On Feb 13, 2007, at 6:15 PM, Luis Ontanon wrote:

> I have a further question:
>
> Can the PMTU of the smallest path of an association shrink between the
> transmission of a bundled packet and its retransmission forcing the
> sender to retransmit the various chunks of a message in more messages?
Sure. And on retransmissions the packets are built depending on what
was received by the peer, so the packet can look different.
>
> i.e. : Can I just track the highest TSN in a packet and look for its
> ack, or should I track every data chunk of each packet separatelly?
No, you have to track each TSN. Have a look at the packet-sctp.c
reassembly code and also at the GUI stuff in gtk/ subdirs, where
also all associations are tracked.
>
> Thanks Again,
> Luis
>
> On 2/13/07, Michael Tuexen <[EMAIL PROTECTED]> wrote:
>> Hi Luis,
>>
>> see my comments in-line.
>>
>> Best regards
>> Michael
>>
>> On Feb 13, 2007, at 2:14 PM, Luis Ontanon wrote:
>>
>>> I need to implement the identification of retransmitted packets and
>>> calculation of RTT to the sctp dissector.
>>>
>>> - Is someone already working on this?
>> I recently committed code for SCTP reassembly. Please have a look  
>> at it.
>> It should be easy to add retransmission detection to it. Just look  
>> for
>> the TSN of the DATA chunks. The (half associations) are detected by
>> the port numbers and the V-tag.
>>> - other than RFC 2960  (sec. 6.3) what other rfcs I must read in  
>>> order
>>> to implement these?
>> Should be OK.
>>> - are there any caveats I must be aware before starting this  
>>> venture?
>> For RTT detection you have to "connect" both directions of the
>> association.
>> Since you can not really use the source/destination address you  
>> have to
>> use a heuristic to combine both half directions.
>>
>> Please let me know if you have further questions.
>>>
>>> Thanks,
>>>
>>> Luis
>>>
>>> --
>>> This information is top security. When you have read it, destroy
>>> yourself.
>>> -- Marshall McLuhan
>>> _______________________________________________
>>> Wireshark-dev mailing list
>>> [email protected]
>>> http://www.wireshark.org/mailman/listinfo/wireshark-dev
>>
>> _______________________________________________
>> Wireshark-dev mailing list
>> [email protected]
>> http://www.wireshark.org/mailman/listinfo/wireshark-dev
>>
>
>
> -- 
> This information is top security. When you have read it, destroy  
> yourself.
> -- Marshall McLuhan
> _______________________________________________
> Wireshark-dev mailing list
> [email protected]
> http://www.wireshark.org/mailman/listinfo/wireshark-dev

_______________________________________________
Wireshark-dev mailing list
[email protected]
http://www.wireshark.org/mailman/listinfo/wireshark-dev

Reply via email to