HP Serviceguard Cluster Configuration for HP-UX 11i or Linux Partitioned Systems, April 2009

Figure 4. A 4—node cluster from two Superdomes with quorum server

Cluster configuration and partitions

Given the configuration requirements described in Rules1 and 2, a few interesting observations can

be made of clusters utilizing partitioning:

• If it is determined that a cluster lock is needed for a particular configuration, the cluster must be

configured so the cluster lock is isolated from failures affecting the cluster nodes. This means that the

lock device must be powered independently of the cluster nodes (including the hardware cabinets

containing the partitions that make up the cluster).

• Clusters wholly contained within two hardware cabinets and that utilize the cluster lock disk for

quorum arbitration are limited to either two or four nodes. This is due to a combination of the

existing Serviceguard rule that limits support of the cluster lock disk to four nodes and Rule 1.

• Cluster configurations can contain a mixture of vPars, nPartitions, virtual machines, and

independent nodes as long as quorum requirements are met.

• For a cluster configuration to contain no single points of failure, it must extend beyond a single

hardware cabinet, and it must comply with both the quorum rules and the Serviceguard

configuration rules described in HP 9000 Enterprise Servers Configuration Guide, Chapter 6 and

the Serviceguard for Linux Order and Configuration Guide. In other words, “cluster in a box” is not

supported.

Cluster in a box

One unique possible cluster configuration enabled by partitioning is called cluster in a box. In this

case, all the OS instances (nodes) of the cluster are running in partitions within the same hardware

cabinet. While this configuration is subject to single points of failure, it may provide adequate

availability characteristics for some applications and is thus considered a supported Serviceguard

configuration. Users must carefully assess the potential impact of a complete cluster failure on their

availability requirements before choosing to deploy this type of cluster configuration.

A cluster in-a-box configuration consisting exclusively of vPars or virtual machines running in a single

nPartition or non-partitioned box is more vulnerable to complete cluster failure than is a cluster made