VCS Basics

A single VCS cluster consists of multiple systems connected in various combinations to shared storage devices. When a system is part of a VCS cluster, it is a node. VCS monitors and controls applications running in the cluster on nodes, and restarts applications in response to a variety of hardware or software faults.

Client application continue operation with little or no downtime. In some cases, such as NFS, this continuation is transparent to high-level applications and users. In other cases, a user might have to retry an operation, such as a web server reloading a page.

This illustration is a typical VCS configuration of four nodes connected to shared storage. Client workstations receive service over the public network from applications running on VCS nodes. VCS monitors the nodes and their services. VCS nodes in the cluster communicate over a private network.

Click the thumbnail above to view full-sized image.

Example of a Four-System VCS Cluster

Multiple Systems

VCS runs in a replicated state on each node in the cluster. A private network enables the nodes to share identical state information about all resources and to recognize active nodes, nodes that are joining or leaving the cluster, and failed nodes. The private network requires two communication channels to guard against network partitions.

Shared Storage

A VCS hardware configuration typically consists of multiple nodes connected to shared storage via I/O channels. Shared storage provides multiple systems with an access path to the same data, and enables VCS to restart applications on alternate nodes when a node fails, which ensures high availability.

The figures below illustrate the flexibility of VCS shared storage configurations. VCS nodes can only access physically-attached storage.

Click the thumbnail above to view full-sized image.

Two Examples of Shared Storage Configurations

LLT and GAB

VCS uses two components, LLT and GAB, to share data over private networks among systems.

• LLT (Low Latency Transport) provides fast, kernel-to-kernel communications, and monitors network connections. The system administrator configures LLT by creating the configuration files /etc/llthosts, which lists all the nodes in the cluster, and /etc/llttab, which describes the local system's private network links to the other nodes in the cluster.
• GAB (Group Membership and Atomic Broadcast) provides the global message
  order required to maintain a synchronized state among the nodes. It monitors disk communications such as the VCS heartbeat utility. The system administrator configures the GAB driver by creating a configuration file (/etc/gabtab).

For more information, see Verifying LLT and GAB Configuration Files.

Two Types of Channels: Network and Shared Disks

For the VCS private network, there are two types of channels available for heartbeating: network connections and heartbeat regions on shared disks. The shared disk region heartbeat channel is used for heartbeating only, not for transmitting information as are network channels. For information on configuring heartbeat regions on shared disks, see Configuring Membership Heartbeat Regions on Disk (optional).

Each cluster configuration requires at least two channels between systems, one of which must be a network connection. The remaining channels may be a combination of network connections and heartbeat regions on shared disks.

This requirement for two channels protects your cluster against network partitioning.
(For more about network partitioning, refer to the VERITAS Cluster Server User's Guide.) VERITAS recommends configuring at least one heartbeat disk region on each I/O chain shared between systems in addition to private network connections.

The following illustration shows a two-system VCS cluster where sysA and sysB have two private network connections and another connection via the heartbeat disk region on one of the shared disks. If one of the network connections fails, two channels remain. If both network connections fail, the condition is in jeopardy, but connectivity remains via the heartbeat disk.

Click the thumbnail above to view full-sized image.

Two Systems Connected by Two Ethernet Connections and a Heartbeat Disk Region