System thresholds, Alert notifications, Resource monitoring – Brocade Network Advisor IP User Manual v12.3.0 User Manual
Page 610
![background image](/manuals/361713/610/background.png)
558
Brocade Network Advisor IP User Manual
53-1003153-01
System Monitor support on Network OS VDX platforms
15
•
Fan — Configures fan settings
•
Power — Configures power settings
•
Temp — Displays the threshold for the temperature sensor component
•
CID-card — Displays the threshold for the CID card component
•
SFP — Displays the threshold for the small form factor pluggable (SFP) device
•
compact-flash — Displays the threshold for the compact flash device
•
MM — Displays the threshold for the management module
•
LineCard — Displays the threshold for the line card
•
SFM — Displays the threshold for the switch fabric module
System thresholds
System Monitor monitors the health of each component and, based on the threshold value, each
component can be in a marginal state or a down state. Possible states for all monitored FRUs are
removed, inserted, on, off, and faulty. A state of none indicates the switch is not configured. If the
FRU is removed, inserted, or goes into a faulty state, System Monitor sends a RASlog message or
an e-mail alert.
Refer to the “RAS System Messages” chapter of the Network OS Message Reference for details
about each RASlog message.
Alert notifications
System Monitor provides event notifications by way of RASlog messages or e-mail alerts, depending
on the configuration.
•
RASlog — If a component is in a marginal state or a down state, System Monitor generates a
RASlog message to alert the user. It also generates a separate RASlog message for the overall
health of the switch.
•
Email — An e-mail alert sends information about a switch event to a specified e-mail address.
The e-mail specifies the threshold and describes the event, much like an error message.
Resource monitoring
System Monitor monitors CPU and memory usage of the system and alerts the user when
configured thresholds are exceeded. When the CPU usage exceeds the limit, a system monitor alert
is triggered. The default CPU limit is 75 percent. When configuring memory, the limit specifies a
usage limit as a percentage of available resources.
When used to configure memory, monitoring the limit value must be greater than the low limit and
less than the high limit. Three thresholds are supported for memory monitoring:
•
High_limit — Specifies an upper usage limit for memory as a percentage of available memory.
This value must be greater than the value set by the -limit parameter. The maximum is 90
percent. When memory usage exceeds this limit, System Monitor generates a CRITICAL RASlog
message. The default is 80 percent.