Chapter 18 Monitoring Components and Services

This chapter contains information about monitoring components using the Application Server Admin Console. This chapter contains the following sections:

• About Monitoring

• Enabling and Disabling Monitoring

• Viewing Monitoring Data

• Using JConsole

About Monitoring

• Monitoring in the Application Server

• Overview of Monitoring

• About the Tree Structure of Monitorable Objects

• About Statistics for Monitored Components and Services

Monitoring in the Application Server

Use monitoring to observe the runtime state of various components and services deployed in a server instance of theApplication Server. With the information on the state of runtime components and processes, it is possible to identify performance bottlenecks for tuning purposes, aid capacity planning, predict failures, do root cause analysis in case of failures, and ensure that everything is functioning as expected.

Turning monitoring on reduces performance by increasing overhead.

You can also use management rules to alert you of potential problems with your server, and optionally take action when the server reaches certain performance thresholds. For more information, see Chapter 19, Configuring Management Rules.

Overview of Monitoring

To monitor the Application Server, perform these steps:

1. Enable the monitoring of specific services and components using either the Admin Console or the asadmin tool.

   For more information on this step, refer to Enabling and Disabling Monitoring.

2. View monitoring data for the specified services or components using either the Admin Console or the asadmin tool.

   For more information on this step, refer to Viewing Monitoring Data.

About the Tree Structure of Monitorable Objects

The Application Server uses a tree structure to track monitorable objects. Because the tree of monitoring objects is dynamic, it changes as components are added, updated, or removed in the instance. The root object in the tree is the server instance name, for example, server. (In the Developer profile, just one server instance is permitted.)

The following command displays the top level of the tree:

asadmin> list --user adminuser  --monitor server
server.applications
server.http-service
server.connector-service
server.jms-service
server.jvm
server.orb
server.resources
server.thread-pools

The following sections describe these sub-trees:

• The Applications Tree

• The HTTP Service Tree

• The Resources Tree

• The Connector Service Tree

• The JMS Service Tree

• The ORB Tree

• The Thread Pool Tree

The Applications Tree

The following schematic shows the top and child nodes for the various components of enterprise applications. The nodes at which monitoring statistics are available are marked with an asterisk (*). For more information, refer to EJB Container Statistics.

Example 18–1 Applications Node Tree Structure

applications
   |--- application1
   |    |--- ejb-module-1
   |    |        |--- ejb1 *
   |    |                  |--- cache (for entity/sfsb) *
   |    |                  |--- pool (for slsb/mdb/entity) *
   |    |                  |--- methods
   |    |                       |---method1 *
   |    |                       |---method2 *
   |    |                  |--- stateful-session-store (for sfsb)*
   |    |                  |--- timers (for s1sb/entity/mdb) *
   |    |--- web-module-1
   |    |        |--- virtual-server-1 *
   |    |                       |---servlet1  *
   |    |                       |---servlet2  *
   |--- standalone-web-module-1
   |    |        |----- virtual-server-2 *
   |    |                       |---servlet3 *
   |    |                       |---servlet4 *
   |    |        |----- virtual-server-3 *
   |    |                       |---servlet3 *(same servlet on different vs)
   |    |                       |---servlet5 *
   |--- standalone-ejb-module-1
   |    |        |--- ejb2 *
   |    |                  |--- cache (for entity/sfsb) *
   |    |                  |--- pool (for slsb/mdb/entity) *
   |    |                  |--- methods
   |    |                       |--- method1 *
   |    |                       |--- method2 *
   |--- application2

The HTTP Service Tree

The nodes of the HTTP service are shown in the following schematic. The nodes at which monitoring information is available are marked with an asterisk (*).

Example 18–2 HTTP Service Schematic (Developer Profile Version)

http-service
         |--- virtual-server-1
              |--- http-listener-1 *
              |--- http-listener-2 *
         |--- virtual-server-2
              |--- http-listener-1 *
              |--- http-listener-2 *

Example 18–3 HTTP Service Schematic (Cluster and Enterprise Profile Version)

http-service *
             |---connection-queue *
             |---dns *
             |---file-cache *
             |---keep-alive *
             |---pwc-thread-pool *
             |---virtual-server-1*
             |        |--- request *
             |---virtual-server-2*
             |        |--- request *

The Resources Tree

The resources node holds monitorable attributes for pools such as the JDBC connection pool and connector connection pool. The following schematic shows the top and child nodes for the various resource components. The nodes at which monitoring statistics are available are marked with an asterisk (*). See JDBC Connection Pools Statistics.

