Using ClusterCATS
Optimizing ClusterCATS

Using Server Failover

The ability to fail over servers that have become unavailable to redundant servers is a cornerstone of any mission-critical application, one that ensures continuous, reliable operation. Server failover is an option to select during installation. If you did not select it, you must reinstall ClusterCATS to enable it.

Static versus ClusterCATS dynamic IP addressing

There are two schemes with which you implement server failover:

• Static IP addressing - when a computer fails, IP address(es) bound to its web server are reassigned to the most available cluster member's web server. When the failed-over server comes back online, it must claim the IP address and then restart again.
• Dynamic IP addressing - ClusterCATS can be configured to dynamically assign IP addresses so that when a server fails, its IP address(es) can be assigned to other servers. When the failed-over server comes back online, ClusterCATS returns the IP addresses to it without conflict.

On Windows clusters, you should use server failover with the ClusterCATS dynamic IP address scheme. To configure ClusterCATS dynamic IP addresses, the IP address associated with the computer name must be different from the IP addresses associated with the websites. ClusterCATS refers to the IP address associated with the computer name as the maintenance address. For more information, see "ClusterCATS Dynamic IP Addressing (Windows only)".

Windows domain controllers

If you use Windows NT Domain server authentication, each web server in a cluster must participate as a member NT server in a domain. Do not set a server in your cluster the primary domain controller (PDC). Server failover will interfere with the function of the PDC. You can set an NT server as a backup domain controller, but this configuration is not recommended.

Copyright © 2002, Macromedia Inc. All rights reserved.

Comments