Bug ID 696201: Anomaly threshold floor calculation (for bins with very low learnt threshold) during signature generation

Last Modified: May 14, 2019

Bug Tracker

Affected Product:  See more info
BIG-IP AFM(all modules)

Known Affected Versions:
13.0.0, 13.0.0 HF1, 13.0.0 HF2, 13.0.0 HF3, 13.0.1, 13.1.0, 13.1.0.1, 13.1.0.2, 13.1.0.3, 13.1.0.4, 13.1.0.5, 13.1.0.6, 13.1.0.7, 13.1.0.8, 13.1.1, 13.1.1.1

Fixed In:
14.0.0, 13.1.1.2

Opened: Nov 27, 2017
Severity: 3-Major

Symptoms

AFM might generate a dynamic signature for those bins that have a very low learnt threshold during the learning phase, if the current traffic rate spikes and increases above the anomaly threshold floor db variable value as specified by l4bdos.anomaly.threshold.floor

Impact

This might cause AFM to generate signatures with higher false positives. This is specifically due to incorrect application of db variable setting 'l4bdos.anomaly.threshold.floor' that should be interpreted as the 'floor' value of learnt thresholds for any bin. So, if the learnt threshold of a bin is lower than this db variable, the baseline threshold of the bin should be set to the db variable for anomaly detection phase.

Conditions

AFM dynamic signature feature is enabled.

Workaround

There is no workaround at this time.

Fix Information

This issue is fixed by making sure that db variable 'l4bdos.anomaly.threshold.floor' is used as the 'floor' value of baseline thresholds for those bins that have a learnt threshold lower than this db variable.

Behavior Change