Home > Articles > Operating Systems, Server > Microsoft Servers

  • Print
  • + Share This
This chapter is from the book

R2 Highlights and Capabilities

The Gartner Group report on OpsMgr 2007, referenced at the beginning of this chapter, notes the product's greatest weakness to be its lack of monitoring non-Microsoft platforms. Historically, Operations Manager 2007's strength is in monitoring the Windows environment and lessens when managing non-Microsoft IT systems and devices, whether using Microsoft or third-party add-ons to provide management capabilities. Enter the R2 release—which incorporates cross-platform monitoring!

At the Microsoft Management Summit (MMS) in 2008, Microsoft's annual conference on managing the Windows platform, Microsoft announced the upcoming enhancement of cross-platform monitoring, also known as X-Plat. Not only was OpsMgr going to change to encompass monitoring of non-Windows platforms, but the code would be open source. To add some humor to what was a seismic shift in direction by the company (pigs would fly!), Microsoft distributed pigs...with wings, an illustration of which is displayed in Figure 1.4.

Figure 1.4

Figure 1.4 Pigs flew when Microsoft announced cross-platform monitoring for OpsMgr.

X-Plat was initially going to be part of a service pack; however, at TechEd-Europe in November 2008, Microsoft announced that X-Plat and other changes would be bundled into an R2 release.

In addition to extending OpsMgr 2007 monitoring capabilities to Unix and Linux environments using the familiar Operations Manager console, R2's integration with System Center Virtual Machine Manager (VMM) 2008 enables maximizing availability of virtual workloads. But that's not all—in addition to getting a facelift with new skins (see Figure 1.5), the R2 release includes numerous other performance and functionality enhancements. R2 also resolves quite a few bugs and incorporates a number of hotfixes—see KB article 971410 at http://support.microsoft.com/kb/971410/ for the list.

Figure 1.5

Figure 1.5 The new skin of the R2 Operations console

Successive chapters of this book cover in detail many of the new features in R2, with the remainder of this chapter discussing those topics not discussed elsewhere.

Importing Management Packs

Both MOM 2005 and OpsMgr 2007 support downloading management packs (MPs) from the System Center Management Pack Catalog (http://pinpoint.microsoft.com/en-US/systemcenter/managementpackcatalog) using a web browser outside the Operations Manager console. New with R2 is the capability to download MPs within the Administration node in the console. To download management packs from the OpsMgr console, perform the following steps:

  1. In the OpsMgr console, navigate to Administration -> Management Packs. From the Actions pane on the right side, select the Download Management Packs task to open the Download Management Packs wizard.
  2. The first screen of the wizard asks you to select a local folder in which to store the downloaded management packs. Click on the Browse button to browse to a folder such as c:\Management Packs.
  3. Now, click the Add button in Figure 1.6 to find the management packs you want to download.
    Figure 1.6

    Figure 1.6 Select the Add button to add management packs to the download list.

  4. When selecting management packs, the search criterion in the View section of the Select Management Packs from Catalog page enables you to select one of the following search options:

    • All management packs in the catalog
    • Updates available for installed management packs
    • All management packs released within three months
    • All management packs released within six months

    For this example, the search is for updates to installed management packs. Figure 1.7 shows the results.

    Figure 1.7

    Figure 1.7 The list of installed management pack updates shows updates for Exchange, System Center, and Windows.

  5. You can expand the categories to see the specific management packs available within each area. Click on the Add button in Figure 1.7 to select specific management packs.

    If there are dependencies on other management packs or versions, you will be alerted to this in the download wizard and can select to download those to resolve the issue. Should a management pack contain potentially harmful content, it will alert you to this as well and allow you to view that content.

  6. Complete your selection, and then select OK to view the download list and begin the download. Figure 1.8 shows the download list for the Hyper-V 2008 management pack.
    Figure 1.8

    Figure 1.8 The list of management packs to download

  7. After downloading the management packs to disk, you can use the Import Management Packs task in the Actions pane to import the management packs into your management group.

