Bug ID 621379: TCP Lossfilter not enforced after iRule changes TCP settings

Last Modified: Oct 01, 2018

Bug Tracker

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

Known Affected Versions:
12.1.0, 12.1.0 HF1, 12.1.0 HF2, 12.1.1, 12.1.1 HF1, 12.1.1 HF2, 12.1.2, 12.1.2 HF1, 12.1.2 HF2, 12.1.3, 12.1.3.1, 12.1.3.2, 12.1.3.3, 12.1.3.4, 12.1.3.5, 13.0.0

Fixed In:
13.1.0, 13.0.0 HF1, 12.1.3.6

Opened: Oct 07, 2016
Severity: 4-Minor

Symptoms

TCP Lossfilter function doesn't work properly, although the first few losses will be properly ignored.

Impact

Sending rate declines due to packet losses improperly interpreted as congestion.

Conditions

TCP profile has ALL of the following settings: mptcp disabled; rate-pace disabled; tail-loss-probe disabled; fast-open disabled; cmetrics-cache-timeout = 0; congestion ctrl is reno, new-reno, high-speed, or scalable; nagle enabled or disabled; rtx_thresh = 3; loss-filter settings are both > 0. an iRule changes any of the above settings except loss-filter.

Workaround

Change any of the conditions above.

Fix Information

Properly handle loss-filter state when switching TCP stacks.

Behavior Change