Wireshark · Wireshark-bugs: [Wireshark-bugs] [Bug 4624] TCP reassembly can call subdissector with incorrect TCP sequence number

[bugzilla-daemon@xxxxxxxxxxxxx]

https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4624

--- Comment #6 from jmmikkel@xxxxxxx 2010-04-30 20:26:19 PDT ---
Thank you for commiting the change!

And I had noticed the bug marked "UNCONFIRMED" earlier so I had gone off and
written code to help confirm it. Since it's already written, might as well
add it here...

I made a dinky "fake" protocol and dissector. I wrote a program to send PDUs
across multiple packets in such a way as to demonstrate the problem. The packet
trace I'm attaching is the result. Dissecting this trace with the attached
packet-fake.c plugin (tshark -V -R 'tcp.len > 0') results in the output I'm
attaching. Consider frame 6:

subdissector called
Fake Protocol
    Relative sequence number: 51
    Message length: 100
    Data: 333435363738393A3B3C3D3E3F404142434445464748494A...
subdissector called
Fake Protocol
    Relative sequence number: 51
    Message length: 50
    Data: 9798999A9B9C9D9E9FA0A1A2A3A4A5A6A7A8A9AAABACADAE...

As the message data differs, these cannot both be sequence number 51.
The second should be 151, and it does print that with my patch.

-- 
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.