Re: [Question] Rexmt Pkts and Triple Duplicate Acks

From: Daikichi Osuga (osuga@mml.yrp.nttdocomo.co.jp)
Date: 04/07/02

  • Next message: Hekkk Hekk: "Re: How to dump packet after filtered"

    Message-ID: <002e01c1dea5$5c7fe420$f23315ac@mmlad.yrp.nttdocomo.co.jp>
    From: "Daikichi Osuga" <osuga@mml.yrp.nttdocomo.co.jp>
    Subject: Re: [Question] Rexmt Pkts and Triple Duplicate Acks
    Date: Mon, 8 Apr 2002 11:30:34 +0900
    
    

    tcptrace recognize duplicate ack simply by same ack sequence number.

    but actual TCP stack recognize duplicate ack by more conditions.
    if advertised window is growing (ie yellow line jump up)
    it isn't recognized as duplicate ack.

    --
    Daikichi Osuga
    

    ----- Original Message ----- From: "Bokyung Wang" <wangbok@cs.pdx.edu> To: <tcptrace@tcptrace.org> Sent: Saturday, April 06, 2002 6:07 AM Subject: [Question] Rexmt Pkts and Triple Duplicate Acks

    > Hello, > > I'd like to know how many packets are retransmitted by triple duplicate acks > (fast rexmt) > and by timeout event. > In the tcptrace output, there are # of rexmt pkts and # of triple duplicate > acks, > but sometimes # of triple duplicate acks are more than # of rexmt pkts. > > I was wondering how does tcptrace count # of triple duplicate acks from > tcpdump input? > > Thanks in advance. > > -bokyung > > > ---------------------------------------------------------------------------- > To unsubscribe, send a message with body containing "unsubscribe tcptrace" to > majordomo@tcptrace.org. >

    ---------------------------------------------------------------------------- To unsubscribe, send a message with body containing "unsubscribe tcptrace" to majordomo@tcptrace.org.



    This archive was generated by hypermail 2b30 : 04/08/02 EDT