Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 20 Sep 2002 19:11:40 -0700 (PDT)
From:      Julian Elischer <julian@elischer.org>
To:        Mike Silbersack <silby@silby.com>
Cc:        net@freebsd.org
Subject:   Re: Tcp question.
Message-ID:  <Pine.BSF.4.21.0209201908080.21069-100000@InterJet.elischer.org>
In-Reply-To: <20020920204805.G6326-100000@patrocles.silby.com>

next in thread | previous in thread | raw e-mail | index | archive | help


On Fri, 20 Sep 2002, Mike Silbersack wrote:

> 
> On Fri, 20 Sep 2002, Julian Elischer wrote:
> 
> > OK so I have 3 machines:
> >
> >
> > A------router--------B-------router--------C
> >
> >
> > if I send data from B to A I see 7MB/sec.
> > if I send data from B to C I see 700KB/sec
> >
> > tcpdump shows some odd behaviour in the slow link:
> > (tcpdump run from (B))
> 
> Could you repost with more of the timestamp intact?  It's really hard to
> read as is.
> 
> > 000221 C.ssh > B.916: . [tcp sum ok] ack 66194 win 24624
> >   nop,nop,timestamp 259781842 16260556> (DF)
> >   (ttl 64, id 18252, len 52)
> > **why wait here**?

My question is:
Why doesn't 'B' send more data here?

> > 003030 C.ssh > B.916: . [tcp sum ok] ack 68930 win 24624
> >   nop,nop,timestamp 259781842 16260556> (DF)
> >   (ttl 64, id 18253, len 52)
> > 000279 C.ssh > B.916: P 3769:3813(44) ack 70298 win 24624
> >   nop,nop,timestamp 259781842 16260556> (DF)
> >   (ttl 64, id 18254, len 96)
> 
> Hm, the TCP stack doesn't look like it thinks that it's waiting... the
> timestamps don't seem to be incrementing.  Of course, since I can't see
> the timestamp, I don't know what waiting means. :)

timestamps have too high granularity to figgure out why a connection is
only getting 700KB/sec
I want to know why there is a 3030 uSec gap (3 whole milisecs)
and tcp on B doesn't send more data.

If you look it has had 4 packets ack'd and it should respond..

maybe the problem is that the netisr hasn't run the tcp stack yet?
The machine isn't really doing anything else except user-space work.

> 
> Have you tried disabling delayed acks just to be doubly sure that the
> delayed ack code isn't causing the problem somehow?
> 
> Also, is C really a FreeBSD box?  I didn't think we used window sizes
> of 24K.
> 
> Mike "Silby" Silbersack
> 
> 
> 
> 
> 
> 


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-net" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0209201908080.21069-100000>