Example 18–4 Resources Schematic

resources
        |---connection-pool1(either connector-connection-pool or jdbc)*
        |---connection-pool2(either connector-connection-pool or jdbc)*

The Connector Service Tree

The connector services node holds monitorable attributes for pools such as the connector connection pool. The following schematic shows the top and child nodes for the various connector service components. The nodes at which monitoring statistics are available are marked with an asterisk (*). See JMS/Connector Service Statistics.

Example 18–5 Connector Service Schematic

connector-service
        |--- resource-adapter-1
        |        |-- connection-pools
        |        |        |-- pool-1 (All pool stats for this pool)
        |        |-- work-management (All work mgmt stats for this RA)

The JMS Service Tree

The JMS services node holds monitorable attributes for pools such as the connector connection pool. The following schematic shows the top and child nodes for the various JMS service components. The nodes at which monitoring statistics are available are marked with an asterisk (*).

Example 18–6 JMS Service Schematic

jms-service
        |-- connection-factories [AKA conn. pools in the RA world]
        |        |-- connection-factory-1 (All CF stats for this CF)
        |-- work-management (All work mgmt stats for the MQ-RA)

The ORB Tree

The ORB node holds monitorable attributes for connection managers. The following schematic shows the top and child nodes for the ORB components. The nodes at which monitoring statistics are available are marked with an asterisk (*). See Statistics for Connection Managers in an ORB.

Example 18–7 ORB Schematic

orb
        |--- connection-managers
        |        |--- connection-manager-1 *
        |        |--- connection-manager-1 *

The Thread Pool Tree

The thread pool node holds monitorable attributes for connection managers. The following schematic shows the top and child nodes for the ORB components. The nodes at which monitoring statistics are available are marked with an asterisk (*). See Thread Pools Statistics.

Example 18–8 Thread Pool Schematic

thread-pools
        |    |--- thread-pool-1 *
        |    |--- thread-pool-2 *

About Statistics for Monitored Components and Services

This section describes the monitoring statistics that are available:

• EJB Container Statistics

• Web Container Statistics

• JDBC Connection Pools Statistics

• JMS/Connector Service Statistics

• Statistics for Connection Managers in an ORB

• Thread Pools Statistics

• Transaction Service Statistics

• Java Virtual Machine (JVM) Statistics

• JVM Statistics in Java SE

• Production Web Container (PWC) Statistics

EJB Container Statistics

The EJB container statistics are described in the following tables:

• Table 18–1

• Table 18–2

• Table 18–3

• Table 18–4

• Table 18–5

• Table 18–6

EJB statistics are described in the following table.

The statistics available for EJB method invocations are listed in the following table.

The statistics for EJB Session Stores are listed in the following table.

The statistics available for EJB pools are listed in the following table.

The statistics available for EJB caches are listed in the following table.

The statistics available for Timers are listed in the following table.

Web Container Statistics

The web container fits into the tree of objects as shown in The Applications Tree. Web container statistics are displayed for each individual web application. Statistics available for the web container for servlets are shown in Table 18–7, and statistics available for web modules are shown in Table 18–8.

Statistics available for web modules are shown in Web Container Statistics.

JDBC Connection Pools Statistics

Monitor JDBC resources to measure performance and capture resource usage at runtime. As the creation of JDBC connections are expensive and frequently cause performance bottlenecks in applications, it is crucial to monitor how a JDBC connection pool is releasing and creating new connections and how many threads are waiting to retrieve a connection from a particular pool.

The statistics available for the JDBC connection pool are shown in the following table.

JMS/Connector Service Statistics

The statistics available for the connector connection pools are shown in Table 18–10. Statistics for Connector Work Management are shown in Table 18–11.

Statistics available for Connector Work Management are listed in the following table.

Statistics for Connection Managers in an ORB

The statistics available for the connection manager in an ORB are listed in the following table.

Thread Pools Statistics

The statistics available for the thread pool are shown in the following table.

Transaction Service Statistics

The transaction service allows the client to freeze the transaction subsystem in order to roll back transactions and determine the transactions that are in process at the time of the freeze. The statistics available for the transaction service are shown in the following table.

Java Virtual Machine (JVM) Statistics

The JVM has monitorable attributes that are always enabled. The statistics available for the JVM are shown in the following table.

