Monitoring Application Performance User and Reference Guide

Transcription

1 Foglight Monitoring Application Performance User and Reference Guide

2 2012 Quest Software, Inc. ALL RIGHTS RESERVED. This guide contains proprietary information protected by copyright. The software described in this guide is furnished under a software license or nondisclosure agreement. This software may be used or copied only in accordance with the terms of the applicable agreement. No part of this guide may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording for any purpose other than the purchaser s personal use without the written permission of Quest Software, Inc. The information in this document is provided in connection with Quest products. No license, express or implied, by estoppel or otherwise, to any intellectual property right is granted by this document or in connection with the sale of Quest products. EXCEPT AS SET FORTH IN QUEST'S TERMS AND CONDITIONS AS SPECIFIED IN THE LICENSE AGREEMENT FOR THIS PRODUCT, QUEST ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED OR STATUTORY WARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL QUEST BE LIABLE FOR ANY DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF PROFITS, BUSINESS INTERRUPTION OR LOSS OF INFORMATION) ARISING OUT OF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF QUEST HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Quest makes no representations or warranties with respect to the accuracy or completeness of the contents of this document and reserves the right to make changes to specifications and product descriptions at any time without notice. Quest does not make any commitment to update the information contained in this document. If you have any questions regarding your potential use of this material, contact: Quest Software World Headquarters LEGAL Dept 5 Polaris Way Aliso Viejo, CA legal@quest.com Refer to our Web site for regional and international office information. Trademarks Quest, Quest Software, the Quest Software logo, Foglight, IntelliProfile, PerformaSure, Spotlight, StealthCollect, TOAD, Tag and Follow, Vintela Single Sign-on for Java, and vfoglight are trademarks and registered trademarks of Quest Software, Inc in the United States of America and other countries. For a complete list of Quest Software s trademarks, please see Other trademarks and registered trademarks are property of their respective owners. Third Party Contributions Foglight contains some third party components. For a complete list, see the License Credits page in Foglight online help. User and Reference Guide December 2012 Version 5.7.1

3 Monitoring Application Performance: User and Reference Guide 3 Table of Contents Introduction to this Guide...5 About Quest Software, Inc Contacting Quest Software... 5 Contacting Quest Support... 6 Introducing Foglight APM...7 Understanding the Foglight APM Implementation... 8 Foglight Requirements for APM Workflows... 8 Recommended Service Structures... 8 Service Level Agreements in Foglight APM... 9 Using the SOC End User Tab...11 Pre-requisites for Monitoring End User Data in the SOC Configuring Trace Analysis for the SOC Exploring the End User Tab Real Users Synthetics Configuring Display Options Understanding Geographical Perspectives...18 Creating a Map Agent Viewing Real End User Activity Editing Map Agent Properties Generating Geographical Perspective Reports Examining Response Times...28 Configuring Request Traces Associating FxV Hit Filter Mappings Exploring Response Time Breakdowns Viewing the Response Time Breakdown Explorer Execution Time History Top Sampled Requests by Execution Time Additional Configuration Options Assigning Proxy Objects Mapping Datasources... 35

4 Monitoring Application Performance: User and Reference Guide 4 Table of Contents Using the SOC for APM Triage Database Layer Application Layer Virtual Machine Creating Custom Drag-and-Drop Dashboards Linking a Custom Dashboard to the Explore Icon for a Service Reference Overview of APM Tiles NET Tile DB2 Tile Host Tile Java Tile JMX Tile MQ Tile Oracle Tile OS Tile Real User Tile SQL Server Tile Sybase Tile Synthetics Tile Multi-Location Synthetics Tile Single-Location Synthetics Tile...81 Virtual Hyper-V VMware Tile Overview of Detail Views Application Server Detail View Host Detail View Real User Performance Detail View...86 Performance Tab Response Time Breakdown Tab SLA Violators Tab Synthetic Result Detail View Synthetic User Performance Detail View Virtual Machine Detail View ESX Server Detail View Topology Object Name Reference Table Index... 95

5 Introduction to this Guide This User and Reference Guide provides an overview of Application Performance Monitoring (APM) with the Foglight Cartridge for Application Operations. It includes conceptual information to help you envision an APM installation with Foglight and an overview of incident management workflows through the Service Operations Console (SOC). This guide is intended for Application Performance Managers who administer the overall design, implementation, and execution of the application performance monitoring strategy. About Quest Software, Inc. Established in 1987, Quest Software (Nasdaq: QSFT) provides simple and innovative IT management solutions that enable more than 100,000 global customers to save time and money across physical and virtual environments. Quest products solve complex IT challenges ranging from database management, data protection, identity and access management, monitoring, user workspace management to Windows management. For more information, visit Contacting Quest Software Mail Web site Quest Software, Inc. World Headquarters 5 Polaris Way Aliso Viejo, CA USA Refer to our Web site for regional and international office information.

