Chat now with support
Chat with Support

Foglight Experience Monitor 5.8.1 - Installation and Administration Guide

Installing and configuring Multi-appliance clusters Configuring the appliance Specifying monitored web traffic Transforming monitored URLs Managing applications Foglight components and the appliance Using the console program Troubleshooting the appliance Appendix: Third party software Appendix: Dell PowerEdge system appliance

Distribution metrics

The Distribution Metrics configuration page, accessed by clicking Configure > Reporting > Distribution Metrics allows you to add and modify a distribution configuration for specific distribution metrics. Distribution metrics are used to display the exact breakdown (or frequency) of individual data points for metrics, such as End-to-End Time and Processing Time.

To configure a distribution you need to choose a name, a metric, and define the units and buckets. In the Name box, enter a self-explanatory name for the distribution configuration. From the Metric list, select the distribution metric. You can only choose from the list of existing distribution metrics.

 
* 
IMPORTANT: Each distribution metric has a default configuration except for the Access Speed Distribution which cannot be customized.

When a distribution metric is edited, the changes apply to all metrics across all categories that use the distribution configuration. You can also create additional distribution configurations and use them when configuring Application Components, Services, SOAP Applications and SOAP Transactions.

For each bucket, enter the start and end time in milliseconds. For example, the first bucket might start at 0 and end at 25. All data points between 0 and 25 milliseconds are counted in this bucket.

Depending upon the level of granularity in which you want to view data, you can define the number of buckets and the range of each bucket to suit your needs. Buckets display the total count of individual data points that fall within the lower and upper ranges defined for the bucket.

Click the Add a Bucket to create additional buckets or to remove the last bucket click Remove Last Bucket.

Figure 30. Distributions: Edit page

 
 
* 
IMPORTANT: Each distribution configuration contains a version number. If a distribution configuration is modified, metrics for that distribution collected prior to the change can no longer be displayed in the web console. If you attempt to view a distribution metric, previously stored in the database in a time frame prior to the last configuration change, a warning message is displayed.
 
* 
NOTE: The maximum number of buckets that can be configured for a distribution is 100.
 
* 
NOTE: The system must be restarted in order for the changes to take effect. For more information, see System restart.

After a distribution is configured, you can view metrics from the related page. For example, the Metric View for Enterprise displays several distributions. The Command Initial Response Time Distribution displays three columns: unit (msecs), count and percentage. The unit column displays the bucket range, the count column displays the number of downloads for each bucket and the percentage column displays the percentage for each bucket of the total number of downloads.

Figure 31. Viewing distributions

Percentile metrics

The Percentiles section, accessed by clicking Configure > Reporting > Options, allows you to change the percentage of events that are included in percentile metric calculations.

Figure 32. Viewing percentile metrics

Percentile metrics are used to verify the quality of service for a specific event that is related to the end-user experience. Some examples of percentile metrics are Page End-to-End Time Percentile and Page Processing Time Percentile.

Percentile metrics indicate the quality of service by reporting the maximum time required for an event type to occur, where the time reported represents the upper threshold for n% of all occurrences of the event over a given time frame.

For example, if the Page End-to-End Time Percentile metric has a value of four seconds, this means n% of the pages downloaded for the monitored site were received by end users in less than four seconds. By default, the percentile level is set to 95%. This means the metrics for the aforementioned events report the time required for an event to complete for the 95th percentile.

The system approximates the time required for n% of all events to occur by using the mean, standard deviation, and the standard normal variable. See “Metric definitions” in the Foglight Experience Monitor Metric Reference Guide for information about how these metrics are calculated.

Changing the percentile level

To change the percentile level, enter a value in the box on the Percentile Metrics configuration page, then click Apply. All percentile metrics are calculated using the new percentile level.

Foglight metrics

The Foglight Metrics configuration page, accessed by clicking Configure > Reporting > Foglight Metrics, allows you to select a subset of location data that is important to your configuration and export to the Cartridge for FxM in Foglight.

Figure 33. Foglight Metrics page

Select from the five metric categories: city, region, country, ISP, and subnet information that you want to become available in Foglight. For each metric category, any associated application and service information is also exported to Foglight.

For example, by selecting the city Reno, NV (US) the following metrics are exported to Foglight:

•
City for Reno
•
Application by City for Reno
•
Service by City for Reno
To configure Foglight metrics:
1
Select the metric category that you want to configure and click its Add link.
For example to add a city click the Add City link.
2
Select a resource from the list or use the search box to look for a specific city.
The resource is added to the City metric list.
3
Continue to add resources to each list that you want exported to Foglight.
4
Click Apply.

