D OC4J Performance Metrics

This appendix covers the following metrics:

JTA Resource Metrics
JCA Metrics
OC4J J2EE Application Metrics
OC4J JMS Metrics
OC4J Task Manager Metrics
Java Object Cache JOC Metrics

D.1 JTA Resource Metrics

Table D-1 shows the JTA resource metrics.

The metric table type is JTAResource.

Table D-1 /oc4j/JTAResource Metrics

Metric withJSR-77 JTA Resource Description Unit

Metric withJSR-77 JTA Resource	Description	Unit
`ActiveCount` jtaresource_active	Total count of active transactions. A consistently high value can indicate a heavy load on a server.	ops
`AverageCommitTime` jtaresource_averageCommit	Average commit time of all transactions. This is the average of the `jtaresource_performTransaction` values, however, this is a mean average so there may be spikes in the system indicating other issues as well.	msecs
`CommittedCount` jtaresource_committed	Total count of transactions which have committed.	ops
`HeuristicCommittedCount` jtaresource_heuristicCommitted	Total count of heuristically committed transactions. A high value suggests the system or application may not be automated enough, for example, too much system administration in general or inadequate handling of transaction architecture, or a particular issue has occurred which required extensive administration. This is due to a subordinate `TransactionManager` and not a resource manager being rollbacked while in the prepared state.	ops
`HeuristicCount` jtaresource_heuristic	Total count of all heuristically rolledback and committed transactions. See comments for `heuristicCommittedCount` and `heuristicRolledbackCount`.	ops
`HeuristicMixedExceptionCount` jtaresource_heuristicMixedException	Total count of `HeuristicMixedExceptions` encountered. A high value can indicate a high number of potentially non-ACID outcomes resulting from inconsistent or inappropriate administrative intervention.	ops
`HeuristicRollbackExceptionCount` jtaresource_heuristicRollbackException	Total count of `HeuristicRollbackExceptions` encountered. A high value suggests the system or application may not be automated enough, for example, too much system administration in general or inadequate handling of transaction architecture, or a particular issue has occurred which required extensive administration. Unlike the `rolledbackDueToAdminCount` metric which indicates adminstrative rollback at the root transaction manager level while a transaction is active, this is due to either a subordinate TransactionManager or resource manager being rolledback while in the prepared state.	ops
`HeuristicRolledbackCount` jtaresource_heuristicRolledback	Total count of heuristically rolledback transactions. A high value here suggests the system or application may not be automated enough for example, too much system administration in general or inadequate handling in transaction architecture, or a particular issue has occurred which required extensive administration. This is due to a subordinate TransactionManager and not a resource manager being committed while in the prepared state.	ops
`IllegalStateExceptionCount` jtaresource_illegalStateException	Total count of `IllegalStateExceptions` encountered. A high value should be rare and should only be possible as a result of prior administrative intervention.	ops
`PerformTransaction` jtaresource_performTransaction	Time from begin to end of the transaction. This is useful as a high-level indicator of performance issues, however, as it is a measure from the begin to the end of the transaction only, anything that occurs within this time could be the cause. For example, application logic, database activity, JMS activity, or transaction processing.	msecs
`RollbackCompletion` jtaresource_rollbackCompletion	Time required for a rollback completion. A high value indicates delays in the rollback calls on resourcemanagers which may be a result of network latency or resource manager issues.
`RollbackExceptionCount` jtaresource_rollbackException	Total count of RollbackExceptions encountered. A high value can indicate an issue in the system (for example, a database is down) which results in performance degradation. This can result from both a direct internal system failure or from the application calling `setRollbackOnly` for some reason. Suggest examining the fine-grained rollback cause counts and logs for the cause of the rollbacks as well as looking into application code which calls `setRollbackOnly`.	ops
`RolledbackCount` jtaresource_rolledback	Total count of transactions which have rolledback. A high value can indicate an issue in the system (for example a database is down) which results in performance degradation.	ops
`RolledbackDueToAdminCount` jtaresource_rolledbackDueToAdmin	Total count of transactions that have rolledback due to administrative action. A high value here suggests the system or application may not be automated enough. For example, too much system administration in general or inadequate handling of transaction architecture, or a particular issue has occurred which required extensive administration.	ops
`RolledbackDueToAppCount` jtaresource_rolledbackDueToApp	Total count of transactions that have rolledback due to the application calling `setRollbackOnly` or `rollback` explicitly. A high value here can occur for any reason, but most often occurs due to some handled exception within an application, for example, SQLException during database update. Suggest looking into application code which calls `setRollbackOnly` or `rollback` to see why it is doing so.	ops
`RolledbackDueToResourceCount` jtaresource_rolledbackDueToResource	Total count of transactions that have rolledback due to and error in an enlisted resource. A high value here can indicate an issue with one or more resource managers, for example, database or the network connection between OC4J and these resources.	ops
`RolledbackDueToTimedOutCount` jtaresource_rolledbackDueToTimedOut	Total count of transactions that have rolledback due to timeout. A high number can indicate any number of issues are causing the transaction, or activity within transactional bounds, to take too long or the timeout value specified is not flexible enough. Suggest looking into what activities within the transactions involved, which may be of a certain type or application, are taking up time or adjust the `transaction-timeout` value in the `transaction-manager.xml` configuration file.	ops
`SecurityExceptionCount` jtaresource_securityException	Total count of `SecurityExceptions` encountered. A high value, or any value getter than 0, can indicate an issue with the identity on the thread executing this.	ops
`SinglePhaseCommitCompletion` jtaresource_singlePhaseCommitCompletion	Time required for a single-phase commit completion. A single phase commit involves committing of a single resource only and therefore no 2PC costs, for example logging, are incurred and a large value here generally indicates an issue with the resource which is being committed, for example network latency to the database, suggest looking closer at the metrics of the resource involved in the commit.
`SystemExceptionCount` jtaresource_systemException	Total count of `SystemExceptions` encountered. A high value here should never occur and indicates a serious failure in the system. Suggest analyzing OC4J and resource manager logs.
`TransactionSuspended` jtaresource_transactionSuspended	Time transaction has been suspended. A high value here indicates that the transactions are being held in a suspended state waiting for return a call in a different or no transactional context, generally from an EJB method call, or during propagation of a transaction context. Suggest analyzing the application to determine what activity is taking place during the suspend or whether there is a network latency in the case of propagation.
`TwoPhaseCommitCompletion` jtaresource_twoPhaseCommitCompletion	Time required for a two-phase commit completion. A high value here indicates delays in the prepare and commit calls on resource managers or the transaction record logging in OC4J. Suggest looking closer at the metrics of the resources involved in the prepare and commit as well as performance settings in for transaction record logging in OC4J found in the `transaction-manager.xml` configuration file.