6 Monitoring Application Performance: User and Reference Guide 6 Introduction to this Guide Contacting Quest Support Quest Support is available to customers who have a trial version of a Quest product or who have purchased a Quest product and have a valid maintenance contract. Quest Support provides unlimited 24x7 access to our Support Portal at From our Support Portal, you can do the following: Retrieve thousands of solutions from our Knowledge Base Download the latest releases and service packs Create, update, and review Support cases View the Global Support Guide for a detailed explanation of support programs, online services, contact information, policies, and procedures. The guide is available at: support.

7 1 Introducing Foglight APM Application Performance Monitoring (APM) refers to a set of capabilities or technologies used by IT staff to support the incident and problem management processes on multi-tier applications. A comprehensive APM implementation generally includes technology that includes the following capabilities: End-user experience monitoring Run-time application topology and execution path discovery, modeling, and display Transaction profiling across the technology stack Resource consumption and event monitoring within the discovered components Analytics Quest s Foglight APM solution supports all of these features, and provides the information to the right users at the right time through a set of role-oriented operational dashboards. Although there is no single organizational alignment or uniformly agreed upon set of titles for the actual practitioners of APM, the following role descriptions indicate the types of functions that typically interact with a Foglight APM installation: Foglight Administrator responsible for configuring and managing Foglight, and for performing administrative tasks. Note With the exception of Foglight Administrator, the roles described here are not roles defined in Foglight. Application Performance Manager responsible for the design, implementation, and execution of the APM strategy. Application Support responsible for ensuring that application services are performing as expected. Identifies and triages service degradations or outages. Platform Specialist or Administrator responsible for the operation of an individual application component (for example, a database, a host, an application server, or a web server). Receives and responds to Foglight alarms on their platform or domain. Application Architect a senior escalation resource who is typically engaged by Application Support for assistance in locating the source of an application issue. Application Developer responsible for the application code. Uses the Foglight transactional performance dashboards to gain understanding of the operational execution and performance of their code.

8 Monitoring Application Performance: User and Reference Guide 8 Chapter 1 Introducing Foglight APM Understanding the Foglight APM Implementation Foglight supports a modular approach to APM that allows customers to implement as many or as few pieces of the APM solution as required. This guide is intended as a comprehensive overview of an end-to-end APM installation strategy, and therefore includes cross-domain views. Before you implement an APM strategy, it is important to perform proper requirement gathering, environment sizing, project planning, and scoping, just as you would with any other application. Quest Professional Services has developed a comprehensive strategy for APM implementation. You can review our process online at: Foglight Requirements for APM Workflows In order to make the best use of this document, ensure that your Foglight installation meets the following requirements before beginning the workflows: Install the Foglight Management Server and ensure it is running. For more information, see the Foglight Installation and Setup Guide set. Install and configure the cartridges required to monitor application components (for example, operating system, application servers, databases). For more information, see the documentation included with the cartridges you are using. Configure End-User transaction in Foglight Experience Monitor (FxM), Foglight Transaction Recorder (FTR), and Foglight Experience Viewer (FxV). For more information, see the documentation for these components. Build services, either through Dependency Mapping or manually, and add the services to the Service Operations Console. For more information, see the topic Creating and Maintaining Services in the Managing Dependency Mapping User Guide. Recommended Service Structures In Foglight, a service is any component or group of components that you want to monitor. If you have the Advanced Operator role, you can create services in the Service Builder dashboard to reflect the components in your monitored environment that are meaningful or interesting to your organization. For more information, see the topic Monitoring Your Services in the Foglight User Help. APM users typically organize their application services into three types: a Top Level Service, Tier Services, and one or more User Services. Each logical tier in the application should have its own service, which includes the application components (for example, databases, application servers, web servers), and the hosts that support the components. The Top Level Service contains all elements in the application. It includes all child services and is added to the Service Operations Console (SOC) for Application Support. The Tier Service is a logical representation of a service component. Foglight organizes data into default tiers, including User, Web, Application, Database, Host, and Agent. Over time, this Tier Service can track operational level agreements.

9 Monitoring Application Performance: User and Reference Guide 9 Chapter 1 Introducing Foglight APM The User Service contains all End-User artifacts, and organizes the transactions that support the application. For most users, a single End User service is sufficient. For example, a typical small web application might have the following service structure: This is a high-level, or tier view of the service structure. This service includes four tiers: user, web, application, and database. This service can also be represented as: This is the Application Infrastructure view. Here you can see the hosts contained in each tier: an Apache server (Web tier), two instances of WebLogic each hosted on a separate application server (App tier), and an Oracle database instance (DB tier). For more information, see Using the SOC End User Tab on page 11. Service Level Agreements in Foglight APM Foglight automatically examines each service and establishes its availability and service level compliance. By default, a service is available and compliant with its Service Level Agreement (SLA) if it does not have any fatal alarms. Quest recommends the following general guidelines for APM: Implement a top-level SLA based on end-user transactional data to allow the Application Support team to become familiar with SLA functions and the data available. For more information, see Defining How the SLA is Calculated for Services on page 10. Defer the implementation of tier-based SLAs until your organization is ready to implement official operational level agreements for the infrastructure and component level teams.

