Clustering NetWorker and monitoring system by NetWorker daemon-specific monitor improves availability of backup operation system on which NetWorker is used.
Clustering NetWorker server, sharing tape device and NetWorker management database (stores information such as when and what data is backed up and NetWorker setting) improves backup availability. Even failure occurs during backup and failover occurs in cluster system, backup can be continued on standby node.
Auto Detection and Recovery of NetWorker Failure
RMS configuration script and NetWorker server daemon detector automatically detects and recover failure on NetWorker and this minimizes time to stop backup service.
The Detector constantly monitors NetWorker, and if it detects a failure, it will notify RMS (Reliant Monitor Services). If the AUTORECOVER attribute is enabled, RMS will start script then perform automatic recovery for the failed NetWorker.
Taking over Backup Service
NetWorker-specific RMS configuration script and detector which monitors NetWorker server daemon cooperate with PRIMECLUSTER and it enables to takeover backup service.
If backup on the OPERATING node cannot be continued, backup operation is auto-switched from the OPERATING node to the STANDBY node. Then, the NetWorker server is automatically started on the STANDBY node (the new OPERATING node) and resumes backup service.
The new OPERATING node enables backup operation without changing NetWorker definition. Also, data that is backed up on the previous OPERATING node can be recovered on the new OPERATING node.
Virtual Client Backup
Data on the shared disk that PRIMECLUSTER manages can be backed up by using a virtual server name (logical node name).
The logical node name is a virtual name related to virtual IP address defined by takeover IP address of PRIMECLUSTER. Moreover, it is a name managed to support virtual NetWorker client, server, and storage node PRIMECLUSTER Wizard for NetWorker.
It is possible to treat NetWorker on the cluster system as a single similar system by using the logical node name. Therefore, it is possible to back up data on a shared disk from the standby node (new operation node) without the definition change of NetWorker even if failover is generated, and the backup data can be recovered from each node.
The logical node name can be specified each user application of the cluster. Therefore, data on a shared each disk can be backed up and recovered by logical node name treating each application as a single system even if in N:1 operation standby form or mutual standby form.
Local Backup of nodes within cluster
Data on a local disk of each node within the cluster can be backed up by using a physical node name of the node within the cluster.
A local disk of the STANDBY node can be backed up because the NetWorker client is operated on all nodes within the cluster. Moreover, the backup operation can be continued without the definition change of NetWorker even if failover is occurred.
Standby operation making about NetWorker client
The standby operation of the NetWorker client is achieved by running a necessary NetWorker client in operation/standby system when the cluster does failover in cooperation with PRIMECLUSTER.
Note
NetWorker Server is a NetWorker Storage Node or NetWorker Client. Therefore NetWorker server includes function of storage node cluster or client cluster.
