Basis Display Types and Views of the Alert Monitor

Purpose

The Alert Monitor provides two system views (Current Status and Open Alerts), with which you can display the status of your monitored systems. In addition, there are also the display types Alert Browser and Detail Data for displaying more detailed information about open alerts or selected monitoring tree elements (MTEs).

Current Status

This system view displays the current data for every MTE. When you begin to check the system, you begin with this view:

Heartbeat alerts are not displayed in this view (see Triggering a Heartbeat Alert if no Values Are Reported).

Open Alerts

The following image shows the Open Alerts view. This view shows where alerts exist that have not yet been analyzed and set to complete. The color indicator for alerts is set in accordance with the most serious alert that has not yet been processed. Switch to this view to see whether problems have occurred since the last time you used the Alert Monitor.

This view does not necessarily correspond to the current status of the system.

Alert Browser

A third display type is the Alert Browser, which you start through the Alert Monitor. The Alert Browser displays a list of the open alerts in a flat hierarchy. To do this, select the MTEs for which you want to display the alerts that have not yet been completed. The Alert Browser tool is therefore well suited to systematically processing all alerts that have not yet been completed.

Technical Views

The Technical Viewsshould be considered separately from the above display types. In technical views, the displayed information is not current values or alerts for nodes, but rather information about properties and methods of the nodes. You can also display this information for individual nodes by choosing the Properties pushbutton, however you can see the relevant information for all nodes of a monitor in the technical views. There are the following technical views:

· Technical View: Info on MTE

· Technical View: Method Allocation

· Technical View: Status Data Collector

· Technical View: Status Autoreaction

· Displaying the Technical View: Threshold Values

· Technical View: Central Performance History

Displaying Detail Data

You can display the detail data for one or more MTEs from both system views of the Alert Monitor. The display of the detail data shows you:

· The most serious alert that has not yet been processed (if one exists)

· For performance attributes, the current performance data, the associated alert threshold values and the combined, smoothed data for the last 30 minutes and 24 hours

· For status and log attributes, the last message

· The general properties of the selected MTEs

Basis Displaying the Technical View: Info on MTE

Use

MTE classes and attribute groups group nodes of the alert monitoring tree by topic and therefore simplify Customizing, as you no longer need to adjust the properties of all affected monitoring tree elements (MTEs), but rather only of the corresponding MTE class or attribute group. You can find out the assigned MTE class and attribute group by selecting an MTE and choosing Properties. This method is very time-consuming for obtaining an overview of the assigned MTE classes and attribute groups of the MTEs of an entire monitor. The technical view Info on MTE exists as an alternative for this purpose. This view displays the following information for the MTEs of the entire monitor:

· MTE class to which the MTE belongs

· Attribute group to which the MTE belongs (for performance, status, and log attributes)

· Short text that is assigned to the MTE class as help (usually activated using the F1 key)

To additionally display the long help text for an MTE, choose the relevant MTE by double clicking it.

· The nodes are always colored green in the technical view Info on MTE; as it contains no alert information.

· If the corresponding general or specific properties of the MTE are not determined by an MTE class or attribute group, but rather individually for this MTE, no class or group is displayed in this view.

Procedure

To activate the Info on MTE technical view, follow the procedure below:

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Expand the monitor set that contains the monitor that you require, and choose Load Monitor.

3. Choose Views ® Info on MTE. The alert tree now no longer displays the alert status and reported values, but information about the MTEs of the monitor.

Basis Displaying the Technical View: Method Allocation

Use

The individual monitoring tree elements (MTEs) of the Alert Monitor are assigned methods. These methods determine the data collection of the MTEs (data collection methods) and the possible reactions to an alert in the MTE (auto-reaction and analysis methods). You can find out the assigned methods by selecting an MTE and choosing Properties. This method is very time-consuming for obtaining an overview of the assigned methods for the MTEs of an entire monitor. The technical view Method Allocation exists as an alternative for this purpose. This view displays the method assignments for the MTEs of the entire monitor:

· Assigned data collection method: If the MTE is assigned an active data supplier, the system will display in the corresponding column, as the active data suppliers are not defined in the monitoring architecture

· Assigned auto-reaction method

· Assigned analysis method

· The nodes are always colored green in the technical view Method Allocation; as it contains no alert information.

· Summary nodes or monitoring objects can always be assigned methods; these assignments only take effect for the subordinate monitoring attributes, which can inherit their method assignment from the superordinate nodes.

