|Oracle® Enterprise Manager Framework, Host, and Services Metric Reference Manual
10g Release 4 (10.2.0.4)
|PDF · Mobi · ePub|
The Oracle Enterprise Manager Framework, Host, and Services Metric Reference Manual (hereafter referred to as the Framework, Host, and Services Metric Reference Manual) lists all the framework, host, and services target metrics that Enterprise Manager monitors. This manual compiles in one place all the framework, host, and services target metric help available online, eliminating the need to have the Grid Control Console up and running.
This preface describes:
This manual contains a chapter for each Enterprise Manager framework, host, and services target for which there are metrics.
The metrics in each chapter are in alphabetical order according to category.
The information for each metric comprises a description, summary of the metric's "vital statistics", data source (if available), and user action. The following list provides greater detail:
Explanation following the metric name. This text defines the metric and, when available, provides additional information pertinent to the metric.
Explains in table format the target version, collection frequency, upload frequency, operator, default warning threshold, default critical threshold, consecutive number of occurrences preceding notification, and alert text for the metric.
How the metric is calculated. In some metrics, data source information is not available.
Suggestions of how to solve the problem causing the alert.
This section provides examples of Metric Summary tables you will see in the Oracle Enterprise Manager Framework, Host, and Services Metric Reference Manual.
|Target Version||Collection Frequency|
|All Versions||Every 15 Minutes|
The following table shows a metric where the server evaluation frequency is the same as the collection frequency.
|Target Version||Evaluation and Collection Frequency||Upload Frequency||Operator||Default Warning Threshold||Default Critical Threshold||Consecutive Number of Occurrences Preceding Notification||Alert Text|
|All Versions||Every 10 Minutes||After Every Sample||>||10000000||12500000||1||Bytes sent by the server are %value%|
The following table shows a metric where the server evaluation frequency is different from the collection frequency.
|Target Version||Server Evaluation Frequency||Collection Frequency||Upload Frequency||Operator||Default Warning Threshold||Default Critical Threshold||Consecutive Number of Occurrences Preceding Notification||Alert Text|
|10.1.0.x||Every Minute||Every 5 Minutes||After Every Sample||>||Not Defined||Not Defined||2||Generated By Database Server|
As previously mentioned, the Metric Summary table is part of the overall metric information. The following table provides descriptions of columns in the Metric Summary table.
|Column Header||Column Definition|
|Target Version||Version of the target, for example, 9.0.2.x and 10.1.0.x. The x at the end of a version (for example, 9.0.2.x) represents the subsequent patchsets associated with that release.|
|Evaluation and Collection Frequency||The rate at which the metric is collected and evaluated to determine whether it has crossed its threshold. The evaluation frequency is the same as the collection frequency.|
|Server Evaluation Frequency||The rate at which the metric is evaluated to determine whether it has crossed its threshold. For server-generated alerts, the evaluation frequency is determined by Oracle Database internals. For example, if the evaluation frequency is 10 minutes, then when the Average File Write Time degrades to the point an alert should trigger, it could be almost 10 minutes before Enterprise Manager receives indication of the alert. This column is present in the Metric Collection Summary table only for Oracle Database 10g metrics.|
|Collection Frequency||The rate at which the Management Agent collects data. The collection frequency for a metric comes from the Enterprise Manager default collection file for that target type.|
|Upload Frequency||The rate at which the Management Agent moves data to the Management Repository. For example, upload every nth collection. The upload frequency for a metric comes from the Enterprise Manager default collection file for that target type. This column is present in the Metric Collection Summary table only when the Upload Frequency is different from the Collection Frequency.|
|Comparison Operator||The comparison method Enterprise Manager uses to evaluate the metric value against the threshold values.|
|Default Warning Threshold||Value that indicates whether a warning alert should be initiated. If the evaluation of the warning threshold value returns a result of TRUE for the specified number of consecutive occurrences defined for the metric, an alert triggers at the warning severity level.|
|Default Critical Threshold||Value that indicates whether a critical alert should be initiated. If the evaluation of the critical threshold value returns a result of TRUE for the specified number of consecutive occurrences defined for the metric, an alert triggers at the critical severity level.|
|Consecutive Number of Occurrences Preceding Notification||Consecutive number of times a metric's value reaches either the warning threshold or critical threshold before a notification is sent.|
|Alert Text||Message indicating why the alert was generated. Words that display between percent signs (%) denote variables. For example,
To reduce the page count in this document, the following abbreviations and acronyms are used:
|Management Agent or Agent||Oracle Management Agent|
|Management Service or OMS||Oracle Management Service|
|Management Repository or Repository||Oracle Management Repository|
A metric is a unit of measurement used to determine the health of a target. It is through the use of metrics and associated thresholds that Enterprise Manager sends out alerts notifying you of problems with the target.
