To enable the Sentinel Log Manager servers for high availability, you require the following:
Redundant, clustered Sentinel Log Manager nodes.
Access to shared data storage.
Virtual IP addresses that can be used to transparently shift from a failed node to another node.
Scripts to start, stop, and monitor the application based on policies defined in the cluster solutions. You can use cluster solutions such as Cluster Resource Agents or LSB init scripts on Linux Enterprise High Availability systems.
There are many packages in the market that enable high availability, testing for Sentinel Log Manager was performed with SUSE Linux Enterprise High Availability (HA) Extension, shared storage RAID drives, and custom scripts. This architecture can be replicated across data centers to ensure availability of everything from the Sentinel Log Manager server to the Collector Managers and Collectors.
If the environment includes Windows-based Collector Managers, high availability for the Windows Collector Managers may be setup using the Microsoft Windows Clustering approach by using the Microsoft Windows Clustering approach Other clustering solutions (for example, Veritas Clustering Services) may be used, but Novell has not tested these at this time. For information on using any of these clustering services, please refer to the respective vendor documentation.
High availability for the event sources should also be considered on a case by case basis because of the wide variety of devices.