10 Monitoring Application Performance: User and Reference Guide 10 Chapter 1 Introducing Foglight APM Defining How the SLA is Calculated for Services When you create a new, or edit an existing, service, you can specify which components of the service are used to determine its overall availability. Note This features requires Foglight Management Server or later. To define how the SLA is calculated: 1 On the Foglight navigation panel, under Dashboards, click Services > Service Builder. 2 To edit an existing service, click the edit icon for the tier. Tip To create a new service, click Add A New Category, and review the Building a Service topic in the Foglight User Help. The Create Basic Information page of the Edit Service Wizard opens. 3 In the Objects used to determine the availability of this service section, select one of the following options: All components added to this service. This is the default option. Only components in selected tiers. Select this option to specify the individual tiers that you want to include. Click the check box beside the tier name to include it. 4 Click Finish.

11 2 Using the SOC End User Tab Service mechanisms in Foglight allow users to organize monitoring components into logical groups. A primary use case for this is for a multi-tier application where objects are organized into service tiers and visualized in the Service Operations Console (SOC). The Foglight Cartridge for Application Operations includes an extension to the SOC that adds visualization for end-user data in the context of application services in Foglight. This information appears on the End User tab of the SOC. Use this tab to quickly identify if any transactions are having issues, to determine the magnitude of impact on the user population, and to view preliminary information that can aid in troubleshooting, including geographic discrepancies and whether the performance problems are in the back end architecture or the client-side browser or network. The following specific end-user monitoring systems can be included: FxM Foglight Experience Monitor. A network appliance that monitors Web transactions from real end users by tapping into the HTTP/HTTPS traffic stream as it transverses the internal network. FxV Foglight Experience Viewer. A transaction repository that stores and indexes every HTTP/HTTPS request/response pair captured from the network. FxV users can search, find, and analyze individual problematic transactions. FTR Foglight Transaction Recorder. A Windows-based application that records and plays back synthetic Web transactions on regular intervals. Note For complete information about FxM, FxV, and FTR, see their product documentation. Pre-requisites for Monitoring End User Data in the SOC In order to make full use of the End User performance monitoring in the SOC, ensure that the following pre-requisites are met: The End User Core Cartridge, the Cartridge for FxM, and the Cartridge for FxV have been installed on the Management Server. Important These cartridges must be installed to enable the End User tab in the SOC, regardless of whether you are collecting data from FxM and FxV.

12 Monitoring Application Performance: User and Reference Guide 12 Chapter 2 Using the SOC End User Tab The Foglight Experience Monitor (FxM) and/or Foglight Transaction Recorder (FTR) are installed and configured to capture end user data and the appropriate End User cartridges are installed and configured to send end user data to Foglight and visualize it in the browser interface. You must create a service that includes FTRResult and/or FxMApplicationResult objects. You can create and edit services using the service builders included with the Cartridge for Application Operations (that is, Build Service from Host Dependencies, Update Service from Dependency Data, or Build Services from Service Suggestions). For information about building services, see Creating and Maintaining Services in the Managing Dependency Mapping User Guide. For Foglight cartridge installation instructions, see the Administration and Configuration Help. Important If you choose to manually create a service using the Foglight Service Builder, you must manually identify and select topology objects. For more information, see Manually Building a Service in the Managing Dependency Mapping User Guide. Configuring Trace Analysis for the SOC In order to make the trace analysis functionality in the Cartridge for Application Operations useful, you need to create hit filters in FxV. A hit filter is a collection of match conditions and actions to be performed when a hit matches those conditions. Every hit filter is evaluated against each hit as it enters an FxV Archiver. Hit filters can be used to detect and alert on any per-hit conditions, to mark interesting hits for later searching, and to manage hit storage. They can also be used to define events within transaction filters. Hit filters are helpful because they provide a human context to the captured hits. For example, it is easier to search and alert on Login Attempts using a hit filter than to search and alert by the server, path, and fields used by the Web application. Hit filters are very powerful in their ability to extract custom fields and update metrics. Using regular expressions, hit filters can pull data from fields, headers, cookies, or from page content. In order to ensure that hit filters are coordinated with the FxMApplicationResult, use the FxMApplicationResult name as a filter criterion in FxV. For procedures describing how to configure hit filters in FxV, see Configuring the Hit Analysis Process in the Foglight Experience Viewer User Guide. For more information about trace analysis in the SOC, see Real User Performance Detail View on page 86. Exploring the End User Tab The End User tab displays a set of tiles representing transactional performance collected from FxM and/or FTR. The type and dynamics of transactional data dictate the number and type of tiles that appear on this tab, along with the Display Options you can configure (see Configuring Display Options on page 16).

