Novell Cluster Services™ allows you to configure up to 32 NetWare servers into high-availability cluster, where resources can be dynamically switched or moved to any server in the cluster. Consolidation of applications and operations on a cluster has benefits such as lower costs, scalability, and increased availability.
For a cluster to work as a high-availability system, the file system, the applications, and services that run on the cluster should be cluster-enabled. SMS supports backup and restore of cluster resources. In addition, the backup session can be automatically recovered in case of a failover or failback condition.
The backup engines have to be modified to be able to support backup and restore of cluster resources. For more details, refer Section 1.1.4, Backing Up and Restoring Data.
SMS supports automatic recovery of backup sessions in failover and failback situations.
If the connection to the TSA is terminated after a failback or failover, the engine attempts to reconnect to the TSA through SMDR. Ensure SMDR and TSA modules are loaded in the failed-over/failed-back cluster nodes.
If a scan is in progress when the connection terminates, the engine should end the scan by calling NWSMTSScanDataSetEnd. After this, the engine should try to reconnect after waiting for a configurable time and then retry at regular intervals until the connection is re-established or the number of retries are expired. The engine should call NWSMConnectToTSA in a loop until the connection is established. After the engine establishes the connection, to continue from the last completely backed up dataset, call NWSMTSScanDataSetContinue keeping all the parameters the same as NWSMTSScanDataSetBegin and pass an additional parameter called cursor. Cursor represents the full path of the last successfully backed up data set. The prerequisite for calling NWSMTSScanDataSetContinue is the successful back up of atleast one data set. The name of the next data set is returned and backup job continues. If there is no last backed-up successful data set, use NWSMTSScanDataSetBegin
NOTE:Recovering backup session on cluster failover or failback is not supported in NetWare versions earlier than NetWare 6. In NetWare 5.x, shared cluster resources are represented as normal resources.
Recovery of a restore session in case of failover or failback is currently not supported.
How the tape should be managed in a failover or failback situation is not specified. Define a scheme that best suits your need.
Also consider the following while modifying the backup engine:
Separate backup jobs are required per cluster resource (Shared Cluster enabled NSS pool for File System resource). Only SLP based discovery mechanism is supported for clusters.