Chatta subito con l'assistenza
Chat con il supporto

Foglight for Java EE Technologies 5.9.7.5 - Application Servers User Guide

Monitoring Application Servers Monitoring Systems Monitoring Servers Monitoring Applications Monitoring Requests Managing Traces Using Object Tracking to Locate a Memory Leak Monitoring Methods Application Servers Monitor Views
JVM view Method Groups view Request Types view Entity EJBs view Message Driven EJBs view Stateful Session EJBs view Stateless Session EJBs view Deployed Applications view JSPs/Servlets components view Resource Adapters components view Web Applications components view Web Services components view .NET views JBoss Services views Oracle Services views Tomcat Services views WebLogic Services views WebSphere Services views JMX Administration dashboard JMX Explorer dashboard
Appendix: Regular Expressions

Using a rule action to collect single traces

As an alternative to manually collecting single traces using the interface, you can add a RequestTracesAction to any rule to collect single traces to assist during issue diagnostics.
 
* 
NOTE: This action was called JavaEETracesAction in previous versions.
For example, you have a request type which is experiencing high response times and you have a rule that sends alarms when it reaches the threshold to inform you of this problem. You then specify the RequestTracesAction to collect single traces on a specific request type to monitor the issue.
To collect single traces using a rule action:
1
On the navigation panel, under Dashboards, click Administration > Rules & Notifications > Rule Management.
2
Click the name of a rule in the Rule Management table.
 
* 
TIP: To shorten the list of rules, select the cartridge that contains the rule from the Cartridge list (for example, ApplicationServers-Java), or if you know the name of the rule, use the Search box.
3
From the menu that opens, select View and Edit.
4
In the Rule Detail dialog box that opens, click Rule Editor.
Figure 25. Rule detail dialog box
The Edit Rule Page opens.
5
Click the Conditions, Alarms & Actions tab.
6
Select a severity level (for example, Warning) and click the Action tab.
7
From the Action menu, select RequestTracesAction.
 
* 
IMPORTANT: This action was called JavaEETracesAction in previous versions.
8
Click Add to add the RequestTracesAction.
The default action parameters are listed, as follows:
•
Request Type — Use this action parameter to specify unique request types.
 
* 
NOTE: The Request Type action parameter is required; all others are optional.
•
Collect Parameters — Use this action parameter to collect all parameters.
•
Collect in full detail — Use this action parameter to indicate whether you want to collect methods in full detail or component detail (the default).
•
Count — Use this action parameter to specify the number of traces being collected.
•
Criteria — Use this action parameter to specify one or more criteria.
Create an expression type Severity Level Variable to specify the criteria. Use any of the arguments listed in the AppServerCollectTraces function table to define the criteria string. For example:
def criteria = [];
criteria.add("responseTimeOver:10000"); // > 10 seconds
//criteria.add("exceptional");
//criteria.add("incomplete");
return criteria;
// comment: response times are in milliseconds
 
•
Duration — Use this action parameter to indicate the time period to collect the number of traces.
•
Is Regular Expression — Use this action parameter if you want to specify the Request Type using regular expressions to match the request type for a single trace collection.
•
Match All criteria — Use this action parameter to limit the trace collection to all or any of the following criteria: The response time of the request is over a specified threshold, The response time of the request is under a specified threshold, The request exits due to an exception, The request is incomplete, or a parameter of the request is set to a specific value.
•
The agent name IDs — Use this action parameter to specify the agentNameIDs property of the agent that you want to collect traces on.
Create an expression type Severity Level Variable to specify the agent. For example, to use a variable to specify an agent in Groovy:
def agentNameID = [];
agentNameID.add(scope.agent.agentNameID);
return agentNameID;
 
•
Track SQL Parameters — Use this action parameter to track SQL bind variable values. The default value is set to Off.
9
In the Value column for Request Type, click Default and specify the value.
Click Change to update the action parameter.
10
Once the action parameters have been updated, click Save All.
For more information about rules, see Getting started: view and edit rule definitions in the Foglight Administration and Configuration Help.

