NiCE PowerHA Management Pack
NiCE IT Management Solutions
NiCE PowerHA Management Pack
NiCE IT Management Solutions
NiCE PowerHA Management Pack
NiCE IT Management Solutions
Monitoring IBM PowerHA clusters using Microsoft SCOM and Azure Monitor SCOM MI
IBM PowerHA monitoring using Microsoft SCOM
Clusters are used to overcome capacity shortages so your IT services stay aligned with your service level agreements. Advanced cluster monitoring will help you prevent business losses by upfront actions on impacted systems.
The NiCE PowerHA Management Pack for Microsoft SCOM will alert you on impacted PowerHA components vital to your business.
Features
PowerHA Discovery
Environments of increasing business demands and their critical applications need to be available 24/7. The NiCE PowerHA Management Pack ensures that any component’s failure does not affect the overall system. It keeps the highly fault-tolerant system cost-effective and efficient. IBM PowerHA consists of numerous objects such as Nodes, Resources, Resource groups, and AIX computers. Each object has its own properties, such as startup policies, service labels, log paths, and identifiers. In a large environment, oversight of important components is easily lost. The ‘out-of-the-box discover’ allows the discovery of all PowerHA objects, including an extensive list of properties for each object. Utilizing the OpsMgr class model, a full PowerHA inventory model is mapped out. With integrations to SC Service Manager, this valuable information is also available with Microsoft System Center Service Manager.
High Availability and Performance
The Management Pack collects detailed data from your PowerHA environment using predefined event conditions and threshold monitors. The smart data and event correlation will increase the IBM PowerHA (HACMP) server availability and performance while lowering the total cost of ownership (TCO). The NiCE PowerHA Management PACK enables service-oriented management, as well as incident management in your PowerHA environment, to support your business-critical processes.
Failover Cluster Detection
In the case of a resource group experiencing a service incident, for example, a node failure, the cluster manager acts upon the failover policies. It initiates a failover of resource groups to another available node. Failover detection rules will evaluate relevant entries in the cluster log as well as using cluster utilities to inform you about the events that occurred during failover.
IBM PowerHA Alerting
Incidents impacting high-available infrastructure components must be processed with high urgency.The NiCE PowerHA Management Pack includes monitors and alert rules to provide immediate, comprehensive information on impacted components. It includes monitoring 10 different cluster availability states, including ‘joining’, ‘barrier’, ‘cbarrier’, and many others. Additionally, if an incorrect configuration is applied to a cluster, the Management Pack raises the correct alert instead. The detailed information given drastically helps to understand issues and to reduce the mean time to recover.
Resource Group Monitoring
PowerHA Resource Groups contain critical cluster resources, typically interfaces and mount points.If a resource group fails or becomes unavailable, resources are impacted. To be on top of all the resource group-impacting events, the NiCE PowerHA Management Pack includes several monitors dedicated to resource and resource group monitoring. Furthermore, it complements the monitoring provided by the Microsoft AIX Management Packs.
SLA Reporting
Mission-critical workloads require consistent performance. Good SLA reporting helps to manage, monitor, and modify resources the right way at any time.