Split logging

To create a separate log file for each sensor, topology agent or business application handler (grouping pattern), which improves the readability of the sensor, topology agent or business application handler logs, you can use split logging. However, split logging can affect TADDM performance.

Sensor log files

To use split logging, set the value of the following property to true in the collation.properties file, if it is not already set (the default value is true): 
com.ibm.taddm.log.split.sensors=true

If this property is not set to true, the default logging for all sensors is put in the $COLLATION_HOME/log/services/DiscoveryManager.log file.

When it is used, split logging applies to the TADDM server and the anchor servers, and the split log files are placed in the following directories:
On the TADDM server
The following directory:
  • $COLLATION_HOME/log/sensors/runid/sensorName-IP.log

    The runid includes the date and time of the discovery. The sensorName-IP (the log file name) includes the sensor name and the IP address of the target system. The following example is a log file on the TADDM server: $COLLATION_HOME/log/sensors/20070621131259/SessionSensor-10.199.21.104.log.

On the anchor server
One of the following directories:
  • ANCHOR_DIR/log/sensors/runid/sensorName-IP.log
  • ANCHOR_DIR\log\sensors\runid\sensorName-IP.log

To set the split log files cleanup properties for TADDM server, see Generic properties.

Because anchor servers run for only short periods of time, if the split log files are accumulating on anchor servers, they must be deleted manually.

Topology agent log files

To use split logging, set the value of the following property to true in the collation.properties file, if it is not already set (the default value is true): 
com.ibm.taddm.log.split.agents=true

If this property is not set to true, the default logging for all topology agents is put in the $COLLATION_HOME/log/services/TopologyBuilder.log file.

When it is used, split logging applies to the TADDM server, and the split log files are placed in the following directory: 
$COLLATION_HOME/log/agents/agentName.log
The agentName (name of the log file) contains the topology agent name. The following example is a log file on the TADDM server: 
$COLLATION_HOME/log/agents/AppDescriptorAgent.log

Business application handler (grouping patterns) log files

To use split logging, set the value of the following property to true in the collation.properties file, if it is not already set (the default value is true): 
com.ibm.taddm.log.split.bizapp=true
The split log files are placed in the following directories:
dist/log/bizapps/<pattern>/<starttime>.log
where:
  • <pattern> is the identifier of the grouping pattern (name-GUID), for example J2EE App pattern-F12AC23451AB3A4FAF58E9187ABF1169.
  • <starttime> is the start time of BizAppsAgent run, for example 1416408057548.

If the com.ibm.taddm.log.split.bizapp property is set to false, all log files are placed in one file, the log/services/PatternsSchedulingService.log file.

Generic properties

The split log files are deleted automatically after a specified period of time. You can use the following generic properties to modify the log retention before cleanup, the cleanup hour and the interval between cleanups.
  • com.ibm.taddm.log.split.retention=7 - the log retention before cleanup specified in days. The default value is 7.
  • com.ibm.taddm.log.split.clean.hour=3 - the cleanup hour specified in the hour of the day. The default value is 3.
  • com.ibm.taddm.log.split.clean.interval=24 - the interval between cleanups specified in hours. The default value is 24.
If you want to modify the log retention property for a specific group, set the following property:
  • com.ibm.taddm.log.split.retention.group=7
where group can have the following values: sensors for sensor log files, agents for topology agent log files and bizapp for the business application handler log files.