ActiveCount

jtaresource_active

Total count of active transactions. A consistently high value can indicate a heavy load on a server.

ops

AverageCommitTime

jtaresource_averageCommit

Average commit time of all transactions. This is the average of the jtaresource_performTransaction values, however, this is a mean average so there may be spikes in the system indicating other issues as well.

msecs

CommittedCount

jtaresource_committed

Total count of transactions which have committed.

ops

HeuristicCommittedCount

jtaresource_heuristicCommitted

Total count of heuristically committed transactions.

A high value suggests the system or application may not be automated enough, for example, too much system administration in general or inadequate handling of transaction architecture, or a particular issue has occurred which required extensive administration. This is due to a subordinate TransactionManager and not a resource manager being rollbacked while in the prepared state.

ops

HeuristicCount

jtaresource_heuristic

Total count of all heuristically rolledback and committed transactions. See comments for heuristicCommittedCount and heuristicRolledbackCount.

ops

HeuristicMixedExceptionCount

jtaresource_heuristicMixedException

Total count of HeuristicMixedExceptions encountered.

A high value can indicate a high number of potentially non-ACID outcomes resulting from inconsistent or inappropriate administrative intervention.

ops

HeuristicRollbackExceptionCount

jtaresource_heuristicRollbackException

Total count of HeuristicRollbackExceptions encountered.

A high value suggests the system or application may not be automated enough, for example, too much system administration in general or inadequate handling of transaction architecture, or a particular issue has occurred which required extensive administration. Unlike the rolledbackDueToAdminCount metric which indicates adminstrative rollback at the root transaction manager level while a transaction is active, this is due to either a subordinate TransactionManager or resource manager being rolledback while in the prepared state.

ops

HeuristicRolledbackCount

jtaresource_heuristicRolledback

Total count of heuristically rolledback transactions.

A high value here suggests the system or application may not be automated enough for example, too much system administration in general or inadequate handling in transaction architecture, or a particular issue has occurred which required extensive administration. This is due to a subordinate TransactionManager and not a resource manager being committed while in the prepared state.

ops

IllegalStateExceptionCount

jtaresource_illegalStateException

Total count of IllegalStateExceptions encountered.

A high value should be rare and should only be possible as a result of prior administrative intervention.

ops

PerformTransaction

jtaresource_performTransaction

Time from begin to end of the transaction.

This is useful as a high-level indicator of performance issues, however, as it is a measure from the begin to the end of the transaction only, anything that occurs within this time could be the cause. For example, application logic, database activity, JMS activity, or transaction processing.

msecs

RollbackCompletion

jtaresource_rollbackCompletion

Time required for a rollback completion.

A high value indicates delays in the rollback calls on resourcemanagers which may be a result of network latency or resource manager issues.

RollbackExceptionCount

jtaresource_rollbackException

Total count of RollbackExceptions encountered.

A high value can indicate an issue in the system (for example, a database is down) which results in performance degradation. This can result from both a direct internal system failure or from the application calling setRollbackOnly for some reason.

Suggest examining the fine-grained rollback cause counts and logs for the cause of the rollbacks as well as looking into application code which calls setRollbackOnly.

ops

RolledbackCount

jtaresource_rolledback

Total count of transactions which have rolledback.

A high value can indicate an issue in the system (for example a database is down) which results in performance degradation.

ops

RolledbackDueToAdminCount

jtaresource_rolledbackDueToAdmin

Total count of transactions that have rolledback due to administrative action.

A high value here suggests the system or application may not be automated enough. For example, too much system administration in general or inadequate handling of transaction architecture, or a particular issue has occurred which required extensive administration.

ops

RolledbackDueToAppCount

jtaresource_rolledbackDueToApp

Total count of transactions that have rolledback due to the application calling setRollbackOnly or rollback explicitly.

A high value here can occur for any reason, but most often occurs due to some handled exception within an application, for example, SQLException during database update.

Suggest looking into application code which calls setRollbackOnly or rollback to see why it is doing so.

ops

RolledbackDueToResourceCount

jtaresource_rolledbackDueToResource

Total count of transactions that have rolledback due to and error in an enlisted resource.

A high value here can indicate an issue with one or more resource managers, for example, database or the network connection between OC4J and these resources.

ops

RolledbackDueToTimedOutCount

jtaresource_rolledbackDueToTimedOut

Total count of transactions that have rolledback due to timeout.

A high number can indicate any number of issues are causing the transaction, or activity within transactional bounds, to take too long or the timeout value specified is not flexible enough.

Suggest looking into what activities within the transactions involved, which may be of a certain type or application, are taking up time or adjust the transaction-timeout value in the transaction-manager.xml configuration file.

ops

SecurityExceptionCount

jtaresource_securityException

Total count of SecurityExceptions encountered.

A high value, or any value getter than 0, can indicate an issue with the identity on the thread executing this.

ops

SinglePhaseCommitCompletion

jtaresource_singlePhaseCommitCompletion

Time required for a single-phase commit completion.

A single phase commit involves committing of a single resource only and therefore no 2PC costs, for example logging, are incurred and a large value here generally indicates an issue with the resource which is being committed, for example network latency to the database, suggest looking closer at the metrics of the resource involved in the commit.

SystemExceptionCount

jtaresource_systemException

Total count of SystemExceptions encountered.

A high value here should never occur and indicates a serious failure in the system. Suggest analyzing OC4J and resource manager logs.

TransactionSuspended

jtaresource_transactionSuspended

Time transaction has been suspended.

A high value here indicates that the transactions are being held in a suspended state waiting for return a call in a different or no transactional context, generally from an EJB method call, or during propagation of a transaction context.

Suggest analyzing the application to determine what activity is taking place during the suspend or whether there is a network latency in the case of propagation.

TwoPhaseCommitCompletion

jtaresource_twoPhaseCommitCompletion

Time required for a two-phase commit completion.

A high value here indicates delays in the prepare and commit calls on resource managers or the transaction record logging in OC4J.

