立即与支持人员聊天
与支持团队交流

Foglight for Application Operations 5.9.8.5 - User Guide

Introducing Foglight APM Monitoring Transactions Examining Response Times Examining Error Breakdowns Viewing Real User Activity from a Geographical Perspective Using the SOC Transactions Tab Using the SOC for APM triage Creating Custom Drag-and-Drop Dashboards APM Tile and View Reference Appendix: Enabling End User Transactions from FxM and FxV

Investigating response time breakdowns

To view the Response Time Breakdown Explorer:
•
On the navigation panel, under Dashboards, click APM > Response Time Breakdown.

The Response Time Breakdown Explorer opens.

Figure 8. Response Time Breakdown Explorer dashboard

This dashboard displays the top sampled requests (that is, those with the longest execution time) for a selected transaction 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.

The timeslice is the collection period during which the currently selected request took place. It is highlighted in the graph on the right, and displayed in the lower left corner of the dashboard.

To view the execution times for a different transaction:
•
Open the navigation panel and select a transaction from the Breakdown for Transactions list.
Figure 9. An example of the transaction list in the navigation panel

For more information, see the following topics:

•
Sampled requests
•
Execution Time History
•
Next steps

Sampled requests

The Response Time Breakdown dashboard is divided into three sections: a list of the sampled requests, a dependency path view that allows you to trace the request through the components, and charts of the historical execution times. Select a request from the list to update the dependency path view.

In the dependency path view:

•
Blue arrows mark the path of the currently selected request. For example, in the previous dashboard image, the selected request (physician-web/physician/viewPatients.action) begins at the end user (Front End Time), then passes through a web server and the MedRec1Server2 component before reaching the database.
•
Dark grey arrows indicate the path taken by at least one request in the current timeslice (for example, between 10:45 and 10:50pm).
Select a trace from the list to select the timeslice.
Figure 10. The timeslice appears in the lower left corner of the dashboard
•
Light grey arrows indicate the path taken by at least one request in the selected time range (for example, in the last four hours).
Response Time Breakdown display options

You can change the number of requests shown, and control whether request breakdowns that include trace data are preferred (that is, shown at the top of the list).

To change the display options:
1
Open the action panel on the right side of the display area.
2
In the Actions section of the General tab, click Display Options.
The Display Options dialog box opens.
Figure 11. Display Options dialog box
3
Type a value in the Show top samples by execution time box. The default is to show the top 1000 sampled requests. Set this value to zero to show all available requests.
4
By default the Prefer requests with traces check box is cleared. Selecting it means that requests in the time range for which traces were captured appear in the list even if there are requests without traces that have longer execution times.
The same limit on the number of sampled requests shown applies. Changing the preferred requests selection does not affect the sort order of the table.
5
Click OK to save your changes.
The Response Time Breakdown dashboard refreshes.

Execution Time History

The charts on the right 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).

Figure 12. The vertical blue bar on the chart indicates the current timeslice

Next steps

•
To drill down for more information about a component, click the component icon in the dependency path view
•
If you are using Foglight APM Appliances, click Find Related Hits to drill down to the Search Hits page, where you can investigate pivots or map the hit. From there you can also replay the content of the hit as the affected end user saw it to determine exactly where an issue occurred. For more information about searching hits, see the Foglight APM User Guide.
or
If you are using FxV, click View Transaction Replay to open the hit and session that match the selected request. From there 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.
•
To review the details of the request, including calls completed and incomplete, response time current, average, and maximum, and exceptional exits, click Request Details. The Java Requests view opens.
 
* 
TIP: 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 Foglight for Application Servers User Guide.
•
To collect traces for a selected request, click Collect Traces. In the Collect Traces dialog box, specify the number of traces to collect, the length of the collection period (in seconds), and whether to collect at the Full Detail instrumentation level with SQL Parameters. If you do not want Full Detail or SQL Parameters, clear the appropriate check box.
 
* 
TIP: For more information about instrumentation levels, see the Foglight for Java EE Technologies Installation Guide or help, and search for “Full detail instrumentation level”.
•
If you have collected traces for a request, click Examine Trace to drill down to the trace details. For more information about the Traces View, see Managing traces in the Foglight for Application Servers User Guide.
相关文档

The document was helpful.

选择评级

I easily found the information I needed.

选择评级