New and Updated Templates

A template serves as a starting point. OpsMgr 2007 provides templates for several object types to help make it easier to create custom objects. R2 brings additions and enhancements to the templates existing in the RTM and SP 1 releases of OpsMgr 2007. The next sections discuss the Process Monitoring Template, Windows Service Management Pack Template, and OLE DB Management Pack Template. You will find the list of Management Pack templates in the Authoring node of the Operations Manager console. This list will vary based on the management packs loaded into your management group. Other templates new to R2 include Unix/Linux LogFile and Unix/Linux Service. To initiate the wizard for any of the Management Pack templates, right-click on that template, choose Add Monitoring Wizard, and then select the monitoring type.

Process Monitoring Template

Prior to R2, there were ways to monitor processes, but they were a bit clunky and required custom monitors and scripts. The authors of System Center Operations Manager 2007 Unleashed have firsthand experience with this, as they wrote a process monitor available with that book! The new Process Monitoring template enables you to specify a number of criteria when monitoring a process, as follows:

  • Defining scenarios both for processes you want and for unwanted processes.
  • Targeting a group; this narrows the scope of the monitor.
  • Specifying the minimum and maximum number of process instances and duration, and choosing to generate an alert if the process runs longer than that duration.
  • Generating an alert if CPU usage or memory usage exceeds a specified threshold, and indicating the number of consecutive samples that should exceed a threshold before generating an alert.

Figure 1.9 shows part of the configuration of a monitor created using this template.

Figure 1.9

Figure 1.9 Using the OpsMgr 2007 R2 Process Monitoring template

Windows Service Management Pack Template

The Windows Service template lets you discover and monitor a Windows service by doing little more than typing in the service name. This template was first available with the RTM version of OpsMgr 2007 and performs the following two functions:

  • Creates a new class (target) for your service
  • Creates a discovery to find instances of that new class

The RTM release had several issues, as follows:

The R2 version enables wildcard entry to select multiple, similarly named services.

OLE DB Management Pack Template

Enhancements to the OLE DB template allow operators to identify the database and set thresholds for connection, query, and fetch times (see Figure 1.10), and to type or paste a custom query to run against the remote OLE DB data source. You can use this template to simulate synthetic transactions from a user perspective.

Figure 1.10

Figure 1.10 Setting Timing Thresholds in the OLE DB Management Pack template

Note that if you created an OLE DB monitor using an earlier release of OpsMgr 2007, you must upgrade it to the new template before saving any additional changes to the monitor. Open the Properties page to upgrade the template automatically; then click OK at the end of the upgrade process.

User Interface

Operations console performance is greatly improved in R2. Here are the areas where improvements are most evident:

  • Opening new views in the monitoring space
  • Pivoting between views
  • Selecting multiple items in the Results view and rendering the Details pane more quickly
  • Indicating that a view is in the progress of loading

Other interface enhancements include the Overrides Summary and a search tool to assist in creating dashboard views:

  • The Overrides Summary View enables you to view all rule and monitor overrides for sealed and unsealed management packs, and you can customize it by grouping items by multiple column headers.
  • When you create a dashboard view, you populate the different panes in the dashboard with individual, existing views. The R2 release provides a search tool you can use to find the views you want quickly. The search also includes views that you have created in My Workspace.

Run As

With Operations Manager 2007 R2, Microsoft added distribution and targeting features for Run As Accounts and Run As Profiles. When you associate a Run As Account with a particular Run As Profile, R2 lets you target the profile to any class available in Operations Manager and see the logical relationship between the two. Both Run As Account distribution and Run As Account targeting must be correctly configured for the Run As Profile to work properly.

You can specify which computers will receive the Run As Account credentials. You can choose to distribute the Run As Account credentials to every agent-managed computer (less secure) or only to selected computers (more secure). As an example, say you have a script that runs as a response to a monitor or a script that performs some level of monitoring against a SQL Server database. Rather than associating the profile with the agent, you can choose to associate the profile with a database instance or with the SQL Server Database Engine for that agent. This enables you to target the profile by group, object, or instance class. Chapter 9 discusses how targeting works. For additional information on Run As Profiles and Run As Accounts, see http://technet.microsoft.com/en-us/library/bb735423.aspx.

