Date: Mon, 18 Nov 2002 20:33:34 +0100 Subject: Re: Does tcptrace measure RTT for every ACK? Message-ID: <20021118193334.GA18529@pusa.informat.uv.es> From: uaca@alumni.uv.es
On Sat, Nov 16, 2002 at 09:10:09PM -0800, Stas Khirman wrote:
> Gentlemen,
>
> Did you considered to use three-way handshake as a good estimation for RTT?
Yes:
What do you think about the following?
suposse the following scenario:
Monitor
[Host A]------------------------------------[Host B]
SYN
t0 ····················t1···················t2
SYN+ACK
t5 ····················t4···················t3
ACK
t6 ····················t7···················t8
then
t4-t1= two way delay from monitor to Host B
t7-t4= two way delay from monitor to Host A
so when we have the following situation
Monitor
[Host A]------------------------------------[Host B]
A sends packet i
))))))))))))))))))))))))>>>>>>>>>>>>>>>>>>>>>>>>>>>>
B acks packet i
((((((((((((((((((((((((<<<<<<<<<<<<<<<<<<<<<<<<<<<<
tcptrace actualy meassures rtt as
sum('>') +sum('<')
in this case making the above calculations we do instead
sum('>') +sum('<') +(t7 -t4)
Note: this makes tha asumtion that t3-t2 and t6-t5 is near zero
The same can be done in the other way
This is based on a white paper I read from packeteer.com
I have just made a patch for tcptrace that does this wich is at
http://despistao.cjb.net:8000/tcptrace/symetric_rtt.diff
Please note that the code is very untested
any comment will be greatly apprecited
regards
Ulisses
Debian GNU/Linux: a dream come true
-----------------------------------------------------------------------------
"Computers are useless. They can only give answers." Pablo Picasso
---> Visita http://www.valux.org/ para saber acerca de la <---
---> Asociación Valenciana de Usuarios de Linux <---
----------------------------------------------------------------------------
To unsubscribe, send a message with body containing "unsubscribe tcptrace" to
majordomo@tcptrace.org.
This archive was generated by hypermail 2b30 : 11/19/02 EST