13 Monitoring Application Performance: User and Reference Guide 13 Chapter 2 Using the SOC End User Tab End user transaction groups are divided into: Real Users (FxM and FxV) Synthetics (FTR) There are a number of options for controlling how end user transactions are grouped. For more information, see Configuring Display Options on page 16. Note You must add a service containing end user objects (FTRResults and/or FxMApplicationResults) to the SOC in order for the End User tab to be visible. To access the End User tab: 1 Log in to the Foglight browser interface. 2 On the navigation panel, under Homes, click Service Operations Console. 3 On the Service Operations Console, click the End User tab.

14 Monitoring Application Performance: User and Reference Guide 14 Chapter 2 Using the SOC End User Tab Real Users When a service includes an FxMApplicationResults model object, it appears on the Real User Tile, providing an at-a-glance view of that transaction s performance. This tile displays the total number of active end users interacting with the monitored application, transaction front-end and back-end response times, and other metrics. These metrics are collected by FxM. The Trace Analysis health status icon, located in the bottom right corner, shows that trace analysis metrics are tracking to their historical norms. Deviations from those norms indicate an anomaly that should be investigated. For more information about the metrics appearing on this tile, see Real User Tile on page 77. Drill down on this tile (click the title bar) to open the Real User Performance Detail View. For more information about the metrics appearing on this view, see page 86. The tabs available in this view depend on the monitoring capabilities of your Foglight installation. For example:

15 Monitoring Application Performance: User and Reference Guide 15 Chapter 2 Using the SOC End User Tab The Response Time Breakdown Tab requires the Cartridge for Java EE Technologies version or later. For more information, see Requirements for Using the Response Time Explorer Dashboard on page 28. The SLA Violators Tab requires FxM and a deployed and configured FxM Map Agent. For more information, see Requirements for Using the Geographical Perspective Dashboard on page 18. Synthetics When a service includes an FTRResults object, the End User tab displays one or more Synthetics tiles. These metrics are collected by Foglight Transaction Recorder (FTR). Depending on the number of tiles and the configured Display Options (see Configuring Display Options on page 16), a collection of Single-location Synthetic Layout or Multi-location Synthetics Layout tiles appear. Single-location Synthetic Layout The Single-Location Synthetics Tile shows summarized data about a synthetic transaction executed from a single location. This type of layout shows the number of script executions and their execution times over the monitored period, along with other metrics. For more information about the metrics appearing on this tile, see page 81. Drill down on this tile (click the title bar) to open the Synthetic Result Detail View. For more information about the metrics appearing on this view, see page 90. Multi-location Synthetics Layout. The Multi-Location Synthetics Tile shows summarized data about a synthetic transaction executed from multiple locations. This type of layout shows the total number of script execution locations and the numbers of locations for each alarm state (Normal, Warning, Critical, and Fatal). For more information about the metrics appearing on this tile, see page 80.

16 Monitoring Application Performance: User and Reference Guide 16 Chapter 2 Using the SOC End User Tab Drill down on this tile (click the title bar) to open the Synthetic User Performance Detail View. For more information about the metrics appearing on this view, see page 91. Drilling down on an individual location displays the Synthetic Result Detail View. For more information about the metrics appearing on this view, see page 90. Configuring Display Options You can customize the layout of the End User tab by selecting the appropriate options in the Tile Display Options dialog box. Use these options to control the total number of tiles, the order in which they appear, and the roll up options for multiple FTR locations. Note These settings apply on a per-user basis. To configure display options: 1 On the Service Operations Console, on the End User tab, click Display Options. The Tile Display Options dialog box opens.

17 Monitoring Application Performance: User and Reference Guide 17 Chapter 2 Using the SOC End User Tab 2 Configure these options, as required. The Enable Grouping option groups user tiles together by transaction name. By default, groups are sorted by their severity first, then by their name. This grouping is not affected by the Sort Order options. The Show at most n tiles or groups setting determines the total number of tiles displayed. Synthetic transaction locations are first rolled up using the rollup criteria below. After that, the total number of tiles is truncated based on the maximum number of tiles that you specify. You can roll up synthetic transactions locations into multi-location FTR (Synthetics) tiles based on these criteria, in the following order: the maximum number of single-location Synthetics tiles, using the Rollup if there are more than n FTR tiles setting, and then the maximum number of locations per script, using the When there are more than n location(s) for a script setting. Use the Sort Order options to sort the tiles by either type (Real End Users first) or alarm severity (Sort by Alarm State). For example, you can configure the End User tab to display a maximum of ten tiles and sort them by the alarm state (highest severity first). 3 Click Save. The Tile Display Options dialog box closes and the End User tab refreshes, reflecting the newly updated display options.