JVM Statistics in Java SE

With Java SE, additional monitoring information can be obtained from the JVM. Set the monitoring level to LOW to enable the display of this additional information. Set the monitoring level to HIGH to also view information pertaining to each live thread in the system. More information on the additional monitoring features for Java SE is available in a document titled Monitoring and Management for the Java Platform, which is available from http://java.sun.com/javase/6/docs/technotes/guides/management/.

The Java SE monitoring tools are discussed at http://java.sun.com/javase/6/docs/technotes/tools/#manage.

The statistics available for class loading in the JVM in Java SE are shown in the following table.

The statistics available for compilation in the JVM in Java SE are shown in the following table.

The statistics available for garbage collection in the JVM in Java SE are shown in the following table.

The statistics available for memory in the JVM in Java SE are shown in the following table.

The statistics available for the operating system in the JVM in Java SE are shown in the following table.

The statistics available for the runtime in the JVM in Java SE are shown in the following table.

The statistics available for ThreadInfo in the JVM in Java SE are shown in the following table.

The statistics available for threads in the JVM in Java SE are shown in the following table.

Production Web Container (PWC) Statistics

Statistics are available for the following PWC components and services for as described in the following tables.

These statistics are only available for the cluster and enterprise profiles.

• Table 18–24

• Table 18–25

• Table 18–26

• Table 18–27

• Table 18–28

• Table 18–29

• Table 18–30

• Table 18–31

Statistics for PWC virtual servers are listed in the following table.

The statistics available for PWC requests are listed in the following table.

The cache information section provides information on how the file cache is being used. Statistics for PWC file caches are listed in the following table.

This section provides information about the server's HTTP-level keep-alive system. The statistics available for PWC Keep Alive are listed in the following table.

The DNS Cache caches IP addresses and DNS names. The server’s DNS cache is disabled by default. A single cache entry represents a singular IP address or DNS name lookup. The statistics available for PWC DNS are listed in the following table.

Statistics for PWC thread pools are listed in the following table.

The Connection Queue is the queue in which requests are held prior to being serviced. Statistics for the connection queue show the number of sessions in the queue and the average delay before the connection is accepted. Statistics for PWC connection queues are listed in the following table.

Statistics for PWC HTTP service are listed in the following table.

Enabling and Disabling Monitoring

This section contains the following topics:

• Configuring Monitoring Levels Using the Admin Console

• To Configure Monitoring Levels Using asadmin

Configuring Monitoring Levels Using the Admin Console

To configure monitoring in the Admin Console:

• For the developer profile, go to Configuration -> Monitoring

• For the cluster and enterprise profiles, go to Configurations -> Configuration -> Monitoring

By default, monitoring is turned off for all components and services. To turn monitoring on, select LOW or HIGH from the combo box. To turn monitoring off, select OFF from the combo box.

For details on configuring monitoring, see the online help available with the Admin Console.

To Configure Monitoring Levels Using asadmin

1. Use the get command to find out what services and components currently have monitoring enabled.

   asadmin> get --user admin-user server.monitoring-service.module-monitoring-levels.*

   Returns:

   server.monitoring-service.module-monitoring-levels. connector-connection-pool = OFF server.monitoring-service.module-monitoring-levels. connector-service = OFF server.monitoring-service.module-monitoring-levels.ejb-container = OFF server.monitoring-service.module-monitoring-levels.http-service = OFF server.monitoring-service.module-monitoring-levels.jdbc-connection-pool = OFF server.monitoring-service.module-monitoring-levels.jms-service = OFF server.monitoring-service.module-monitoring-levels.jvm = OFF server.monitoring-service.module-monitoring-levels.orb = OFF server.monitoring-service.module-monitoring-levels.thread-pool = OFF server.monitoring-service.module-monitoring-levels.transaction-service = OFF server.monitoring-service.module-monitoring-levels.web-container = OFF

2. Use the set command to enable monitoring.

   For example, to enable monitoring for the HTTP service:

   asadmin> set --user admin-user server.monitoring-service.module-monitoring-levels.http-service=LOW

   To disable monitoring, use the set command and specify OFF for the monitoring level.

Viewing Monitoring Data

• Viewing Monitoring Data in the Admin Console

• Viewing Monitoring Data With the asadmin Tool

Viewing Monitoring Data in the Admin Console

In the developer profile, to view monitoring data, go to Application Server -> Monitor.