· If no analysis or auto-reaction method is assigned to the MTE, is displayed in the corresponding column here.

Procedure

To activate the Central Method Allocation technical view, follow the procedure below:

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Expand the monitor set that contains the monitor that you require, and choose Load Monitor.

3. Choose Views ® Method Allocation. The alert tree now no longer displays the alert status and reported values, but information about the methods assigned to the MTEs of the monitor.

Basis Displaying the Technical View: Status Data Collector

Data suppliers deliver values that are displayed in the Alert Monitor. They each belong to the individual system components and create monitoring objects that report values to the monitoring architecture. These values are displayed in the monitor sets.

It can be useful to check the status of the data collection methods if problems occur in the monitoring architecture. You can find out the most important information about the assigned data collection methods on one hand by choosing Properties, if you choose the data collection method in the Methods tab page by double clicking it. On the other hand, you can find out the definition and runtime status of the data collection method of an MTE by choosing Display Details ()

This method is very time-consuming for obtaining an overview of the data collection methods and their status for the MTEs of an entire monitor. The technical view Status Data Collector exists as an alternative for this purpose.

In this technical view, you can also manually start the data collection method for any MTE to which a data collection method is assigned. To do this, choose the relevant MTE by double clicking it. This is useful, for example during test phases or for MTEs for which you require the current value, even if it is not yet time for the next start of the method.

Note that when you start a data collection method manually, it runs under your user name and therefore also under your authorizations. As the method runs under the SAPSYS user when it is started automatically, and this user has only very few authorizations, you can assume from a successful test with a manual start that the method will also run without problems during normal operation.

Procedure

To activate the Status Data Collector technical view, follow the procedure below:

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Expand the monitor set that contains the monitor that you require, and choose Load Monitor.

3. Choose Views ® Status Data Collector. The alert tree now no longer displays the alert status and reported values, but information about the data collection methods of the MTEs of the monitor.

Result

The view displays the following information about the data collection methods that are assigned to the individual MTEs of the monitor:

· Assigned data collection method: If the MTE is assigned an active data supplier, the system will display in the corresponding column, as the active data suppliers are not defined in the monitoring architecture

· Type of the data collection method; depending on the definition of the method, the following types are possible:

Type	Description
[Running in agent]	The method is periodically executed in a CCMS agent
[Running in Autoabap]	The method is executed periodically in the dialog process.
[Running in Background]	The method is periodically executed in the background process as a job.
[Kernel function]	The method is automatically executed in the C kernel.
[???]	The method type is not yet known; this type is only displayed for a short type after the method is created.

· Definition status of the method:

Type	Description
PRESET	The method has just been created; the properties of the method are still determined by the settings in the data supplier. After a maximum of five minutes, the system reads the current settings of the method from the active properties variant and changes the definition status.
DBSET WPSET	The settings of the active properties variant are already included in the properties of the method in this status. However, as it is not in the CHECKED definition status, you should check whether the method is released as a data collection method.
CHECKED	The method is released and can therefore be executed.
CEN_CHECKED	The method is executed in a CCMS agent; as all settings for the method are set in the agent, do not change these settings manually.

· Runtime status of the method:

Type	Description
Ready	The method is ready for the next execution; however, this is not yet due.
Run required	The method is to be executed, however the responsible dispatcher has not yet started the method.
Running	The method is currently running; the color of the corresponding attribute is yellow.
Error Fatal Error	An error occurred during the last run of the method. The difference between the two statuses is that a method in the status Error is started again at the next scheduled run, a method in the status Fatal Error is not. A method in the status Fatal Error is only started again after a warm start of the monitoring segment (see Resetting the Segment to WARMUP Status). The color of the corresponding attribute is red. You can also reset the method status by selecting the corresponding MTE, choosing Display Details (), and, on the following screen, choosing Edit ® Method Status ® Reset ® Select Data Collection Method.

The time specification after the runtime status shows the time since which the method has been in this status.

· Frequency of the method call in seconds (Cycle):

The color of the nodes is not passed upward in the alert monitoring tree. Monitoring objects and summary nodes therefore always have the color Green.

Basis Displaying the Technical View: Status Autoreaction CCMS

In the monitoring architecture, you can react automatically to an alert using auto-reaction methods. These methods are automatically started in the case of an alert and by default are executed in the system in which the alert occurs. Almost no assignments are made in the standard SAP system; however, there are several predefined auto-reaction methods in the monitoring architecture that you can assign to any MTE classes.