18 3 Understanding Geographical Perspectives The Geographical Perspective dashboard displays the real end user activity for the top 100 locations for a selected application component based on performance problems and errors, total traffic, response time, and other factors. It provides a list of performance-related measurements (such as, Show locations with the most: Hits or Show locations with the highest: Hit Response Times ) that help you quickly and easily find patterns and problems that have geographical features. Requirements for Using the Geographical Perspective Dashboard Foglight Experience Monitor (FxM) Appliance FxM Agent configured and actively collecting data Foglight Management Server, version or later Cartridge for FxM and End-User Core Cartridge installed on the Management Server Cartridge for Application Operations, version 5.7 or later FxM Map Agent deployed and configured

19 Foglight: Monitoring Application Performance Guide 19 Chapter 3 Understanding Geographical Perspectives Creating a Map Agent To visualize real end user activity on the map, you must first configure a map agent. A map agent monitors a single FxM Appliance. It must be paired with an existing Foglight Real User Monitoring for Web agent to import geographical information into Foglight. You can only deploy one map agent per Foglight Real User Monitoring for Web agent instance. To create a map agent: 1 On the navigation panel, under Dashboards, click End User > Geographical Perspective. 2 The Geographical Perspective for ApplicationName dashboard opens in the display area. 3 Select an FxM Application from the FxM Application Result Table located at the bottom of the navigation panel. 4 On the Geographical Perspective dashboard, click Create a map agent to import geographic data. The Map Agent Setup wizard opens. The map agent requires an existing FxM agent that is active and collecting data from a Foglight Experience Monitor (FxM) Appliance.

20 Foglight: Monitoring Application Performance Guide 20 Chapter 3 Understanding Geographical Perspectives Important If your FxM Appliance is at or near capacity, or if it is experiencing packet drops, it is recommended that you do not deploy the map agent. The map agent causes a small but measurable impact on the Appliance. 5 Verify that your FxM Appliance and agent are in a healthy state and click Next. 6 On the Advanced Agent Settings page, specify the agent settings. a Select the granularity levels at which you want the map agent to import data. You can select any combination of: Hourly, Daily, Weekly, or Monthly. By default, only Hourly and Daily are selected.

21 Foglight: Monitoring Application Performance Guide 21 Chapter 3 Understanding Geographical Perspectives b Specify the location where the map agent is deployed. By default, this is the same Agent Manager that is hosting the FxM agent. Select an alternate location from the drop-down list if necessary. 7 Click Finish. 8 On the Geographical Perspectives for ApplicationName dashboard, click Import geographical data to import the data for that application from FxM to Foglight. A message box opens. 9 Review the health of your Appliance. Note If the load on your FxM Appliance is too high due to other geographic data collections that are currently running, you can deactivate those collections to reduce the load. Click the text link: Click here to deactivate or reactivate geographical data import for other applications. 10 Click OK. The map agent imports geographical data for the selected application. The dashboard refreshes, displaying the relevant end user activity. Note This process may take some time, depending on the amount of data being imported. If there is no data available for your selected time period, the aggregation interval and timeslice fields will appear empty. If necessary, you can edit the map agent properties through the Agent Status dashboard. For more information, see Editing Map Agent Properties on page 25.

22 Foglight: Monitoring Application Performance Guide 22 Chapter 3 Understanding Geographical Perspectives Viewing Real End User Activity The Geographical Perspective dashboard is divided into three panes. The left-hand pane contains lists of: available aggregation intervals, timeslices, and measurements. The center pane displays either a map, if the Map tab is selected, or a list of results in a tabular format, if the List tab is selected. The right-hand pane provides graphical results from the measurements, and the option to drill down for more information. Note If you are using a version Management Server, only the List tab is available. The Map tab requires Management Server or later. If you are using Internet Explorer (IE) 8 or earlier, only the List tab is available. To view real end user activity in a geographical perspective: 1 On the navigation panel, under Dashboards, click End User > Geographical Perspective. 2 Select an FxM application component from the FxM Application Result Table at the bottom of the navigation panel. 3 Select a Measurement from the list to view the related transaction data. The map and right pane update based on your selection. For example: select Show locations with the highest: Page Response Times. The map updates to indicate the locations that are experiencing the highest average page response times in the selected timeslice. The right pane displays a chart of the page processing time for the entire application (all locations), and the top five locations are listed in a table below the graph and timeslice selector. Double-click anywhere on the map to zoom in on that location. You can also use the zoom tool in the map pane to zoom in or out. You can move the map in any direction by clicking and holding the left mouse button and dragging the cursor.

