Raritan Computer COMMANDCENTER NOC - Manual (Page 138)

126 COMMANDCENTER NOC ADMINISTRATOR GUIDE
on demand. The remaining item is determining when performance metrics have reached a point at
which they should be acted upon. And with the CC-NOC’s capability of managing performance
thresholds, that’s easy too!
Thresholding
An exciting new feature significantly improved with the CC-NOC is threshold alerts. This
allows the CC-NOC to notify you of potential problems pro-actively, before they occur, based
on performance metrics gathered by the CC-NOC through SNMP and WMI.
How it works
The CC-NOC gathers performance data directly from managed devices using SNMP, and
through a proxy system for Windows Servers, and 5 workstations, using a CC-NOC appliance.
Each time the data is collected or reported, the CC-NOC compares certain data points against
configurable threshold values. Click the Admin tab, Network Management, Configure
Performance Thresholds for details on how to configure thresholds. If the value is higher or
lower, depending on the type of threshold, than the threshold value, an event will be generated.
This event can be configured for notification through the notification configuration option under
the Admin tab.
There are four key pieces to threshold monitoring: type, value, trigger, and rearm.
Type
A threshold can be a high or low threshold. A high threshold means that an event will be
generated if the actual value is higher than the threshold value. Conversely, a low threshold will
cause an event to be generated when the actual value reported is lower than the threshold value.
Value
Value refers to the point at which the threshold is exceeded, whether it is a low or high threshold.
For instance, if the value for a high threshold is 80, any reported value over 80 will exceed the
threshold.
Trigger
The trigger is the number of times in a row the threshold must be exceeded before an event is
generated. If a trigger is set at 3, for example, the threshold must be exceeded for three
consecutive reports before an event is generated.
Rearm
After an event is generated, no further events will be generated for the same threshold until the
rearm value is reached. The rearm value protects you from a flurry of events when the value
bounces around the threshold value—just above and just below. The rearm value must be
reached in the opposite direction of the threshold type. For example, if a disk drive had a high
threshold of 90% utilization and no rearm value, an application that wrote a temporary file and
deleted it on exit might cause the utilization to bounce between 89% and 90%. The rearm value
protects you from this condition by enforcing that the utilization problem must be addressed, as
opposed to temporarily repaired. Until the utilization drops below 80%, you will not receive
another notification.