User agent transformations

The User Agents page, accessed by clicking Configure > Reporting > User Agents, allows you to create, modify, or delete rules that determine how user agent HTTP headers are transformed before being recorded in the User Agents metric category. The HTTP headers are sent by the browser to inform the web server about the identity of the browser. Whenever the system encounters a user agent string that does not match any of the defined rules, the original user agent string is recorded. By default, the page displays a number of preconfigured transformation rules that can be removed or modified by an Administrator.

Figure 34. User Agents page

The User Agents metric category gives a breakdown of user agents that accessed the monitored site from a variety of perspectives. These metrics can show you which user agent had the longest page processing time.

Figure 35. Example of Metric Analysis page

The list of user agents displayed are retrieved from the HTTP headers of incoming session traffic. The appliance uses transformation rules that take the user agent string, and change it into a form that is more easily understood by end users.

For example:
Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 6.0)

would be transformed into:
Explorer 7.0

To add a user agent transformation rule:
1
Navigate to the Configure > Reporting > User Agents page and click Add User Agent Rule.
The User Agents > Edit page is displayed.
2
In the Find box, type a string or regular expression that will be used to match against the user-agent HTTP header.
Alternatively, click Select User Agent to select from a list of user agents that the appliance has encountered so far.
This list above is populated with both transformed and non-transformed user agent strings.
a
Search for a specific user agent by typing in a query using the Search text box.
b
If you need to search by Perl regular expression, check the Interpret as a regular expression option.
c
Click Go.
d
Select a User Agent from the returned list.
e
Close the Resource List.
3
In the Display box, type a string that the system will use as the resource name for user agents matching the Find string.
4
Select the Interpret find string as a regular expression check box if you want the regular expression implemented in Perl.
If you select this option, matching will be performed in a case-sensitive manner. Perl regular expressions can be made case-insensitive by adding the following characters at the beginning of the string “(?!)”. If you do not select this option matching is performed in a case-insensitive manner. The system will match the Find string against any part of the user agent string.
5
Select the Extract version following the find string check box if you would like the user agent version number to appear with the agent name in the web console.
 
* 
NOTE: By default, this check box is not selected.
When this option is selected the match is based on the digits that appear after the search string. When there are no digits after the search string, within three characters, the system assumes that there is no version.
For example, if a transformation rule’s Find value is mozilla, and its Display value is Mozilla, Mozilla 4.0 and Mozilla 5.0 will appear as separate entries in the user agents list.
6
Select Applies only to major version numbers check box if the transformation rule you are creating should only be applied to user agent versions that fall within the defined range entered in the from and to boxes.
7
Click the Test link to test the Find string against the current list of User Agents.
 
* 
NOTE: Results displayed may include some strings that have already been transformed by existing User Agent rules.
8
Click OK to save the user agent transformation rule.
The new user agent transformation rule takes effect immediately.

All existing transformation rules are listed on the User Agents page. To edit a user agent transformation rule, click its corresponding Edit link to display the User Agents > Edit page, where you can modify the string the rule matches against incoming user agent strings, the display name in the web console, or how user agent version numbers are handled.

To remove an individual transformation rule, select its corresponding check box, then click Delete.

To reset the default set of transformation rules, click Reset User Agent Rules (located at the bottom of the page). The current set of rules is dropped and then the default rule set is initialized.

Using multiple transformation rules with different user agent version ranges

The Has Version and the Applies only to major version numbers check boxes can be configured differently in multiple user agent transformation rules to differentiate user agent versions ranges whose performance you wish to evaluate separately or differently.

For example, the appliance’s default handling of the Mozilla® web browser as user agent is as follows:

Rule A:
Find: mozilla
Display: Netscape
Has Version: yes
Applies only to major version numbers: from 0 to 4

Rule B:
Find: mozilla
Display: Mozilla
Has Version: yes
Applies only to major version numbers: n/a

If mozilla is found in the user agent, it is displayed as Mozilla, along with its version number unless is it found as version 0 through 4, in which case, it is displayed as Netscape along with its version number.

Use multiple overlapping transformation rules to distinguish or consolidate user agent types and versions as required in order to effectively evaluate your web applications.

 
* 
NOTE: The user agent transformation rules are applied in the order that they appear on the Configure > Reporting > User Agents page. If you want a user agent transformation rule, that you add, to have priority over others, you need to use the up and down arrows to move the position of the rule in the list.
Related Documents

The document was helpful.

Select Rating

I easily found the information I needed.

Select Rating