23 Foglight: Monitoring Application Performance Guide 23 Chapter 3 Understanding Geographical Perspectives The fill-level of each location indicator on the map shows how close that location is to the maximum value in the current timeslice. In this example, Georgia has the highest response time, so its location indicator appears full. 4 Select a timeslice from the Timeslice list, or use the forward or back arrow buttons to step through the available timeslices. Tip The vertical blue bar on the graph indicates the currently selected timeslice. 5 Hover your cursor over a location indicator to open a quick view of metrics for that location. 6 Click an individual location indicator to view a graph of the average page response time for that location in the right pane. The Similar Locations list in the right pane updates to show locations that are ranked near the selected location in the current category.

24 Foglight: Monitoring Application Performance Guide 24 Chapter 3 Understanding Geographical Perspectives that are also exhibiting the performance issues that the selected location is experiencing. Click any location name in the list to view the graph for that location. 7 Click Location Detail to drill down to detailed information for the selected location. The Location Detail view provides a table of measurements in context that helps you understand how the selected location is behaving relative to other locations during the selected timeslice and aggregation period. Measurements for the selected location are sorted in order of severity. The value in the Highest Rank column indicates the highest ranking that the selected location attained at any point during the selected time range. The time range may span multiple timeslices. For example, at some time during the selected time range, New York had the highest page response time of all locations. Therefore, for the measurement Page Response Time, New York s Highest Rank is 1 (first). The value in the Current Rank column indicates the selected location s current ranking for the selected timeslice. This may or may not be the same as its highest rank, depending on when the peak value for the location occurred. In this example, New York is ranked second for the Current Rank for Page Response Time. That is, during the currently selected timeslice, New York has the second highest page response time. 8 Click any location s name in either the Similar Locations list or the Top Ranked Location column to open a detail view for that location for comparison. 9 To view the transaction data for the selected location in the FxM Appliance, click More Details.

25 Foglight: Monitoring Application Performance Guide 25 Chapter 3 Understanding Geographical Perspectives Editing Map Agent Properties After you have created a map agent, as described in Creating a Map Agent on page 19, you can edit its properties from the Agent Status dashboard. For example, you may want to change the available aggregation granularity, or change the scheduled collection interval. To edit map agent properties: 1 On the navigation panel, under Dashboards, click Administration > Agents > Agent Status. 2 Select the map agent in the list and click Edit Properties. Tip Map agents have the type MapFxMAgent and, by default, their names begin with Geo@. 3 Click Modify the private properties for this agent. The settings become editable. 4 Ensure that the settings for the FxM Appliance database IP address and password match those that the FxM agent is using to access the database. 5 The Associated FxM Agent Name must match the name of the FxM agent exactly. If this box is blank, or if the value does not match that of the FxM agent, the map agent cannot access the FxM data to populate the geographical perspectives. 6 The Locations Per Measurement setting determines the number of locations from which real end user activity is collected for display on the Geographical Perspective dashboard. By default, this value is 100.

26 Foglight: Monitoring Application Performance Guide 26 Chapter 3 Understanding Geographical Perspectives 7 The Maximum SQL Retries setting controls the number of times the map agent tries to reconnect to an appliance that is unreachable before stopping data collection. The default is twelve attempts. 8 The five True/False settings allow you to specify the granularity of the data aggregation. Click True for each of the FxM reporting intervals you want to make available: five minutes, hourly, daily, weekly, or monthly. By default, the five minute, weekly, and monthly intervals are disabled (set to False). Important If you have a large number of application components, using the five minute interval can cause a heavy load on the FxM Appliance, negatively impacting performance. 9 The FxM Application Collection list controls which of the available application components the map agent collects data from. To edit the default list, click Edit. Select or clear the check boxes for the applications and click Save Changes. or To create a new list without affecting the default list, click Clone. Type a name for the cloned list and click OK. To edit the cloned list, click Edit. Select or clear the check boxes for the applications and click Save Changes. 10 The Data Collection Scheduler setting controls how frequently the map agent checks the FxM Appliance for data. By default, the collection interval is fifteen (15) minutes. Note The agent imports all data for the selected granularities up to one hour old regardless of the collection interval. To edit the default schedule, click Edit. To create a new schedule without affecting the default schedule, click Clone. 11 When you are finished setting the agent properties, click Save. Generating Geographical Perspective Reports There are a number of reports specific to geographic perspectives. For example, you can generate reports on top locations by errors and violations, by response time, or by user volume. To access these reports: Click Reports at the top right corner of the Geographical Perspectives dashboard.