Using Groovy to collect single traces

You can also use a Groovy script to collect single traces. In Groovy, request types are specified by a regular expression, so the same set of single trace criteria can be used for a group of request types.
When you have a group of request types for which you want to collect single traces using the same criteria, you set the criteria using a Groovy script called from the command line (for example, using fglcmd). The collected single traces are stored on the Foglight Management Server, and displayed in the Application Servers Monitor > Requests view for users to review. Use the following function to trigger the collection of requested traces.
AppServerCollectTraces function
 
* 
NOTE: This function was called JEECollectTraces in previous versions. Any existing Groovy scripts using JEECollectTraces continue to work. You do not need to update them.
This function is used to collect traces in Groovy. Set all of the following arguments:
 
Table 16. Required arguments for the AppServerCollectTraces Groovy function.
Argument
Description
requestType: String
Request type to collect trace.
isRegEx: Boolean
Request type for which to collect a trace defined as regular expression (true/false).
tracePeriod: Long
Duration (in seconds) of the request to trace.
count:Integer
The number of traces to collect.
CollectInFullDetail:Boolean
Collect in Full Detail (true/false).
For details about instrumentation levels, see Appendix: Application Methods in the Foglight for Java EE Technologies Installation Guide.
trackSQLParams:Boolean
Track SQL Parameters (true/false).
matchALL:Boolean
Match all the criteria (true/false).
criteria:List
A list of criteria. See the next table for details.
CollectParameters:Boolean
Collect all trace parameters (true/false).
agents:List
List of agents to collect traces on. If the list is empty or null, then traces will be collected on all agents.
Criteria are specified as a list of one or more strings.
 
Table 17. Criteria for the AppServerCollectTraces Groovy function
Format
Example
all
all
none
none
responseTimeOver:t
ReponseTimeOver:600001
responseTimeUnder:t
ResponseTimeUnder:10001
exceptional
exceptional
incomplete
incomplete
traceParameterContains:k:v
traceParameterContains:name:Bob
traceParameterDoesNotContain:k:v
traceParameterDoesNotContain:name:Tom
traceParameterIs:k:v
traceParameterIs:user:Bob
traceParameterIsNot:k:v
traceParameterIsNot:user:Tom
traceParameterStartsWith:k:v
traceParameterStartsWith:accountnumber:0
traceParameterEndsWith:k:v
traceParameterEndsWith:accountnumber:9
1 Criteria are in milliseconds.
For example:
// Collect 5 traces in full detail for request "GET /samples/test.do" for 15 minutes
// only if response time is over 1 minute and user is Bob for the agent
def criteria = [];
criteria.add("traceParameterIs:user:Bob");
criteria.add("responseTimeOver:60000");
def agentNameIDs = [];
agentNameIDs.add(appServer.monitoringAgent.agentNameID);
// appServer is the server topology object
AppServerCollectTraces("GET /samples/test.do", false, 900, 5, false, true, true, criteria, true, agentNameIDs);
 

Viewing traces

After you have configured trace collection, you can review the trace details through the Traces view.
To view single trace details:
1
On the navigation panel, under Dashboards, click Application Servers > Monitor.
2
On the Application Servers Monitor dashboard, click the Requests tile.
3
In the Traces column of the Request Types table, click the number of traces for the request.
Figure 26. The number of traces appears in the Traces column.
Figure 27. The Traces view opens, displaying both Aggregated Traces and Single Traces.
4
Optional — To view aggregated trace details, click a server name in the Aggregated Traces table, or to view single trace details, click a server name in the Single Traces table. The Trace Diagnosis view opens.

Traces view

Use the Traces view to review individual traces of a selected request. This view is divided into four sections: Aggregated Traces table, Single Traces table, Component Technology Breakdown, and Performance charts.
From the Traces view, you can set a breakdown of execution time by selecting Tiers and Component Technologies. By setting any of these options, a selected single trace or an aggregate trace can be used as a comparison benchmark for all the traced metrics. Once specified, both the list view and the tree view display the comparison in the Average Execution Time and Average Exclusive Time columns, and the rest of the metrics comparisons are shown at the bottom of the page, in the method details area.
To learn more about how to set a breakdown, see Viewing breakdowns .
Related Documents

The document was helpful.

Seleziona valutazione

I easily found the information I needed.

Seleziona valutazione