In the cluster and enterprise profiles, to view monitoring data for a stand-alone instance, go to Stand-Alone Instances -> Instance -> Monitor. To view monitoring data for a clustered instance, go to Clusters -> Cluster -> Instance -> Monitor.

You can select and view monitoring data for JVM, server, applications, thread pools, HTTP service, transaction service, log statistics, and call flow statistics. A diagram showing how these components and services are organized is shown in About the Tree Structure of Monitorable Objects.

For details on viewing or configuring monitoring, see the online help available with the Admin Console.

For more information on the attributes for each component or service, refer to About Statistics for Monitored Components and Services.

Viewing Monitoring Data With the asadmin Tool

This section contains the following topics:

• To Use the asadmin monitor Command to View Monitoring Data

• To Use the asadmin get and list Commands to View Monitoring Data

• Understanding and Specifying Dotted Names

• Examples of the list and get Commands

• Examples for the list --user admin-user --monitor Command

• Examples for the get --user admin-user --monitor Command

• To Use the PetStore Example

• Expected Output for list and get Commands at All Levels

To Use the asadmin monitor Command to View Monitoring Data

asadmin has two ways of viewing monitoring data. The first is to use the monitor command. This command prints out the commonly-monitored statistics, and has options for filtering out statistics and capture the output in a Comma Separated Values (CSV) file.

1. To view monitoring data, use the monitor command, and specify the type of monitoring data: httplistener, keepalive, filecache, connectionqueue, jdbcpool, jvm, threadpool, servlet, connection, connectorpool, endpoint, entitybean, messagedriven, statefulsession, statelesssession, httpservice, or webmodule.

   For example, to view data for jvm on server, enter the following:

   asadmin>monitor --type jvm --user adminuser server

   JVM Monitoring UpTime(ms) HeapSize(bytes) current min max low high count 327142979 0 531628032 0 45940736 45940736

2. To view monitoring data and send the output to a CSV file, use the filename option. For example:

   asadmin> monitor --type jvm --filename myoutputfile --user adminuser server

To Use the asadmin get and list Commands to View Monitoring Data

The monitor command is useful in most situations. However, it does not offer the complete list of all monitorable objects. To view all monitorable data using the asadmin tool, use the asadmin list and asadmin get commands followed by the dotted name of a monitorable object, as follows.

1. To view the names of the objects that can be monitored, use the asadmin list command.

   For example, to view a list of application components and subsystems that have monitoring enable for the server instance, type the following command in a terminal window:

   asadmin> list --user adminuser --monitor server

   The preceding command returns a list of application components and subsystems that have monitoring enabled, for example:

   server.resources server.connector-service server.orb server.jms-service server.jvm server.applications server.http-service server.thread-pools

   For further examples using the list command, refer to Examples of the list and get Commands. For further information on the dotted names you can use with the list command, refer to Understanding and Specifying Dotted Names.

2. To display monitoring statistics for an application component or subsystem for which monitoring has been enabled, use the asadmin get command.

   To get the statistics, type the asadmin get command in a terminal window, specifying a name displayed by the list command in the preceding step. The following example attempts to get all attributes from a subsystem for a specific object:

   asadmin> get --user adminuser --monitor server.jvm.*

   The command returns the following attributes and data:

   server.jvm.dotted-name = server.jvm server.jvm.heapsize-current = 21241856 server.jvm.heapsize-description = Provides statistical information about the JVM's memory heap size. server.jvm.heapsize-highwatermark = 21241856 server.jvm.heapsize-lastsampletime = 1080232913938 server.jvm.heapsize-lowerbound = 0 server.jvm.heapsize-lowwatermark = 0 server.jvm.heapsize-name = JvmHeapSize server.jvm.heapsize-starttime = 1080234457308 server.jvm.heapsize-unit = bytes server.jvm.heapsize-upperbound = 518979584 server.jvm.uptime-count = 1080234457308 server.jvm.uptime-description = Provides the amount of time the JVM has been running. server.jvm.uptime-lastsampletime = 1080234457308 server.jvm.uptime-name = JvmUpTime server.jvm.uptime-starttime = 1080232913928 server.jvm.uptime-unit = milliseconds

   For further examples using the get command, refer to Examples of the list and get Commands. For further information on the dotted names you can use with the get command, refer to Understanding and Specifying Dotted Names.

