Probe: Hot/Cold Interface Counters (Fabric Interfaces)

Purpose

This probe determines hot/cold interface counters. It determines if interface counters are hot (too high) or cold (too low). A given interface (considering only leaf fabric interfaces) is considered to be in a hot state if its average counter value is greater than "Max". A given interface (considering only leaf fabric interfaces) is considered to be in a cold state if its average counter value is less than "Min". If such undesired state is observed for more-than "Threshold Duration" over the last "Duration" period, an anomaly is raised. Distinct anomalies are raised for hot and cold states. If more than "Max Hot Interface Percentage" percent of interfaces on a given device are hot, we raise an anomaly. If more than "Max Cold Interface Percentage" percent of interfaces on a given device are cold, we raise an anomaly. Finally, the last "Anomaly History Count" anomaly state-changes are stored for observation.

Source Processor

leaf interface traffic (Interface Counters)

Purpose: wires in interface traffic samples (measured in bytes per second) from each spine facing interface on each leaf.

Output Stage: leaf_int_traffic

Set of traffic samples (for each spine-facing interface on each leaf). Each set member has the following keys to identify it: system_id (id of the leaf system, usually serial number), interface (name of the interface), role (role of the interface, such as 'fabric').

Additional Processor(s)

leaf interface tx avg (Periodic Average)

Purpose: Calculate average traffic during period specified by average_period facade parameter. Unit is bytes per second.

Input Stage: leaf_int_traffic

Output Stage: leaf_int_tx_avg

Set of traffic average values (for each spine-facing interface on each leaf). Each set member has the following keys to identify it: system_id (id of the leaf system, usually serial number), interface (name of the interface), role (role of the interface, such as 'fabric').

interface sum per device (Sum)

Purpose: Sum average traffic for all interface under consideration per device.

Input Stage: leaf_int_tx_avg

Output Stage: if_counter_sum_per_device

Set of numbers, each indicating the total average traffic for all interface under consideration per device, expressed in bytes per second. Each set member has the following key to identify it: system_id (id of the leaf system, usually serial number).

interface sum per device per link role (Sum)

Purpose: Sum average traffic for all interface under consideration per device, per interface role.

Input Stage: leaf_int_tx_avg

Output Stage: if_counter_sum_per_device_role

Set of numbers, each indicating the total average traffic for all interface under consideration per device, expressed in bytes per second. Each set member has the following keys to identify it: system_id (id of the leaf system, usually serial number), role (role of the interface, such as 'fabric').

live leaf interface cold (Range)

Purpose: Evaluate if the average traffic on spine facing interfaces on each leaf is within acceptable range. In this case acceptable range means larger than min facade parameter (in bytes per second unit).

Input Stage: leaf_int_tx_avg

Output Stage: live_leaf_int_cold

Set of true/false values, each indicating if traffic averages for each spine-facing interface on each leaf is within acceptable range. Each set member has the following keys to identify it: system_id (id of the leaf system, usually serial number), interface (name of the interface) role (role of the interface, such as 'fabric'). Samples unit is bytes per second.

live leaf interface hot (Range)

Purpose: Evaluate if the average traffic on spine-facing interfaces on each leaf is within acceptable range. In this case acceptable range is between 0 and max facade parameter (in bytes per second unit).

Input Stage: leaf_int_tx_avg

Output Stage: live_leaf_int_hot

sustained cold leaf interface (Time in State)

Purpose: Evaluate if the average traffic spine facing interfaces on each leaf has been outside acceptable range, (as defined by 'live leaf interface cold' processor) for more than 'threshold_duration' seconds during the last 'total_duration' seconds. These two parameters are part of facade specification.

Input Stage: live_leaf_int_cold

Output Stage: cold_leaf_int

Set of true/false values, each indicating if the traffic average for each spine-facing interface on each leaf has been in 'cold' range for more than specified period of time. Each set member has the following keys to identify it: system_id (id of the leaf system, usually serial number), interface (name of the interface) role (role of the interface, such as 'fabric'). Samples unit is bytes per second.

sustained hot leaf interface (Time in State)

Evaluate if the average traffic spine facing interfaces on each leaf has been outside acceptable range, (as defined by 'live leaf interface hot' processor) for more than 'threshold_duration' seconds during the last 'total_duration' seconds. These two parameters are part of facade specification.

Input Stage: live_leaf_int_hot

Output Stage: hot_leaf_int

Set of true/false values, each indicating if the traffic average for each spine-facing interface on each leaf has been in 'hot' range for more than specified period of time. Each set member has the following keys to identify it: system_id (id of the leaf system, usually serial number), interface (name of the interface) role (role of the interface, such as 'fabric'). Samples unit is bytes per second.

system percent cold (Match Percentage)

Purpose: Calculate percentage of interfaces that are cold on any given device under consideration.

Input Stage: cold_leaf_int

Output Stage: system_perc_cold

Set of numbers, each indicating the the percentage of cold interfaces on any given device under consideration. Each set member has the following key to identify it: system_id (id of the leaf system, usually serial number).

system percent hot (Match Percentage)

Purpose: Calculate percentage of interfaces that are hot on any given device under consideration.

Input Stage: hot_leaf_int

Output Stage: system_perc_hot

Set of numbers, each indicating the the percentage of hot interfaces on any given device under consideration. Each set member has the following key to identify it: system_id (id of the leaf system, usually serial number).

device cold (Range)

Purpose: Evaluate if the percentage of cold interfaces on a specific device is outside the acceptable range, where acceptable range in his case means less than 'max_cold_interface_percentage', which is a facade parameter.

Input Stage: system_perc_cold

Output Stage: device_cold_anomalous

Set of boolean values, each indicating if the the percentage of cold interfaces on any given device was out of acceptable range. Each set member has the following key to identify it: system_id (id of the leaf system, usually serial number).

device hot (Range)

Purpose: Evaluate if the percentage of hot interfaces on a specific device is outside the acceptable range, where acceptable range in his case means less than 'max_hot_interface_percentage', which is a facade parameter.

Input Stage: system_perc_hot

Output Stage: device_hot_anomalous

Set of boolean values, each indicating if the the percentage of hot interfaces on any given device was out of acceptable range. Each set member has the following key to identify it: system_id (id of the leaf system, usually serial number).

For more information about this probe, from the blueprint, navigate to Analytics > Probes, click Create Probe, then select Instantiate Predefined Probe from the drop-down list. Select the probe from the Predefined Probe drop-down list to see details specific to the probe.