27 Foglight: Monitoring Application Performance Guide 27 Chapter 3 Understanding Geographical Perspectives The report wizard prompts you to specify the input parameters (such as time range, application, and granularity) and properties (such as report name and export format). You can also schedule the report to run at regular intervals. For more information about a specific report, run the report wizard from the Geographical Perspectives dashboard.

28 4 Examining Response Times The Response Time Explorer dashboard displays the sampled requests with the longest execution times for a selected end user session during a selected timeslice. The requests are listed by execution time. Use this dashboard to quickly see where request bottlenecks are occurring, and where in the application infrastructure the time is being spent. With the Timeslice navigator, you can view the historical average response times in five minute intervals. Requirements for Using the Response Time Explorer Dashboard Foglight Management Server, version or later, with the following cartridges installed and configured: Cartridge for Application Operations, version or later Cartridge for Java EE Technologies, version or later

29 Foglight: Monitoring Application Performance Guide 29 Chapter 4 Examining Response Times Cartridge for FxM, version or later Cartridge for FxV, version or later End-User Core Cartridge, version or later Foglight Experience Monitor (FxM) Appliance Foglight Experience Viewer (FxV) Appliance, version or later Configuring Request Traces The Cartridge for Java EE Technologies provides the ability to trace requests. You must configure request tracing through the Java EE Agent recording.config property file. For detailed information about configuring request tracing, see the Managing Java EE Systems: Installation and Configuration Guide. In particular, you must enable the IndividualRequestBreakdowns option in order for the request details to be matched to the appropriate end user results. To enable IndividualRequestBreakdowns: 1 On the navigation panel, click Administration > Agents > Agent Properties. 2 In the Namespace > Type column, expand the JavaEE module (click the + sign). 3 Select recording.config. 4 Locate the section that begins with: # Rules for the collection of details on sampled requests. 5 Configure the request detail rules to include IndividualRequestBreakdowns. For example: RequestDetailRules = { /.*medrec.*/ : ("IndividualRequestBreakdowns"), }; RequestDetailDefault = ("IndividualRequestBreakdowns"); 6 Click Save. Associating FxV Hit Filter Mappings In order to present a comprehensive view of requests as they relate to end user results, you must create a link from the FxV hit filter to the Java request and to the FxM (end user) application result. For detailed information about configuring FxV hit filters and FxM application results, see the documentation included with the Foglight Experience Viewer and Foglight Experience Monitor Appliances respectively. The response time breakdown only applies to the specific requests that pass through an instrumented application server. The Front End/Back End time breakdown shown on the Real User Tile applies to complex multi-hit page objects that match the FxM filter. If there is a substantial difference in displayed response time between the Real User Tile and the response time breakdown, the difference is attributable to non-java page elements.

30 Foglight: Monitoring Application Performance Guide 30 Chapter 4 Examining Response Times By default, if no mapping is provided and an existing hit filter has the exact same name as the transaction, that filter is used. The application result is mapped automatically. To associate a hit filter to an application result: 1 On the Service Operations Console (SOC), click the title bar of the Real User Tile of the application result you want to work with. 2 The Real User Performance Detail View opens, with the Response Time Breakdown tab displayed. 3 Click Associate Hit Filter Mappings. The Associate FxV Hit Filter Mappings dialog box opens. 4 Click Add. The Associate FxV Hit Filter Wizard opens. 5 Select the hit filter that matches the application result. 6 Optional click Next to configure proxy objects. For more information about assigning proxy objects, see Assigning Proxy Objects on page 35.

31 Foglight: Monitoring Application Performance Guide 31 Chapter 4 Examining Response Times 7 Click Finish. The associated hit filter appears in the list in the Associate Hit Filter Mappings dialog box. Exploring Response Time Breakdowns The Real User Performance Detail View provides an at-a-glance overview of response time breakdowns for a selected real user transaction group. To view the Response Time Breakdowns: 1 On the Service Operations Console (SOC), click the title bar of any real user transaction group that has an associated hit filter. Tip Real user transaction groups that are configured for response time breakdown views are indicated with the icon at right: The Real User Performance Detail View opens. 2 Click the Response Time Breakdown tab. Here you can quickly check the execution time for a request through the various components that are included in the transaction group. Note The execution times in this view are aggregated for the current timeslice. The upper portion of the tab provides a list of components with average execution times for the current timeslice and a sparkline that visualizes the average execution time through that component for the selected time range.

