Table Of Contents
Alerts
Alert Types
NAM Alarms
Threshold Violations
System Events
Alert Severity Color Codes
Displaying Alerts
Filtering Alerts
Displaying Alert Details
Suspending and Resuming Alerts
Alerts
The Cisco PVM Alert Viewer displays a list of alerts occurring over time or based on specific filtering criteria. You can filter the list of alerts based on time period, source, type, severity, or cause, as well as view individual alert details.
This chapter contains the following sections:
•
Alert Types
•
Alert Severity Color Codes
•
Displaying Alerts
•
Displaying Alert Details
•
Suspending and Resuming Alerts
Note
The functions discussed in this chapter apply to Cisco PVM Administrators only.
Alert Types
Alerts consist of three general types:
•
NAM alarms—violations of criteria set in an external interface and sent through SNMP.
•
Threshold violations—violations of criteria set in Cisco PVM.
•
System events—notifications of licensing errors, timeouts, and system health information.
For every data aggregation period, the system performs the following functions for each statistic:
1.
Aggregates all the data collected for a specific metric (measurement) during the specified aggregation period
2.
Compares the results with the Dynamic Threshold value calculated based on the deviation from the baseline that corresponds to the severity level assigned to the specific Threshold
3.
Raises an alert if the result is greater than the percentage deviation assigned to the Threshold.
System-generated alerts include system health events. Cisco PVM monitors the following metrics:
•
CPU utilization of each CPU on the system
•
RAM utilization
•
Disk utilization for each of the locally and remotely mounted file systems and
•
Tablespace utilization for all tablespaces in the Cisco PVM database.
This section contains the following topics:
•
NAM Alarms
•
Threshold Violations
•
System Events
NAM Alarms
NAM-generated alerts result from a violation of criteria configured for the device in either the NAM Traffic Analyzer or a third-party SNMP application. The violations are read from the RMON MIB, and include violations for:
•
Rising Threshold Crossed
•
Falling Threshold Crossed.
All NAM alerts are listed as minor in the Cisco PVM alert viewer.
Threshold Violations
Threshold-generated alerts result from violations of user-defined criteria set in Cisco PVM. These alerts appear when a Threshold is violated for an assigned traffic metric with data aggregated for any data source in a specified Data Source Group. Administrators assign specific metrics, Data Source Groups, and alert severity levels using Setup > Thresholds.
Administrators can define two types of Thresholds in Cisco PVM:
•
Dynamic Thresholds—the Cisco PVM server performs automatic baselining for each statistic-specific attribute and metric combination based on previous data collection. User-defined percentage deviations from the baseline are translated into alerts, increasing in severity with the degree of deviation.
•
Fixed Thresholds—the system generates an alert once a user-defined minimum value for a specific metric has been exceeded.
Note
For a discussion of Threshold calculations, see Threshold Setup, page 2-31.
System Events
Cisco PVM monitors system health and utilization metrics for CPUs, RAM, disks, and database tables. System events that generate alerts include:
•
Success or failure of resource inventory import status.
•
NAM connectivity failure or reconnect.
•
NAM SNMP timeout.
•
NAM Application Response Time (ART) configuration change.
•
NAM general configuration change.
•
Switch/Router connectivity failure or reconnect.
•
Switch/Router general configuration change.
•
Switch/Router SNMP timeout.
Alert Severity Color Codes
Cisco PVM displays a color-coded icon next to each alert in the list indicating severity. Table 6-1 shows the color assigned to each severity level:
Table 6-1 Alert Severity Color Codes
Severity
|
Color
|
Critical
|
Red
|
Major
|
Orange
|
Minor
|
Yellow
|
Warning
|
Cyan
|
Cleared
|
Green
|
Indeterminate
|
Blue
|
Information
|
Gray
|
Note
A Cleared alert has either had its Threshold adjusted, or the current traffic level no longer violates the Threshold level of the previous alert. A given alert might go through all the severity levels unless it is cleared quickly or the severity level set in Thresholds is adjusted.
Displaying Alerts
Cisco PVM displays a paginated, continuous list of performance alerts in descending order by date, and allows you to filter alerts by time period and other criteria. The display time period defaults to the last hour.
Note
The maximum number of alerts that the GUI will display is 1,000. An alert will remain in the GUI list until it falls to greater than the last 1,000 alerts logged in the system.
Step 1
On the Cisco PVM dashboard, click Alerts.
The Alerts window appears, showing the alerts for the hour (see Figure 6-1).
Tip
You can sort the list in ascending or descending order by clicking any of the column headers.
Figure 6-1 Alerts Window
Table 6-2 describes the fields available in the Alerts Window.