Suggest looking closer at the metrics of the resources involved in the prepare and commit as well as performance settings in for transaction record logging in OC4J found in the transaction-manager.xml configuration file.

D.2 JCA Metrics

Table D-2 shows the JCA metrics.

The metric table type is jca_connection_stats.

Table D-2 oc4j/application/OracleASjms/JCAmetrics Metrics

Metric	Description	Unit
`closeCount.count`	Number of connection handles closed count.	ops
`createCount.count`	Number of connection handles created count.	ops
`poolName.value`	Name of connection pool value.	pool name
`useTime.avg`	Time spent using a connection.	time
`useTime.completed`	Time spent using a connection.	ops
`useTime.maxTime`	Time spent using a connection.	time
`useTime.minTime`	Time spent using a connection.	time
`useTime.time`	Time spent using a connection.	time
`waitTime.avg`	Time spent waiting for a connection to be available.	time
`waitTime.completed`	Time spent waiting for a connection to be available.	ops
`waitTime.maxTime`	Time spent waiting for a connection to be available.	time
`waitTime.minTime`	Time spent waiting for a connection to be available.	time
`waitTime.time`	Time spent waiting for a connection to be available.	time

Table D-3 shows the JCA connection pool stats metrics.

The metric table type is jca_connection_pool_stats.

Table D-3 /oc4j/jca_connection_pool_stats Metrics

Metric	Description	Unit
`closeCount.count`	Number of ManagedConnections closed.	ops
`createCount.count`	Number of ManagedConnections created.	ops
`errorCount.count`	Number of connection error events.	ops
`expiredCount.count`	Number of expired connections removed from pool.	ops
`freePoolSize.maxValue`	Number of free connections in the pool.	connections
`freePoolSize.minValue`	Number of free connections in the pool.	connections
`freePoolSize.value`	Number of free connections in the pool.	connections
`inactivityTimeout.value`	Configuration parameter: timeout for unused connections.	time
`inactivityTimeoutCheck.value`	Configuration parameter: when to check for unused connections.
`initial-capacity.value`	Configuration parameter: number of connections to be pre-created by the pool.	ops
`invalidCount.count`	Number of invalid connections removed from pool.	ops
`maxPoolSize.value`	Configuration parameter: maximum number of connections.	connections
`minPoolSize.value`	Configuration parameter: minimum number of connections.	connections
`poolSize.maxValue`	Size of connection pool.	connections
`poolSize.minValue`	Size of connection pool.	connections
`poolSize.value`	Size of connection pool.	connections
`requestTimeoutCount.count`	Number of failed connection requests due to timeout.	ops
`scheme.value`	Scheme configuration parameter: connection pooling scheme.
`useTime.avg`	Time spent using a connection	time
`useTime.completed`	Time spent using a connection	ops
`useTime.maxTime`	Time spent using a connection	time
`useTime.minTime`	Time spent using a connection	time
`useTime.time`	Time spent using a connection	time
`waitTime.avg`	Time spent waiting for a connection to be available	time
`waitTime.completed`	Time spent waiting for a connection to be available	ops
`waitTime.maxTime`	Time spent waiting for a connection to be available	time
`waitTime.minTime`	Time spent waiting for a connection to be available	time
`waitTime.time`	Time spent waiting for a connection to be available	time
`waitTimeout.value`	Configuration parameter: timeout waiting for a connection in `fixed_wait` scheme
`waitingThreadCount.active`	Number of threads waiting for connection.	ops
`waitingThreadCount.avg`	Number of threads waiting for connection active.
`waitingThreadCount.completed`	Number of threads waiting for connection active.	ops
`waitingThreadCount.maxActive`	Number of threads waiting for connection active.	threads
`waitingThreadCount.maxTime`	Number of threads waiting for connection active.	time
`waitingThreadCount.minTime`	Number of threads waiting for connection active.	time
`waitingThreadCount.time`	Number of threads waiting for connection active.	time

D.3 OC4J J2EE Application Metrics

This section lists the OC4J J2EE application related metrics.

This section covers the following metrics:

Web Module Metrics
Web Context Metrics
OC4J Servlet Metrics
OC4J JSP Metrics
OC4J EJB Metrics
OC4J OPMN Info Metrics
OC4J Work Management Pool Metrics

D.3.1 Web Module Metrics

There is one set of metrics for each Web module within each J2EE application.

Table D-4 shows the web module metrics.

The metric table type is oc4j_web_module.

Table D-4 OC4J/application/WEBs Metrics

Metric	Description	Units
`parseRequest.active`	Current number of threads trying to read/parse AJP or HTTP requests
`parseRequest.avg`	Average time spent to read/parse requests	msecs
`parseRequest.completed`	Number of web requests that have been parsed	ops
`parseRequest.maxActive`	Maximum number of threads trying to read/parse AJP or HTTP requests	threads
`parseRequest.maxTime`	Maximum time spent to read/parse requests	msecs
`parseRequest.minTime`	Minimum time spent to read/parse requests	msecs
`parseRequest.time`	Total time spent to read/parse requests from the socket	msecs
`processRequest.active`	Current number of threads servicing web requests
`processRequest.avg`	Average time spent servicing web requests	msecs
`processRequest.completed`	Number of web requests processed by this application	ops
`processRequest.maxActive`	Maximum number of threads servicing web requests	threads
`processRequest.maxTime`	Maximum time spent servicing a web request	msecs
`processRequest.minTime`	Minimum time spent servicing a web request	msecs
`processRequest.time`	Total time spent servicing this application's web requests	msecs
`resolveContext.active`	Current number of threads trying to create/find the servlet context
`resolveContext.avg`	Average time spent to create/find the servlet context	msecs
`resolveContext.completed`	Count of completed context resolves	ops
`resolveContext.maxActive`	Maximum number of threads trying to create/find the servlet context	threads
`resolveContext.maxTime`	Maximum time spent to create/find the servlet context	msecs
`resolveContext.minTime`	Minimum time spent to create/find the servlet context	msecs
`resolveContext.time`	Total time spent to create/find the servlet context. Each web module (WAR) maps to a servlet context	msecs

D.3.2 Web Context Metrics

Table D-5 shows the web context metrics. There is one set of web context metrics for each Web context module within each J2EE application.

The metric table type is oc4j_context.

