Using Multiple Tools

I l @ ve RuBoard

Vendors realize that no single product can provide a total solution that fits every need of an enterprise. And because you probably won't find one tool or enterprise management platform that fits all of your fault management needs, several vendors provide integration tools. This features provides you with more flexibility in selecting the tools that best fit your monitoring needs. This means that you can select best-in-class monitoring tools and manage them from your enterprise management framework platform of choice.

Many of the monitoring tools integrate with enterprise management frameworks, such as IT/O, Unicenter TNG, and Tivoli TME. This section covers a few of the integration tools, but not all of them.

There are many other combinations of integration tools. For example, Oracle's OEM product is integrated with IT/O such that each tool can be launched from the other. IT/O can launch OEM in context. Also, OpenView SMART Plug-Ins work with MeasureWare, PerfView, and IT/O.

IT/Operations and the Network Node Manager

OpenView IT/O includes the Network Node Manager (NNM). NNM can be used separately, however, integration provides all of the NNM capabilities from within the IT/O environment, and also includes more system and problem management capabilities.

IT/O provides more managed node capabilities, while NNM provides mostly monitoring capabilities through standards such as SNMP. With IT/O agents , you can do more than just monitor managed nodes. You can manage them, too. With IT/O, you can set up more sophisticated message conditions and configure automatic or operator-initiated actions to run on the managed nodes. IT/O also provides more flexibility to add additional monitoring to your managed nodes.

In both NNM and IT/O, colored icons on the graphical map are used to reflect node status. In NNM, node status is usually determined by the network status of the system, including the status of the network interface cards. In IT/O, node status is determined not only by the network status of the system, but also by the IT/O alarm status for that node.

IT/O and PerfView

PerfView can be integrated into the IT/O Application Bank. With this integration, you can easily select a managed node, then use PerfView to graph metrics collected on the selected node. Additionally, templates can be configured so that if you receive a PerfView alarm condition in the IT/O Message Browser, you can run an operator-initiated action that will invoke PerfView to create a graph showing metrics related to the alarm condition.

If you have the MeasureWare Agent installed on an IT/O managed node, it will detect the IT/O agent running and automatically start sending alarms to it.

BMC PATROL and IT/O

PATROL does not require a special management console. Instead, its server-based PATROL agents can send monitoring information via SNMP traps to management platforms such as HP OpenView NNM and IT/O, CA UniCenter TNG, and Tivoli TME. This level of integration requires an additional product called PATROLVIEW, and there are special versions for OpenView, Cabletron SPECTRUM, Remedy AR, Sun's Solstice Domain Manager, and Tivoli TME. The PATROL product can be integrated with Unicenter TNG through a special product called PatroLink.

You need to install PATROLVIEW on the IT/O management station. This process installs and sets up message groups, message templates, node groups, threshold monitors , and application icons.

One of the unique things about PATROL's integration with IT/O is that PATROL agents from many systems, not just IT/O managed nodes, can send events to IT/O. PATROLVIEW includes a PATROL Event Translator server, the PET server, which translates PATROL events in IT/O events. The PATROL agents send events to the PET server on the IT/O managed node. The PET server processes and forwards the events to the IT/O agent.

The PET server runs on one or more IT/O managed nodes. It receives and processes events from one or more PATROL agents, then forwards the data to the IT/O agent that processes the events and forwards them, if necessary, to the IT/O console. You can also install the PET server on more than one managed node. This is useful if one of your PET servers goes down. If you have only one PET server configured and it goes down, you will lose the connection to all of the PATROL agents that send events to the PET server and on to IT/O.

If you want access to the PATROL console or agent configuration utilities from your IT/O management station, these tools must be installed there as well.

PLATINUM ProVision and IT/O

PLATINUM provides the ProVision Adapter for HP OpenView. This provides you with a single point of integration to ProVision. The adapter automatically enables all ProVision tools from within the IT/O environment. The adapter comes with IT/O templates and monitors that get configured into IT/O at installation time.

Using the PLATINUM IT/O Event Adapter, ProVision tools send events through the POEMS event manager to the IT/O agent at the IT/O managed node.

PLATINUM also provides an MIB that can be used by any SNMP-ready management framework to receive events as SNMP traps from ProVision tools. The MIB can be loaded into NNM, IT/O, or AIX NetView, for example.

PLATINUM ProVision is also tightly integrated with Tivoli TME. PLATINUM has integrated their products into TME to provide a common interface to the PLATINUM products from the TME environment.

EMS and OpenView NNM or IT/O

EMS provides templates for receiving events into OpenView NNM and IT/O. With EMS, you can configure events to be sent as SNMP traps. If the system where you are configuring EMS monitors is an IT/O managed node, you can configure an event to be sent as an opcmsg . In addition to configuring EMS notification targets, you also need to configure your management station to receive, recognize, and display events.

For NNM, you need to load the EMS-specific trap configuration file. You also need to make sure the management station is listed as a trap destination on the monitored system.

For IT/O, you need to configure the templates into IT/O. Templates include message conditions, automatic actions to gather additional event data, and instructions about each event. Once configured at the management station, EMS templates need to be assigned and distributed to managed nodes.

Also available for IT/O is an Application Bank utility called monvols . With monvols , you can configure monitoring of all logical or physical disk volumes on an HP-UX managed node in a single operation. Event notifications are sent to the IT/O Message Browser.

OpenView NNM and IT/O templates for receiving EMS events are freely available, along with the EMS Developer's Kit and other tools at the Hewlett-Packard software Web site at http://www.software.hp.com , under High Availability. monvols is also available at this Web site.

I l @ ve RuBoard


UNIX Fault Management. A Guide for System Administrators
UNIX Fault Management: A Guide for System Administrators
ISBN: 013026525X
EAN: 2147483647
Year: 1999
Pages: 90

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net