Bad "bad" retransmit recovery in TCP

From: Guohan Lu (lguohan00_at_mails.tsinghua.edu.cn)
Date: 01/25/04

  • Next message: Andre Oppermann: "Re: Bad "bad" retransmit recovery in TCP" 
    To: <freebsd-net@freebsd.org>
Date: Mon, 26 Jan 2004 05:17:27 +0800

    hi,
        I noticed in my TCP trace that there are many bad "bad" retransmit recoveries. That is the packet get lost,
    the timeout happens. However, when the lost packet are acknowledged, the cwnd of sender is not 2,
    but resumes the value before timeout. This kind of behavior happens when the RTT is very small,
    around several ms, no more than 10ms.
        It seems to me that following code is run when the acknowledgement for the lost packet is received.
      if (tp->t_rxtshift == 1 && ticks < tp->t_badrxtwin) {
       tp->snd_cwnd = tp->snd_cwnd_prev;
       tp->snd_ssthresh = tp->snd_ssthresh_prev;
       tp->snd_nxt = tp->snd_max;
       tp->t_badrxtwin = 0; /* XXX probably not required */
      }
       I wonder if t_badrxtwin is not correctly set in this situation? (when the rtt is very small)
       
    Guohan Lu

    _______________________________________________
    freebsd-net@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-net
    To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"

  • Next message: Andre Oppermann: "Re: Bad "bad" retransmit recovery in TCP"

    Relevant Pages

    • Re: timeout & retransmission of TCP packets
      ... > doubt in retransmission & timeout concepts in TCP connections. ... RTT optimizer = plus ... The destination receiver can also "request" a re-transmission by the ... window before receiving _any_ acknowledgement. ...
      (comp.os.linux)
    • Re: Bad "bad" retransmit recovery in TCP
      ... > recoveries. ... That is the packet get lost, the timeout happens. ... > It seems to me that following code is run when the acknowledgement ... > for the lost packet is received. ...
      (freebsd-net)