As of SAP Web Application Server 6.10, you can also define central auto-reaction methods. The auto-reaction methods are not started in the system, in which the alert occurs, but rather in the central monitoring system. In this way, it is possible for reactions to events that occur in monitored components to be performed immediately in a central location.

If problems occur during the execution of auto-reaction methods, it can be useful to check the status of the auto-reaction methods. You can find out the definition and runtime status of an MTE by choosing Display Details (). This method is very time-consuming for obtaining an overview of the auto-reaction methods for the MTEs of an entire monitor. The technical view Status Autoreaction exists as an alternative for this purpose.

In this technical view, you can also manually start the auto-reaction method for any MTE to which an auto-reaction method is assigned. To do this, choose the relevant MTE by double clicking it. This is especially useful during test phases.

Note that when you start an auto-reaction method manually, it runs under your user name and therefore also under your authorizations. As the method runs under the SAPSYS user when it is started automatically, and this user has only very few authorizations, you can assume from a successful test with a manual start that the method will also run without problems during normal operation.

Procedure

To activate the Status Autoreaction technical view, follow the procedure below:

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Expand the monitor set that contains the monitor that you require, and choose Load Monitor.

3. Choose Views ® Status Autoreaction. The alert tree now no longer displays the alert status and reported values, but information about the auto-reaction methods of the MTEs of the monitor.

Result

The view displays the following information for the active monitor about the auto-reaction methods that are assigned to the individual MTEs:

· Assigned auto-reaction method

· Type of the auto-reaction method; depending on the definition of the method, the following types are possible:

Type	Description
[Central Autoreaction]	The method is defined as a central auto-reaction method.
[Running in Autoabap]	The method is executed in the dialog process.
[Running in Background]	The method is executed in the background process as a job.
[???]	The method type is not yet known; this type is only displayed for a short type after the method is created.

· Definition status of the method:

Type	Description
PRESET	The method has just been created; the properties of the method are still determined by the settings in the data supplier. After a maximum of five minutes, the system reads the current settings of the method from the active properties variant and the definition status is changed.
DBSET CEN_SET WPSET	The settings of the active properties variant are already included in the properties of the method in this status. However, as it is not in the CHECKED definition status, you should check whether the method is released as an auto-reaction method.
CHECKED	The method is released and can therefore be executed as a local auto-reaction method.
CEN_CHECKED	The method is released as a central auto-reaction method.

· Runtime status of the method:

Type	Description
Ready	The method is ready for the next execution; however, this is not yet due.
Run required	The method is to be executed, however the responsible dispatcher has not yet started the method.
Running	The method is currently running; the color of the corresponding attribute is yellow.
Error Fatal Error	An error occurred during the last run of the method. The difference between the two statuses is that a method in the status Error is started again at the next scheduled run, a method in the status Fatal Error is not. A method in the status Fatal Error is only started again after a warm start of the monitoring segment (see Resetting the Segment to WARMUP Status). The color of the corresponding attribute is red. You can also reset the method status by selecting the corresponding MTE, choosing Display Details (), and, on the following screen, choosing Edit ® Method Status ® Reset ® Select Auto-Reaction Method.

· The time specification after the runtime status shows the time since which the method has been in this status.

· The color of the nodes is not passed upward in the alert monitoring tree. Monitoring objects and summary nodes therefore always have the color Green.

Basis Displaying the Technical View: Threshold Values CCMS

Threshold values dictate which reported values generate an alert for the node for performance attributes (see Properties of Performance Attributes). To display these threshold values, select the corresponding node and choose Properties. This method is very time-consuming for obtaining an overview of the threshold values for the MTEs of an entire monitor. The technical view Threshold Values exists as an alternative for this purpose. This view displays the following information for the MTEs of the entire monitor:

· Attribute group to which the MTE belongs

· Threshold values of the MTE for which a yellow and a red alert are generated; also the threshold values at which the alert level is lowered again.

· Unit of the reported performance values

· Comparison value that specifies how the value of the performance attribute is calculated from the reported value, such as the last reported value, the average of the last few minutes (smoothing), or the average of the last full quarter hour/hour (average).

· Information, about whether an alert is to be generated when the value of the performance attribute falls below the threshold value or when the value exceeds the threshold value

To additionally display the long help text for an MTE, choose the relevant MTE by double clicking it.

· The nodes are always colored green in the technical view Threshold Values; as it contains no alert information.

