지금 지원 담당자와 채팅
지원 담당자와 채팅

Foglight APM for Real User Experience 5.9.9 - User Guide

Getting started with Foglight APM Monitoring transactions Viewing real user activity from a geographical perspective Assessing real user experiences during key sequences Monitoring the performance of web sites and endpoints Searching APM data Replaying sessions, hits, and sequences Visualizing search results Creating custom drag-and-drop dashboards APM tile and view reference

Getting started with Foglight APM

You can start assessing your real user experiences of your monitored web sites and web applications using any of the APM dashboards. Depending on your job function, however, you may be more interested in some results than others. The following table lists common getting started tasks, the dashboard to use, and the people who primarily use the dashboard.
 
* 
TIP: If you are not seeing data for a web site or web application, ask the person with the Application Performance Manager function in your organization to consider monitoring the site/application.
 
Table 1. Common “getting started” tasks
Task
APM dashboard
Job functions
For information, see
Monitor real user activity on your sites and applications. Investigate alarms and any tiles with a non-normal status.
Transactions
•
Application Support
•
Platform Specialist or Administrator
 
Monitoring transactions
Find out the locations where your site or application performs well/poorly or experiences high/low volume.
NOTE: Displays data only when an APM Administrator enables mapping on selected hit analyzers.
Geographical Perspective
•
Application Support
•
Application Architect
 
Viewing real user activity from a geographical perspective
Assess the health of your sites and applications as well as all the monitored endpoints (server + port). Investigate alarms and any tiles with a non-normal status.
Web Sites and Endpoints
•
Application Support
•
Developers
•
Network Administrators
Monitoring the performance of web sites and endpoints
If a real user reports having issues with your web site, you can search for the session and replay it to see what your user experienced.
Search
•
Application Support
•
Application Architect
 
Searching APM data
Monitor critical paths in your sites and applications.
NOTE: Displays data only when an APM Administrator defines sequence analyzers to identify the series of events to track.
Sequences
•
Any
Assessing real user experiences during key sequences

Monitoring transactions

A transaction occurs when a user or synthetic robot makes a request to a monitored application. Depending on the components in your installation, Foglight APM can capture and correlate transactional activity at multiple layers including:
•
Real/synthetic user behavior
•
Web requests and responses
•
Web servers and application servers
•
Application code execution
•
Database I/O
•
Operating systems
•
Physical and virtual servers
With the Transactions dashboard, you can examine transactional data in the following ways:
•
Fully-linked end-user to database transactions
•
End-user transactions only (synthetic or real) aggregated by location or other details
•
Application layer transactions only (Java™ EE or .NET®)
The tiles on the Transactions dashboard represent different transactional views. Use the table below to quickly determine the type of transaction, then review Investigating real user APM transactions for an overview of how to drill down for more information.
Transaction type icons
 
Table 2. Transaction type icons and links to full descriptions
Icon
Transaction Type
Foglight APM for Real User Experience transaction.
For more information, see Real User (APM) Performance detail view.
Foglight Experience Monitor/Viewer End User transaction.
For more information, see Real User (FxM) Performance detail view.
Foglight for Java EE Technologies or Foglight for Microsoft .NET transaction.
For more information, see Application Server detail view.
Foglight Transaction Recorder Synthetic User transaction.
For more information, see Synthetic Result detail view.
Sequence.
For more information, see Sequence tile.
Pivot - Browsers.
For more information, see Pivot tiles.
Pivot - Content Types.
For more information, see Pivot tiles.
Pivot - Endpoints.
For more information, see Pivot tiles.
Pivot - Locations.
For more information, see Pivot tiles.
Pivot - OSes.
For more information, see Pivot tiles.
For example, in the following image, the Physician group of tiles represents a real user perspective with pivot data. The performance summary on each tile includes an alarm state overview and other key metrics. For a full description of the metrics displayed on each type of tile, see APM tile and view reference.
 
* 
NOTE: If some of the transactional data that you want to investigate is not available in your installation, you may need to install and configure additional Foglight components, such as Foglight for Java EE Technologies.
Figure 2. The Transactions dashboard in Foglight.
For more information, see the following topics:
•
Investigating real user APM transactions
•
Investigating application server transactions
•
Configuring display options

Investigating real user APM transactions

The Transactions dashboard enables you 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.
 