Understanding and Specifying Dotted Names

In the asadmin list and get commands, specify the dotted name of monitorable objects. All child objects are addressed using the dot (.) character as separator, thus these are referred to as dotted names. If a child node is of singleton type, then only the monitoring object type is needed to address the object, otherwise a name of the form type.name is needed to address the object.

For example, http-service is one of the valid monitorable object types and is a singleton. To address a singleton child node representing the http-service of instance server, the dotted name is:

server.http-service

Another example, application, is a valid monitorable object type and is not a singleton. To address a non-singleton child node representing, for example, the application PetStore, the dotted name is:

server.applications.petstore

The dotted names can also address specific attributes in monitorable objects. For example, http-service has a monitorable attribute called bytesreceived-lastsampletime. The following name addresses the bytesreceived attribute:

server.http-service.server.http-listener-1.
  bytesreceived-lastsampletime

The administrator is not expected to know the valid dotted names for asadmin list and get commands. The list command displays available monitorable objects, while the get command used with a wildcard parameter allows the inspection of all available attributes on any monitorable object.

The underlying assumptions for using the list and get commands with dotted names are:

• Any list command that has a dotted name that is not followed by a wildcard (*) gets as its result the current node’s immediate children. For example, list --user adminuser --monitor server lists all immediate children belonging to the server node.

• Any list command that has a dotted name followed by a wildcard of the form .* gets as its result a hierarchical tree of children nodes from the current node. For example, list --user adminuser --monitor server.applications.* lists all children of applications and their subsequent child nodes and so on.

• Any list command that has a dotted name preceded or followed by a wildcard of the form *dottedname or dotted * name or dotted name * gets as its result all nodes and their children matching the regular expression created by the provided matching pattern.

• A get command followed by a .* or a * gets as its result the set of attributes and their values belonging to the current node to be matched.

For more information, read Expected Output for list and get Commands at All Levels.

Examples of the list and get Commands

This section contains the following topics:

• Examples for the list --user admin-user --monitor Command

• Examples for the get --user admin-user --monitor Command

Examples for the list --user admin-user --monitor Command

The list command provides information about the application components and subsystems currently being monitored for the specified server instance name. Using this command, you can see the monitorable components and subcomponents for a server instance. For a more complete listing of list examples, see Expected Output for list and get Commands at All Levels.

Example 1

asadmin> list --user admin-user --monitor server

The preceding command returns a list of application components and subsystems that have monitoring enabled, for example:

server.resources
server.orb
server.jvm
server.jms-service
server.connector-service
server.applications
server.http-service
server.thread-pools

It is also possible to list applications that are currently monitored in the specified server instance. This is useful when particular monitoring statistics are sought from an application using the get command.

Example 2

asadmin> list --user admin-user --monitor server.applications

Returns:

server.applications.adminapp
 server.applications.admingui
server.applications.myApp

For a more comprehensive example, see To Use the PetStore Example.

Examples for the get --user admin-user --monitor Command

This command retrieves the following monitored information:

• All attribute(s) monitored within a component or subsystem

• Specific attribute monitored within a component or subsystem

  When an attribute is requested that does not exist for a particular component or subsystem, an error is returned. Similarly, when a specific attribute is requested that is not active for a component or subsystem, an error is returned.

  Refer to Expected Output for list and get Commands at All Levels for more information on the use of the get command.

Example 1

Attempt to get all attributes from a subsystem for a specific object:

asadmin> get --user admin-user --monitor server.jvm.*

Returns:

server.jvm.dotted-name= server.jvm
server.jvm.heapsize-current = 21241856
server.jvm.heapsize-description = Provides statistical information about 
   the JVM's memory heap size.
server.jvm.heapsize-highwatermark = 21241856
server.jvm.heapsize-lastsampletime = 1080232913938
server.jvm.heapsize-lowerbound = 0
server.jvm.heapsize-lowwatermark = 0
server.jvm.heapsize-name = JvmHeapSize
server.jvm.heapsize-starttime = 1080234457308
server.jvm.heapsize-unit = bytes
server.jvm.heapsize-upperbound = 518979584
server.jvm.uptime-count = 1080234457308
server.jvm.uptime-description = Provides the amount of time the JVM has 
   been running.
server.jvm.uptime-lastsampletime = 1080234457308
server.jvm.uptime-name = JvmUpTime
server.jvm.uptime-starttime = 1080232913928
server.jvm.uptime-unit = milliseconds

Example 2

Attempt to get all attributes from a Java EE application:

asadmin> get --user admin-user --monitor server.applications.myJavaEEApp.*

Returns:

No matches resulted from the wildcard expression.
CLI137 Command get failed.

