[e2e] TCP/TCPDUMP/Why do I see such a thing happening ?

[Spencer Dawkins]

Hi, Kalyan,

It would help me to know whether you're measuring this at the
sender, at a different on the same LAN as the sender, at the
same LAN as the receiver, or at the receiver...

Spencer

--- Greg Minshall <minshall at acm.org> wrote:
> > 1048542209.180726 111.111.11.111.1326 > 222.22.222.222.80: .
> ack 
> >   55132 win 63712
> > 1048542209.180726 222.22.222.222.80 > 111.111.11.111.1326: .
> 
> >   56568:56580(12) ack 2514 win 64651
> 
> >why transfer 56568:56580, (Out-of-order ??) when asked for 
> >55132:56568 above and why only 55132:56568<<
> 
> i didn't see a reply to this, but i would say there is a bug
> in the TCP of the 
> sender.  my memory cells are unfortunately leaky, but it looks
> familiar.  i 
> suspect someone on the list will recognize it.
> 
> 
> also...
> 
> >why P is set in the below transfer
> > 1048542209.180726 222.22.222.222.80 > 111.111.11.111.1326: P
> >   94228:95295(1067) ack 2514 win 64651
> 
> presumably because that was the last of the application data
> 222... had to 
> transmit, so it was hinting to 111... that now might be a good
> time to send 
> the accumulated data up to the receiving application.
> 
> cheers, Greg
>