Intel 480T, Руководство пользователя

Страница: 287 / 367

C H A P T E R 1 5 Status Monitoring and Statistics
285
Alarms
The Alarms group provides a versatile, general mechanism for
setting threshold and sampling intervals to generate events on any
RMON variable. Both rising and falling thresholds are supported,
and thresholds can be on the absolute value of a variable or its delta
value. In addition, alarm thresholds may be autocalibrated or set
manually.
Alarms inform you of a network performance problem and can
trigger automated action responses through the Events group.
Events
The Events group creates entries in an event log and/or sends SNMP
traps to the management workstation. An event is triggered by an
RMON alarm. The action taken can be configured to ignore it, to
log the event, to send an SNMP trap to the receivers listed in the trap
receiver table, or to both log and send a trap. The RMON traps are
defined in RFC 1757 for rising and falling thresholds.
Effective use of the Events group saves you time. Rather than
having to watch real-time graphs for important occurrences, you
can depend on the Event group for notification. Through the SNMP
traps, events can trigger other actions, providing a mechanism for
an automated response to certain occurrences.
Configuring RMON
RMON requires one probe per LAN segment, and standalone
RMON probes have traditionally been expensive. Therefore, Intel’s
approach has been to build an inexpensive RMON probe into the
agent of each system. This allows RMON to be widely deployed
around the network without costing more than traditional network
management. The switch accurately maintains RMON statistics at
the maximum line rate of all of its ports.
For example, statistics can be related to individual ports. Also,
because a probe must be able to see all traffic, a stand-alone probe
must be attached to a nonsecure port. Implementing RMON in the
switch means that all ports can have security features enabled.