There are no monitorable attributes exposed at the Java-EE-application level, therefore this reply displays.

Example 3

Attempt to get a specific attribute from a subsystem:

asadmin> get --user admin-user --monitor server.jvm.uptime-lastsampletime

Returns:

server.jvm.uptime-lastsampletime = 1093215374813

Example 4

Attempt to get an unknown attribute from within a subsystem attribute:

asadmin> get --user admin-user --monitor server.jvm.badname

Returns:

No such attribute found from reflecting the corresponding Stats 
interface: [badname]
CLI137 Command get failed.

To Use the PetStore Example

The following example illustrates how the asadmin tool might be used for monitoring purposes.

A user wants to inspect the number of calls made to a method in the sample PetStore application after it has been deployed onto the Application Server. The instance onto which it has been deployed is named server. A combination of the list and get commands are used to access desired statistics on a method.

1. Start the Application Server and the asadmin tool.

2. Set some useful environment variables to avoid entering them for every command:

   asadmin> export AS_ADMIN_USER=admin AS_ADMIN_PASSWORD=admin123 asadmin>export AS_ADMIN_HOST=localhost AS_ADMIN_PORT=4848

3. List monitorable components for instance server:

   asadmin> list --user adminuser --monitor server*

   Returns (output will be similar to):

   server server.applications server.applications.CometEJB server.applications.ConverterApp server.applications.petstore server.http-service server.resources server.thread-pools

   The list of monitorable components includes thread-pools, http-service, resources, and all deployed (and enabled) applications.

4. List the monitorable subcomponents in the PetStore application (-m can be used instead of --monitor):

   asadmin> list -m server.applications.petstore

   Returns:

   server.applications.petstore.signon-ejb_jar server.applications.petstore.catalog-ejb_jar server.applications.petstore.uidgen-ejb_jar server.applications.petstore.customer-ejb_jar server.applications.petstore.petstore-ejb_jar server.applications.petstore.petstore\.war server.applications.petstore.AsyncSenderJAR_jar server.applications.petstore.cart-ejb_jar

5. List the monitorable subcomponents in the EJB module signon-ejb_jar of the PetStore application:

   asadmin> list -m server.applications.petstore.signon-ejb_jar

   Returns:

   server.applications.petstore.signon-ejb_jar.SignOnEJB server.applications.petstore.signon-ejb_jar.UserEJB

6. List the monitorable subcomponents in the entity bean UserEJB for the EJB module signon-ejb_jar of the PetStore application:

   asadmin> list -m server.applications.petstore.signon-ejb_jar.UserEJB

   Returns (with dotted name removed for space considerations):

   server.applications.petstore.signon-ejb_jar.UserEJB.bean-cache server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods server.applications.petstore.signon-ejb_jar.UserEJB.bean-pool

7. List the monitorable subcomponents in the method getUserName for the entity bean UserEJB in the EJB module signon-ejb_jar of the PetStore application:

   asadmin> list -m server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods.getUserName

   Returns:

   Nothing to list at server.applications.petstore.signon-ejb_jar. UserEJB.bean-methods.getUserName. To get the valid names beginning with a string, use the wildcard "*" character. For example, to list all names that begin with "server", use "list server*".

8. There are no monitorable subcomponents for methods. Get all monitorable statistics for the method getUserName.

   asadmin> get -m server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods.getUserName.*

   Returns:

   server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.executiontime-count = 0 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.executiontime-description = Provides the time in milliseconds spent during the last successful/unsuccessful attempt to execute the operation. server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.executiontime-lastsampletime = 1079981809259 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.executiontime-name = ExecutionTime server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.executiontime-starttime = 1079980593137 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.executiontime-unit = count server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-count = 0 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-description = Provides the number of times an operation was called, the total time that was spent during the invocation and so on. server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-lastsampletime = 1079980593137 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-maxtime = 0 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-mintime = 0 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-name = ExecutionTime server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-starttime = 1079980593137 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-totaltime = 0 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.methodstatistic-unit = server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumerrors-count = 0 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumerrors-description = Provides the total number of errors that occured during invocation or execution of an operation. server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumerrors-lastsampletime = 1079981809273 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumerrors-name = TotalNumErrors server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumerrors-starttime = 1079980593137 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumerrors-unit = count server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumsuccess-count = 0 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumsuccess-description = Provides the total number of successful invocations of the method. server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumsuccess-lastsampletime = 1079981809255 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumsuccess-name = TotalNumSuccess server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumsuccess-starttime = 1079980593137 server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.totalnumsuccess-unit = count

