Managing NerveCenter - Understanding NerveCenter - Role in Network Management Strategy - Using Multiple NerveCenter Servers -
Standalone Operation      Integration with NMPs for Node Information

Using Multiple NerveCenter Servers

Because one NerveCenter server can inform another NerveCenter server or management platform of a network condition, it's possible to set up NerveCenter servers at remote sites that notify a centrally located NerveCenter server or management platform of the noteworthy network conditions at those remote sites.

Distributed NerveCenter Servers

This is a reliable solution because the remote NerveCenter servers use TCP/IP to notify the centrally located NerveCenter server of network conditions and retransmit messages as necessary to ensure their delivery.

There are a couple of advantages to this type of setup:

• Only a small amount of data is transmitted over the WAN. Any bandwidth intensive monitoring is conducted on a LAN and is managed by a remote NerveCenter server.
• The remote NerveCenter servers can be run in lights-out mode. Being able to run NerveCenter lights-out means that:
  • NerveCenter runs as a Windows service or as a UNIX daemon
  • You can monitor and configure NerveCenter from a remote location
  • You can modify all NerveCenter parameters without shutting NerveCenter down
  • No display or operators are required at a site
• The central NerveCenter can further correlate and filter conditions across remote NerveCenter Server domains

Integration with Network Management Platforms

A network management platform (NMP) is an operations and problem-management solution for use in a distributed multi-vendor environment. Intelligent distributed agents on managed nodes monitor system and application log files and SNMP data. The agents apply filters and thresholds to monitored data and forward messages about conditions of interest to a central management station. When the management station receives these messages, it can automatically take corrective action -- such as broadcasting a command to a set of systems -- or an operator can initiate this response.

You can integrate NerveCenter with the following network management platforms:

• CA Unicenter TNG
• Hewlett Packard OpenView IT/Operations
• Hewlett Packard OpenView Network Node Manager
• Micromuse Netcool/OMNIbus
• Tivoli Systems TME

Additionally, with OpenView Network Node Manager, you can direct NerveCenter to take its node information from the management platform and configure NerveCenter to take over all polling activity and event processing. See the later section, Integration with NMPs for Node Information, for more information.

You can integrate your NerveCenter installation with the NMP so that the NMP can send messages to NerveCenter for correlation or processing. After the messages arrive, NerveCenter correlates the conditions described in these messages with related conditions -- from the NMP or from other sources -- and can respond with any of its alarm actions, as appropriate. In addition, NerveCenter can send a message to an NMP in response to any network condition, whether the condition was originally detected by the NMP or not.

NMPs alone can detect a condition and invoke an action in response. However, you must integrate the NMP with NerveCenter if you want to:

• Correlate conditions detected by the NMP on different devices
• Correlate different types of conditions detected by the NMP on the same device
• Correlate conditions detected by the NMP with other types of events or conditions on the same device or across different devices