uCosminexus Service Platform, Overview

[Contents][Glossary][Index][Back][Next]

1.4.2 Redundant configuration of the HCSC server using cluster software

Cluster software is a program that implements switching of a system including server programs with the aim of improving the system's reliability and availability. The service platform can link with the following cluster software:

In the execution environment of the service platform, you can combine two HCSC servers that have the same contents by integrating with the cluster software to set up an environment of a more redundant cluster configuration (cold standby configuration).

The setup information of two HCSC servers configuring the cluster must be the same. The HCSC components to be deployed on these servers must have the same contents.

The following figure shows the cluster configuration of HCSC servers:

Figure 1-15 Cluster configuration of HCSC servers

[Figure]

In the execution environment of the service platform, you can set up a system in which the executing and standby nodes operate on a 1-to-1 basis (1-to-1 node switching system), by integrating with the cluster software.

If an error occurs on the executing node of a 1-to-1 node switching system, the cluster software detects the error and then automatically switches over to the standby node to continue applications. The following figure shows the switching of nodes when an error occurs.

Figure 1-16 Node switching when an error occurs

[Figure]

The following is the flow of operations from the occurrence of an error until the switching of nodes.

  1. An error occurs on the HCSC server on the executing node.
  2. The cluster software detects the error on the HCSC server on the executing node.
  3. The network to one of the machines on the HCSC server on the executing node is disconnected.
  4. The cluster software switches the network to one of the machines on the HCSC server on the standby node.
  5. The cluster software starts the HCSC server on the standby node.

You must specify the settings in the cluster software to enable the HCSC server at the standby node to start automatically in the event of a failure.