Table D-5 OC4J/application/Webs/context Metrics

Metric	Description	Unit
`resolveServlet.time`	Total time spent to create/locate servlet instances (within the servlet context). This includes the time for any required authentication.	msecs
`resolveServlet.completed`	Total Number of lookups for a servlet by OC4J	ops
`resolveServlet.minTime`	Minimum time spent to create/locate the servlet instance (within the servlet context)	msecs
`resolveServlet.maxTime`	Maximum time spent to create/locate the servlet instance (within the servlet context)	msecs
`resolveServlet.avg`	Average time spent to create/locate the servlet instance (within the servlet context)	msecs
`sessionActivation.active`	Number of active sessions	ops
`sessionActivation.time`	Total time in which sessions have been active	msecs
`sessionActivation.completed`	Number of session activations	ops
`sessionActivation.minTime`	Minimum time a session was active	msecs
`sessionActivation.maxTime`	Maximum time a session was active	msecs
`sessionActivation.avg`	Average session lifetime	msecs
`service.time`	Total time spent servicing requests. The service metrics for the servlet include any time spent in the calls to the database. If you need to determine just the oc4j service time, subtract the appropriate execution times.	msecs
`service.completed`	Total number of requests serviced	ops

D.3.3 OC4J Servlet Metrics

Table D-6 shows the servlet metrics. There is one set of servlet metrics for each servlet in each Web module within each J2EE application.

The metric table type is oc4j_servlet.

Table D-6 OC4J/application/WEBs/context /SERVLETS/servlet Metrics

Metric	Description	Units
`service.active`	Current number of threads servicing this servlet	threads
`service.avg`	Average time spent in servicing the servlet	msecs
`service.completed`	Total number of calls to service()
`service.maxActive`	Maximum number of threads servicing this servlet	threads
`service.maxTime`	Maximum time spent on a servlet's service() call	ops
`service.minTime`	Minimum time spent on a servlet's service() call	msecs
`service.time`	Total time spent on the servlet's service() call	msecs

D.3.4 OC4J JSP Metrics

D.3.4.1 JSP Runtime Metrics

Table D-7 shows the JSP metrics. There is one set of JSP metrics for each Web context for each J2EE application.

The metric table type is oc4j_jspExec.

Table D-7 OC4J/application/WEBs/context /JSP Metrics

Metric	Description	Unit
`processRequest.time`	Time spent processing requests for JSPs Only used for Context/Application name	msecs
`processRequest.completed`	Number of requests for JSPs processed by this application	ops
`processRequest.minTime`	Minimum time spent processing requests for JSPs	msecs
`processRequest.maxTime`	Maximum time spent processing requests for JSPs	msecs
`processRequest.avg`	Average time spent processing requests for JSPs	msecs
`processRequest.active`	Current number of active requests for JSPs	ops

D.3.4.2 JSP Metrics

Table D-8 shows the JSP metrics. There is one set of metrics for each JSP in each Web module.

The metric table types are oc4j_jsp(threadsafe=true) and oc4j_jsp(threadsafe=false).

To list these metrics using dmstool, enclose the metric table type in quotation marks.

For example:

dmstool -table "oc4j_jsp(threadsafe=true)"

Table D-8 OC4J/application/WEBs/context /JSPjsp_name Metrics

Metric	Description	Unit
`activeInstances.value`	Number of active instances. Only used when threadsafe=false	instances
`availableInstances.value`	Number of available (that is, created) instances. This value is only provided when threadsafe=false.	instances
`service.active`	Current number of active requests for the JSP
`service.avg`	Average time spent servicing the JSP	msecs
`service.completed`	Number of requests for JSPs processed by this JSP	ops
`service.maxTime`	Maximum time spent servicing the JSP	msecs
`service.minTime`	Minimum time spent servicing the JSP	msecs
`service.time`	Time to serve a JSP (that is, actual execution time of the JSP)	msecs

D.3.5 OC4J EJB Metrics

D.3.5.1 OC4J EJB Session Bean Metrics

Table D-9 shows the EJB Session bean metrics showing information on each session bean.

The metric table type is oc4j_ejb_session_bean.

Table D-9 OC4J EJB Session Bean Metrics

Metric	Description	Unit
`session-type.value`	Provides information on the session type: `Stateless` or `Stateful`	String
`transaction-type.value`	Provides information on the transaction type: `Container` or `Bean`	String

D.3.5.2 EJB Entity Bean Metrics

Table D-10 shows the entity bean metrics. Oracle Application Server provides a set of these metrics for each type of bean in each EJB jar file in each J2EE application.

The metric table type is oc4j_ejb_entity_bean.

Table D-10 OC4J/application/EJBs/ejb-jar-module/ejb-name Metrics

Metric	Description	Unit
`transaction-type.value`	Possible values: `container` or `bean`
`session-type.value`	Possible values: `stateful` or `stateless`
`bean-type.value`	Possible values: `session` or `entity bean`
`exclusive-write-access.value`	Possible values: `true` or `false`
`isolation.value`	Possible values: serializable, `uncommitted`, `committed`, `repeatable_read`, `none`, `DB-determined` The value is DB-determined when the isolation attribute is omitted.
`persistence-type.value`	Possible values: `container or bean` or

D.3.5.3 EJB Method Metrics

Table D-11 shows the EJB method metrics. There is one set of EJB method metrics for each method within each type of EJB bean.

The metric table type is oc4j_ejb_method.

The client.* metrics show values for the actual implementation of the method. The wrapper.* metrics show values for the wrapper that was automatically generated for the method.