IIS 7 and ASP.NET 64-Bit Apps

R2 adds support for monitoring Internet Information Services (IIS) 7, without having to enable the backward compatibility APIs or legacy management features. Operations Manager 2007 R2 also discovers and monitors 64-bit ASP.NET applications running on IIS servers.

Large-Scale Monitoring of URLs

R2 now supports up to 2,000 URL monitors per management server. The overall capacity of URL monitors increases commensurately for the management group.

Maintenance Mode

R2 streamlines the process of placing a computer and all its related objects into maintenance mode, as follows:

  • You can put an entire computer into Maintenance mode—This automatically puts the Health Service and the associated Health Service watcher into maintenance mode, which suppresses all alerts on that computer.
  • Management pack authors can specify what should be included in an alert and how it is formed when a monitor comes out of maintenance mode.
  • Alerts are generated on HealthService heartbeat failures after exiting out of maintenance mode.

32-Bit Performance Counter Support on 64-Bit Systems

When running 32-bit applications on a 64-bit operating system (OS), unless the pre-R2 OpsMgr agent is 32-bit, you cannot monitor those applications. However, if you are running the 32-bit agent, you will not be able to monitor 64-bit applications or the OS itself. OpsMgr 2007 R2 incorporates a module change, allowing management packs to query the 32-bit registry hive on a 64-bit Windows installation to perform discoveries successfully. This change is also available in a SP 1 hotfix.

Configuration Manager (ConfigMgr) 2007 is one example of a 32-bit application that has trouble being monitored in a 64-bit environment prior to OpsMgr 2007 R2. The ConfigMgr 2007 management pack for OpsMgr 2007 R2 is updated to take advantage of the new discovery method. ConfigMgr 2007 SP 2 will also have native 64-bit performance counters added.

Web Console (Health Explorer)

R2 adds the Health Explorer to the Web console, giving the Web console operational parity with the Operations Manager console for monitoring capabilities. Figure 1.11 displays the Web Health Explorer for one of the servers in the domain used in this book: Odyssey.com.

Figure 1.11

Figure 1.11 Viewing the Health Explorer using the R2 Web console

Notification Subscription Wizard

R2 totally restructures the notification feature, making it easier to set up notifications. The Notifications folder in the Administration node contains folders for Channels, Subscribers, and Subscriptions:

  • A notification is sent by channels; examples are SMTP, instant messaging, or command.
  • Subscribers are the entities that receive a notification; an example is user@odyssey.com.
  • Subscriptions describe what the subscriber will be notified of and the schedule for those notifications.

As before, you can create new channels and subscribers prior to configuring a subscription. However, the New Subscription Wizard enables creating new channels and subscribers if the one you want to use does not already exist. Figure 1.12 displays the Notification Subscription Wizard. After specifying the notification criteria, you can create new subscribers or select existing ones.

Figure 1.12

Figure 1.12 The R2 Notification Subscription Wizard

The next step of the wizard enables you to create a channel.

R2 also provides the capability to create a new subscription directly from an alert and to add the parameters from the selected alert to an existing subscription. This significantly improves efficiency and reduces human error in remembering and incorrectly typing the required string!

Service Level Monitoring

You can use the Service Level Tracking Wizard to define thresholds known as Service Level Objectives (SLOs), enabling you to classify monitors and rules to compare the availability and performance of monitored applications. In the Authoring node of the Operations console, navigate to Management Pack Objects -> Service Level Tracking to define thresholds for a Service Level Agreement (SLA) and create SLOs to represent this SLA. Figure 1.13 displays the Service Level Objectives page of the Service Level Tracking Wizard:

  • The threshold definition for Availability is created using a Monitor State SLO (see Figure 1.14).
  • The Threshold definition for Performance is created with a Collection Rule SLO (see Figure 1.15).
Figure 1.13

Figure 1.13 Adding a Service Level Objective while creating a SLA

Figure 1.14

Figure 1.14 Define the thresholds to track.

