Show TOC

Procedure documentationEditing Metric Types

Procedure

The templates delivered by SAP contain a lot of metric types for the four categories of the E2E Monitoring and Alerting Infrastructure (MAI). The categories are:

  • Availability

  • Performance

  • Exception

  • Configuration

  • Self-Monitoring

    Note Note

    The category Self-Monitoring contains information about MAI itself. Information about the status of the managed objects is kept separate from information about MAI.

    End of the note.

If you have based custom templates on an SAP default template, the custom template contains these metric types and their attributes. To list the metric types, choose the template and go to the Metrics tab.

To change the attributes of a metric type, go to change mode, select the metric type, and choose Change Settings. The most important settings and their meaning are described below.

Note Note

This procedure is in the Create and Edit Custom Templates section of the Using Templates process. Perform the remaining steps of the procedure as well.

End of the note.
Activating and Deactivating Metric Types

Not all metric types of SAP templates are active by default. Values are not collected and displayed in System & Application Monitoring for inactive metric types. If some metric types of a product are missing, check whether they are flagged as inactive in the list, and if so, activate them. You can similarly deactivate metric types which you do not want to monitor.

To activate or deactivate metrics in a custom template, set the flag in the list of metric types, and save.

Note Note

If all metric types of an event are inactive, the event is inactive. The applies similarly to alerts.

End of the note.
Change Metric Type Description

SAP delivers metric types with a meaningful description. You can then create an additional custom description to document company-specific information about the metric type, such as responsibilities or process information.

To create a custom description for a metric type, proceed as follows:

  1. Select a metric type and choose Change Settings, in change mode.

  2. In Metric Type Details, choose the Overview tab.

  3. Choose Create Custom Description, and create the description.

  4. Save your changes.

Specify Data Usage

The metric types whose values are determined for the E2E Monitoring and Alerting Infrastructure, can be sent to two data users:

  • Event calculation engine

    Active metric types are monitored by System & Application Monitoring. The Event Calculation Engine (ECE) calculates status, events and alerts from the values collected and the configuration settings in the alerting directory. If you want to use this metric type, set this flag.

    Note Note

    Only deactivate the flag if you do not need a metric type in the system monitoring of technical monitoring, but do want to save its history data in the SAP NetWeaver Business Warehouse (see below). Such metrics are displayed in gray in technical monitoring. If you do not need either data users, deactivate the entire metric type (see above).

    End of the note.

  • SAP NetWeaver Business Warehouse

    Active metric types are saved in the SAP NetWeaver Business Warehouse as historical data, so that you can display the development of the value over time. There are applications for this in System & Application Monitoring.

    If you want to save the history of the metrics in BW, make the required settings under Granularity. The BW Mapping Rule is technical. Don't change it for a metric type:

    • Short No aggregation, only the data of the last 24 hours, in the highest possible temporal resolution.

    • Medium Like Short, but with hourly aggregates, which are saved for 30 days by default.

    • Long Like Medium, but with daily, weekly and monthly aggregates. This setting is required for the metric types which are displayed in interactive reporting.

    • Optimized for Metric Monitor Default delivered setting, which corresponds to Medium. You can configure individually, if the system load to save in BW is too great, but only do so in consultation with SAP Support.

    You should not usually need to change this setting, except in the following special cases:

    • You do not use the history data in BW for some metric types. In this case, clear the Send Values to SAP NetWeaver Business Warehouse flag for these metric types.

    • You have added your own reports to reporting. In this case, assign the Granularity value Long to the metric types.

    • You use the metric monitor to monitor the long-term development of selected metric types, over more than the last month. In this case too, assign the Granularity value Long to the metric types.

Specify Threshold Values and Type

The ECE needs calculation rules, to be able to determine the status of a metric from its collected values. To make these settings, select a metric type, and choose Change Settings in change mode, and go to the Thresholds tab. Make the changes and save.

These rules are distinguished by threshold types. For a detailed description of the types, choose a type from the Threshold Types and the Note pushbutton to its left. Here is just a rough explanation of the groups of types:

Threshold type

Description

  • Already Rated

  • Already Rated (Best rating of last n measurements)

The status here is already assigned in the collector of the metric value, not in the ECE. You can choose between two types for which one measurement, or a specified minimum number of measurements, with the specified status, trigger an event.

  • Info Only

Use this threshold type if the metric value is to be displayed in System & Application Monitoring, but no explicit rating (Red/Yellow/Green) is necessary.

  • Counter Threshold

  • Delta Threshold

  • Numeric Threshold (Green/Red)

  • Numeric Threshold (Green/Yellow/Red)

  • Numeric Threshold (GtoY/YtoR/RtoY/YtoG)

  • Numeric Threshold (High water mark)

  • Range Threshold

These threshold types represent different ways of handling numeric thresholds. They differ in details such as the existence of a yellow status, existence of hysteresis, or the number of values required for the status to change.

For more information, call the help text mentioned above, for the type.

  • Text Threshold

  • Text Threshold (Regular Expression)

These threshold types search for character strings in metric types with a text value. The text is searched for this character string in a specified sequence.

Each threshold type has threshold values or other parameters which determine the status of the metric for each value, for example the threshold value from which a response time has the status red. You can adjust these values to your requirements, to ensure that events or alerts are created when a problem occurs, to which you must react.

It is less often necessary to change the threshold type. Reasons why it could be necessary to change the threshold type include:

  • You want to display the value of a metric in System & Application Monitoring, but do not want it to trigger events or alerts. In this case, change the type to Info only.

  • You use a metric of type Already Rated, but want an event to only be generated when there are several such values. In this case, change the type to Already Rated (Best rating of last n measurements).

  • You want a metric with numerical threshold values to only be able to have the statuses red and green, because your processes only recognize the distinction between „OK“ and „not OK“. In this case, change the type to Numeric threshold (Green/Red).

  • You want to switch between threshold types with and without hysteresis (see the description of the type Numeric threshold (GtoY/YtoR/RtoY/YtoG)). The type Numeric threshold (Green/Yellow/Red) without hysteresis is easier to follow, the type with hysteresis prevents flickering statuses when the measured value varies around the threshold value.

Note Note

You can also create your own metrics, in a custom template. Go to Expert Mode and choose Start of the navigation path Create Next navigation step Metric End of the navigation path.

Monitoring Log Files with the Monitoring and Alerting Infrastructure contains an example.

End of the note.