Metric	Description	Unit
`client.active`	Current number of threads accessing the actual implementation of this method	ops
`client.avg`	Average time spent inside the actual implementation of this method	msecs
`client.completed`	Number of requests for beans processed by this application	ops
`client.maxActive`	Maximum number of threads accessing the actual implementation of this method	ops
`client.maxTime`	Maximum time spent inside the actual implementation of this method	msecs
`client.minTime`	Minimum time spent inside the actual implementation of this method	msecs
`client.time`	Time spent inside the actual implementation of this method	msecs
`ejbPostCreate.active`	Current number of threads executing `ejbPostCreate`	ops
`ejbPostCreate.avg`	Average time spent in `ejbPostCreate`	msecs
`ejbPostCreate.completed`	Number of times this `ejbPostCreate` has been called	ops
`ejbPostCreate.maxTime`	Maximum time spent in `ejbPostCreate`	msecs
`ejbPostCreate.minTime`	Minimum time spent in `ejbPostCreate`	msecs
`ejbPostCreate.time`	Time spent in the `ejbPostCreate` method (entity beans)	msecs
`trans-attribute.value`	Transaction attribute. Possible values: `NotSupported`, `Supports`, `RequiresNew`, `Mandatory`, and `Never`
`wrapper.active`	Current number of threads accessing the automatically generated wrapper method
`wrapper.avg`	Average time spent inside the automatically generated wrapper method	msecs
`wrapper.completed`	Number of requests for beans processed by this application	ops
`wrapper.maxActive`	Maximum number of threads that access the wrapper	ops
`wrapper.maxTime`	Maximum time spent inside the automatically generated wrapper method	msecs
`wrapper.minTime`	Minimum time spent inside the automatically generated wrapper method	msecs
`wrapper.time`	Time spent inside the automatically generated wrapper method. Note: Not all the wrapper methods invoke the actual bean implementation at runtime (for example, create method in a stateless bean). This means that the time spent in the wrapper code could be less than the time spent in the bean implementation	msecs

D.3.5.4 EJB Stateless Bean Metrics

Table D-12 shows the EJB stateless bean metrics.

The metric table type is oc4j_ejb_stateless_bean.

Table D-12 OC4J EJB Stateless Bean Metrics

Metric	Description	Unit
`pooled.count`	Number of Pooled Instances count	count
`pooled.maxValue`	Number of Pooled Instances	ops
`pooled.minValue`	Number of Pooled Instances	ops
`pooled.value`	Number of Pooled Instances	ops
`ready.count`	Number of Ready Instances	count
`ready.maxValue`	Number of Ready Instances	ops
`ready.minValue`	Number of Ready Instances	ops
`ready.value`	Number of Ready Instances
`session-type.value`	Session type

D.3.5.5 EJB Stateful Bean Metrics

Table D-13 shows the EJB stateful bean metrics.

The metric table type is oc4j_ejb_stateful_bean.

Table D-13 OC4J EJB Stateful Bean Metrics

Metric	Description
`passive.count`	Number of Passivated Instances	count
`passive.maxValue`	Number of Passivated Instances	ops
`passive.minValue`	Number of Passivated Instances	ops
`passive.value`	Number of Passivated Instances	ops
`ready.count`	Number of Ready Instances	count
`ready.maxValue`	Number of Ready Instances	ops
`ready.minValue`	Number of Ready Instances	ops
`ready.value`	Number of Ready Instances	ops
`session-type.value`	Session type
`transaction-type.value`	Transaction

D.3.5.6 EJB Message-Driven Bean Metrics

Table D-14 shows the message-driven bean metrics.

The metric table type is oc4j_ejb_message-driven_bean.

Table D-14 OC4J EJB Message-driven Bean Metrics

Metric	Description	Units
`applicationExceptionCount.count`	Number of application exceptions thrown	count
`failedMessageDeliveryCount.count`	Number of failed message deliveries	count
`messageDelivery.avg`	Message delivery attempts	time
`messageDelivery.completed`	Message delivery attempts	ops
`messageDelivery.maxTime`	Message delivery attempts	time
`messageDelivery.minTime`	Message delivery attempts	time
`messageDelivery.time`	Message delivery attempts	time
`messageEndpointCount.value`	Number of message endpoints	ops
`messageEndpointType.value`	Message endpoint type	Class name
`pooled.count`	Number of Pooled Instances	count
`pooled.maxValue`	Number of Pooled Instances	ops
`pooled.minValue`	Number of Pooled Instances	ops
`pooled.value`	Number of Pooled Instances	ops
`ready.count`	Number of Ready Instances	count
`ready.maxValue`	Number of Ready Instances	ops
`ready.minValue`	Number of Ready Instances	ops
`ready.value`	Number of Ready Instances	ops
`startTime.value`	The MDB available time for service	time
`successfulMessageDeliveryCount.count`	Number of successful message deliveries	count
`systemExceptionCount.count`	Number of SystemExceptions thrown	count
`transaction-type.value`	Transaction value

D.3.6 OC4J OPMN Info Metrics

Table D-15 shows the OC4J OPMN information metrics.

The metric table type is oc4j_opmn.

Table D-15 OC4J OPMN Information Metrics

Metric	Description	Unit
`default_application_log.value`	Specifies the default application log file path.
`ias_cluster.value`	Specifies the Oracle Application Server cluster name.	String
`ias_instance.value`	Specifies the Oracle Application Server instance name.	String
`jms_log.value`	Specifies the JMS log file path.	String
`oc4j_instance.value`	Specifies the OC4J instance ID.	String
`oc4j_island.value`	Specifies the OC4J island ID.	String
`opmn_group.value`	Specifies the OPMN group ID.	String
`opmn_sequence.value`	Specifies the OPMN sequence ID.	String
`rmi_log.value`	Specifies the RMI log file path name.	String
`server_log.value`	Specifies the application server log file path.	String

D.3.7 OC4J Work Management Pool Metrics

Table D-16 shows the OC4J Work management pool metrics.

The metric table type is oc4j_workManagementPool.

Table D-16 OC4J Work Management Pool Metrics

Metric	Description	Unit
`idleThreadCount`	Number of idle threads in the pool. This is a current thread pool state metric.	threads
`keepAlive`	Time before idle threads are removed from available pool. This is a configuration value metric	milliseconds
`maxPoolSize`	Maximum number of threads in the pool. This is a configuration value metric	threads
`maxQueueSize`	Maximum queue size. This is a configuration value metric.	work_requests
`minPoolSize`	Minimum number of threads in the pool. This is a configuration value metric.	threads
`queueFullEvent`	Number of work submission failures due to full queue. This is a current thread pool state metric.	ops
`queueSize`	Current queue size. This is a current thread pool state metric.	work_requests
`totalThreadCount`	Total number of threads in the pool. This is a current thread pool state metric.	threads
`workStartDuration`	Duration between work accepted and work started events. This is a current thread pool state metric. Waiting time is defined as the time period between the work submission is accepted and the execution of the work starts. This metric measures the duration between a work request submission is accepted by the pool and the time when a thread is allocated from the thread pool to run the work. If a thread is readily available, this would measure the processing overhead of the threadpool in finding an available thread and setting up the proper context for processing the work. If all available threads are busy handling other work requests, this time would also include the queuing time.	ops

D.4 OC4J JMS Metrics

OC4J JMS metrics are organized into metric tables and fall into two categories:

JMS API-level metrics: collected on objects visible to the JMS API (for example, connections, sessions, producers, consumers, and browsers). JMS API-level metrics are collected and maintained only for Web and EJB clients (application clients also collect API-level metrics, but do so in their own JVM; these metrics are not available on the OC4J JMS server).
JMS Server-level metrics: collected by the OC4J JMS server and maintained independent of client-state. JMS Server-level metrics are collected and maintained for all types of clients: Application, Web, and EJB.

