Using Central Monitoring Administration, you can view and manage thresholds.
To view the global thresholds, perform the following steps:
Expand the monitoring solutions list and select the monitor type you want to create a global threshold for.
The following table lists the options you can set:
|Absolute thresholds||Metrics that show unambiguous behavior such as up/down, availability, or capacity violation are good candidates for Absolute thresholds.|
|Signature thresholds||Signature thresholds are more appropriate for metrics that are more transient, such as response time, packet per second, and so on, where the data can exhibit big swings against a generally upward/downward trend.|
|Abnormality thresholds||Abnormality thresholds generate informational events when key metrics go into exceptional states. These events become useful during troubleshooting scenarios using Probably Cause Analysis (PCA), but are otherwise ignored.|
|Severity||All thresholds||Severity of the event created:|
Minor, Major, or Critical (Absolute and Signature thresholds)
Information (Abnormality thresholds)
|Duration||All thresholds||The duration (in minutes) of delay from when the threshold is violated to when an event is generated. The same duration is required to close an event, starting with the first data point where the threshold is no longer violated.|
|Outside Baseline||All thresholds||The baseline to associate with the threshold. This helps to reduce false events associated with latest trends. |
For example, response time may be higher during peak time. In this case, associating the response time with appropriate baseline will cover any false event. If you select Auto Baseline, an event is generated. Other options are Hourly, Daily, Weekly, Hourly and Daily, and All Baselines.
|Comparison||Absolute thresholds||This is a reference point for comparison of the actual measurements with the defined threshold values. The supported operators for comparison are >, <, ==, >=, and <=. For example, if you select the >= operator for an attribute and set the threshold value to 90%, events are generated when the threshold value is greater than or equal to 90%.|
The values for which events are triggered. For example:
|Predict||Absolute thresholds||Select this option to generate predictive events.|
|Auto Close||Absolute thresholds||Clear this option if you do not want to automatically close the triggered event, when the triggered event returns to normal state. This is useful for events that have a very short duration.|
|Above or Below||Signature and Abnormality thresholds|
Reference point for the following parameters:
|Advanced options for Signature and Abnormality thresholds|
|Minimum Sampling Window||Signature and Abnormality thresholds||Define minimum window of time to keep raw data points that will be considered for intelligent threshold computation. Data points outside this range are usually discarded.|
|Threshold||Signature and Abnormality thresholds||The threshold field is used as an additional constraint, which must be met for the signature event to be generated. Usually, a signature event is generated if the data values go outside the baseline range. When the threshold field is set, it additionally requires the data to be above or below the threshold value. This is useful to filter out insignificant abnormalities.|
|Absolute Deviation||Signature and Abnormality thresholds||Deviation is a way of introducing tolerance into the baseline values. If there is a small change in baseline that is not significant for reporting purposes, deviation can be set to ignore this small change. For example, if baseline low value is 40, baseline high value is 100, and deviation is set to 2, then a signature event will be generated when the average of all data values inside the threshold duration goes above 102 or below 38. The deviation value is additive (not multiplicative) and is given in the same units being measured by the particular monitor attribute (that is #,%, bytes, milliseconds, and so on). The default value for absolute deviation is 5 units.|
|Percentage Deviation||Signature and Abnormality thresholds||Percent Deviation has the same meaning as the existing deviation field, except that the actual value of the deviation is expressed in terms of a percent of the baseline value (rather than an absolute value). The default value for percentage deviation is 5%.|