---

What Are the Foundation Services?

The Netra HA Suite Foundation Services are a suite of highly available and distributed software services that run either on the Solaris™ Operating System (Solaris OS) or Carrier Grade Linux distributions. Refer to the Netra High Availability Suite 3.0 1/08 Release Notes for information about which Solaris OS releases and which Linux distributions are supported for use with the Netra HA Suite software.The Foundation Services enable you to deploy applications in a continuous-availability environment. You can use the Netra HA Suite Foundation Services to create a highly available, dynamically scalable cluster of distributed nodes, and to augment existing highly available frameworks.

The following figure illustrates a basic Foundation Services cluster.

The concepts of clusters, server nodes (master and vice-master nodes), and client nodes (diskless and dataless nodes) are described in Cluster Model.

The Foundation Services are designed to do the following:

• Allow for hardware replacement, upgrade, and diagnostics without incurring system outage

• Provide highly available services (running on server nodes) to applications (in general, running on client nodes)

• Give HA-unaware applications a simple means for becoming HA aware

---

High-Level View of the Foundation Services

The following figure shows a high-level view of the services and application programming interfaces (APIs) provided by the Netra HA Suite Foundation Services product.

The Foundation Services offer the following services:

• A reliable network communication service provided by the Carrier Grade Transport Protocol (CGTP). CGTP limits the consequences of single network failure by duplicating the communication on a minimum of two links. For more information, see Chapter 3.

• A Cluster Membership Manager (CMM) to provide a global view of the cluster. The Cluster Membership Manager determines which nodes are members of the cluster. It assigns the roles and attributes of nodes, detects the failure of nodes, and notifies clients of changes to the cluster. A heartbeat mechanism detects node failure. For more information, see Chapter 5.

• A Reliable File Service (RFS) to ensure that data is accessible to applications, even in the event of hardware or software failure. Reliable File Service uses mounted file systems, IP replication of disk-based data or dual-hosted disks, and IP address failover of the master role. For more information, see Chapter 6.

• A Reliable Boot Service (RBS) to ensure the boot of diskless nodes regardless of software or hardware failures. For more information, see Chapter 7.

  
  

  ---

  Note - Diskless nodes are not supported on the Linux OS, therefore, Reliable Boot Service is not supported on clusters running Linux.

  ---

  
  

• An External Address Manager (EAM) to allow external access to the cluster on highly available IPv4 or IPv6 addresses. For more information, see Chapter 8.

  
  

  ---

  Note - IPv6 addresses are not supported on Linux.

  ---

  
  

• A Node State Manager (NSM) with user-provided scripts executed to react to role changes, when the node becomes master or vice-master. For more information, see Chapter 9.

• A Process Monitor Daemon (PMD) to survey Netra HA Suite daemons, some operating system daemons (on the Solaris OS only), and some companion product daemons. If any of the monitored daemons fail, the PMD initiates a recovery response. For more information, see Chapter 10.

• A Node Management Agent (NMA) to monitor cluster statistics. The Node Management Agent can initiate a switch to the backup node, change some error recovery responses, and listen for notifications of some cluster events.

  The Node Management Agent is compliant with the Java Management Extensions (JMX) and based on the Java Dynamic Management Kit. For more information, see Chapter 11.

  
  

  ---

  Note - The Node Management Agent is supported only on clusters running the Solaris Operating System.

  ---