Each OC4J JMS metric table (metric table type) contains metrics for instances of the same type; different instances have unique names. For each instance in a metric table, a set of metrics is collected. The names for metrics in each instance are unique IDs that OC4J JMS generates.

Instances may have one or more metrics whose value is the name of another metric instance. For example, the JMS session instances contain metrics that point to the parent containing JMS connection instance. You can use the pointers to navigate through the metrics.A parent metric instance usually includes a counter metric indicating the number of child metrics of a certain type that have been created. Child metric instances may appear and disappear as the underlying objects are created and destroyed; the counter keeps track of the total number of such instances that were created during the lifetime of the parent.

Note:

Oracle Application Server JMS metrics are available only for OC4J JMS (thus, metrics are not available for OJMS).

D.4.1 JMS Metric Tables

OC4J JMS metrics are divided into three types, based on how they are updated:

CTOR Metrics: Metrics that are set in the constructor or initialization routine of the associated JMS object, and are never changed during the lifetime of the object.
Normal Metrics: Object level state metrics that are updated as soon as the associated state of the JMS object changes.
Lazy Metrics: these state metrics are updated lazily, that is, not as soon as the underlying metric value changes, but only periodically (these are typically server store metrics and are updated each time the store is cleaned up of expired messages).

Table D-17 shows a summary of the organization of the OC4J JMS metric tables.

Table D-17 OC4J JMS Metric Tables

JMS Metric Table Type	Parent Table Type	Number of Instances	Description
JMSConnectionStats	JMSStats	1 per JMS connection	Statistics for the JMS connections active in this server
JMSDestinationStats	JMSStats	1 per permanent JMS destination	Statistics for each permanent JMS destination known to the OC4J JMS server
JMSDurableSubscriberStats	JMSStats	1 per JMS durable subscriber	Statistics for each JMS durable subscription known to this server
JMSMessageBrowserStats	JMSSessionStats	1 per JMS queue browser	Statistics for the JMS queue browsers in this server
JMSMessageConsumerStats	JMSSessionStats	1 per JMS message consumer	Statistics for the JMS consumers active in this server
JMSMessageProducerStats	JMSSessionStats	1 per JMS message producer	Statistics for the JMS producers active in this server
JMSPersistenceStats	JMSDestinationStats	1 per server-side persistent destination	Statistics for operations on the persistence file for each persistent destination
JMSRequestHandlerStats	JMSStats	1 per remote JMS connection	Statistics for the request handler thread servicing a remote JMS connection.
JMSSessionStats	JMSConnectionStats	1 per JMS session	Statistics for the JMS sessions active in this server
JMSStats	none	1	Statistics for the OC4J JMS Server
JMSStoreStats	JMSDestinationStats JMSTemporaryDestinationStats	1 per server-side message store	Statistics for each message store (one per queue, one per subscription per topic) on the OC4J JMS server
JMSTemporaryDestinationStats	JMSStats	1 per temporary JMS destination	Statistics for each temporary JMS destination known to the OC4J JMS server

D.4.2 JMS Stats Metric Table

Table D-18 shows the JMS Stats metrics.

The metric table type is JMSStats.

Table D-18 JMSStats Metric Table

Metric	Description	Update	Unit
`activeConnections`
`activeHandlers`
`address`	The hostname(s) from which the JMS server accepts remote connections	ctor	string
`closeConnection`
`closeConsumer`
`commit`
`connections`
`createConsumer`
`deqMessage`
`enqMessage`
`host.value`	The explicit hostname on which the OC4J JMS server is running.	ctor	string
`listMessages`
`messageCommitted`
`messageCount`
`messageDequeued`
`messageDiscarded`
`messageEnqueued`
`messageExpired`
`messagePagedIn`
`messagePagedOut`
`messageRecovered`
`messageRolledback`
`oc4j.jms.checkPermissions`
`oc4j.jms.debug`	Value of the `oc4j.jms.debug` OC4J JMS control knob	ctor	bool
`oc4j.jms.forceRecovery`	Value of the `oc4j.jms.forceRecovery` OC4J JMS control knob	ctor	bool
`oc4j.jms.j2ee14`
`oc4j.jms.listenerAttempts`	Value of the `oc4j.jms.listenerAttempts` OC4J JMS control knob	ctor	int
`oc4j.jms.maxOpenFiles`	Value of the `oc4j.jms.maxOpenFiles` OC4J JMS control knob	ctor	int
`oc4j.jms.messagePoll`	Value of the `oc4j.jms.messagePoll` OC4J JMS control knob	ctor	msecs
`oc4j.jms.noDms`	Value of the `oc4j.jms.noDms OC4J` JMS control knob	ctor	bool
`oc4j.jms.noJmx`
`oc4j.jms.pagingThreshold`
`oc4j.jms.printStackTrace`
`oc4j.jms.reconnectAttempts`
`oc4j.jms.reconnectWait`
`oc4j.jms.rememberALLXids`
`oc4j.jms.saveAllExpired`	Value of the `oc4j.jms.saveAllExpired` OC4J JMS control knob	ctor	bool
`oc4j.jms.serverPoll`	Value of the `oc4j.jms.serverPoll OC4J` JMS control knob	ctor	msecs
`oc4j.jms.socketBufsize`	Value of the `oc4j.jms.socketBufsize` OC4J JMS control knob	ctor	int
`peekMessage`
`pendingMessageCount`
`port.value`	The TCP/IP port on which the JMS server listens for incoming connections	ctor	int
`registerConnection`
`rollback`
`startTime`	`System.currentTimeMillis()` when the OC4J JMS server was started	ctor	msecs
`stats`
`storeSize`
`taskManagerInterval`	The scheduling interval of the OC4J task manager (and the scheduling interval for the OC4J JMS expiration task)	ctor	msecs

D.4.3 JMS Request Handler Stats

Table D-19 shows the JMS Request Handler Stats.

