Performance Network Alarms

This article lists the Performance: Network alarms of a role, with what each alarm monitors and its factory defaults. Network alarms cover the data flow through the interface card. They focus primarily on processing rates as information is routed. This should not be confused with server rates or statistics.

The Performance: Network category is on the Alarms tab of a role (Configure > System Configuration > Roles). To turn these alarms on and change their limits, times, and profiles, see Configure Alarms in a Role.

In the example alarm configuration called out below: If % Net Utilization is 30% or higher for 120 seconds a Yellow level alarm is triggered. If % Net Utilization is 50% or higher for 240 seconds a Red level alarm is triggered.

Performance Network Alarms and Defaults

The Default column shows the factory values. A role can change them.

Alarm	Description	Default
Bandwidth (Megabits)	Monitors the estimated current bandwidth of the network interface measured in megabits per second. For interfaces that do not vary in bandwidth or for those where no accurate estimation can be made, this value is the nominal bandwidth.	Off. Yellow 100 Mb for 120 s, Red 100 Mb for 240 s.
Net Broadcasts / sec	Monitors the rate of receipt of network packets whose destination address is a broadcast address. All nodes on the network segment receive packets sent to the broadcast address. Communications between network routers regarding the network topology, advertisements from various systems, and certain protocols used in translating names to addresses are all examples of broadcast packet use. High broadcast packet rates are especially troublesome because they consume network bandwidth and resources (CPU, memory, and I/O) on all systems on the network segment (since they all receive such packets). Lower is better.	On. Yellow 20 for 120 s, Red 40 for 240 s.
Net Kilobytes / sec	Monitors the total number of bytes received per second on a network interface. The total received bytes are counted independent of the number of packets in which the data is received. Network bytes per second are useful in measuring the traffic load on a network interface. Lower is better.	Off. Yellow 1250 for 120 s, Red 1500 for 240 s.
Net Output Queue	Monitors the length of the network output queue in packets. This is an instantaneous count, which is not averaged over time.	On. Yellow 1.5 for 120 s, Red 2 for 240 s.
Net Packets / sec	Monitors the total number of packets received per second on a network interface. Each physical packet counts as one frame, independent of the number of bytes contained in the packets. Network packets per second are useful in measuring the traffic load on a network interface. Lower is better.	On. Yellow 1000 for 120 s, Red 1500 for 240 s.
% Net Utilization	Monitors the percentage of network bandwidth in use on the network interface. Most networks are fully saturated at less than 100% utilization. For example, as Ethernet networks become congested, collisions occur far more frequently, degrading performance and causing rapid saturation. Lower is better.	On. Yellow 30 for 120 s, Red 50 for 240 s.
% TCP Retransmitted	Monitors the percentage of sent TCP segments that were retransmitted because they contain one or more bytes that were sent before.	On. Yellow 5 for 120 s, Red 10 for 240 s.