ServerAssist can collect the value of any performance counter for Windows, or any numeric OID for a Linux system or network hardware. Furthermore, application monitoring creates a number of synthetic performance values describing the behaviour of the application.
The collected performance data can be displayed on the performance chart, but this requires visual analysis to see important changes in the values of collected items. ServerAssist can monitor the collected performance data, and respond when the value for a performance item crosses a configured threshold.
For any performance value, two thresholds divide the value into three or two categories, depending on whether a normal value is below or above the thresholds, or in between them.
Should a value cross one of the thresholds, a notification is generated. You can either be told immediately, or limit the notifications by allowing brief changes to be ignored, provided that they self-correct in a few minutes.
While any theshold can be configured manually, policies make applying standard thresholds quick and simple.
Integrated With Core Health
Any performance item value can be selected to be part of the system's core health monitoring, with the fixed thresholds used as the basis for the state of the item.
Fixed thresholds are very useful in situations where there are known limits for a performance value. However, it is common that a rapid change in a value, where it deviates significantly from previous values, while not being a problem in and of itself, nevertheless indicates a situation that warrants investigation.
For example, if a system with a 1TB disk had 500GB free, and in the space of a few minutes the free space dropped to 300GB, there is still plenty of space remaining, but the change is probably unexpected and would merit examination.
ServerAssist's dynamic thresholds cater for this situation, notifying you when a value changes in an unusual way.
pre-defined limits and unexpected changes are both covered