinder-volume), and moving an active-passive service must be performed using Pacemaker, which ensures that the correct stop-start sequence is followed.

All the systemd resources are independent and are expected to be able to withstand a service interruption, with the result that you will not need to manually restart any service (such as openstack-nova-api.service) if you restart galera. When orchestrating your HA deployment entirely in director, the templates and puppet modules used by director ensure that all services are configured and launched correctly, particularly for HA. In addition, when troubleshooting HA issues, you will need to interact with services using both the HA framework and the systemctl command.