OpsManJam

If the Microsoft Offices Sharepoint Server 2007 Management Pack is installed, Event ID: 21405 is logged on all computers that do not have Sharepoint Installed.  This mp creates a new class, group, and overrides to target the discovery only at the computers with the MOSS specific registry keys. Just import the Workaround MP after the Sharepoint MP has been imported (or import both at the same time).  You should also download the MOSSWSSWorkArounds doucment (http://www.opsmanjam.com/OpsManJam%20Library/Management%20Packs/MOSSWSSWorkArounds.docx)

James Harper

This is Steve Wilson's scripting guide for discovery. This guide focuses on using scripts for discovery workflows, and includes a sample management pack.

Steve Wilson

The article is mainly targeted to Partners and MP authors who want to create MPs that discover SNMP devices components. Familiarity and knowledge of writing basic SNMP monitoring MP is assumed.

Daniel Okine

Customers, who have made an investment in Microsoft System Center Operations Manager 2007 to provide accurate analysis of the performance of their System Center Configuration Manager 2007 infrastructure, also need to monitor aspects of operating system deployment (OSD) not provided out of the box by the System Center Configuration Manager 2007 management pack.  This management pack monitors the following specific actions, which are not captured in the current version of the System Center Configuration Manager 2007 MP:

Matt Goedtel

Customers, who have made an investment in Alachisoft NCache to provide high-performance caching solution for .NET applications, also need to proactively monitor the health and availability of the solution.  This management pack monitors the following specific actions:
• Monitors the health of the NCache Windows service
• Monitors for events generated by the application that indicate availability issues.
• Monitors key performance counters and collects data for performance views/reports.

Matt Goedtel

By default, System Center Operations Manager (OpsMgr) 2007 only monitors services declared in installed management packs and/or declared through the Windows Service monitoring template. This is also tied to the fact that OpsMgr is a specialized monitoring tool, trying to alert not on every service in the system, but only on those deemed critical and/or noteworthy enough to alert in the console.

However, there are some customers who might want to receive alerts when any automatic service is actually not started. There are a few examples going into direction, when you look at the OpsMgr community:

- This example explains how to use a management pack generated by the Windows Service monitoring template, and apply a wildcard to them:
http://blogs.technet.com/brianwren/archive/2008/03/07/using-wildcards-with-the-windows-service-template.aspx
- This other example shows how to have a single rule to cycle through all services and set a moitor state to critical when at least one automatic service is not started:
http://social.technet.microsoft.com/Forums/en/operationsmanagergeneral/thread/5ef23055-0b61-440f-a7c5-bd7db74fd0f4

This MP is based on the previous script, which works as is but has a limitation: when you install it on a default Windows Server 2003 or Windows Server 2008 machine, it will almost surely fire an alert about well-known services : “Performance Logs and Alerts” in Windows Server 2003 ; “Software Protection” and “Shell Hardware Detection” in Windows Server 2008.

So this MP implements some overrides mechanisms, to make sure you can discard specific services from the search. Using those overrides, you can:
- Make sure alerts are not generated for well-known services discussed earlier
- Discard alerts for services already monitored by other management packs and/or Windows Service monitoring templates

Bruno Saille

This sample management pack shows how a timed script can map an application pool to its current worker process and create a custom performance counter telling the current CPU usage. This script is freely based on IISAPP.VBS, which is a script built into the operating system, to list application pools and their worker processes. Since OpsMgr 2007 discovers IIS 6.0 application pools by default, all there was left to do was to attach this timed script to the “IIS 2003 Application Pool” type.

“Note : This MP was mainly created to demo how to map W3WP processes to “known” applications names. For better alerting, it could be enhanced to use average performance counters samples, and not a single performance counter sample.”

Bruno Saille

MPAuthor sample management pack.  Enable monitors and rules only during business hours.  Complete description is provided in the properties of the management pack.

MPAuthor

MPAuthor sample management pack.  Return mutiple values from a single property bag and monitor all with a single monitor type.  Complete description is provided in the properties of the management pack.

MPAuthor

Example of SNMP discovery. It uses a data source type composed of Scheduler, SnmpScanProbe and FilteredClassSnapshotDataMapper to probe network devices and determine whether or not the devices are the Windows SNMP Service. If so, they are discovered as instances of a custom class named Microsoft.SNMP.Service. To test the MP, install the Windows SNMP Service, configure the service to accept SNMP packets from the management server and use Discovery Wizard to discover it as a network device. Then, import the MP. The discovery runs an expression against SnmpVarBind[6] to determine whether or not it’s the Windows SNMP Service.

Michael Sadoff

MPAuthor sample management pack.  Process and reformat data from a data source using a VBScript script.  Complete description is provided in the properties of the management pack.

MPAuthor

The MP is designed to avoid accidental deletion of discovered process instances if a monitored process quits unexpectedly. In the MP, if a process that was discovered in the past quits unexpectedly, and the discovery runs again, the instance will not be deleted from the topology. Instead, a monitor will detect that the process is not running and alert the user. To do this, a VBScript initializes WSMAN.Automation to discover process instances and sets the IsSnapshot property for the discovery item to False. If necessary, the user can override IsSnapshot via the discovery to delete instances that are no longer used.

Michael Sadoff