Dell POWEREDGE M1000E - Manual (Page 28)

8 Fabric Watch Administrator’s Guide
53-1002752-01
Logical switch support
1
Switch policies
Switch policies are a series of rules that define specific health states for the overall switch. Fabric
OS interacts with Fabric Watch using these policies. Each rule defines the number of types of errors
that transitions the overall switch state into a state that is not healthy. For example, you can specify
a switch policy so that if a switch has two port failures, it is considered to be in a marginal state; if it
has four failures, it is in a down state.
You can define these rules for a number of classes and field replaceable units, including ports,
power supplies, and flash memory.
See “Switch status policy planning” on page 82 for information on configuring switch policies.
See Chapter 10, “Fabric Watch Reports,” for information on viewing the current switch policies
using the Switch Status Policy report.
Logical switch support
Fabric Watch can monitor the switch health on eight logical switches. You can configure thresholds
and notifications for ports that belong to a particular logical switch. Each logical switch has its own
Fabric Watch configuration and triggers notifications based on its local configuration.
Fabric Watch supports port movement from one logical switch to another. Whenever a port is
moved, thresholds associated with the port are deleted from which the logical switch from which
the port was moved, and created for the logical switch to where the port is moved.
A logical interswitch link (LISL) is the logical portion of the physical connection that joins base
switches. You can enable or disable port thresholds and create thresholds for state changes on
LISLs, but Fabric Watch does not support other threshold areas such as link loss or signal loss for
LISLs as it does for normal E_Ports.
Threshold monitoring using SNMP tables
Understanding the components of SNMP makes it possible to use third-party tools to view, browse,
and manipulate Brocade switch variables remotely. Every Brocade switch and director supports
SNMP.
When an event occurs and its severity level is at or below the set value, the Event Trap traps
(swFabricWatchTrap), are sent to configured trap recipients.
Once the switch status policy changes, Fabric Watch sends a connUnitStatusChange SNMP trap.
Any Fabric Watch RASLOG is converted into an swEventTrap.
Refer to the Fabric MIB Reference for information about the following:
• Understanding SNMP basics
• How to enable or disable the sending of traps from the various MIBs
• SNMP trap bit mask values
• Loading Brocade management information bases (MIBs)