The metric table type is JMSRequestHandlerStats.

Table D-19 JMSRequestHandlerStats Metrics

Metric	Description	Update	Unit
`address.value`	The hostname from which the remote connection originates (may be an implicit, special address)	ctor	string
`connectionID.value`	The ID of the `JMSConnectionStats` instance	ctor	string
`host.value`	The explicit hostname from which the remote connection originates	ctor	string
`port.value`	The TCP/IP port from which the remote connection originates	ctor	int
`startTime.value`	`System.currentTimeMillis()` when the request handler was started	ctor	string

D.4.4 JMS Connection Stats

Table D-20 shows the JMS Connection Stats.

The metric table type is JMSConnectionStats.

Table D-20 JMSConnectionStats Metrics

Metric	Description	Update	Unit
`address.value`	The implicit hostname of the remote JMS server host for this connection as specified in the connection factory used to create this connection; set only for non-local connections.	ctor	string
`clientID.value`	The administratively configured (for ctor) or programmatically set (for normal) clientID for this connection	ctor/normal	string
`domain.value`	The JMS domain ("queue", "topic", or "unified") of this connection	ctor	string
`exceptionListener.value`	The stringified name of the current exception listener for this connection	normal	string
`host.value`	The explicit hostname of the remote JMS server host for this connection; set only for non-local connections	ctor	string
`isLocal.value`	"true" if and only if the JMS connection is local to the OC4J JMS server in the same JVM	ctor	boolean
`isXA.value`	"true" if and only if the connection is in XA mode	ctor	boolean
`port.value`	The remote JMS server port for this connection; set only for non-local connections	ctor	int
`startTime.value`	`System.currentTimeMillis()` when this connection was created	ctor	msecs
`user.value`	The user identity for this connection	ctor	string
method-name	An interval timer metric (PhaseEvent Sensor) for every major method call in this connection object.	normal

D.4.5 JMS Session Stats

Table D-21 shows the JMS Session Stats.

The metric table type is JMSSessionStats.

Table D-21 JMSSessionStats Metrics

Metric	Description	Update	Unit
`acknowledgeMode.value`	The acknowledge mode of this session. The valid modes are: `AUTO_ACKNOWLEDGE`, `CLIENT_ACKNOWLEDGE`, `DUPS_OK_ACKNOWLEDGE`, and `SESSION_TRANSACTED`.	ctor	string
`domain.value`	The JMS domain ("queue", "topic", or "unified") of this session	ctor	string
`isXA.value`	"true" if and only if the session is in XA mode	ctor	boolean
`sessionListener.value`	The stringified name of the current distinguished listener for this session	normal	string
`startTime.value`	`System.currentTimeMillis()` when this session was created	ctor	msecs
`transacted.value`	"true" if and only if the session is transacted	ctor	boolean
`txid.value`	The integer count of the current local transaction associated with this session; the counter is increment each time a local transaction is committed/rolledback; not set for non-transacted session	normal	int
`xid.value`	The Xid of the current distributed transaction associated with this session; set to a null/empty string when in a local transaction mode; not set if the session never participates in a global transaction	normal	string
method-name	An interval timer metric (PhaseEvent Sensor) for every major method call in this session object	normal

D.4.6 JMS Message Producer Stats

Table D-22 shows the JMS Producer Stats.

The metric table type is JMSProducerStats.

Table D-22 JMSProducerStats Metrics

Metric	Description	Update	Unit
`deliveryMode.value`	The current delivery mode of this producer. The valid delivery mode values are: `PERSISTENT` and `NON_PERSISTENT`.	normal	string
`destination.value`	The name of the identified destination for this producer; null/empty for an unidentified producer	ctor	string
`disableMessageID.value`	The value is `true` when message IDs are disabled for the producer	normal	boolean
`disableMessageTimestamp.value`	The value is `true` when message timestamps are disabled for the producer	normal	boolean
`domain.value`	The JMS domain ("queue", "topic", or "unified") of this producer	ctor	string
`priority.value`	The current priority of this producer	normal	int
`startTime.value`	`System.currentTimeMillis()` when this producer was created	ctor	msecs
`timeToLive.value`	The current timeToLive of this producer	normal	msecs
method-name	A phase timer (PhaseEvent Sensor) metric for every major method call in this producer object	normal

D.4.7 JMS Message Browser Stats

Table D-23 shows the JMS Browser Stats.

The metric table type is JMSBrowserStats.

Table D-23 JMSBrowserStats Metrics

Metric	Description	Update	Unit
`destination.value`	The name of the destination for this browser	ctor	string
`selector.value`	The message selector for this browser; null/empty string if unspecified	ctor	string
`startTime.value`	`System.currentTimeMillis()` when this browser was created	ctor	msecs
method-name	An interval timer metric (PhaseEvent Sensor) for every major method call in this browser object; calls to "hasMoreElements" and "nextElement" are made on individual enumeration objects, but counted as PhaseEvents in the browser object to simplify data collection, multiple enumerations can be active on the same browser	normal

D.4.8 JMS Message Consumer Stats

Table D-24 shows the JMS Message Consumer Stats.

The metric table type is JMSMessageConsumerStats.

Table D-24 JMSMessageConsumerStats

Metric	Description	Update	Unit
`destination.value`	The name of the destination for this consumer	ctor	string
`domain.value`	The JMS domain ("queue", "topic", or "unified") of this consumer	ctor	string
`messageListener.value`	The stringified name of the current message listener for this consumer	normal	string
`name.value`	The name of the durable subscriber for this consumer; set only for durable topic subscriptions	ctor	string
`noLocal.value`	The noLocal setting of a subscription; set only for topic consumers	ctor	boolean
`selector.value`	The message selector for this consumer; null/empty string if unspecified	ctor	string
`startTime.value`	`System.currentTimeMillis()` when this consumer was created	ctor	msecs
method-name	An interval timer metric (PhaseEvent Sensor) for every major method call in this consumer object	normal

D.4.9 JMS Durable Subscription Stats

Table D-25 shows the JMS Durable Subscription Stats.

The metric table type is JMSDurableSubscriptionStats.