Thresholds are boundary values against which monitored metric values are compared. For example, for each disk device associated with the Disk Utilization (%) metric, you can define a different warning and critical threshold. Some of the thresholds are predefined by Oracle, others are not.
A threshold is reached.
An alert has been cleared.
The availability of a monitored service changes. For example, the availability of an application server changes from up to down.
A specific condition occurs. For example, an alert is triggered whenever an error message is written to a database alert log file.
See Also:See the Oracle Enterprise Manager Concepts manual and the Enterprise Manager online help for additional information about metrics, thresholds, and alerts
Out of the box, Enterprise Manager comes with thresholds for critical metrics. Warning and critical thresholds are used to generate an alert, letting you know of impending problems so that you can address them in a timely manner.
To better suit the monitoring needs of your organization, you can edit the thresholds provided by Enterprise Manager and define new thresholds. When defining thresholds, the key is to choose acceptable values to avoid unnecessary alerts, while still being notified of issues in a timely manner.
You can establish thresholds that will provide pertinent information in a timely manner by defining metric baselines that reflect how your system runs for a normal period of time.
The metrics listed on the Edit Thresholds page are either default metrics provided by Oracle or metrics whose thresholds you can change.
The Specifying Multiple Thresholds functionality allows you to define various subsets of data that can have different thresholds. By specifying multiple thresholds, you can refine the data used to trigger alerts, which are one of the key benefits of using Enterprise Manager.The key in specifying multiple thresholds is to determine how the comparison relates to the metric threshold as a whole. What benefit will be realized by defining a more stringent or lax threshold for that particular device, mount point, and so on?For example, using the Average Disk I/O Service Time metric, you can define warning and critical thresholds to be applied to all disks (sd0 and sd1), or you can define different warning and critical thresholds for a specific disk (sd0). This allows you to adjust the thresholds for sd0 to be more stringent or lax for that particular disk.
To access metrics in the Grid Control Console, use the All Metrics page associated with a particular target by doing the following:
From the Grid Control Console, choose the target.
On the target's home page, click All Metrics in the Related Links section.
On the All Metrics page, choose the metric of interest. If you need help, click Help. The help for that metric displays.
In the unlikely situation that a metric does not report a correct value, you need to determine if the problem is related to the:
Metric providing the wrong values or failing with an error, or
If the problem is after the Management Agent in the execution flow of the metric, that is, the metric value is correct but, for some reason, the data is not reaching the Oracle Management Service.
To aid you in this determination, Oracle provides the Metric Browser; a troubleshooting tool that can be used with Enterprise Manager to see the raw data being collected by the Management Agent.
When enabled, the Metric Browser can be accessed using a web browser, for example, Netscape, Firefox, and Internet Explorer, using a URL of the form:
Note:You can determine the protocol (http or https), the host name, and the Management Agent port that should be used from the output of the following command (run on the Management Agent host):
<agent_home>/bin/emctl status agent
The Management Agent URL, listed in the output to that command, needs only to have browser placed between emd and main.
By default, the Metric Browser is disabled. When the Metric Browser is disabled, you receive the following error:
HTTP Error 403 - Forbidden if the metric browser has not been enabled.
Follow these steps to enable the Metric Browser.
The Metric Browser is enabled by setting the enableMetricBrowser property in the Management Agent's emd.properties file. The location of that file depends on the type of Management Agent you are working with:
For the Grid Control (central|standalone) Management Agent, the file is:
For a clustered (RAC) Management Agent install, the file is:
For the Database Control Management Agent, the file is:
For Application Server Control Management Agent, the file is:
Make a backup copy of the emd.properties file.
Edit the file and locate the line that reads:
#To enable the metric browser, uncomment the following line #This is a reloadable parameter # #enableMetricBrowser=true
Uncomment the line: #enableMetricBrowser=true, so that it reads:
Reload the Management Agent Configuration using the command:
<AGENT_HOME>/bin/emctl reload agent
After reloading the Management Agent, the Metric Browser will be enabled and therefore accessible using a browser.
Running the metric collection outside the Management Agent is specific to each metric and requires a firsthand knowledge of each specific metric. Each metric has its own method of collecting its data and some metrics cannot be run standalone because they are calculated from other metrics.
An example of running the metric collection outside the Management Agent is the command line.