<div dir="ltr">Thanks. My problem is solved.<div>It seems to me paping can sample for the RTT over this particular port, not the whole connection setup time. Maybe it only record for the sync time but I am not fully sure. From the latency result, it is quite close to RTT of the link.</div>
<div><br><div>-Mo</div></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/12/5 Toke Høiland-Jørgensen <span dir="ltr"><<a href="mailto:toke@toke.dk" target="_blank">toke@toke.dk</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">Dong Mo <<a href="mailto:montedong@gmail.com">montedong@gmail.com</a>> writes:<br>
<br>
> I also found this to sample tcp latency performance:<br>
> <a href="https://code.google.com/p/paping/" target="_blank">https://code.google.com/p/paping/</a><br>
<br>
</div>Right; note that that measures time to establish a new connection, not<br>
the delay of packets within an already established connection. My guess<br>
would be that each established connection would have a new source port,<br>
thus hashing into a new bin.<br>
<div class="im"><br>
> Does fq_codel's flow classification supports external classifier<br>
> filter? Say I only want to hash on dst port in my case.<br>
<br>
</div>I'm pretty sure it supports the normal classifier (tc filter). That<br>
overrides the built-in hashing completely, but IIRC you can do your own<br>
hashing in tc filter :)<br>
<span class="HOEnZb"><font color="#888888"><br>
-Toke<br>
</font></span></blockquote></div><br></div>