Properties for XPI Service: AF Core

adapterMonitor.monitoringMode

Possible values: cpa or registered

Default: cpa

This property specifies how the adapter monitor tries to identify the available adapters. If the value is set to cpa, it checks which adapter type metadata schemas are available in the CPA Cache service.

The list of available adapters can also be generated. This list is based on the registered adapters of the AF Core service adapter monitor.

This property must not be changed.

administration.expireActivationStatusEntriesAt

Type: Time of day

Default: 02:30:00

This property specifies the time of day at which the daily cleanup job for history entries for the channel activation status runs.

administration.keepActivationStatusHistory

Value type: Value in days

Default: 180

This property specifies how long history entries for the channel activation status should be kept in the database. Entries older than the specified number of days are deleted by a daily cleanup job.

administration.processHistoryLength

Value type: Integer

Default: 25

This property configures the number of process status entries kept in memory for each channel in communication channel monitoring.

administration.processHistoryCutoff

Value type: Value in minutes

Default: 60

This property configures the time duration after an error has happened that the channel will show green status again.

administration.useCache

Value type: Boolean

Default: true

This property specifies if the information whether a channel is started or stopped is cached in memory to optimize performance.

This property must not be changed.

af.datasource

Default: jdbc/SAP/BC_XI_AF

This property must not be changed.

alerting.maxConnections

Value type: Integer

Default: 5

This property specifies the maximum number of connections from the adapter framework to the central monitoring server.

alerting.connectionTimeout

Value type: Value in milliseconds

Default: 600000 (10 minutes)

This property specifies the timeout for an existing connection from the adapter framework to the central monitoring server.

alerting.events.chunkLength

Value type: Integer

Default: 200

This property specifies the chunk length of a sequence of messages.

More information: Maintaining Alert Engine Processes

alerting.events.highVolumeThresholdInterval

Value type: Value in milliseconds

Default: 4000

This property specifies the threshold interval for high volume. If for two consecutive chunks the sending time is not within the threshold interval, the alert engine switches to emergency mode.

More information: Maintaining Alert Engine Processes

alerting.maxWaitTime

Value type: Value in milliseconds

Default: 5000

This property specifies the maximum time a call waits for an available connection. After the specified time, the call fails and an exception is raised.

alerting.suppressInterval

Value type: Value in milliseconds

Default: 50000

This property defines a time interval in which alerts with the same error are suppressed. Only one alert is generated. If more errors occur with the same error source, they do not generate alerts. This reduces the amount of alerts in a system.

alerting.target

Value type: 0,1

Default: 0

This property specifies whether the message-based alerting is switched on (1). If the value is set to 0, the classic alerting function is active.

alerting.isActive

Value type: Boolean

Default: true

This property specifies whether the alerting is active (true) for the runtime engine.

More information: Configuring Component-Based Message Alerting

clusterCommunicationTimeoutMsec

Value type: Value in milliseconds.

Default: 15000 (15 seconds)

It specifies the timeout value for the cluster communication in AS Java. Even though the default value of 15 seconds should be sufficient for the adapter monitoring functionality, this value may be increased in cases where you are experiencing cluster timeout exceptions of the cluster communication mechanism.

idmapRemover.checkInterval

Value is in minutes.

Default: 1440 (i.e. once a day)

The AF Core service tries to remove persisted IDMapper entries with expired persist durations from the database at system startup and then repeatedly after the configured time interval.

Jarm.[x].enabled

Value type: Boolean

Default: true

These properties specify for which components and adapters within the adapter framework the performance measurement (JARM) should be performed.

This property must not be changed.

license.aggregation.DB.access

Type: Integer

Default: 1

This property specifies the persistence layer to be used, where the default value 1 stands for database.

license.aggregation.hourly.weighting

Type: Integer

Default: 4

This property is used to speed up the hourly aggregation of message payload data for license auditing. It specifies how often the aggregation should run within one hour.

The default value 4, for example, results in an aggregation run every 15 minutes.

license.aggregation.hourly.procNum

license.aggregation.daily.procNum

license.aggregation.monthly.procNum

Type: Integer

Default: 1

These properties specify the number of times an aggregation is to run for each aggregation period.

The default value 1 results in one aggregation run every hour, day, or month. You can increase this value if there is more runtime data available than can be handled by just one aggregation run in the given period of time.

license.aggregation.recurrence.interval

Type: Integer (millisecond)

Default: 300000

This property specifies the time interval (for example, every 5 minutes), after which the system checks for scheduled (hourly, daily, or monthly) aggregations.

All detected scheduled aggregations are executed.

messaging.connectionDefinition

This property specifies a list of connection definitions with the following form:

(name=global, messageListener=localejbs/AFWListener, exceptionListener= localejbs/AFWListener,, pollInterval=60000, pollAttempts=60, Send.maxConsumers=5, Recv.maxConsumers=5, Call.maxConsumers=5, Rqst.maxConsumers=5)

In the SAP PI installation environment the message listener bean names must not be changed at any point.

The global property set defines the defaults for all adapter connections (queues). The name of the connection is retrieved from the CPA cache. You can set specific values by adding a property set for the respective adapter connection.

The pollInterval and pollAttempts parameters define how often and with which retry interval the messaging system tries to retrieve the listener beans from JNDI at system startup. This is required because the messaging system is implemented as a Java service and the message and exception listener beans are implemented as part of Java applications, which are started in a second phase during server startup. In the case of a very slow server startup, it may be necessary to increase the pollInterval or pollAttempts values in order to successfully start the four queues for sending and receiving the synchronous and asynchronous messages.

The maxConsumer parameters define the number of threads to be used to process messages in parallel for each queue type and for each connection definition.

More information: Java Service Properties that May Require Changes .

profile.performance.cache.size

Type: Integer

Default: 500

This property specifies the maximum number of cache objects for performance data.

profile.performance.runtime

Type: Boolean

Default: false

This property enables the measurement of performance data for the Adapter Engine.

profile.runtime

Type: Boolean

Default: true

This property enables the message overview function.

profile.runtime.interval

Type: Integer (minutes)

Default: 15

This property specifies the minutes for the periodical execution of the aggregation job for the message overview.

profile.runtime.sync.statistics

Type: Boolean

Default: true

This property enables or disables synchronization during system startup of the Message Overview message status data for "hanging" messages with message status information from messaging system store.

profile.runtime.sync.statistics.lastMinutes

Type: Integer (minutes)

Default: 1440 minutes (24 hours)

When the message overview synchronization runs, it synchronizes the message overview status only for the messages in scheduled status, which are processed in the latest "lastMinutes" minutes.

transactionManager

Default for AS Java: ts

JNDI name of the AS Java transaction manager.

This property must not be changed.

updateCPACacheJobInterval

Default: 1440 (minutes) The task will be executed after 1440 minutes (1 day).

If communication.type for XPI Service: CPA Cache Service is set to new , then updateCPACacheJobInterval is to schedule a task to automatically recover missing CPA objects on any cluster node.

More information: Properties for XPI Service: CPA Cache