· If the properties of the MTE are not specified by the attribute group, but are defined individually for this MTE, the view displays instead of the attribute group.

Procedure

To activate the Threshold Values technical view, follow the procedure below:

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Expand the monitor set that contains the monitor that you require, and choose Load Monitor.

3. Choose Views ® Threshold Values. The alert tree now no longer displays the alert status and reported values, but the threshold values for the MTEs of the monitor.

Basis Displaying the Technical View: Central Performance History CCMS

he Central Performance History of the monitoring architecture allows you to save performance values of the monitoring architecture long-term, and to output these values in reports to compare the current performance data with its earlier development.

The system only collects values for the MTE classes that you have selected for this purpose. You do this by assigning MTE classes to Collection and Reorganization Schemata. These schemata contain information about how long, and in which resolution, performance values should be collected. To obtain an overview of this type of which data is collected in the Central Performance History, you required the collection and reorganization schemata that are assigned to the individual nodes.

The view displays the following information:

· Assigned collection and reorganization schema

· Resolution of the collected performance values

· Collection method (see Collecting and Reorganizing Performance Values)

· Time zone of the data collection (if explicitly specified)

Procedure

To activate the Central Performance History technical view, follow the procedure below:

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Expand the monitor set that contains the monitor that you require, and choose Load Monitor.

3. Choose Views ® Collect./Reorg.Schema Assignment, or choose Display Assignments of the Central Performance History (). The alert tree now no longer displays the alert status and reported values, but the assigned Collection and Performance Schemata.

Processing Alerts CCMS Basis

Purpose

Alerts are assigned methods using the associated monitoring tree elements (MTEs). A method can be a report, a function module, an SAP transaction, or a URL that is to executed as a reaction to an alert. You should differentiate between auto-reaction methods that the system automatically executes as a reaction to an alert (see also Selected Methods of the Alert Monitor, such as an Automatic Alert Notification), and analysis methods that you start manually when analyzing an alert (see Starting Methods).

After you have analyzed an alert and corrected its cause, you no longer need to display the alert in the Alert Monitor: you can complete the alert. This means that it is deleted from the Alert Monitor and saved to an Alert Database. You can display completed alerts from the Alert Database. The Alert Database is also monitored using the Alert Monitor: if the size of the database reaches a defined limit, an alert is triggered. To limit the size of the database, you can reorganize completed alerts.

By default, an alert represents a problem message. You can, however, activate the generation of “green alerts” for selected nodes, that is, an alert of this type for an all-clear or for an explicit message that an action was successful (see Working with All-Clears (Green Alerts)).

See also:

Changing Properties and Method Assignments

Starting Methods CCMS Basis

Use

Monitoring tree elements (MTEs) in the Alert Monitor and alerts in the Alert Browser are assigned methods. You can call these methods to investigate the status of an alert. For example, you could analyze a component of the SAP system with an analysis method, integrate additional monitoring functions into the monitoring architecture with data collection methods, and use auto-reaction methods if you want the system to react automatically to an alert.

Prerequisites

To be able to start all three method types from the monitor display, you must first activate the maintenance functions. If the maintenance functions are deactivated, you can only start analysis methods. The reason for this is that the manual start of an analysis method is part of the everyday work of an administrator, while the manual start of other methods is only necessary more rarely. Possible scenarios are:

· To obtain the current measured value for an attribute immediately, start the associated data collection method manually. This can be especially useful if the method is only called infrequently (for example, if it is called as a job).

· To test an auto-reaction method, start the associated auto-reaction method manually.

Procedures

You can start the methods from the alert monitoring tree, from the Technical Views or the Alert Browser.

Data collection and auto-reaction methods run in the monitoring architecture either in dialog under the SAPSYS user (which has only very restricted authorizations) or in the background under the user that created the corresponding job. If you start methods manually, these methods run in dialog under your user. The method could therefore behave differently to the way it behaves when started automatically by the monitoring architecture. Therefore, if you want to, for example, test new methods, do not rely solely on the results that you obtain from manual starts of the methods.

Starting a Method from the Alert Monitoring Tree

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Choose a monitor from the monitor sets.

3. Choose Extras ® Activate Maintenance Functions.

4. Select an MTE in the monitor.

5. Choose Edit ® Node (MTE) ® Start Method ® Start Analysis Method or Start Data Collection Method or Start Auto-Reaction Method. You can also start the analysis method by double clicking the node.

Starting a Method from the Technical Views

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Choose a monitor from the monitor sets.