9. To also get a specific statistic, such as execution time, use a command such as the following:

   asadmin> get -m server.applications.petstore.signon-ejb_jar. UserEJB.bean-methods.getUserName.executiontime-count

   Returns:

   server.applications.petstore.signon-ejb_jar.UserEJB.bean-methods. getUserName.executiontime-count = 1

Expected Output for list and get Commands at All Levels

The following tables show the command, dotted name, and corresponding output at each level of the tree.

server.applicationsserver.thread-poolsserver.
resourcesserver.http-serviceserver.transaction-
serviceserver.orb.connection-managersserver.orb.
connection-managers.orb\.Connections\.Inbound\.
AcceptedConnectionsserver.jvm

The following table shows the command, dotted name, and corresponding output for the applications level.

appl1app2web-module1_warejb-module2_jar...

The following table shows the command, dotted name, and corresponding output for stand-alone modules and enterprise applications at the applications level.

CreateCount_CountCreateCount_
DescriptionCreateCount_
LastSampleTimeCreateCount_
NameCreateCount_
StartTimeCreateCount_
UnitMethodReadyCount_
CurrentMethodReadyCount_
DescriptionMethodReadyCount_
HighWaterMarkMethodReadyCount_
LastSampleTimeMethodReadyCount_
LowWaterMarkMethodReadyCount_
NameMethodReadyCount_
StartTimeMethodReadyCount_
UnitRemoveCount_CountRemoveCount_
DescriptionRemoveCount_
LastSampleTimeRemoveCount_
NameRemoveCount_StartTimeAttribute 
RemoveCount_Unit

The following table shows the command, dotted name, and corresponding output for the HTTP Service level.

The following table shows the command, dotted name, and corresponding output for the thread pools level.

The following table shows the command, dotted name, and corresponding output for the resources level.

The following table shows the command, dotted name, and corresponding output for the transaction service level.

The following table shows the command, dotted name, and corresponding output for the ORB level.

The following table shows the command, dotted name, and corresponding output for the JVM level.

Using JConsole

This section contains the following topics:

• Securing JConsole to Application Server Connection

• Prerequisites for Connecting JConsole to Application Server

• Connecting JConsole to Application Server

• Connecting JConsole Securely to Application Server

Administration (management and monitoring) of the Application Server is based on JMX Technology. This means that the managed components are represented as MBeans in the MBeanServer running in the Application Server's JVM.

Java SE 5 enhances management and monitoring of the JVM by including a Platform MBean Server and by including MBeans to configure the JVM. Application Server leverages these enhancements and registers its MBeans with the Platform MBean Server. Thus a JMX Connector Client gets a unified view of JVM MBeans as well as Application Server MBeans.

To view all the MBeans, Application Server provides a configuration of the Standard JMX Connector Server called System JMX Connector Server. As part of Application Server startup, an instance of this JMX Connector Server is started. Any compliant JMX connector client can connect to the server using this Connector Server.

Java SE also provides tools to connect to an MBean Server and view MBeans registered with it. JConsole is one such popular JMX Connector Client and is available as part of the standard Java SE distribution. For more information on JConsole, see http://java.sun.com/javase/6/docs/technotes/guides/management/jconsole.html

When you configure JConsole with Application Server, Application Server becomes the JMX Connector's server end and JConsole becomes the JMX Connector's preferred client end. Connecting JConsole to Application Servershows how to make a successful connection .

Securing JConsole to Application Server Connection

There are subtle differences in how to connect to Application Server, or any JMX Connector Server end, based on the transport layer security of the connection. If the server end is secure (guarantees transport layer security), there is a little more configuration to be performed on the client end.

• By default, the developer profile of Application Server is configured with a non-secure System JMX Connector Server.

• By default, cluster and enterprise profiles of Application Server are configured with a secure System JMX Connector Server.

