The various features of the Data Collector Service Monitor are configurable in the Portal.

Use the following procedure to configure the processing features of the Data Collector Service Monitor.

1. From the Portal, select Portal Management > CX Settings > Data Collector.
2. Set the configuration settings for the features.

   The following table lists the Data Collection Service Monitor features, configuration settings, and default values.

   Table 1. Service Monitor features and the associated configuration setting in Portal Management

   Setting in Portal Management	Default value	Description
   Service Monitor	Enabled	When enabled, all service monitor functionality occurs according to the configured Service Monitor settings described in the remaining columns of this table.. When disabled, none of the features or functions of Service Monitor occur (including the writing of system information to the log file).
   Service Monitor - Alert when a dimension count of new values in past hour is > [count]	1000	A warning level alert is generated if a dimension has had at least this many new values added in the past hour. When the warning level is reached, it might be time to adjust or disable the dimension, since well-defined dimensions should not grow this quickly.
   Service Monitor - Alert when a fact table grows by > [count] rows in an hour	50,000	A warning level alert is generated if fact data (otherwise known as event data or reporting data) in the data table has grown by at least this many rows in the past hour.
   Service Monitor - Alert when fact data > [hours] old is being aggregated	2	A warning level alert is generated if fact data (otherwise known as event data or reporting data) currently being aggregated by the Data Collector fact aggregation process is at least this many hours older than current hour.
   Service Monitor - Alert when logged errors > [count]	2	If the count of errors detected by the service monitor during the check time-frame (as specified in Service Monitor - Check errors/warnings from current time through [minutes] back) is more than this number, and the difference between the min and max timestamp of those errors is at least the number of minutes in the configured time span (as specified in Service Monitor - Errors/Warnings must span at least [minutes] to trigger alert), an alert is triggered. Also, the Data Collector Logging Level will be raised to ALL if it is currently lower than that.
   Service Monitor - Alert when logged warnings > [count]	2	If the count of warnings detected by the service monitor during the check time frame (as specified in Service Monitor - Check errors/warnings from current time through [minutes] back) is more than this number, and the difference between the min and max timestamp of those errors is at least the number of minutes in the configured time-span (as specified in Service Monitor - Check errors/warnings from current time through [minutes] back), an alert will be triggered. Also the Data Collector Logging Level will be raised to ALL if it is currently lower than that.
   Service Monitor - Check errors/warnings from current time through [minutes] back	60	Each time Service Monitor checks the log for errors and warnings (every x minutes configured in Service Monitor - Check frequency [minutes]), it will check from current time back this many minutes.
   Service Monitor - Check frequency [minutes]	30	Execute all enabled service monitor checks this frequently.
   Service Monitor - De-activate a fact when table grows by > [count] rows in an hour	75,000	If a fact aggregation table has grown by at least this many rows in the past hour, it will be set to inactive and a warning level alert is generated. The fact must be re-activated via the Event Manager.
   Service Monitor - Display messages with alert(s) in portal to Admin level users	Enabled	Triggered alerts will display upon portal login to Admin level users in the Announcement section in the lower right-hand corner of page. These announcements are set to expire after 2 hours.
   Service Monitor - Display messages with alert(s) in portal to all users	Disabled	Triggered alerts will display upon portal login to any level of portal user in the Announcement section in the lower right-hand corner of page. These announcements are set to expire (stop displaying upon login) after 2 hours.
   Service Monitor - Email address for Data Collector Service Monitor alerts	no default value	Service Monitor will email warnings/errors according to the configured settings in Service Monitor - Email top [count] of log entries per DC component, to the email address specified in this property. Additionally, the email is sent to the configured Tealeaf administrator email address (in Portal Management > CX Settings > Miscellaneous), which may be the same or different than this one.
   Service Monitor - Email top [count] of log entries per DC component	10	If the number of warnings or errors logged according to the configured settings triggers an email, the first [count] warning or error entries for each component (for example, Data Collection, Dimension Extraction, Fact Aggregation, Data Manager, etc) will be included in the email generated by the Data Collector Service Monitor.
   Service Monitor - Errors/Warnings must span at least [minutes] to trigger alert	10	Within the check time-frame (as specified in Service Monitor - Check errors/warnings from current time through [minutes] back) the earliest and latest errors or warnings must be at least this many minutes apart to trigger an alert. This is to prevent alerts occurring due to conditions like a service restart, in which it is normal to see several warnings and possibly errors occur very close together as connections are closed and threads are terminated.

3. Configure Service Monitor Log Data - Days Retain.

   Number of days to retain the Data Collector Service Monitor data. The default value is 30.

   You can find this configuration setting on the Data Retention page at Portal Management > CX Settings > Data Retention.