Wireshark-bugs: [Wireshark-bugs] [Bug 8629] bad tcp checksum not detected
Date: Fri, 10 May 2013 14:49:06 +0000

Comment # 11 on bug 8629 from
(In reply to comment #10)
> It's not clear to me how others feel about this, but I was thinking that a
> bad checksum shouldn't necessarily prevent reassembly from occurring as long
> as the TCP preference to "Allow subdissector to reassemble TCP streams" is
> selected.

If the user explicitly request to verify the checksum, should not we prevent
reassembly if it is invalid (like a real TCP stack would act I guess)? If the
data was really altered on the wire, I assume the TCP stack will request a
retransmission and Wireshark should get a valid TCP fragment later in the
capture.
Trying to reassemble a corrupted stream will most probably leas to a malformed
packet error or something similar for upper layers, which might raise invalid
bug reports.
Maybe the TCP option should be renamed from "Validate the TCP checksum if
possible" to "Always validate the TCP checksum (will fail if offload is used)
before reassembly" or something similar.


You are receiving this mail because:
  • You are watching all bug changes.