32 Foglight: Monitoring Application Performance Guide 32 Chapter 4 Examining Response Times The lower portion of the tab provides the same average execution time information in a dependency path view, with the average execution time through each component displayed below the icon for the component. The dark grey arrows indicate the path taken by at least one request in the current timeslice (for example, between 3:20 and 3:25pm). The light grey arrows indicate the path taken by at least one request in the selected time range (for example, in the last four hours). For example, in the previous image, you can see that ms were spent on MedRec1Server1, and only 79.9 ms were spent on MedRec1Server2. The breakdown information makes it clear that MedRec1Server1 is causing a bottleneck for requests. Click a component icon to view a chart of the average historical response time for that component based on the selected time range. This chart provides a comparison of the average execution time for the selected component (in blue), overlaid on the total average execution time for the transaction (in grey). The vertical blue bar on the chart indicates the current timeslice. Tip If you click a datasource component icon, the details view also includes a Map to Application Component button. For more information, see Mapping Datasources on page 35. Click More Details to drill down to a detailed view of the selected application component. For example, from an application server component, you can access the Java server details view.

33 Foglight: Monitoring Application Performance Guide 33 Chapter 4 Examining Response Times Viewing the Response Time Breakdown Explorer Click Explore Breakdowns on the Real User Performance detail view to drill down for more information. The Response Time Breakdown Explorer dashboard opens. This dashboard displays the top sampled requests (that is, those with the longest execution time) for a selected transaction (a mapped FxMApplicationResult) during a selected timeslice. The requests are listed by execution time, along with an indication of where in the path the bottleneck occurred, and the time the request was executed. Execution Time History The charts on the left-hand side provide a comparison of the average execution time for the component (in blue), overlaid on the total average execution time for the transaction (in grey). The vertical blue bar on the chart indicates the current timeslice. Use the Timeslice navigator to step through the selected time range (for example, four hours) in five minute intervals. To view the execution times for a different transaction: Open the navigation pane and select a transaction from the Breakdown for Applications list.

34 Foglight: Monitoring Application Performance Guide 34 Chapter 4 Examining Response Times Top Sampled Requests by Execution Time The right-hand side of the dashboard provides a list of the sampled requests with the top (longest) execution times, and a dependency path view that allows you to trace the request through the components. Select a request from the list to update the dependency path view. Blue arrows mark the path of the currently selected request. For example, in the image above, the selected request (medrec/admin/viewnewlyreigsteredpatients.action) begins at the end user, passes through the medrecweb proxy object, and reaches the MedRec1Server2 component. It does not reach the database. The dark grey arrows indicate the path taken by at least one request in the current timeslice (for example, between 3:20 and 3:25pm). The light grey arrows indicate the path taken by at least one request in the selected time range (for example, in the last four hours). Click a component s host name to drill down for more information about that host. From the Response Time Breakdown Explorer you can: View Transaction Replay Opens the hit and session that match the selected request in the FxV Appliance. Here you can replay the content of the hit as the affected end user saw it to determine exactly where an issue occurred. For more information about the Hit Inspector, see the Foglight Experience Viewer User Guide. Request Details Opens the Java EE Requests view, which allows you to review the details of the request, including calls completed and incomplete, response time current, average, and maximum, and exceptional exits. You can also collect traces from this view (click the Action icon for the request and select Collect Traces). For more information, see Monitoring Requests in the Monitoring Custom Applications User and Reference Guide. Examine Java Trace If you have collected traces for a request, you can drill down to the trace details.

35 Foglight: Monitoring Application Performance Guide 35 Chapter 4 Examining Response Times For more information about the Traces View, see Managing Traces in the Monitoring Custom Applications User and Reference Guide. Additional Configuration Options There are additional configuration options for exploring response time breakdowns that may be useful in some instances. For example, if you are associating hit filter mappings to a transaction that contains a proxy object, review the topic Assigning Proxy Objects. If your dependency path view contains a datasource that needs to be mapped to a specific component, review the topic Mapping Datasources. Assigning Proxy Objects In some cases, requests may pass through one (or more) proxy objects, such as a firewall, load balancer, or router, before they reach the application server tier. If you know that this is the case in a transaction group for which you want to view response times and map a hit filter, you can create a proxy path. To select a proxy path: 1 In the Real User Performance Detail View, click Associate Hit Filter Mappings. Note If the transaction is mapped to the default hit filter, you must first remove the default hit filter, then add it manually in order to assign proxy objects. 2 In the Associate Hit Filter Mappings dialog box, click Add. 3 Select the FxV Hit Filter from the list. Click Next. 4 On the Proxy Path Selection page, select the option that accurately reflects the transaction path you are trying to map: No hosts, One host, or Multiple hosts between the transaction and application tier. 5 Click Next. 6 Select the proxy object from the list. Tip 7 Click Next. If you know the name of the object, you can use the search box to quickly locate it. 8 On the Summary page, click Finish. 9 The list in the Associate Hit Filter Mappings dialog box updates. Close the dialog box. Mapping Datasources By default, datasources appear on the Response Time Breakdown tab with limited information. If you know the proper type of a datasource (for example, if it is an Oracle database), you can map the datasource to a specific application component. A correct mapping provides access to drill down information, such as host or database metrics.