3. Choose Extras ® Activate Maintenance Functions.

4. To start a data collection method manually, choose Views ® Status Data Collector. Choose the node for which you want to start the associated data collection method by double-clicking it, and confirm the following dialog window (see also Displaying the Technical View: Status Data Collector).

5. To start an auto-reaction method manually, choose Views ® Status Data Autoreaction. Choose the node for which you want to start the associated auto-reaction method by double-clicking it, and confirm the following dialog window (see also Displaying the Technical View: Status Autoreaction).

Starting a Method from the Alert Browser

...

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.

2. Start a monitor from the monitor sets.

3. Select an MTE in the monitor display.

4. Choose Open Alerts and then Display Alerts.

5. Select and alert, and choose Edit ® Node (MTE) ® Start Method ® Start Analysis Method or Start Auto-Reaction Method. You can also start the analysis method by double clicking the node.

See also:

Defining, Releasing, and Transporting Methods

Completing Alerts CCMS Basis

When you set an alert to completed status, it is removed from the active alerts that are displayed in the Alert Monitor and the Alert Browser. Set alerts to completed as soon as you have solved the underlying problem or have determined that the alert was triggered by a temporary problem and can be ignored.

When you set an alert to completed, it is saved in the Alert Database. The alert is retained there until you delete it by reorganizing completed alerts.

Procedure

1. Choose CCMS ® Control/Monitoring ® Alert Monitor, or call transaction RZ20.
2. Expand the monitor set that contains the monitors that you require and choose Start Monitor.
3. Call the Open Alerts view for this monitor.
4. Start the Alert Browser by selecting a node in the monitor and choosing Display Alerts.
5. In the Alert Browser, select the Alerts that you want to set to completed.
6. Choose Complete Alerts.

See Also:

Automatically Complete Alerts

Automatically Complete Alerts CCMS Basis

An alert is first stored in the monitoring segment of the shared memory. When you complete the alert, the alert is deleted from the monitoring segment and stored in the alert database. In this way, you can ensure that there is always enough place in the monitoring segment. This is especially true for complex IT landscapes with a large number of monitoring tree elements (MTEs). In the Spacesubtree of the CCMS Selfmonitoring monitor, you can check whether there is still enough free space in the monitoring segment.

If there is no more space in the monitoring segment for additional alerts, alerts are automatically completed, using several levels:

· First, the alerts that are beyond the maximum number of alerts to be stored of the MTE responsible are automatically completed (see General Properties of Monitoring Tree Elements).

· If this is not sufficient, alerts of all MTEs of the monitoring segment are automatically completed, starting with those that are above the maximum number of alerts to be stored; if necessary, an increasing percentage of alerts within this number is also automatically completed.

To simplify this task, you can automatically complete alerts. You can select the conditions under which alerts should be automatically completed:

· You can automatically complete alerts which are older than a certain number of days.

· You can automatically complete alerts when the free storage space in the monitoring segment falls below the threshold value for a yellow alert. Beginning with the oldest alerts, the system completes the required number of alerts so that the performance attribute for the free storage space is assigned the color green again. You can find the attribute (FreeAlertSlots) in the Space subtree of the CCMS Selfmonitoringsapurl_li monitor.

You can also ensure that there is enough free storage space in the monitoring segment by reducing the number of reported alerts for each MTE (see General Properties of Monitoring Tree Elements).

Procedure

To automatically complete alerts, you must set the parameters of the CCMS_Space_Collect method. This method is responsible for determining the free storage space in the monitoring segment. Follow the procedure below:

...

1. Choose CCMS ® Configuration ® Alert Monitor, or call transaction RZ21.

2. The system displays the Monitoring: Properties and Methods screen. In the Methods group box, select Method Definitions and choose Display Overview.

3. The system displays a list of method definitions. Choose the method CCMS_Segment_Space_Collect.

4. The system displays the Monitoring: Methods screen. Choose the Parameters tab page. The parameters have the following meaning:

Parameter	Possible Values	Description
CMPL_ALERT_AFTER_DAY		If a value higher than zero is entered here, alerts that are older than days are automatically completed.
CMPL_ALERT_IF_QUOTA	or X	If the value X is entered here, alerts are completed if the free storage space in the monitoring segment falls below the threshold value for a yellow alert.

5. Choose change mode using the Display « Change () pushbutton. Enter the desired parameter values and save your entries.

Result

From now on, alerts are automatically completed in accordance with your settings.