Figure 1.15

Figure 1.15 Define the Collection Rule for the Service Level Objective.

After creating these thresholds, the Service Level Tracking Summary Report will compare SLO thresholds with monitored data from the OpsMgr Reporting database.

Reporting Enhancements

R2 adds a number of long-awaited and requested features for reporting users and authors, as follows:

  • Service Level Tracking Reporting—The "Service Level Monitoring" section discussed how R2 provides Service Level Tracking. After creating thresholds, you can run the tracking reports comparing thresholds with state and performance data collected by the system. To run these reports, navigate to the Microsoft Service Level Report Library under the Reporting node in the OpsMgr console. The Service Level Tracking Report compares SLOs with the monitored data. The Service Level Objective Detail Report is very flashy, and is displayed in Figure 1.16.
    Figure 1.16

    Figure 1.16 Service Level Tracking Objective Detail Report

  • Saving Reports to a Management Pack—You can save reports to an existing management pack, enabling you to share a report using a specific set of parameters. See http://technet.microsoft.com/en-us/library/dd789017.aspx for details. As mentioned in this reference, a management pack with a saved report to another management group using a different data warehouse requires you to save the report again for it to function properly.
  • Report Object Picker—Object picker enhancements allow searching and filtering by name or class of object, making it easier to find the objects needed to retrieve data. Filter settings are applied under the Options button for both the Add Objects and Add Group functions, and the feature enables you to select only certain (meaningful) classes of objects. Figure 1.17 illustrates this using a report in the SQL Server 2008 management pack. The Options button was selected in the previous dialog, bringing up the SQL 2008 DB Engine as the only valid object to select for this report.
    Figure 1.17

    Figure 1.17 The R2 Report Object Picker automatically displays the objects of interest for you to select.

    This feature has the most value when the list of supported classes is defined in a management pack. However, only newer releases of management packs take advantage of this capability.
  • Authoring Console Support—The Authoring console now includes a Reporting space, exposing all available reporting elements.
  • Enhanced Most Common Events Report—R2 adds multi-selection parameters to this report.

Recalculating and Resetting Monitor States

The Health Explorer includes the ability to force the agent to recalculate the health of a particular monitor. One of the most common uses for this is when you want to confirm that the action that you took actually fixed the problem. If the management pack is correctly implemented, all you need to do is select the unit monitor in Health Explorer and click on the "Recalculate Health" button.

Health Explorer also includes a Reset Health button. The behavior of these buttons has varied during the different updates of OpsMgr 2007, and has pretty much been a moving target:

  • At OpsMgr 2007 RTM, the interface only had a Reset button, which actually did a Recalculate, but only if on-demand detection was defined by the management pack. It did not do a Reset (for further information, see Chapter 14, "Monitoring with Operations Manager," in System Center Operations Manager 2007 Unleashed).
  • SP 1 added a Recalculate button. The Recalculate button had a similar issue to the Reset button in RTM, as it actually worked only when there was on-demand detection. The Reset button worked as expected, and changed the monitor state to healthy regardless of its current state (even if it was already in a healthy state).
  • R2 continues to have both the Reset and Recalculate buttons, and Recalculate works more accurately than it did in SP 1.

Power Consumption Monitoring

A new Power Management Library MP enables you to monitor power consumption for each computer or a group of computers. Monitored systems must be running Windows Server 2008 R2 or Windows 7. Additional information is available at http://technet.microsoft.com/en-us/library/dd789061.aspx. See http://www.windowsservercatalog.com/results.aspx?&chtext=&cstext=&csttext=&chbt ext=&bCatID=1333&cpID=0&avc=10&ava=0&avq=30&OR=1&PGS=25&ready=0 for the current list of supported hardware.

  • + Share This
  • 🔖 Save To Your Account