why would congestion window validation prevents to have larger window at the moment (time = 6.25s) when fast recovery exits since everything outstanding when entering fast retransmit is cumulatively acked? (you may not be able to see this from the sequence plot but there's a green ack dot below the blue sack dots around time = 6.25s) you can verify this by looking at output from tcpdump at http://www.cs.berkeley.edu/~shelleyz/temp/cold_switch100_7.out so TCP should be able to send however many segments allowed by the current value of cwnd, which should not be 2. thanks, shelley ----- Original Message ----- From: <kuznet@ms2.inr.ac.ru> To: "Shelley Qian Zhuang" <shelleyz@cs.BErkeley.EDU> Cc: <linux-net@vger.kernel.org> Sent: Tuesday, July 30, 2002 1:08 PM Subject: Re: Strange Fast Retransmit/Recovery Exit Behavior > Hello! > > > congestion avoidance after fast retransmit/recovery, and the window size > > (cwnd) is set to 2 segments rather than half of the cwnd before entering > > fast retransmit. > > Apparently, congestion window validation prevents to have larger > window at the moment. This is normal when lots of segments was lost > and when connection had full RCV.WND. > > > > sequence # vs. time plot at > > http://www.cs.berkeley.edu/~shelleyz/temp/roam_mipv6_100ms_sack.eps > > Sigh... next time, please, make plain straight tcpdump, better binary. > Pictures of this kind are almost unreadable. > > Alexey > - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
