This section familiarizes you with the way to use the Red Hat Cluster Suite (RHCS) software for
creating failover clusters from Parallels servers.
In This Chapter
Cluster Configuration Options ................................................................................. 9
Configuring Shared Storage for Failover Clusters ..................................................... 10
Creating and Configuring a Failover Cluster ............................................................. 14
Cluster Configuration Options
You can create a cluster of two or more servers running Parallels Server Bare Metal to provide a
high degree of availability for your servers. A typical failover cluster implementation must meet the
following requirements:
• The cluster configuration is as follows:
• A shared SCSI storage device (iSCSI or Fibre Channel) is used as a cluster storage.
• The /vz partition is located on the shared SCSI storage device and formatted as ext3.
• Clustered file systems (GFS v1, GFS v2, and NFS) are not used.
• One and the same version of the Parallels Server Bare Metal software is installed on all servers
in the cluster.
• Only one /vz partition is created per clustered service.
• The following cluster resources are available in the cluster: the /vz filesystem, the
/etc/init.d/vz-cluster script, and the IP address resource.
C HAPTER 3
Deploying a Failover Cluster
Komentarze do niniejszej Instrukcji