From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yinglin Sun Subject: Re: [TCP_CA_CWR] Causes for entering TCP_CA_CWR state with 0 retransmissions Date: Tue, 14 Jul 2009 15:12:16 -0400 Message-ID: <46bf9ede0907141212x7a7a84f8l1882ffefbaf21e14@mail.gmail.com> References: <46bf9ede0907141138q3b665aefv46a18591c3cfc2ce@mail.gmail.com> <20090714.114211.254914419.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org To: David Miller Return-path: Received: from mail-ew0-f226.google.com ([209.85.219.226]:65410 "EHLO mail-ew0-f226.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753815AbZGNTMR convert rfc822-to-8bit (ORCPT ); Tue, 14 Jul 2009 15:12:17 -0400 Received: by ewy26 with SMTP id 26so3522709ewy.37 for ; Tue, 14 Jul 2009 12:12:16 -0700 (PDT) In-Reply-To: <20090714.114211.254914419.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: Hi David, I use the total number of retransmissions reported by tcp_info. The field name is tcpi_total_retrans. From the kernel source code, I found that this number is for both fast retransmit and timeout retransmissions. On the other hand, if fast retransmission happened, ca_state should be TCP_CA_Recovery or TCP_CA_Disorder, but it's TCP_CA_CWR. Thanks. Yinglin On Tue, Jul 14, 2009 at 2:42 PM, David Miller wrot= e: > From: Yinglin Sun > Date: Tue, 14 Jul 2009 14:38:07 -0400 > >> I have a question about tcp_ca_state "TCP_CA_CWR". =A0Under what >> circumstances does it enter this state? >> I'm doing some experiments and have trace from tcp_info. From tcp_in= fo >> trace, I found that many places show ca_state is 2 (TCP_CA_CWR). Und= er >> this state, the congestion window size decreases but the total numbe= r >> of retransmissions is still 0. > > What kind of "retransmissions"? =A0I bet the counter you are looking > at is counting timeout based retransmissions, which means it isn't > including the most predominant type of retransmissions, which are > fast retransmissions which are triggered in response to ACKs. >