Skip to content
CA Application Performance Management - 9.7
Documentation powered by DocOps

Configure the System

Last update August 16, 2018

You can configure the system:

Grant non-Admin Users Access to Workstation

After you Install or Upgrade the Workstation as an administrator, manually grant access to the non-administrator users to use the workstation. Use the Windows explorer folder sharing properties to grant access the the Workstation.

Configure the Clock Skew Tolerance

The clock skew tolerance is the variation of the clock setting adjustments that are typically done in small increments. For example, setting a slightly fast running computer clock back a few seconds account for the small alterations you see in the logs. The variations in this delay cause the clock to timestamp the same metric at different times. These variations cause the system to get two metric values for the overlapping time period.

Turning back time is something that can happen both explicitly (the operator changes the clock time), or implicitly using automatic clock synchronizations. Automatic clock synchronization is likely to occur with large installations.

Enterprise Manager addresses clock skew tolerance one of two ways:

  • For large clock adjustments the system issues an error and exits.
  • For small backwards clock adjustments, a warning is issued and the harvest cycle is stopped until the timestamp is as far along as it was previously.
    The value of this threshold is set at a default of 60 seconds, but can be configured. A negative clock adjustment of less than seconds results in a warning and temporary harvest suspension, while anything greater results in a system exit.

If the Enterprise Manager detects that your system clock has been set back, the Enterprise Manager delays the harvest cycle until the clock catches up to its previous time. A system clock set back is a common occurrence when a network time service resets a fast CPU clock. A warning message usually appears for this delay. If you want to suppress warning messages, set the optional introscope.enterprisemanager.clock.skewsuppresswarn property. The value of this property is the number of milliseconds the metric harvest is delayed before it is considered serious enough to be logged.

Note: For more information, see KB000045841 Impact of leap seconds on APM.

Follow these steps:

  1. Stop the Enterprise Manager.
  2. Go to <EM_Home>\config directory and open the IntroscopeEnterpriseManager.properties file.
  3. Add the following property, where <XXX> is a number in seconds: introscope.enterprisemanager.clock.skewtoleranceseconds=<XXX>
  4. (Optional) Set the introscope.enterprisemanager.clock.skewsuppresswarn property. The default value is 0 milliseconds.
    Note: This value cannot be set to more than the introscope.enterprisemanager.clock.skewtoleranceseconds property, or it suppresses the warnings and results in a system exit.
  5. Start the Enterprise Manager
    The new settings are applied.

Set the Maximum Data Point Retrieval Property

Set the maximum data point retrieval property for the Command Line Workstation (CLW), Workstation, and WebView. A data point is the value of one metric for a specified period. This Enterprise Manager clamp property limits the maximum number of metric data points retrieved from SmartStor during the execution of a batch of metric queries.

The Enterprise Manager returns the full data for at least the first query in a batch. After you execute a query, you cannot stop it. However, if a query exceeds the maximum data point that is set, subsequent queries in the batch do not execute. Therefore, the clamp can cause missing values in a table that comprise many values produced by a single query and that are executed together in a batch. This property applies to both a clustered Collector or a standalone Enterprise Manager. The introscope.enterprisemanager.query.datapointlimit property is hot configurable.

Note: If you change this clamp property in a cluster, you must change it for all Collectors in the cluster.

Important: If you do not set this property, the following problems can occur:

  • Enterprise Manager overload, which slows CA APM system performance and causes a memory error.
  • Too much data is extracted. Large queries can cause the Enterprise Manager to stop responding.


Follow these steps:

  1. Go to the <EM_Home>\config directory and open the IntroscopeEnterpriseManager.properties file.
  2. Add this property:
    introscope.enterprisemanager.query.datapointlimit=<a data point limit>
    Where the data point limit sets a limit on the data points retrieved from SmartStor per query batch. 

    Important! If you do not set this property, the query is unlimited.

  3. Save the IntroscopeEnterpriseManager.properties file. 
    The setting is applied when the Enterprise Manager checks the property file during a hot deploy. The hot deploy occurs approximately every 60 seconds.

Limit the Data Points Returned by a Query

Set the introscope.enterprisemanager.query.returneddatapointlimit property as a CLW, Workstation, or WebView performance measure. This Enterprise Manager clamp property limits the number of data points that are returned from a batch of metric queries. A data point is the value of one metric for a specified period.

The number of data points that a query returns depends on these factors:

  • Number of matching metrics.
  • Time range that you request.
  • Resolution of the SmartStor data that depends on the configuration of the SmartStor data tiers.

