OpenSolaris™ Bible

Chapter 16. Clustering OpenSolaris for High Availability

IN THIS CHAPTER

• Introduction to high-availability clusters

• Overview of Open High Availability Cluster

• Setting up a cluster

• Using a cluster

• Advanced cluster administration

• Making custom services highly available

• Disaster recovery with Open High Availability Cluster

OpenSolaris is a solid, highly robust operating system. As described in Chapters 12 and 13, OpenSolaris contains considerable built-in support for reliability and availability in the form of predictive self-healing. Because of hardware and software failures, however, there are limits to the level of availability that a single physical machine can provide, regardless of the robustness of the underlying operating system. To increase the availability of your system beyond the capabilities of a single machine, you can cluster two or more machines together. This chapter describes the Open High Availability Cluster software that you can use to tightly couple multiple machines running OpenSolaris for augmented availability of the system as a whole.

Introduction to High-Availability Clusters

Before delving into the details of the Open High Availability (HA) Cluster software, it's helpful to understand why high availability is important, and why it sometimes makes sense to cluster for high availability.

First, downtime is costly. If you're using OpenSolaris only as a desktop or laptop operating system, you probably aren't too concerned with occasional downtime; but if you're using OpenSolaris ...