KACE Systems Management Appliance 13.2 Common Documents - KACE Service Desk Administrator Guide

Working with monitoring profiles

Monitoring profiles describe the criteria for creating an alert, by identifying text to search for in the device's log and associating that text with a defined alert level.

The appliance provides a set of default profiles for log monitoring of devices with supported operating systems, and also for SNMP trap devices. Beyond that, you can modify existing monitoring profiles, create your own profiles, and upload profiles created by other users. In addition, you have access to standard Log Enablement Packages (LEPs) to enable application and threshold monitoring.

The available monitoring profiles are listed on the Monitoring Profiles page

Tip:	TIP: To display only the log monitoring profiles, in the top-right corner, click View By > Type > Log. To display the monitoring profiles for SNMP trap devices, click View By > Type > SNMP Trap.

As an example, the default profile for creating alerts for Mac OS X devices indicates that /var/log/system.log is the log that the monitoring function scans, looking for text that would trigger an alert. The following table describes the default search text in the Include Text field and the associated alert levels.

Text searched for in log	Alert level
critical	Critical
error	Error
fatal	Error
fail	Error
appliance monitor alert	Error
warn	Warning
unavailable	Warning

You can add other alerts customized to your operational needs.

The default profiles cover the following supported operating systems:

•

•

•

•

•

Oracle Enterprise Linux

•

Red Hat Enterprise Linux

•

•

•

•

For devices with Linux operating systems, there are several different log paths for MySQL and Apache logs, depending on the version of the OS. See Profile log paths for MySQL and Apache.

For Agentless devices that are monitored using the SNMP trap mechanism, you need to provide trap message formats and expressions to capture the specific trap elements. See Configure SNMP trap messages and alerting criteria.

In the Log Enablement Packages list page, Quest publishes a base set of Windows Reliability and Performance Monitor (PerfMon) templates and non-Windows open-source Perl scripts, so that users can extend their monitoring capability and identify system and application performance issues. These templates and scripts are available so that users do not have to create them from scratch. Monitoring on the appliance works without these additional templates and scripts, but the profiles that are created from the templates and scripts are helpful if you want to do performance threshold monitoring.

In addition, for convenience, there is a default profile that can be used if you download optional Windows Reliability and Performance Monitor (PerfMon) templates to managed Windows Server 2003 devices. See Set up a Windows Server 2003 device with an ITNinja monitoring Log Enablement Package (LEP).

Edit a profile

You can change, add, or remove alert criteria and log paths for any existing profile.

If you want to use an existing profile as a starting point for creating a profile, see Create a new profile using a default profile as a template.

To identify events that you want raised as alerts, use strings or regular expressions in Include Text to specify the appropriate message content. For instance, if you enter the string, Physical memory, an alert is raised for every message with that exact string.

To cover multiple possibilities, you can use a regular expression. For example, if you want alerts for any drive mount point that has drive errors, in the form, “Drive /dev/[any drive mount point] has drive errors”, you can use Drive /dev/[a-z]{1,} has drive errors in Include Text. Alerts are raised for any messages that contain "Drive /dev/" followed by any word of any length containing the characters a-z, followed by "has drive errors".

You can exclude specific events from being raised as alerts if you find them unnecessary or distracting. To filter the alerts you do not want to receive, you use Exclude Text to indicate the content that identifies an unwanted alert. You can use Exclude Text to filter whole categories of alerts, or use Exclude Text in conjunction with Include Text to refine a subset of an alert category. See Examples of Include Text and Exclude Text for monitoring profiles.

Go to the Profiles list page:

Log in to the appliance Administrator Console, https://appliance_hostname/admin. Or, if the Show organization menu in admin header option is enabled in the appliance General Settings, select an organization in the drop-down list in the top-right corner of the page next to the login information.

On the left navigation bar, click Monitoring, then click Profiles.

Select the check box for the existing profile that you want to edit, and select Choose Action > Edit to display the Profile Detail page.

Optional: Change or modify the Name and Description of the profile to indicate the edits.

NOTE: If you are editing one of the default profiles, you cannot make any change to the Add Automatically To field.

Make changes to the Criteria settings, according to your needs.

◦

Change Include Filter (SNMP traps only) or Include Text (all other monitoring profiles).

On the line with the include search text (or filter for SNMP traps) you want to change, click the Edit button:

Type the new search text or filter.

◦

Optional. Change Exclude Filter (SNMP traps only) or Exclude Text (all other monitoring profiles).