The Enterprise Manager returns the full data for at least the first query in a batch. After you execute a query, you cannot stop it. However, if a query exceeds the maximum data point that is set, subsequent queries in the batch do not execute. Therefore, the clamp can cause missing values in a table comprised of many values that are produced by a single query and executed together in a batch. This property applies to both a clustered Collector or a standalone Enterprise Manager. The introscope.enterprisemanager.query.returneddatapointlimit property is hot configurable.

This clamp differs from introscope.enterprisemanager.query.datapointlimit because it limits the number of data points returned from the execution of a batch of queries. The introscope.enterprisemanager.query.datapointlimit property limits the number of data points fetched from SmartStor during the execution of a batch of queries. The two are not the same because of the retrieved data-point aggregation.

For example, you set up a single metric query for one hour of data with a resolution of one minute. If the data exists in SmartStor with a resolution of 15 seconds, the query matches 240 data points. However, because you asked for a one-minute resolution, the 15-second data points are aggregated into one-minute data points. Therefore, the query returns only 60 data points.

For any given metric, the query returns either the full set of data that is requested or nothing. You never see partial data for a metric. You see either all the data or nothing. The clamp may reduce the number of metrics for which data is returned.

Important! If you do not set this property in a memory-scarce environment, large data queries can cause an Enterprise Manager memory error.


Follow these steps:

  1. Go to the <EM_Home>\config directory and open the IntroscopeEnterpriseManager.properties file.
  2. Add this property:
    introscope.enterprisemanager.query.returneddatapointlimit= <a return datapoint limit>
    Where the data point limit sets the limit of data points that are sent from SmartStor per query.

    Important! If you do not set this property, the query is unlimited.

  3. Save the IntroscopeEnterpriseManager.properties file.
    The setting is applied when the Enterprise Manager checks the property file during a hot deploy. The hot deploy occurs approximately every 60 seconds.

Increase the Number of Query Results Returned

When RangeQuery, PrefixQuery, WildcardQuery, or FuzzyQuery queries are performed they are expanded before the search is performed. For example, if the indexed documents contain the terms "car" and "cars," the query "ca*" expands to "car OR cars" before the search occurs. This might cause a system error if this number exceeds the default value of the term threshold, which is 2048. If this occurs, a TooManyClauses exception will be returned.

Follow these steps:

  1. Go to <EM_Home>\config directory and open the IntroscopeEnterpriseManager.properties file.
  2. Add the following property with a higher value than its default value of 2048: introscope.enterprisemanager.lucene.search.BooleanQuery.SetMaxClauseCount
    The setting is applied when the Enterprise Manager checks the property file during a hot deploy, which occurs about once every 60 seconds.

Configure APM Status Console Clamps and Important Events

You can configure threshold values for Introscope clamp, CA CEM clamp, and Important Event properties in the apm-events-thresholds-config.xml file. When these property thresholds are met or exceeded, the Enterprise Manager notifies the APM Status console. Clamp information displays in the Active Clamps tab. Important Events information displays in the Important Events table view.

Follow these steps:

  1. Go to <EM_Home>/config directory and back up the apm-events-thresholds-config.xml file.

    Important! Do not disregard backing up the file. If a configured file gets corrupted, the property values revert to their defaults and all your configurations are lost.
  2. Configure the properties as appropriate for your environment.

    Note: Configure only the values in the threshold value fields. Do not change any other fields such as clamp ID, importantEvent ID, and description because Introscope uses these fields for system monitoring.
  3. Save and close apm-events-thresholds-config.xml.
    The Enterprise Manager applies your configuration settings during a hot deploy, which occurs about once every 60 seconds.
    If the Enterprise Manager finds a syntax error in the apm-events-thresholds-config.xml file, it takes the following actions:
    • Notifies the APM Status console to display an apm.events.thresholds.config.error important event in the Important Events table view.
    • Reverts all the clamp and Important Event property values to their default values.
  4. If necessary, troubleshoot the apm-events-thresholds-config.xml file.
    1. Open the apm-events-thresholds-config.xml file.
    2. Review the file for syntax errors and make any needed corrections.
    3. Save and close the file.
    4. If you cannot find or fix the apm-events-thresholds-config.xml file error, replace the corrupt apm-events-thresholds-config.xml file with the backup file you saved in Step 1.
    5. Repeat steps 2 and 3.
  5. If you configure Enterprise Managers across a cluster, perform steps 1 through 4 on all Collectors, the MOM, and CDV, if deployed.
    All the Enterprise Managers are configured with Introscope and CA CEM clamp and Important Event properties. The apm-events-thresholds-config.xml file is deployed and clamp and Important Event clamps activate when property threshold levels are met or exceeded.

Was this helpful?

Please log in to post comments.