* 
TIP: The Service Operations Console includes a Transaction tab that displays transactions as services. The investigative workflow is similar to that described here.
To investigate a transaction:
1
On the navigation panel, under Dashboards, click APM > Transactions.
2
Click the down arrow in the Show transactions in box and select the transaction group you want to view from the list.
 
* 
TIP: Use the Tile Display Options, located in the action panel on the right, to control how many tiles are displayed and how data is rolled up for viewing. For more information, see Configuring display options.
3
Locate a transaction group that is experiencing errors. For example, users in the Physician transaction group are experiencing poor performance. The health status icon indicates Warning status . The transaction has not yet experienced a complete failure, but some component needs attention.
Figure 3. A transaction group, with the warning status illustrated
 
4
Click the name of the transaction in the title bar of the tile to drill down for more information.
The transaction detail view opens.
 
 
* 
TIP: The detail view, and the tabs available, depend on the type of transaction selected. For example, drilling down on a Real User (APM) tile opens a performance detail view (as shown), while drilling down on a pivot result opens a summary view with tiles for each type of pivot you have defined (see Pivot Breakdown detail view for details).
The Hits tab and Pages tab summarize the pivots available from the hit analyzers configured for the transaction.
5
Click the Pages tab. Here you can see that the Warning status has been triggered by the Service Level Attainment for Back End dropping below an acceptable threshold.
Figure 4. An example of the Pages tab of a transaction detail view
6
Click Find Related Hits to investigate further.
The Search Hits page opens. From here, you can continue your investigation using the Search dashboards.
 
* 
NOTE: The execution times in this view are aggregated for the current timeslice.
For more information about the tiles, detail views, and metrics, see APM tile and view reference.

Investigating application server transactions

The Transactions dashboard allows you to visualize application-specific transaction data (that is, Java™ EE or .NET® transactions) with or without end user data. Use this dashboard to quickly identify if any transactions are having issues, 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.
 
* 
TIP: The Service Operations Console includes a Transaction tab that displays transactions as services. The investigative workflow is similar to that described here.
In order to populate the Transactions dashboard or the SOC Transactions tab with data from Foglight for Java EE Technologies or Foglight for Microsoft .NET (version 5.9.1 or later), ensure that you used the appropriate procedures.
To investigate an application server transaction:
1
On the navigation panel, under Dashboards, click APM > Transactions.
2
Click the down arrow in the Show transactions in box and select the transaction group you want to view from the list.
 
* 
TIP: Use the Tile Display Options, located in the action panel on the right, to control how many tiles are displayed and how data is rolled up for viewing. For more information, see Configuring display options.
3
Locate an application server transaction group.
Figure 5. An example of a bottleneck appearing on a transaction group tile
In this example, MedRec1Server1 is causing a bottleneck. It is performing poorly, as indicated by the warning icon ().
4
Click the name of the transaction (/physician-web/physician (phys...)) in the title bar of the tile to drill down for more information.
The transaction detail view opens.
5
On the Response Time Breakdown tab you can see the execution time for a request as it passes 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 illustrates the average execution time during the selected time range.
•
The lower portion of the tab provides the same average execution time information in a dependency path view. The average execution time for each component is displayed below the icon for the component.
•
Application servers are clustered in the dependency view. The list of components provides a tree view of servers included in the cluster.
•
The dark grey arrows (Paths in the current timeslice) indicate the path taken by at least one request in the current timeslice (that is, the most recent sample period reported by the agent).
•
The light grey arrows (Paths in current time range) indicate the path taken by at least one request in the time range selected with the zonar (for example, the last four hours).
•
A yellow exclamation mark indicates that an error has occurred within the marked component during the current timeslice.
Here you can see that the execution time for the application server MedRec1Server1 is longer than it should be to meet service levels.
6
From here you can:
•
Click Explore Breakdowns to open the Response Time Breakdown Explorer.
•
Click Explore Errors to open the Error Breakdown Explorer.
•
Click the Application Infrastructure tab to investigate the application components. This tab displays a tile for each component. In this example, there is a host (medrecapp1.example.domain) and an application server (MedRec1Server1).
Here you can see that the application server has a warning related to one request.
Figure 6. Application server with a warning related to one request
7
Click the name of the application component in the title bar of the tile to drill down for more information.
For example, click medrecapp1.example.domain to investigate the host performance in the Host detail view. Or click MedRec1Server1 to investigate the requests in the Application Server detail view.
For more information about the tiles, detail views, and metrics, see APM tile and view reference.
관련 문서

The document was helpful.

평가 결과 선택

I easily found the information I needed.

평가 결과 선택