Managing the data anomaly configurations

The Data anomaly page is used to detect the metric data anomalies. When the anomalies are detected, alarms are sent to the Event Viewer based on the anomaly configurations. The anomaly configurations are flexible and highly customizable.

Procedure

Searching and refreshing the data anomaly configurations

  • Click Administration > Anomaly detection > Data anomaly.
  • Click the Refresh (refresh button) icon to display the latest set of data anomaly configurations.
  • Click the Search (search button) icon and provide a search string to filter a specific data anomaly configuration from the table.

Create a data anomaly configuration

  • Click Create new and enter the following details in the Data anomaly page.
    Field Details
    Job name Provide a name for the anomaly job.
    Script Select the script that you want to run from the list. You can select the default script or the user-defined script. These user-defined scripts are uploaded from the Automation scripts page. See Managing automation scripts.
    Baseline period in days Provide any value between >= 7 and <=30. It is considered as the baseline period for calculating the metric value. For example, if you specify 2, two days data is taken excluding the current and its previous day.
    Resource type Select the Resource type.
    Resource filter
    Note: It is an optional configuration.

    If there is a failure to raise alerts, Resources for the selected Resource type that has KPI values that match the condition are grouped based on first property that is used in the Resource filter field.

    1. Click Condition editor.
    2. Select properties that are associated with Focal Resource type that is selected previously.

      You can see the formula based on the selected properties in the Formula Editor.

    3. Use the following operators to build the formula based on the selected properties:
      • Logical Or (||)
      • Logical And (&&)
      • Equal to (==)
      • Not equal to (!=)
      • Greater than (>)
      • Greater than equal to (>=)
      • Less than (<)
      • Less than equal to (<=)
      Note: Currently, <, <=, >, >= operators are not working as expected in filtering the Resource types. Also, if the expression in the Condition editor contains these characters, the expression is not parsed for filtering.
    4. Click Clear to remove the formula from the editor.
    5. Click Add.

      You can see the formula in the Resource filter field on the Create new page.

    For example, if you selected interface as the Resource type, you might want to use the property and its value as ifIndex='5'. You can create more complex formulas with the available properties in this filter.
    Note: Keep the values within single quotation marks.
    Schedule name Select the schedule name from the following options:
    • EVERY15MIN
    • EVERYDAY
    • EVERYDAYBYHOUR10
    • EVERYDAYBYHOUR2
    • EVERYHOUR
    • EVERYHOURBYMIN10
    Note: Use EVERY15MIN schedule for better performance.
    Time aggregator The following options are available:
    • Average
    • Min
    • Max
    Based on the selection, data for the selected metric is aggregated for the specified baseline period.
    Select metric Select a metric for which you want to analyze the data anomaly.
    Timezone Select the time zone.
    Threshold criteria pane
    Property name Select a property from the list. Properties are populated in the list based on the Resource type selection.
    Property value Provide a value for the property in the text box. Click inside the text area to see the formula. You can build it further.
    (+)(Add Weekly average AND OR Deviation condition) Click + icon to the following fields:
    • Condition
    • Comparison operator
    • Value
    Condition Select either Aggregated value or Deviation % condition.
    Comparison operator Select a comparison operator from the list.
    Value Provide a value for the selected condition.
    Logical operator Select a logical operator. You have AND, OR, None operators. You can select another condition from the Condition field and provide a value for it from the Value field.
    For example, you can build a formula as follows:
    [{ standardDeviation==0.5}]
    Note: It is mandatory to use == operator.
    Add more If you want to add multiple formulas by using multiple properties, follow these steps:
    1. Click outside the curly brackets of the previously added condition.
    2. Click Add more.
    3. Define a new Threshold criteria.

    Click Reset to reset the formulas and start over.

    Enrich properties (Optional) pane
    Resource type The selected Resource type is available in the field and it cannot be edited.
    Properties
    Select the additional properties to be enriched with the Threshold summary. The selected properties appear as follows:
    resource.Vendor,resource.displayName,resource.lastUpdate,resource.ranMarketDesc
    These properties are displayed in the Event Viewer after you add them in the Properties pane.
    Note: For more information, see Configuring metric anomaly detection.
    Note: _active property cannot be added in Tivoli Netcool/OMNIbus schema. If the _active property is used in a job, you might see an error.

Editing a job

  • To update an existing job, click the Edit icon (Edit icon) in the Action column on Batch analytics page.

    You can see the Update job page.

  • Make the changes as needed and click Update, and then confirm.
    The changes are written to the selected job.
  • Click Save as to save the changes to a new job.

Deleting a job

  • Select a single entry or multiple entries from the table and click the (Delete icon) icon to delete and click Confirm.
  • Click the Run (Run icon) icon for a job in the table for which you want the enabled job to run.
  • Click the History (History icon) icon in Action pane for a job to see the run history of the job in the History page.

    Expand the twistie see the details of the entry in the History page.

  • Click the More actions (More actions icon) icon to enable or disable a job.
    Note: By default, the created job is enabled.
    To stop a job, select the job and click Disable.

Export the Data anomaly jobs

  • Select the items from the grid that you want to export to a JSON file and click Export (export button) and save the Data anomaly.json file.
  • Open the Data anomaly.json file in text editor and copy the content in the CONFIG section of the script. Save the constant in a separate file and name it as <Config.json>. Use this file in the Automation jobs page.
    For example, the contents of the <Config.json> file can be as follows:
    [{
                "focalResource": "hoTypeMSC",
                "focalType": "resourceType",
                "jobname": "Test",
                "metricName": "hoTypeMSC.Successful.Basic.Incoming.Handover.Requests",
                "metricType": "raw",
                "periodunitday": "Yesterday",
                "periodunitdayhour": "",
                "periodunitdayvalue": 4,
                "propertiesEnrich": "",
                "resourceFilter": "resource.tenant == 'base'",
                "thresholdCriteria": "[{ vendor == 'Huawei', aggregated_value > 10   }]",
                "timeaggr": "Average",
                "timezone": "GMT"
            }]