Table 6-2 Alerts Window Field Descriptions
Field
|
Type
|
Description
|
From Date
|
Field
|
Displays the starting date and time of the alerts in the list, and accepts the date selection from the pop-up calendar.
|
Calendar
|
Icon
|
Displays a pop-up calendar for selection of date and time.
|
To Date
|
Field
|
Displays the ending date and time of the alerts in the list, and accepts the date selection from the pop-up calendar.
|
Log Type
|
Drop-down list
|
Allows selection of the type of violation for filtering the list:
• Generic
• Unknown
• Rising Threshold Crossed
• Falling Threshold Crossed
• Cisco PVM System Health
• ARCHIVE
• PURGE
|
Description
|
Text box
|
Allows entry of characters (including the wildcard) for filtering the list of events.
|
Severity
|
Drop-down list
|
Allows selection of the severity type for filtering the list:
• Indeterminate
• Critical
• Major
• Minor
• Warning
• Cleared
• Information
|
Cause
|
Drop-down list
|
Allows selection of the cause of the violation for filtering the list:
• Generic
• Unknown
|
Clear
|
Button
|
Clears all of the fields in the filter area for performing a new filter operation.
|
Filter
|
Button
|
Filters the list of alerts based on one or more criteria entered in the filter area.
|
Severity
|
Column header
|
Lists the severity level next to a color code and links to the Alert Detail window for the individual alert.
|
Date
|
Column header
|
Displays the date and time the violation was generated.
|
Log Type
|
Column header
|
Displays the type of violation that created the alert:
• Generic
• Unknown
• Rising Threshold Crossed
• Falling Threshold Crossed
• Cisco PVM System Health
• ARCHIVE
• PURGE
• ART
|
Description
|
Column header
|
Displays the device, metric, Threshold, or system check that generated the alert.
|
Statistic
|
Column header
|
Displays the type of traffic (statistic) upon which the violation is based.
|
Log Source Type
|
Column header
|
Displays the originating source of the type of alert log created for the violation:
• Switch
• NAM
• Cisco PVM
• Generic or
• Unknown
|
Filtering Alerts
The Alerts list can be filtered based on time interval, log type, alert severity, description, or cause.
You can filter the list using any or all of the fields at the top of the Alerts window.
Step 1
Click Alerts in the Cisco PVM dashboard.
The Alerts window displays a paginated list of all of the alerts generated in the hour, in descending order by date and time.
Step 2
Select the start date by clicking the calendar icon next to the From field.
Step 3
Select the end date by clicking the calendar icon next to the To field.
Step 4
Select the log type from the Log Type drop-down list.
Step 5
Enter descriptive characters in the Description box, using the percent symbol (%) as a wildcard to broaden your search.
Step 6
Select the severity level from the Severity drop-down list.
Step 7
Select the alert cause from the Cause drop-down list.
Step 8
Click Filter.
The window displays the Alerts list containing only the alerts that match the filter criteria.
Tip
If you want to begin filtering with a new set of criteria, click Clear to reset all of the fields in the filter area to blank. To display the original, unfiltered list of Alerts, click Alerts under Generate Reports on the left side of the window.
Displaying Alert Details
Step 1
Click the link in the Severity column of the alert you wish to view. The Alert Detail window (Figure 6-2) displays the individual alert's description along with the traffic type (statistic), database information, and additional information indicating what constraint has been violated.
Figure 6-2 Alert Detail
Table 6-3 describes the fields in the Alert Details window.
Table 6-3 Alert Details Field Descriptions
Field
|
Description
|
Log Id
|
The sequential number of the alert as it appears in the database.
|
Log Type
|
The type of alert generated:
• Generic
• Unknown
• Rising Threshold Crossed
• Falling Threshold Crossed
• System Health
• Purge
• Archive
|
Date
|
The date and time the alert was generated.
|
Severity
|
The severity level of the alert:
• Indeterminate
• Critical
• Major
• Minor
• Warning
• Cleared
• Information
|
Statistic
|
The type of traffic (statistic) upon which the violation is based.
|
Cause
|
Either Generic (known to the system) or Unknown.
|
Managed Object Id
|
The classification of the managed object as it appears in the database.
|
Managed Object Name
|
The name of the managed object as it appears in the database.
|
Description
|
The device, metric, Threshold, or system check that generated the alert.
|
Log Content
|
Detailed information about the actual alert as contained in the database, such as:
• the baseline value at the time of the alert.
• the value that was actually violated (based on severity level, baseline, and standard deviation calculations).
• the Data Source Group assigned to a Threshold (for Threshold violations only).
• the Data Source that generated the traffic.
• the traffic metric monitored, such as bytes, packets, or errors.
• device-specific identifier.
|
Back [button]
|
Closes the Alert Details and returns to the Alerts window.
|
Step 2
Close the details by clicking Back.
The system returns to the Alerts window.
Suspending and Resuming Alerts
Administrators can suspend alerts for individual thresholds defined in Setup. Under the Setup GUI, thresholds can be disabled if you no longer need or want to view traffic-related violations in the Alert Viewer. The threshold definitions remain in the system, and they can be re-enabled if desired. Disabled thresholds do not generate alerts.
With access to a NAM's external interface (NAM Traffic Analyzer), you can make configuration adjustments that also affect whether alerts appear in the Cisco PVM alerts viewer. Such alerts are sent through SNMP to the Alert Viewer directly from the NAM, and are not managed in Cisco PVM.
Note
For a complete discussion of thresholds and how they relate to alerts, see Threshold Setup, page 2-31. See the NAM Traffic Analyzer documentation for information on adjusting NAM alarm settings and trap destinations.
Follow these steps to disable or enable thresholds:
Step 1
Click Setup in the Cisco PVM dashboard.
Step 2
Click Thresholds in the Setup navigation menu.
Step 3
Check the box(es) next to the threshold(s) you want to enable or disable.
Step 4
Click:
•
Disable to suspend the alerts related to the specific threshold(s) or,
•
Enable to resume alerts.
The Thresholds window refreshes to reflect the new status of the selected thresholds in the Status column.