Bug ID 523854: TCP reset with RTSP Too Big error when streaming interleaved data

Last Modified: Sep 13, 2023

Affected Product(s):
BIG-IP CGN, LTM(all modules)

Known Affected Versions:
11.5.3, 11.6.0, 12.0.0, 12.1.0 HF1, 12.1.0 HF2, 12.1.1 HF1, 12.1.1 HF2, 12.1.2 HF1, 12.1.2 HF2

Fixed In:
12.1.0, 12.0.0 HF1, 11.6.1, 11.5.4

Opened: May 18, 2015

Severity: 3-Major

Related Article: K35305250

Symptoms

RTSP connection containing interleaved streams is aborted mid-stream, causing loss of data. This occurs when there is packet loss and retransmission due to an unreliable connection. A RST is sent by BIG-IP with cause "Too big". There is an RTSP profile parameter Maximum Header Size. When the RTSP filter receives a burst of reassembled stream data that exceeds this size, it aborts with that RST cause. When this parameter is raised above the value of parameter Maximum Queued Data, that parameter is exceeded and the RST cause is "Hudfilter abort". When both parameters are raised much higher, an abort is less likely, but can still occur with cause "Out of memory" (which is a false report as the system is not out of memory).

Impact

RTSP traffic is interrupted or dropped TCP session is reset with a cause of "Too Big" or "Hudfilter abort".

Conditions

RTSP profile configured. Interleaved stream. Packet retransmissions due to an unreliable connection.

Workaround

Set both the Maximum Header Size and Maximum Queued Data values to a value greater than 64 KB. This reduces the likelihood of failure, but is only a partial workaround.

Fix Information

RTSP interleaved traffic passes reliably, even over an unreliable connection experiencing packet retransmission.

Behavior Change

Guides & references

K10134038: F5 Bug Tracker Filter Names and Tips