| Purpose |
This probe calculates ECMP imbalance on fabric ports.
A given set of ECMP links (only calculated on leaf-to-spine links),
identified by common system_id, is determined to be imbalanced if the
standard-deviation of the tx_bytes counter (averaged periodically over the
specified period) for the involved leaf-interfaces is above "Max Standard
Deviation".
If such imbalance is observed for more-than "Threshold Duration" over the
last"Duration" time period, we raise an anomaly.
The last "Anomaly History Count" anomaly state-changes are stored for
observation.
If more-than "Max Imbalanced Systems" systems are imbalanced, we raise a
distinct anomaly.
We maintain for inspection the number of imbalanced systems over the last
"System Imbalance History Count" samples.
|
| Source Processor |
| leaf fabric 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_fabric_int_traffic |
Set of traffic samples (for each spine-facing
interface on each leaf). Each set member has the
following keys to identify it: label (human-readable
name of the leaf), system_id (id of the leaf system,
usually serial number), interface (name of the
interface).
|
|
|
| Additional Processor(s) |
| leaf fabric interface traffic avg (Periodic Average) |
Purpose: Calculate average traffic during period specified by
average_period facade parameter. Unit is bytes per second.
Input Stage: leaf_fabric_int_traffic
| Output Stage: leaf_fabric_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: label (human-readable
name of the leaf), system_id (id of the leaf system,
usually serial number), interface (name of the
interface).
|
|
| leaf fabric interface std-dev (Standard Deviation) |
Purpose: calculate standard deviation for a set consisting of
traffic averages for each spine-facing interface on a given
leaf. Grouping per leaf is achieved using 'group_by' property
set to 'system_id'.
Input Stage: leaf_fabric_int_tx_avg
| Output Stage: leaf_fab_int_std_dev |
Set of values, each indicating standard deviation (as
a measure of ECMP imbalance) for traffic averages
for each spine-facing interface on a given leaf.
Each set member has system_id key to identify leaf
whose ECMP imbalance the value represents.
|
|
| std-dev percentage (Ratio) |
Input Stage: leaf_fab_int_std_dev
Output Stage: std_dev_percentage
|
| live ecmp imbalance (Range) |
Purpose: Evaluate if standard deviation between spine-facing
interfaces on each leaf is within acceptable range. In this case
acceptable range is between 0 and std_max facade parameter (in
bytes per second unit).
Input Stage: std_dev_percentage
| Output Stage: live_ecmp_imbalance |
Set of true/false values, each indicating if standard
deviation (as a measure of ECMP imbalance) for
traffic averages for each spine-facing interface on
a given leaf is within acceptable range. Each set
member has system_id key to identify leaf whose ECMP
imbalance the value represents.
|
|
|
| sustained ecmp imbalance (Time in State) |
Purpose: Evaluate if standard deviation between spine-facing interfaces on
each leaf has been outside acceptable range, (as defined by 'live ecmp
imbalance' processor) for more than 'threshold_duration' seconds during last
'total_duration' seconds. These two parameters are part of facade
specification.
Input Stage: live_ecmp_imbalance
Output Stage: system_imbalance
|
| systems imbalanced count (Match Count) |
Purpose: Count how many systems have ecmp imbalance anomaly true at any
instant in time.
Input Stage: system_imbalance
| Output Stage: system_imbalance_count |
Number of systems with ecmp imbalance.
|
|
| imbalanced system count out of range (Range) |
Purpose: Evaluate if the number of imbalanced systems is within acceptable
range, which in this instance means less than 'max_systems_imbalanced' value
which is a facade parameter.
Input Stage: system_imbalanced_count
| Output Stage: imbalanced_system_count_out_of_range |
Boolean indicating if the number of imbalanced systems is within
accepted range, i.e. less than 'max_systems_imbalanced" which is
a facade parameter.
|
|
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.
