netdev
[Top] [All Lists]

Re: Linux TCP/IP stack

To: Rami Cohen <ramic@xxxxxxxxxxxxx>
Subject: Re: Linux TCP/IP stack
From: Andi Kleen <ak@xxxxxx>
Date: Thu, 26 Jul 2001 19:47:35 +0200
Cc: Andi Kleen <ak@xxxxxx>, Rabeeh Khoury <rabeeh@xxxxxxxxxxxxx>, netdev@xxxxxxxxxxx
In-reply-to: <3B602365.AAF773CF@galileo.co.il>; from Rami Cohen on Thu, Jul 26, 2001 at 04:04:21PM +0200
References: <3B5FFCD7.1070602@galileo.co.il> <20010726144045.56795@colin.muc.de> <3B602365.AAF773CF@galileo.co.il>
Sender: owner-netdev@xxxxxxxxxxx
On Thu, Jul 26, 2001 at 04:04:21PM +0200, Rami Cohen wrote:
> Hi Andi,
> Thanks for your comments and your help.
> I'm aware to RFC 793 statement regarding to ACK.
> My problem is that these piggyback ACK can cause duplicate ACK and therefor 
> invoke the
> Fast Retransmit mechanism.
> Consider the following situation when the sender send TCP segment and the 
> receiver send
> ACK.
> Then the sender send another 3 (or more) segments, and meanwhile it receives 3
> piggyback ACKs that have been sent by the receiver (i.e. the receive send 3 
> TCP frame
> before it receives the 3 packets that has been sent by the Sender).
> The sender can assume that these ACKs are duplicate ACK and therefor it 
> enters into
> Fast Retransmit although it shouldn't.
> 
> Can you help me with this issue.

The dupack counter only counts acks which do not carry any data ("pure acks")


-Andi

<Prev in Thread] Current Thread [Next in Thread>