Bug ID 690793: TMM may crash and dump core due to improper connflow tracking

Last Modified: Nov 07, 2022

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, 12.1.3.6, 13.1.0, 13.1.0.1, 13.1.0.2, 13.1.0.3, 13.1.0.4, 13.1.0.5

Fixed In:
14.0.0, 13.1.0.6, 12.1.3.7

Opened: Oct 25, 2017
Severity: 2-Critical
Related AskF5 Article:
K25263287

Symptoms

In rare circumstances, it is possible for the embedded Packet Velocity Acceleration (ePVA) chip to try to process non-ePVA connflows. Due to this improper internal connflow tracking, TMM can crash and dump core.

Impact

TMM crashes and dumps core. A redundant unit will fail over. Traffic may be impacted while TMM restarts.

Conditions

This issue can occur on any system equipped with an ePVA and configured with virtual servers that make use of it to accelerate flows. While no other conditions are required, it is known that modifying a FastL4 virtual server to Standard while the virtual server is processing traffic is very likely to cause the issue.

Workaround

It is not recommended to switch virtual server profiles while running traffic. To change virtual server profiles, it is recommended to halt traffic to the system first. However, this does not eliminate entirely the chances of running into this issue.

Fix Information

The system now checks for HSB flow status update data and prevents false positive matches to virtual servers with non-FastL4 profiles.

Behavior Change