On the line with the text (or filter for SNMP traps) you want to change in order to exclude certain alerts, click the Edit button:

Type the new exclude text or filter.

◦

If the provided search text is case sensitive, select Yes in the Case-sensitive drop-down list.

◦

SNMP traps only. Create a Service Desk ticket automatically each time the appliance receives a specific SNMP alert.

▪

On the line containing an SNMP include and exclude filter (as configured), in the Create Ticket column, click Select Queue, and select a ticket queue that you want to use to create a Service Desk ticket. The appliance will create a Service Desk ticket in the specified ticket queue when it receives an alert resulting from the specific include filter. The device associated with the alert will appear selected in the Service Desk ticket. The name and summary of the event that triggered the SNMP alert will appear in the ticket details. For more information about Service Desk tickets, see Managing Service Desk tickets, processes, and reports.

◦

Change the alert Level.

On the line with the alert level you want to change, click the Edit button:

In the Level drop-down list, select the level from among the five choices: Critical, Error, Warning, Info, and Recovered.

◦

Add an alert Criteria.

On the Criteria category header, click the Add button:

Set the level, include text, exclude text (optional), and case sensitivity.

Click Save at the bottom of the page.

NOTE: You can return a default profile to factory settings for its operating system by using the Reset to Factory Settings button at the bottom of the page.

Configure SNMP trap messages and alerting criteria

You can configure SNMP trap messages and the alerting criteria using the Profiles page.

•

Enable SNMP trap monitoring on the appliance. For more information, see Configure security settings for the appliance

•

Enable monitoring on your SNMP devices. For more information, see Enable monitoring for one or more devices

SNMP (Simple Network Management Protocol) is a protocol for monitoring managed devices on a network. This protocol is supported by Dell Open Manage and many third-party products. When you enable this feature on the appliance, and the related devices are also enabled for monitoring, the appliance can receive SNMP traps from the monitored Agentless devices using SNMP connections, such as printers, projectors, and routers.

SNMP traps are messages initiated by network devices and sent to the trap receiver on the appliance. For example, a router can send a message when its power supply fails. Or, a printer initiates a message when it runs out of paper. The appliance receives these traps and generates alerts when certain pre-defined thresholds are reached.

You can configure SNMP trap messages and the alerting criteria using the Profile Detail page.

You can include or exclude certain events from being detected, as needed.

Go to the Profiles list page:

On the left navigation bar, click Monitoring, then click Profiles.

Complete one of the following steps.

◦

To create a new SNMP trap profile, select Choose Action > New> SNMP Trap Profile.

◦

To edit an existing SNMP trap profile, select it in the list, and Choose Action > Edit.

◦

To duplicate an existing SNMP trap profile, select it in the list, and Choose Action > Duplicate and Edit.

The Profile Detail page appears.

Optional: Change or modify the Name and Description of the profile to indicate the edits.

NOTE: If you are editing one of the default profiles, you cannot make any change to the Add Automatically To field.

Make changes to the Trap Message Formats settings, according to your needs.

For example: %Td (%Tn => %To) %Vz

You can use the following elements in your SNMP trap message:

Element	Description
%Aa	The agent address.
%Ah	The agent host name.
%d	Local day.
%m	Local month.
%y	Local year.
%h	Local hour.
%i	Local minute.
%s	Local second.
%u	Unix timestamp.
%Td	Trap description.
%Tm	Trap MIB (management information base).
%Tn	Trap name.
%To	Trap OID (object ID).
%Tt	Trap type (0-5 Generic; 6 - Enterprise).
%Tv	Trap version (Inform, Trap v1, v2, or v3).
%Vd#	Variable binding description (where '#' is a number representing the element's position in the sequence).
%Vn#	Variable binding name (where '#' is a number representing the element's position in the sequence).
%Vo#	Variable binding OID (where '#' is a number representing the element's position in the sequence).
%Vt#	Variable binding type (where '#' is a number representing the element's position in the sequence).
%Vv#	Variable binding value (where '#' is a number representing the element's position in the sequence).
%Vz	Shows all variable bindings (Name: Value, Name: Value, Name: Value). If a Name is missing (due to a missing MIB file), the OID is displayed instead.

Specify one or more alert levels, as needed.

The following alert levels are available: Critical, Error, Warning, Info, and Recovered.

◦

To add an alert level, under Criteria, click

to add a new alert level.