• The protocol used for communication is RMI/JRMP. If security is enabled for the JMX Connector, the protocol used is RMI/JRMP over SSL.

  ---

  Note –

  RMI over SSL does not provide additional checks to ensure that the client is talking to the intended server. Thus, there is always a possibility, while using JConsole, that you are sending the user name and password to a malicious host. It is completely up to the administrator to make sure that security is not compromised.

  ---

When you install a developer profile domain on a machine such as appserver.sun.com, you will see the following in the Domain Administration Server (DAS) domain.xml file:

<!- – The JSR 160 "system-jmx-connector" – –><jmx-connector accept-all="false" address="0.0.0.0" auth-realm-name="admin-realm" enabled="true" name="system" port="8686" protocol="rmi_jrmp" security-enabled="false"/><!- – The JSR 160 "system-jmx-connector" – –>

The security-enabled flag for the JMX Connector is false. If you are running the cluster or enterprise profile, or if you have turned on security for the JMX Connector in the developer profile, this flag is set to true.

<!- – The JSR 160 "system-jmx-connector" – –><jmx-connector accept-all="false" address="0.0.0.0" auth-realm-name="admin-realm" enabled="true" name="system" port="8686" protocol="rmi_jrmp" security-enabled="true"/>...</jmx-connector><!- – The JSR 160 "system-jmx-connector" – –>

Prerequisites for Connecting JConsole to Application Server

The JConsole setup has two parts: a server end and a client end. For this example, the Application Server domain is installed on a machine called appserver.sun.com, which is a powerful Solaris server. This is the server end.

The client end also has an installation of Application Server. Let us assume that the client end is a Windows machine with Java SE 6.0 and Application Server installed.

The Application Server installation is needed on the client end only when your Application Server domain has security enabled on the remote machine (the default for cluster and enterprise profiles). If you just want to administer an Application Server developer profile domain on the Solaris machine above, you do not need the Application Server installation on this client machine.

If the server and client ends are on the same machine, you can use localhost to specify the host name.

Connecting JConsole to Application Server

This procedure describes connecting JConsole to Application Server without security enabled on the JMX Connector. By default, security is not enabled on Application Server for the developer profile.

1. Start the domain on appserver.sun.com.

2. Start JConsole by running JDK_HOME/bin/jconsole.

3. In the Connect to Agent tab of JConsole, enter user name, password, host name and port (8686, by default).

   The user name refers to the administration user name and password refers to the administration password of the domain.

4. Click Connect.

   In the JConsole window you will see all your MBeans, VM information etc., in various tabs.

Connecting JConsole Securely to Application Server

This procedure describes how to connect JConsole to Application Server with security enabled on the JMX Connector. By default, security is enabled on Application Server cluster or enterprise profiles. Use this procedure if you have security enabled on the developer profile's JMX Connector.

1. Install Application Server on the client machine (where JConsole is installed).

   The only reason you need this is to let JConsole know where the server certificate of the Domain Administration Server that you trust is located. To obtain that certificate, invoke at least one remote asadmin command and to do that, you need the local installation of Application Server.

2. Start the Application Server on appserver.sun.com.

   Since this is a cluster or enterprise domain, the system JMX Connector server is secure. To enable security on the developer profile JMX Connector, see the Admin Console online help.

3. From the local Application Server installation, run install-dir\bin\asadmin list --user admin --secure=true --host appserver.sun.com --port 4848 (where 4848 is the server's administration port).

   Though asadmin list command is chosen for this example, you can run any remote asadmin command. You are prompted to accept the certificate sent by the DAS of appserver.sun.com.

4. Press y to accept the certificate sent by the DAS on appserver.sun.com.

   The server's certificate is stored in a file called .asadmintruststore in your home directory on the client machine.

   ---

   Note –

   This step is not required if your server machine and client machine is the same. That is, if you are running JConsole also on appserver.sun.com.

   ---

5. Let JConsole know the trust store location by using the following JConsole command:

   JDK-dir\bin\jconsole.exe -J-Djavax.net.ssl.trustStore="C:\Documents and Settings\user\.asadmintruststore"

6. Start JConsole by running JDK_HOME/bin/jconsole

7. In the Connect to Agent tab of JConsole, enter user name, password, host name and port (8686, by default).

   The user name refers to the administration user name and password refers to the administration password of the domain.

8. Click Connect.

   In the JConsole window you will see all your MBeans, VM information etc., in various tabs.