Table D-25 JMSDurableSubscriptionStats Metrics

Metric	Description	Update	Unit
`clientID.value`	The clientID associated with this durable subscriptions	ctor	string
`destination.value`	The name of the topic for this durable subscription	ctor	string
`isActive.value`	"true" if and only if the durable subscription is currently active (being used by a consumer)	normal	boolean
`name.value`	The user-provided name of the durable subscription	ctor	string
`noLocal.value`	The noLocal flag for this durable subscription	ctor	boolean
`selector.value`	The JMS message selector for this durable subscription	ctor	string

D.4.10 JMS Destination Stats

Table D-26 shows the JMS Destination Stats metrics

The metric table type is JMSDestinationStats.

Table D-26 JMSDestinationStats Metrics

Metric	Description	Update	Unit
`domain.value`	JMS domain, "queue"or "topic", of the destination	ctor	string
`name.value`	The configured name of the destination. As defined in `jms.xml`	ctor	string
`locations.value`	A comma-delimited list of JNDI names bound to the destination. As defined in `jms.xml`	ctor	string
method-name	An interval timer metric (PhaseEvent Sensor) for every major method call in the destination object	normal

D.4.11 JMS Temporary Destination Stats

Table D-27 shows the JMS Temporary Destination Stats.

The metric table type is JMSTempoaryDestinationStats.

Table D-27 JMSTemporaryDestinationStats Metrics

Metric	Description	Update	Unit
`connectionID.value`	The ID of the JMSConnectionStats instance from which this temporary destination was created	ctor	string
`domain.value`	JMS domain, for example "queue" or "topic", of the destination	ctor	string
method-name	An interval timer metric (PhaseEvent Sensor) for every major method call in the destination object	normal

D.4.12 JMS Store Stats

Table D-28 shows the JMS StoreStats metric table.

The metric table type is JMSStoreStats.

Table D-28 JMSStoreStats Metric

Metric	Description	Update	Unit
`destination.value`	A pretty-printed name of the JMS destination associated with this message store	ctor	string
`messageCount.value`	Total number of messages contained in this store	lazy	int
`messageDequeued.count`	Total number of message dequeues (transacted or otherwise)	normal	ops
`messageDiscarded.count`	Total number of message discarded after the rollback of an enqueue	normal	ops
`messageEnqueued.count`	Total number of message enqueues (transacted or otherwise)	normal	ops
`messageExpired.count`	Total number of message expirations	normal	ops
`messagePagedIn.count`	Total number of message bodies paged in	normal	ops
`messagePagedOut.count`	Total number of message bodies paged out	normal	ops
`messageRecovered.count`	Total number of messages recovered (either from a persistence file, or after the rollback of a dequeue)	normal	ops
`pendingMessageCount.value`	Total number of messages part of an enqueue/dequeue of an active transaction	lazy	int
`storeSize.value`	Total size, in bytes, of the message store.	lazy	bytes
method-name	An interval timer metric (PhaseEvent Sensor) for every major method call in the message store object	normal

The following identity holds:

messageCount = messageRecovered + messageEnqueued - messageDequeued - messageDiscarded - messageExpired

If a message is both enqueued and dequeued in the same transaction, the messageEnqueued and messageDequeued events occur, but the messageRecovered and messageDiscarded events do not.

D.4.13 JMS Persistence Stats

Table D-29 shows the JMS Persistence Stats.

The metric table type is JMSPersistenceStats.

Table D-29 JMSPersistenceStats Metrics

Metric	Description	Update	Unit
`destination.value`	A pretty-printed name for the JMS destination associated with this persistence file	ctor	string
`holePageCount.value`	The number of 512b pages currently free in this file	normal	int
`isOpen.value`	"true" iff the persistence file descriptor is currently open (for LRU caching)	normal	boolean
`lastUsed.value`	`System.currentTimeMillis()` when this persistence file was last used (for LRUcaching)	normal	msecs
`persistenceFile.value`	The absolute path name of the persistence file used for this persistent destination. This value differs depending on the operating system where OC4J is running.	ctor	string
`usedPageCount.value`	The number of 512b pages currently in use in this file	normal	int
method-name	An interval timer metric (`PhaseEvent` Sensor) for every major method call in the persistence file object	normal

D.5 OC4J Task Manager Metrics

Table D-30 shows the OC4J Task Manager metrics.

The metric table type is oc4j_task.

Table D-30 OC4J_taskManager Metrics

Metric	Description	Unit
`interval.value`	Shows how often the task should run. The task manager executes all the tasks in a round-robin fashion. If the interval is zero, then the task manager executes the task when it is selected in the round robin.	msecs (Milliseconds)
`run().active`	Number of active threads.	threads
`run().avg`	Average time for the taskmanager to run the task	msecs
`run().completed`	Number of times the taskmanager has run the task.	ops
`run().maxActive`	Maximum number of active tasks.	threads
`run().maxTime`	Maximum time for the task to run.	msecs
`run().minTime`	Minimum time for the task to run.	msecs
`run().time`	Total time spent running the task manager	msecs

D.6 Java Object Cache JOC Metrics

Table D-31 shows the top level Java Object Cache metrics.

The metric table type is joc.

Note:

The JOC metrics are only visible when the DMS element of the javacache.xml configuration file is set to the value true.

Table D-31 JOC Java Object Cache Metrics

Metric	Description	Unit
`disk_Size.value`	Total number of bytes of disk consumed by objects in the cache	bytes
`memory_object_count.value`	Total number of objects in the cache	bytes
`memory_size.value`	Total number of bytes of memory consumed by objects in the cache	bytes
`response_q_size.value`	Response Queue size	ops
`task_count.value`	Total number of async tasks	ops
`time_q_size.value`	Time Queue size	ops
`worker_thread_count.value`	Total number of worker threads	threads

Table D-32 shows the java object cache region metrics.

The metric table type is java_cache_region.

Table D-32 Java Cache Region Metrics

Metric	Description	Unit
`disk_Count.value`	Total number of objects in the region on disk
`disk_Size.value`	Total number of bytes of disk consumed by objects in the region
`disk_average_load_time.value`	The average load time for objects in the region
`memory_average_load_time.value`	The average load time for objects in the region
`memory_object_access_count.value`	Total number of access of objects in the region
`memory_object_count.value`	Total number of objects in the region
`memory_size.value`	Total number of bytes of memory consumed by objects in the region