From: Dominik Kaspar <dokaspar.ietf@gmail.com>
To: "Ilpo Järvinen" <ilpo.jarvinen@helsinki.fi>
Cc: Alexander Zimmermann <alexander.zimmermann@comsys.rwth-aachen.de>,
Netdev <netdev@vger.kernel.org>,
Yuchung Cheng <ycheng@google.com>,
Carsten Wolff <carsten@wolffcarsten.de>,
John Heffner <johnwheffner@gmail.com>,
Eric Dumazet <eric.dumazet@gmail.com>,
Lennart Schulte <Lennart.Schulte@comsys.rwth-aachen.de>,
Arnd Hannemann <arnd@arndnet.de>
Subject: Re: Linux TCP's Robustness to Multipath Packet Reordering
Date: Mon, 20 Jun 2011 14:52:35 +0200 [thread overview]
Message-ID: <BANLkTin6fsc=4GUY+1UKsLEbgzeybx7FHg@mail.gmail.com> (raw)
In-Reply-To: <alpine.DEB.2.00.1106201336450.17529@wel-95.cs.helsinki.fi>
Hi Ilpo,
> Where did you get this idea of reneging?!?
I observed that my scenario of a retransmitted packet overtaking the
original somehow causes TCP to enter the "Loss" state although no RTO
was caused. And since the Loss state seems to be only entered due to
RTO timeout or SACK reneging, I got the idea that reneging must be
occurring.
> Reneging has nothing to do with DSACKs,
> instead it is only detected if the cumulative ACK stops to such
> boundary where the _next_ segment is SACKed (i.e., some reason
> the receiver "didn't bother" to cumulatively ACK for that too). ...
> That certainly does not happen (ever) for out of window DSACKs.
You are right. If I turn off DSACK, the same thing happens: TCP enters
the Loss state without timeouts occurring. Isn't that a sign of
reneging happening? What else can it be?
Dominik
next prev parent reply other threads:[~2011-06-20 12:52 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-25 10:37 Linux TCP's Robustness to Multipath Packet Reordering Dominik Kaspar
2011-04-25 11:25 ` Eric Dumazet
2011-04-25 14:35 ` Dominik Kaspar
2011-04-25 15:38 ` Eric Dumazet
2011-04-26 16:58 ` Dominik Kaspar
2011-04-26 17:10 ` Eric Dumazet
2011-04-26 18:00 ` Dominik Kaspar
2011-04-26 20:16 ` John Heffner
2011-04-26 21:27 ` Dominik Kaspar
2011-04-27 9:57 ` Carsten Wolff
2011-04-27 16:22 ` Dominik Kaspar
2011-04-27 16:36 ` Alexander Zimmermann
2011-06-21 11:25 ` Ilpo Järvinen
2011-06-21 11:34 ` Carsten Wolff
2011-06-21 11:46 ` Ilpo Järvinen
2011-04-27 16:48 ` Eric Dumazet
2011-04-27 17:39 ` Yuchung Cheng
2011-04-27 17:53 ` Alexander Zimmermann
2011-04-27 19:56 ` Dominik Kaspar
2011-04-27 21:41 ` Yuchung Cheng
2011-04-28 6:11 ` Alexander Zimmermann
2011-06-19 15:22 ` Dominik Kaspar
2011-06-19 15:38 ` Alexander Zimmermann
2011-06-19 16:25 ` Dominik Kaspar
2011-06-20 10:42 ` Ilpo Järvinen
2011-06-20 12:52 ` Dominik Kaspar [this message]
2011-06-21 11:35 ` Ilpo Järvinen
2011-04-26 20:43 ` Eric Dumazet
2011-04-26 21:04 ` Dominik Kaspar
2011-04-26 21:08 ` Eric Dumazet
2011-04-26 21:16 ` Dominik Kaspar
2011-04-26 21:17 ` Eric Dumazet
2011-04-25 12:59 ` Carsten Wolff
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='BANLkTin6fsc=4GUY+1UKsLEbgzeybx7FHg@mail.gmail.com' \
--to=dokaspar.ietf@gmail.com \
--cc=Lennart.Schulte@comsys.rwth-aachen.de \
--cc=alexander.zimmermann@comsys.rwth-aachen.de \
--cc=arnd@arndnet.de \
--cc=carsten@wolffcarsten.de \
--cc=eric.dumazet@gmail.com \
--cc=ilpo.jarvinen@helsinki.fi \
--cc=johnwheffner@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=ycheng@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).