◦

To edit an existing alert level, in the row containing the alart level, click

For each level, specify its Include and/or Exclude expressions, and also indicate if the expressions are case-sensitive. These expressions allow you to include or exclude certain events from being detected.

The syntax for Include and Exclude expressions is as follows:

<Field_Type> {TRAP_OID|TRAP_NAME|TRAP_DESCRIPTION|TRAP_TYPE|TRAP_MIB|VARBIND} {=|!=|>|<|>=|<=} <Field_Value> [<AND|OR> <Condition_A>] [<AND|OR> <Condition_B>] ...

For example:

◦

TRAP_OID = “.1.3.6.1.4.1.8072.2.3.2.1”: An alert is generated when the trap OID contains ".1.3.6.1.4.1.8072.2.3.2.1".

◦

TRAP_NAME = "acctngFileFull" AND VARBIND = "acctngFileName|ABC": An alert is generated when the trap name contains "acctngFileFull" and if one of the trap's variable bindings is "acctngFileName" with a value of "ABC".

Click Save at the bottom of the page.

NOTE: You can return a default profile to factory settings for its operating system by using the Reset to Factory Settings button at the bottom of the page.

Create a new profile using a default profile as a template

You can copy a default or existing monitoring profile and edit the copy to create a new profile.

You are not limited to one profile for each device. You can create additional profiles that generate different alerts and bind the profiles to devices that already have one or more profiles associated with them.

Go to the Profiles list page:

On the left navigation bar, click Monitoring, then click Profiles.

Select the check box for the existing profile that you want to start with as a template, and select Choose Action > Duplicate and Edit to display the Profile Detail page.

Rename the profile and modify its description.

Optional: Change or modify the Name and Description of the profile to indicate the edits.

NOTE: If you are editing one of the default profiles, you cannot make any change to the Add Automatically To field.

For the log path, use the path appropriate for the operating system or application.

The path can be the basic one for the operating system, as shown in the table.

Oracle Enterprise Linux

/var/log/messages

Red Hat Enterprise Linux

/var/log/messages

Solaris

/var/adm/messages

SUSE Enterprise Linux

application for Windows Application

NOTE: You must use the Full Name of the event log, as it appears in the properties for that log. To ensure you have the correct Full Name, open the Event Viewer. Expand Windows Logs, right-click the event log and select Properties. Use the version of the Full Name that appears in the field in the Log Properties dialog.

Microsoft-Windows-TaskScheduler/Operational for Windows Task Scheduler Operational

Alternatively, you can enter a path that defines a log that contains data beyond the basic event logs. For instance, if you had an application on SUSE that sends its data to a specific log such as /var/log/<myapplog>, you can use that path in a new profile, and define the search text and alert level as described in this procedure.

For devices with Linux operating systems, there are a number of different log paths for MySQL and Apache logs, depending on the version of the OS. See Profile log paths for MySQL and Apache.

NOTE: Only one log path can be defined in a profile. You must create multiple profiles for multiple logs.

Make changes to the Criteria settings, according to your needs.

◦

Change Include Text.

On the line with the include search text you want to change, click the Edit button:

Type the new search text, and, if necessary, select Yes in the Case-sensitive drop-down list.

Click Save at the right of the row.

◦

Optional: Change Exclude Text.

On the line with the text you want to change in order to exclude certain alerts, click the Edit button:

Type the new exclude text, and, if necessary, select Yes in the Case-sensitive drop-down list.

Click Save at the right of the row.

◦

Change alert Level.

On the line with the alert level you want to change, click the Edit button:

In the Level drop-down list, select the level from among the five choices: Critical, Error, Warning, Info, and Recovered.

Click Save at the right of the row.

◦

Add an alert.

On the Criteria category header, click the Add button:

Set the level, search text, and case sensitivity, and click Save at the right of the row

Repeat for as many alerts as you want to add.

Optional: Reorder the new alert criteria using the Drag button:

Click Save at the bottom of the page.

The profile is available to be assigned to a device on that device's Monitoring Detail page.

请选择您的产品：

为向您提供更好的服务，请填写'Purpose of your Chat'（联系目的）：

针对您的问题建议的解决方案

KACE Systems Management Appliance 13.2 Common Documents - KACE Service Desk Administrator Guide

Working with monitoring profiles

Working with monitoring profiles

Edit a profile

Edit a profile

Configure SNMP trap messages and alerting criteria

Configure SNMP trap messages and alerting criteria

Create a new profile using a default profile as a template

Create a new profile using a default profile as a template