Identifying that there is a problem (or impending problem) with an application or service

Infrastructure Management can identify potential problems before customers are impacted. Infrastructure Management can help reduce the number of outages, as well as improve the meantime between failures, thus increasing customer satisfaction, and keeping and maintaining service-level agreements. The longer Infrastructure Management collects data in your environment, the more accurate Infrastructure Management becomes, allowing you to identify problems even more quickly and accurately.

This topic focuses on monitoring VMware vSphere; however, you can apply this process to any environment or combination of environments supported by Infrastructure Management.

Key concepts

Before following the procedures in this topic, you should be familiar with the following concepts:

• The basics: Understanding how events are generated
• Getting started with thresholds

Components required

This topic requires the following Infrastructure Management components. It is recommended that you read the overviews for each component before continuing:

Infrastructure Management Server

Infrastructure Management Integration Service

PATROL Agent

Infrastructure Management single-server deployment architecture

Infrastructure Management administrator console

Infrastructure Management operator console

TrueSight console

Roles required

The following roles are required to complete the procedures in this topic:

• Infrastructure Management Administrator
• PATROL Administrator
• Virtual Infrastructure Administrator
• Infrastructure Management Operator (IT Operations user)

Before you begin

Before you begin these procedures you must have installed the following Infrastructure Management components:

• TrueSight Infrastructure Management Server
• Integration Service - integrates Infrastructure Management with PATROL and other applications that provide events and data to the TrueSight Infrastructure Management Server
• cell

For detailed instructions about installing these components, see Infrastructure Management single-server deployment architecture . After successfully installing these components, continue with the following tasks.

Tasks

The following tasks are required to identify that there is a problem or impending problem in your environment.

Installing the PATROL Agent and Infrastructure Management Performance Management - VMware vSphere Monitoring

Required roles: BMC PATROL Administrator and Infrastructure Management Administrator

From the TrueSight console, create a monitor installation package that includes the PATROL Agent and PATROL for VMware vSphere monitoring solution, by following these steps:

For detailed instructions, see Creating deployable packages. When you are creating your monitor installation package, you will select VMware vSphere and PATROL Agent for your operating system from the Available list. As part of the package creation process, you must include the host name and port where the Integration Service is running (the default port is 3183), as well as a Central Monitoring Administration (CMA) tag. The CMA tag will be assigned to the PATROL Agent to help Infrastructure Management assign monitoring policies that you will create later.

Download and install the monitor installation package you just created on the server where VMware vSphere is installed. For details about installing the monitor installation package, see Deploying packages to PATROL Agents using the TrueSight console.

After the monitor installation package is downloaded and installed, the PATROL Agent automatically is connected to the Integration Service, and the Integration Service automatically is connected to the TrueSight Infrastructure Management Server.

Configuring Infrastructure Management to warn you when there is a problem or impending problem with VMware vSphere

Required roles: BMC PATROL Administrator, Infrastructure Management Administrator, and Virtual Infrastructure Administrator

To configure Infrastructure Management to warn you when there is a problem or impending problem with VMware vSphere, you must understand the concepts of Key Performance Indicators (KPIs), baselines, and thresholds, as described Baselines. To set up thresholds for the various metrics (attributes) of VMware vSphere, you can use the TrueSight console to create a policy. This policy can be applied to any PATROL Agent that has a matching Central Monitoring Administration (CMA) tag.

For instructions on setting up a policy, see Defining a monitoring policy. To monitor VMware vSphere, the following steps apply:

1. Define general policy details.
   When you are defining the general policy details, ensure that the CMA tag that you enter for the policy matches the CMA tag you entered when installing the PATROL Agent that you installed to monitor VMware vSphere.
2. Add the VMware vSphere monitor type.
3. Configure the thresholds for the various attributes of the VMware vSphere monitor types. 

Verifying that data is being brought into Infrastructure Management

Required role: Infrastructure Management Operator

When the Integration Service connects to the TrueSight Infrastructure Management Server, events will begin to display in the operator console. This does not mean that data from the PATROL Agent and monitoring solutions (also called Knowledge Modules) is actually being brought into the TrueSight Infrastructure Management Server. The events displaying in the operator console could be events generated by the TrueSight Infrastructure Management Server as it monitors itself.

To verify that data is being brought into Infrastructure Management, log into the operator console. In the Main drawer of the Navigation pane, click Abnormalities. Ensure that the devices that you have added are displaying in the in the Device View tab in the right pane.

Select one or more of the devices in the Device View tab and click Show Graphs. If data is being brought into Infrastructure Management, data points will be visible on the graph that is displayed. Depending on the polling frequency of the attributes that you are monitoring, it may take 5 to 10 minutes for data to be displayed in the graph.

Watching for incoming events in the Infrastructure Management operator console

Required role: Infrastructure Management Operator

After you have configured your monitor thresholds and established reliable baselines, the events coming into the Infrastructure Management operator console will warn you when there is a problem or impending problem with one of your monitored devices or services. Event severity levels are one of the ways to help you prioritize what events you handle first.